(19) United States
`(12) Patent Application Publication (10) Pub. No.: US 2003/0166588A1
`(43) Pub. Date:
`Sep. 4, 2003
`Iversen et al.
`
`US 2003O166588A1
`
`(54)
`
`(76)
`
`(21)
`(22)
`
`(60)
`
`SPLCE-REGION ANTISENSE
`COMPOSITION AND METHOD
`
`Inventors: Patrick L. Iversen, Corvallis, OR
`(US); Robert Hudziak, Blodgett, OR
`(US)
`Correspondence Address:
`PERKINS COE LLP
`P.O. BOX 21.68
`MENLO PARK, CA 94026 (US)
`Appl. No.:
`09/848,868
`Filed:
`May 4, 2001
`Related U.S. Application Data
`Provisional application No. 60/202,376, filed on May
`4, 2000.
`
`Publication Classification
`
`(51) Int. Cl." ......................... A61K 48/00; CO7H 21/04;
`A61K 31/675; CO7D 413/14
`(52) U.S. Cl. ............................. 514/44; 514/81; 536/23.1;
`544/81
`
`(57)
`
`ABSTRACT
`
`AntiSense compositions targeted againstan mRNA sequence
`coding for a Selected protein, at a region having its 5' end
`from 1 to about 25 base pairs downstream of a normal splice
`acceptor junction in the preprocessed mRNA, are disclosed.
`The antisense compound is RNase-inactive, and is prefer
`ably a phosphorodiamidate-linked morpholino oligonucle
`otide. Such targeting is effective to inhibit natural mRNA
`Splice processing, produce Splice variant mRNAS, and
`inhibit normal expression of the protein.
`
`

`

`Patent Application Publication
`
`Sep. 4, 2003 Sheet 1 of 2
`
`US 2003/0166588A1
`
`

`

`Patent Application Publication
`
`Sep. 4, 2003 Sheet 2 of 2
`
`US 2003/0166588A1
`
`N
`
`"
`{"
`
`to
`N
`
`7
`
`Fig. 2A-A
`
`O P.
`
`"
`
`z -
`y X
`
`"
`
`r
`N
`Fig. 2B-B
`
`N
`Z=P-X
`)
`2
`
`Y
`
`"
`
`N
`
`N
`)
`Z=P-X
`'''
`
`N
`
`

`

`US 2003/0166588A1
`
`Sep. 4, 2003
`
`SPLCE-REGION ANTISENSE COMPOSITION
`AND METHOD
`0001. This application claims priority to U.S. provisional
`application serial no. 60/202,376, filed May 4, 2000, which
`is hereby incorporated by reference in its entirety.
`
`FIELD OF THE INVENTION
`0002 The present invention relates to therapeutic com
`positions and methods for inhibiting expression of full
`length proteins in cells, and in particular to antisense com
`positions targeted against an MRNA sequence having its 5'
`end 1 to about 25 base pairs downstream of a normal splice
`acceptor junction in a preprocessed mRNA. Such targeting
`is effective to inhibit natural mRNA splice processing and
`produce Splice variant mRNAS.
`
`BACKGROUND OF THE INVENTION
`Inhibition of protein expression by antisense tar
`0.003
`geting of DNA or RNA coding for the protein has been the
`Subject of extensive Study. Many reported procedures have
`employed phosphorothioate-linked oligonucleotides, which
`are charged, nuclease-resistant analogs of native DNA. The
`antisense mechanism involved is based on the activation of
`RNase, which cleaves the target nucleic acid to which the
`oligomer is bound. While these compounds have shown high
`activity, they also tend to show high levels of Side effects, i.e.
`by cleavage of non-target RNA or by non-antisense mecha
`nisms, Such as nonspecific binding to proteins.
`0004 Another class of antisense oligomers, termed
`RNase-inactive, do not promote cleavage of bound RNA and
`are believed to act by Sterically blocking the molecular
`machinery from transcribing, processing, or translating the
`target Sequence. While these compounds tend to produce
`fewer Side reactions, Such as nonselective cleavage, than
`phosphorothioate oligomers, it has generally been necessary
`to target Specific regions of RNA, Such as the AUG Start
`codon, for Successful inhibition. More recently, targeting of
`the splice acceptor junction of nuclear (unspliced) RNA by
`RNase-inactive oligomerS has been reported. Kole and
`Dominski (U.S. Pat. No. 5,665,593) reported suppression of
`missplicing of B-globin RNA, in order to combat variants of
`B-thalassemia which result from Such aberrant Splicing. In
`this case, the aberrant Splice junction was targeted, to direct
`Splicing back to the normal Site. R V Giles et al., Antisense
`& Nucleic Acid Drug Dev. 9:213-220 (1999), targeted a
`Splice junction to induce missplicing of c-myc mRNA. In
`each of these cases, the region targeted is still Somewhat
`restricted, in that the antisense oligomer spans the intron/
`exon splice junction of the pre-mRNA. Due to the advan
`tages accorded by the use of uncharged, RNase-inactive
`oligonucleotides, a demonstration of further flexibility in
`targeting would be quite useful.
`
`SUMMARY OF THE INVENTION
`0005. In one aspect, the invention provides an antisense
`compound, and a corresponding method of inhibiting nor
`mal Splicing of preprocessed RNA in a eukaryotic cell, by
`contacting the cell with Such an antisense compound. The
`compound is characterized by:
`0006 (a1): an uncharged morpholino backbone;
`0007 (a2): a base-sequence length of between 12
`and 25 nucleotide bases, and
`
`0008 (a3): a base sequence that is complementary to
`a target region of a Selected preprocessed mRNA
`coding for a Selected protein, where the 5' end of the
`target region is 1-25 bases downstream of a normal
`Splice acceptor Site in the preprocessed mRNA, and
`having the properties that:
`0009 (b1): the compound is taken up by eukary
`otic cells,
`0010 (b2): the compound hybridizes to the target
`region of preprocessed mRNA in Such cells, and
`0.011) (b3): the compound so hybridized to the
`target pre-mRNA prevents Splicing at the normal
`acceptor Splice Site, Such that the Splice mecha
`nism proceeds to a downstream Splice acceptor
`Site in the preprocessed mRNA, producing a splice
`variant processed MRNA with a truncated coding
`Sequence.
`0012. In more specific embodiments, the 5' end of the
`target region is 2-20 bases, or 2-15 bases, downstream of the
`normal splice acceptor Site. The length of the targeting
`compound is preferably about 15 to 20 nucleotide bases.
`0013 In one embodiment, the compound has intersubunit
`linkages Selected from the group consisting of the Structures
`presented in FIGS. 2AA-2EE. In preferred embodiments,
`the linkages are Selected from a phosphorodiamidate linkage
`as represented at FIG.2B-B, where X=NH, NHR, or NRR',
`Y=O, and Z=O, and an alternate phosphorodiamidate link
`age as represented at FIG. 2B-B, where X=OR, Y=NH or
`NR, and Z=O. R and R' are groups which do not interfere
`with target binding. Preferably, R and R' are independently
`Selected from alkyl and polyalkyleneoxy (e.g. PEG,
`(CHCHO),), or a combination thereof. The alkyl/poly
`alkyleneoxy chain may be Substituted, preferably at the
`distal terminus, by a group Selected from hydroxy, alkoxy,
`amino, alkylamino, thiol, alkanethiol, halogen, OXO, car
`boxylic acid, carboxylic ester, and inorganic ester (e.g.
`phosphate or Sulfonate). Preferably, the chain (independent
`of Substituents) is from 1 to 12 atoms long, and more
`preferably is from 1 to 6 atoms long. In Selected embodi
`ments, R and R' are independently methyl or ethyl. In one
`embodiment, X=N(CH), Y=O, and Z=O.
`0014 NRR' may also represent a nitrogen heterocycle
`having 5-7 ring atoms Selected from nitrogen, carbon, oxy
`gen, and Sulfur, and having at least as many carbon ring
`atoms as non-carbon ring atoms. Examples include morpho
`line, pyrrolidine, piperidine, pyridine, pyrimidine, pyrazine,
`triazine, triazole, pyrazole, pyrrole, isopyrrole, imidazole,
`Oxazole, imidazole, isoxazole, and the like.
`0015. When the downstream splice acceptor site is a
`whole multiple of three bases downstream of the normal
`Splice acceptor Site, the Splice variant mRNA has a coding
`sequence in frame with that of the processed mRNA when
`it is normally spliced.
`0016. The protein is preferably selected from the group
`consisting of myc, my b, rel, foS, jun, abl, bcl, p53, an
`integrin, a cathedrin, a telomerase, hCG, a receptor protein,
`a cytokine, a kinase, HIV rev, human papilloma virus, and
`human parvovirus B19. In selected embodiments, the pro
`tein is selected from myc, myb, abl, p53, hCG-Bsubunit,
`androgen receptor protein, and HIV-1 rev.
`
`

`

`US 2003/0166588A1
`
`Sep. 4, 2003
`
`0.017. In further selected embodiments, the selected pro
`tein has multiple distinct binding regions, as in most tran
`Scription factors, and the truncated coding Sequence codes
`for a variant protein in which one Such binding region is
`disabled. Preferably, the variant protein is a dominant nega
`tive protein. One example is human c-myc, where the
`variant protein is an N-terminal truncated c-myc. In this
`embodiment, the antisense compound employed has a base
`Sequence Selected from the group consisting of SEQ ID
`NOs: 16 through 32 herein. The variant protein may also be
`a C-terminal altered c-myc, in which case the antisense
`compound employed can be an 18- to 20-mer having a base
`Sequence which is a contiguous Sequence Selected from SEQ
`ID NO:34; e.g. SEQ ID NO: 33.
`0.018. In additional exemplary embodiments, the selected
`protein and the corresponding antisense base Sequence(s)
`targeting its pre-mRNA are Selected from the group consist
`ing of:
`0019 (a) human chorionic gonadotropin, B subunit:
`a contiguous 18- to 20-nucleotide Sequence Selected
`from SEQ ID NO: 15; e.g. SEQ ID NO: 14;
`0020 (b) human androgen receptor: a contiguous
`18- to 20-nucleotide sequence selected from SEQ ID
`NO:9 or SEQ ID NO: 13; e.g. SEQ ID NO: 8 or 12,
`respectively;
`0021 (c) human p53: a contiguous 18- to 20-nucle
`otide sequence selected from SEQ ID NO: 36; e.g.
`SEQ ID NO:35;
`0022 (d) human abl: a contiguous 18- to 20-nucle
`otide sequence selected from SEQ ID NO:38; e.g.
`SEQ ID NO:37; and
`0023 (e) HIV-1 rev: a contiguous 18- to 20-nucle
`otide sequence selected from SEQ ID NO: 41; e.g.
`SEO ID NO: 40.
`0024. These and other objects and features of the present
`invention will become more fully apparent when the fol
`lowing detailed description of the invention is read in
`conjunction with the accompanying drawings.
`
`BRIEF DESCRIPTION OF THE DRAWINGS
`0.025
`FIG. 1 shows several preferred subunits having
`5-atom (A), six-atom (B) and Seven-atom (C-E) linking
`groupS Suitable for forming polymers, and
`0.026
`FIGS. 2A-A to 2E-E show the repeating subunit
`Segment of exemplary morpholino oligonucleotides, desig
`nated A-A through E-E, constructed using Subunits A-E,
`respectively, of FIG. 1.
`
`DETAILED DESCRIPTION OF THE
`INVENTION
`0027)
`I. Definitions
`0028. The terms below, as used herein, have the follow
`ing meanings, unless indicated otherwise:
`0029) “Antisense” refers to an oligomer having a
`Sequence of nucleotide bases and a Subunit-to-Subunit back
`bone that allows the antisense oligomer to hybridize to a
`target Sequence in an RNA by Watson-Crick base pairing, to
`form an RNA:oligomer heteroduplex within the target
`
`Sequence, typically with an mRNA. The oligomer may have
`exact Sequence complementarity to the target Sequence or
`near complementarity. These antisense oligomerS may block
`or inhibit translation of the mRNA, and/or modify the
`processing of an mRNA to produce a splice variant of the
`mRNA.
`0030 AS used herein, the terms “compound”, “agent',
`"oligomer' and "oligonucleotide' may be used interchange
`ably with respect to the antisense oligonucleotides of the
`invention.
`0031 AS used herein, a “morpholino oligomer” refers to
`a polymeric molecule having a backbone which Supports
`bases capable of hydrogen bonding to typical polynucle
`otides, wherein the polymer lacks a pentose Sugar backbone
`moiety, and more specifically a ribose backbone linked by
`phosphodiester bonds which is typical of nucleotides and
`nucleosides, but instead contains a ring nitrogen with cou
`pling through the ring nitrogen. Exemplary Structures for
`antisense oligonucleotides for use in the invention include
`the morpholino subunit types shown in FIGS. 1A-E, with the
`linkages shown in FIGS. 2A-A to 2E-E. Such structures are
`described, for example, in Hudziak et al., AntiSense Nucleic
`Acid Drug Dev. 6, 267-272 (1996) and Summerton and
`Weller, Antisense Nucleic Acid Drug Dev. 7, 187-195
`(1997).
`0032 Subunit A in FIG. 1 contains a 1-atom phospho
`rous-containing linkage which forms the five atom repeat
`ing-unit backbone shown at A-A in FIG. 2, where the
`morpholino rings are linked by a 1-atom phosphoamide
`linkage.
`0033) A preferred morpholino oligonucleotide is com
`posed of morpholino subunit structures of the form shown in
`FIG. 2B-B, where the structures are linked together by
`phosphorodiamidate linkages, joining the morpholino nitro
`gen of one Subunit to the 5' exocyclic carbon of an adjacent
`Subunit, and Pi and P are purine or pyrimidine base-pairing
`moieties effective to bind, by base-specific hydrogen bond
`ing, to a base in a polynucleotide. In preferred embodiments,
`the linkages are Selected from a phosphorodiamidate linkage
`as represented at FIG.2B-B, where X=NH, NHR, or NRR',
`Y=O, and Z=O, and an alternate phosphorodiamidate link
`age as represented at FIG. 2B-B, where X=OR, Y=NH or
`NR, and Z=O. R and R' are groups which do not interfere
`with target binding. Preferably, R and R' are independently
`Selected from alkyl and polyalkyleneoxy (e.g. PEG,
`(CH2CH2O)"), or a combination thereof. (An example of
`such a combination would be-(CH2)(CHCHO)). The
`alkyl/polyalkyleneoxy chain may be Substituted, preferably
`at the distal terminus (i.e. the terminus not connected to the
`oligomer backbone), by a group Selected from hydroxy,
`alkoxy, amino, alkylamino, thiol, alkanethiol, halogen, OXO,
`carboxylic acid, carboxylic ester, and inorganic ester (e.g.
`phosphate or Sulfonate). Preferably, the chain (independent
`of Substituents) is from 1 to 12 atoms long, and more
`preferably is from 1 to 6 atoms long. In Selected embodi
`ments, R and R' are independently methyl or ethyl. In one
`embodiment, X=N(CH), Y=O, and Z=O. NRR' may also
`represent a nitrogen heterocycle having 5-7 ring atoms
`Selected from nitrogen, carbon, oxygen, and Sulfur, and
`having at least as many carbon ring atoms as non-carbon
`ring atoms. Examples include morpholine, pyrrolidine, pip
`eridine, and pyridine.
`
`

`

`US 2003/0166588A1
`
`Sep. 4, 2003
`
`0034) Subunits C-E in FIG. 1 are designed for 7-atom
`unit-length backbones as shown for C-C through E-E in
`FIG. 2. In Structure C, the X moiety is as in Structure B and
`the moiety Y may be a methylene, sulfur, or preferably
`oxygen. In Structure D the X and Y moieties are as in
`Structure B. In Structure E, X is as in Structure B and Y is
`O, S, or NR. In all subunits depicted in FIGS. 1A-E, Z is O
`or S, and P, or P, is adenine, cytosine, guanine or uracil.
`0035 A“nuclease-resistant” oligomeric molecule (oligo
`mer) is one whose backbone is not Susceptible to nuclease
`cleavage.
`0.036 AS used herein, an oligonucleotide or antisense
`oligomer “specifically hybridizes to a target polynucleotide
`if the oligomer hybridizes to the target under physiological
`conditions, with a Tm substantially greater than 37 C.,
`preferably at least 50° C., and typically 60° C.-80° C. or
`higher. Such hybridization preferably corresponds to Strin
`gent hybridization conditions, selected to be about 10 C.,
`and preferably about 50° C. lower than the thermal melting
`point (Tim) for the specific sequence at a defined ionic
`Strength and pH. At a given ionic strength and pH, the Tm
`is the temperature at which 50% of a target Sequence
`hybridizes to a complementary polynucleotide.
`0037 Polynucleotides are described as “complementary'
`to one another when hybridization occurs in an antiparallel
`configuration between two Single-Stranded polynucleotides.
`A double-Stranded polynucleotide can be “complementary'
`to another polynucleotide, if hybridization can occur
`between one of the strands of the first polynucleotide and the
`Second. Complementarity (the degree that one polynucle
`otide is complementary with another) is quantifiable in terms
`of the proportion of bases in opposing Strands that are
`expected to form hydrogen bonds with each other, according
`to generally accepted base-pairing rules.
`0038 A “RNase-inactive” or “RNase-incompetent” oli
`gonucleotide or oligonucleotide analog is one which acts via
`an RNase-independent mechanism, unlike RNase-active oli
`gonucleotides, Such as phosphorothioates. They are believed
`to function by sterically blocking target RNA formation,
`nucleocytoplasmic transport or translation, and are thus also
`referred to as “steric blockers'. This class includes, for
`example, methylphosphonates, morpholino oligonucle
`otides, as described herein, peptide nucleic acids (PNA's),
`and 2'-O-allyl or 2'-O-alkyl modified oligonucleotides.
`0039. In a “peptide nucleic acid”, the deoxyribose phos
`phate units of an oligonucleotide backbone are replaced with
`polyamide linkages. Proper backbone Spacing is attained by
`the use of 2-aminoethylglycine units, with a nucleotide base
`attached to each 2-amino group via a methylenecarbonyl
`group. A “2'-O-allyl (or alkyl) modified oligonucleotide' is
`an oligoribonucleotide in which the 2 hydroxyl is converted
`to an allyl or alkyl ether. The alkyl ether is typically a methyl
`ether.
`0040 “Alkyl” refers to a fully saturated acyclic monova
`lent radical containing carbon and hydrogen, which may be
`branched or a Straight chain. Examples of alkyl groups are
`methyl, ethyl, n-butyl, t-butyl, n-heptyl, and isopropyl.
`“Lower alkyl” refers to an alkyl radical of one to six carbon
`atoms, and preferably one to four carbon atoms, as exem
`plified by methyl, ethyl, isopropyl, n-butyl, isobutyl, and
`t-butyl.
`
`0041 A“truncated” protein or coding sequence has some
`portion of the normal protein or Sequence removed from one
`or the other terminus, from an internal region, or a combi
`nation of the above.
`0042 An amino-truncated (N-truncated) or carboxy
`truncated (C-truncated) protein is one having an abnormal or
`deleted amine terminus or carboxy terminus, respectively,
`arising from translation of a splice variant mRNA.
`II. AntiSense Compounds
`0043)
`0044) In accordance with the present invention, it has
`been discovered that an antisense compound having from 12
`to 25 nucleotides, including a targeting base Sequence that is
`complementary to a target region of a Selected preprocessed
`mRNA coding for a selected protein, where the 5' end of the
`target region is 1 to 25 bases downstream, preferably 2 to 20
`bases downstream, and more preferably 2 to 15 bases
`downstream, of a normal splice acceptor Site in the prepro
`cessed mRNA, is effective to inhibit splicing at the normal
`Splice acceptor Site and thus produce splice variant mRNA,
`leading to truncated or otherwise aberrant versions of the
`Selected protein upon translation. Advantages of this Strat
`egy are Set forth below.
`004.5 The antisense compound employed in the present
`invention is one that does not activate RNase H. RNase-H
`active oligomers, of which phosphorothioate oligonucle
`otides are the most prominent example, operate primarily by
`a mechanism in which the target mRNA is cleaved. RNase
`incompetent oligomers, on the other hand, are believed to
`act by a steric blocking mechanism. Such compounds
`include morpholino oligomers, PNA's (peptide nucleic
`acids), methylphosphonates, and 2'-O-alkyl or -allyl modi
`fied oligonucleotides, all of which are known in the art. The
`preferred antisense oligomers (compounds) of the present
`invention are morpholino oligomers, which are composed of
`morpholino subunits of the form shown in U.S. Pat. Nos.
`5,698,685, 5,217.866, 5,142,047, 5,034,506, 5,166,315,
`5,521,063, and 5,506,337, all of which are incorporated
`herein by reference. The Synthesis, Structures, and binding
`characteristics of morpholino oligomers are detailed in these
`patents. In a morpholino oligomer, (i) the morpholino groups
`are linked together by uncharged phosphorus-containing
`linkages, one to three atoms long, joining the morpholino
`nitrogen of one Subunit to the 5' exocyclic carbon of an
`adjacent Subunit, and (ii) the base attached to the morpholino
`group is a purine or pyrimidine base-pairing moiety effective
`to bind, by base-specific hydrogen bonding, to a base in a
`polynucleotide. The purine or pyrimidine base-pairing moi
`ety is typically adenine, cytosine, guanine, uracil or thymine.
`Preparation of Such oligomers is described in detail in U.S.
`Pat. No. 5,185,444 (Summerton and Weller, 1993), which is
`hereby incorporated by reference in its entirety. AS shown in
`the reference, Several types of nonionic linkages may be
`used to construct a morpholino backbone.
`0046 Such morpholino oligomers have shown high bind
`ing affinity for RNA targets, and the uncharged backbone
`favors uptake into cells and reduces non-specific binding
`interactions, relative to charged analogS Such as phospho
`rothioates. They have been shown to provide significantly
`improved activity and Selectivity in inhibiting translation of
`targeted Sequences in comparison to phosphorothioate oli
`gonucleotides. See, for example, Summerton et al., Anti
`sense & Nucleic Acid Drug Dev. 7(2): 63-70, Apr 1997. The
`
`

`

`US 2003/0166588A1
`
`Sep. 4, 2003
`
`morpholino oligomers have very high nuclease resistance
`and good water Solubility, making them good candidates for
`in vivo use. Efficient uptake by cells in vivo is demonstrated
`in co-owned and copending application Serial no. 09/493,
`427 and the corresponding PCT Pubn. No. WO 0044897. As
`described therein, morpholino oligonucleotides having
`phosphoramidate linkages formed heteroduplexes with tar
`get RNA, which are protected in this duplex state from
`nuclease degradation. Such a duplex is expelled from the
`cell, and the target RNA can later be detected in a body fluid
`Sample from the Subject. These results demonstrated that the
`morpholino oligomers (i) migrate to and enter cells in the
`body and (ii) bind with high affinity, via Watson-Crick
`base-pairing, to target nucleic acid regions.
`0047 Exemplary backbone structures for antisense oli
`gonucleotides of the invention include the B-morpholino
`subunit types shown in FIG. 1A-E, as described above. It
`will be appreciated that a polynucleotide may contain more
`than one linkage type.
`0.048. A preferred morpholino oligonucleotide is com
`posed of morpholino subunit structures of the form shown in
`FIG. 2B-B, where the structures are linked together by
`phosphorodiamidate linkages, joining the morpholino nitro
`gen of one Subunit to the 5' exocyclic carbon of an adjacent
`Subunit, P, and P are purine or pyrimidine base-pairing
`moieties effective to bind, by base-specific hydrogen bond
`ing, to a base in a polynucleotide. In preferred embodiments,
`the linkages are Selected from a phosphorodiamidate linkage
`as represented at FIG.2B-B, where X=NH, NHR, or NRR',
`Y=O, and Z=O, and an alternate phosphorodiamidate link
`age as represented at FIG. 2B-B, where X=OR, Y=NH or
`NR, and Z=O. R and R' are groups which do not interfere
`with target binding. Preferably, R and R' are independently
`Selected from alkyl and polyalkyleneoxy (e.g. PEG,
`(CH2CH2O),), or a combination thereof. The alkyl/poly
`alkyleneoxy chain may be Substituted, preferably at the
`distal terminus, by a group Selected from hydroxy, alkoxy,
`amino, alkylamino, thiol, alkanethiol, halogen, OXO, car
`boxylic acid, carboxylic ester, and inorganic ester (e.g.
`phosphate or Sulfonate). Preferably, the chain (independent
`of Substituents) is from 1 to 12 atoms long, and more
`preferably is from 1 to 6 atoms long. In Selected embodi
`ments, R and R' are independently methyl or ethyl. In one
`embodiment, X=N(CH), Y=O, and Z=O. NRR' may also
`represent a nitrogen heterocycle having 5-7 ring atoms
`Selected from nitrogen, carbon, oxygen, and Sulfur, and
`having at least as many carbon ring atoms as non-carbon
`ring atoms. Examples include morpholine, pyrrolidine, pip
`eridine, pyridine, pyrimidine, pyrazine, triazine, triazole,
`pyrazole, pyrrole, isopyrrole, imidazole, oxazole, imidazole,
`isoxazole, and the like.
`0049. The solubility of the antisense compound, and the
`ability of the compound to resist precipitation on Storage in
`Solution, can be further enhanced by derivatizing the oligo
`mer with a Solubilizing moiety, Such as a hydrophilic oli
`gomer, or a charged moiety, Such as a charged amino acid or
`organic acid. The moiety may be any biocompatible hydro
`philic or charged moiety that can be coupled to the antisense
`compound and that does not interfere with compound bind
`ing to the target Sequence. The moiety can be chemically
`attached to the antisense compound, e.g., at its 5' end, by
`well-known derivatization methods. One preferred moiety is
`a defined length oligo ethylene glycol moiety, Such as
`
`triethyleneglycol, coupled covalently to the 5' end of the
`antisense compound through a carbonate linkage, via a
`piperazine linking group forming a carbamate linkage with
`triethyleneglycol, where the Second piperazine nitrogen is
`coupled to the 5'-end phosphorodiamidate linkage of the
`antisense. Alternatively, or in addition, the compound may
`be designed to include one a Small number of charged
`backbone linkages, Such as a phosphodiester linkage, pref
`erably near one of the ends of the compound. The added
`moiety is preferably effective to enhance solubility of the
`compound to at least about 30 mgs/ml, preferably at least 50
`mgS/ml in aqueous medium.
`0050. The compound is designed to hybridize to the
`target Sequence under physiological conditions with a T.
`substantially greater than 37 C., e.g., at least 50 C. and
`preferably 60 C-80 C. Although the compound is not
`necessarily 100% complementary to the target Sequence, it
`is effective to stably and Specifically bind to the target
`Sequence Such that expression of the target Sequence, is
`modulated. The appropriate length of the oligomer to allow
`Stable, effective binding combined with good Specificity is
`about 8 to 40 nucleotide base units, and preferably about
`12-25 base units. Mismatches, if present, are leSS destabi
`lizing toward the end regions of the hybrid duplex than in the
`middle. Oligomer bases that allow degenerate base pairing
`with target bases are also contemplated, assuming base-pair
`Specificity with the target is maintained.
`0051 III. Selection of Target Sequences
`0.052 A. RNA Splicing: Background
`0053. The processing of nuclear RNA following tran
`scription is observed in virtually all living cells. The mam
`malian genome contains genes that make transcripts of
`approximately 16,000 bases in length containing 7 to 8
`exons. The process of Splicing reduces the length of the
`mRNA to an average of 2,200 bases. The initial transcript is
`referred to as heterologous nuclear RNA (hnRNA) or pre
`mRNA. Processing of hnRNA involves an aggregate of
`approximately 20 proteins, referred to collectively as the
`Spliceosome, which carries out Splicing and transport of
`mRNA from the nucleus. The SpliceoSome does not appear
`to Scan from a common direction for all transcripts, introns
`may be removed in a reproducible order but not in a
`directional order. For example, introns 3 and 4 may be
`removed first, followed by removal of introns 2 and 5,
`followed by removal of introns and 6. The order of intron
`removal is not predictable a priori of observation. The
`Sequence recognition for processing is Small, Suggesting that
`errors or multiplicity of processing sites can be anticipated,
`and, in fact, as more genes are investigated, more variation
`in processing of hnRNA has been observed.
`0054.
`In preprocessed mRNA, the two-base sequence
`motifs at exon/intron junctions are invariant. The upstream
`(5') splice donor (SD) junction is of the form exon-/GT
`intron, while the downstream (3) splice acceptor (SA)
`junction is of the form intron-AG/exon. The flanking bases
`are not invariant; however, the base immediately upstream
`of the splice acceptor AG sequence is C about 80% of the
`time.
`
`

`

`US 2003/0166588A1
`
`Sep. 4, 2003
`
`.
`
`.
`
`TACTAAC .
`
`0.055 The current understanding of intron sequence rec
`ognition is as follows:
`0056) 5'-exonAGay
`G100T100)A(62)A(68)G84)T(63)
`CsoAloo Goo/exon-NN. . . -3'
`0057 splice donor (SD) splice acceptor (SA)
`0.058. The numbers in parenthesis represent the approxi
`mate (when<100%) percent utilization of a base at a site.
`The bold A in the middle of the intron is the site (branch
`point) at which the G from the splice donor forms a branched
`2'-5'-Structure referred to as the lariat. The Sequence indi
`cated (TACTAAC) is the consensus observed in yeast. The
`consensus
`Sequence
`in
`mammalian
`cells
`is
`PyNPysoPysz,PuzsaPY7s, where Py represents a pyri
`midine (U/T or C) and Pu represents a purine (A or G). The
`A at the branch point is invariant, and is typically found
`about 12 to 50 bases upstream of the AG site. A pyrimidine
`
`0067 Prior art methods have utilized strategy 2 or 4
`(targeting of SD or SAjunction). See, for example, RV Giles
`et al., cited above, in which a morpholino antisense oligomer
`Spanning a Splice acceptor Site in the c-myc gene is
`described.
`0068 Experiments in support of the present invention
`found that targeting Strategies 4 and 5 above, directed to the
`Splice acceptor, were more reliably effective than Strategies
`1 and 2, directed to the Splice donor.
`0069. For example, studies were carried out with rat
`CYP3A2 pre-mRNA targeted in vivo (whole animal). Ani
`mals were injected i.p. with 100 ug PMO (as shown in FIG.
`2B-B, where Y, and Z are oxygen and X is N(CH)) in
`phosphate buffered saline. The diminished rate of microso
`mal metabolism of erythromycin O-demethylase was moni
`tored to reflect the expected phenotype caused by the
`antisense inhibition. As shown, the splice donor (SD) tar
`geting was less effective than the splice acceptor (SA)
`Strategy.
`
`ANTISENSE
`SEQUENCE ( / indicates splice
`STRATEGY junction)
`
`SEQ ID NO : ERDEM & of control
`
`Control Saline
`
`SD
`
`SA
`
`3'-AAGAGATGGC/CACTCACTGG-5'
`
`3'-GGAAATATC/TGAACCTTGGG-5'
`
`100 + 10.2 (N = 7)
`
`--
`
`4
`
`5
`
`94.7
`
`5. 4 (N
`
`3)
`
`865
`
`3. 8 (N = 3)
`
`rich region (not shown above) is also generally found near
`the 3' end of the intron, about 10-15 bases upstream of the
`AG site.
`0059) Once the spliceosome forms the lariat, two trans
`esterification reactions take place: 1) the 2'-OH of the branch
`point A in the intron to the 5'-phosphate of the SD intron G,
`and 2) the 3'-OH of the SD exon G to the 5'-phosphate of the
`first base of the SA exon. The removed intron is rapidly
`degraded in most cases, and the joined exons are now
`referred to as mature mRNA, which is transported out of the
`nucleus for translation into proteins by the ribosome.
`0060 B. Targeting Strategies
`0061
`Various approaches could be taken to targeting the
`processing (splicing) of mRNA by antisense oligomers. The
`following sites could be targeted:
`0.062
`1. The exon-onlv portion of the SD exon
`y p
`(upstream of SD site), to interfere with SD processing.
`0063. 2. The SD exon/intronjunction, to interfere with
`SD processing.
`0064 3. The site of lariat formation within the intron,
`to block Spliceosomal recognition upstream of the SA
`CXO.
`0065. 4. The SA intron/exonjunction, to interfere with
`SA processing.
`0066 5. The exon-only portion of the SA exon (down
`Stream of SA site), to interfere with SA processing.
`
`0070 Experiments with oligomers antisense to c-myc
`mRNA were conducted in cultured rat NRK cells, evaluating
`the inhibition of cellular DNA synthesis by monitoring
`incorporation of tritiated thymidine. Sequences were derived
`from Genbank Acc. No. YO0396 (rat) and J00120 (human),
`targeting the Splice acceptor region at the beginning of exon
`2 (with the exception of SEQ ID NO: 1, which targets the
`splice donor). The rat and human sequences indicated are
`highly homologous in this region. The oligomers listed in
`Table 1, below, were screened for antiproliferative effects
`using Several proliferation assayS. Primary fibroblasts from
`two species, rat and human, NRK and WI-38, respectively,
`were used for the Screening.
`0.071) Data shown in Table 1 employed NRK cells.
`0072) Data from
`Hlthymidine incorporation assays
`using 20 uM PMO (as shown in FIG. 2B-B, where Y and
`Z are oxygen and X is N(CH)) are presented in Table 1.
`Percentages refer to Hthymidine incorporation relative to
`the vehicle (HO) control. Therefore, the lower the number,
`the greater the antiproliferative effect. It can be seen that all
`oligomers tested exhibited at least Some antiproliferative
`activity. The extent of the inhibitory activity compared
`favorably with the antiproliferative drug Taxol (Paclitaxel,
`Bristol-Myers Squibb, Princeton, N.J.) at 32% of control. As
`10%-20% of the cells are not affected by the scrape loading
`procedure and will, therefore, contribute to the residual
`Hthymidi