Synthesis and characterisation of polyamine-poly(ethylene glycol) constructs for DNA binding and gene delivery

Bioorganic & Medicinal Chemistry 8 (2000) 1779±1797

Synthesis and Characterisation of Polyamine±Poly(ethylene

glycol) Constructs for DNA Binding and Gene Delivery

Shane W. Garrett,^yOwen R. Davies, David A. Milroy, Pauline J. Wood,

Colin W. Pouton and Michael D. Threadgill*

Department of Pharmacy and Pharmacology, University of Bath, Claverton Down, Bath BA2 7AY, UK

Received 7 February 2000; accepted 21 March 2000

AbstractÐImproved non-viral vector systems are needed for ecient delivery of DNA to target cell nuclei in gene therapy. A series

of linear polyamine±poly(ethylene glycol) (PEG) constructs has been synthesised by reaction of appropriately Boc-protected thermine

derivatives with o-methoxyPEG oxiranylmethyl ethers. Constructs carrying 1±3 MeOPEG units and 0, 2 or 4 N-methyl groups

have been prepared by this method. H₂N(CH₂)₃NBoc(CH₂)₃NBoc(CH₂)₃NHBoc was prepared eciently by mono-tri¯uoro-

acetylation of thermine, attachment of Boc and removal of the tri¯uoroacetyl group in one pot. A similar process gave

H₂N(CH₂)₃NBoc(CH₂)₃NBoc(CH₂)₃NH₂. BocMeN(CH₂)₃NHMe was alkylated by 1,3-dibromopropane to give BocMeN(CH₂)₃-

NMe(CH₂)₃NMe(CH₂)₃NMeBoc. A cyanoethylation/reduction sequence extended H₂N(CH₂)₃NBoc(CH₂)₃NBoc(CH₂)₃NH₂to

give H₂N(CH₂)₃NBoc(CH₂)₃NBoc(CH₂)₃NBoc(CH₂)₃NBoc(CH₂)₃NH₂, which was converted to its mono- and di-MeOPEG550

derivatives. Deprotection gave the linear polyamine±MeOPEG constructs. A branched triamine±poly(ethylene glycol) construct

was prepared by acylation of (BocHN(CH₂)₃)₂N(CH₂)₃NH₂with o-methoxyPEG 550 chloroformate, followed by deprotection. A

cyanoethylation/reduction/protection sequence from (H₂N(CH₂)₃)₂N(CH₂)₃NHBoc gave a protected pentamine. Alkylation with

Br(CH₂)₅CONH(CH₂)₂NHBoc, deprotection, acylation with MeOPEG chloroformate and deprotection gave a pentamine±MeO-

PEG construct in which the MeOPEG is attached through a linker to the central amine. The linear hexamine construct carrying

MeOPEG550 at only one terminus was the most eective DNA-interactive member of the two series in an ethidium displacement

Introduction

toxicity have also been proposed.²Stimulation of anti-

tumour immune responses through gene therapy has

been investigated.^3,4

Gene therapy is of growing interest in the therapy of sev-

eral diseases. In some diseases, there is a defect in a speci®c

gene and the therapeutic strategy is to supply a gene that

will lead to biosynthesis of the appropriate functional

protein. In some other cases, the strategy is to deliver a

gene that encodes a protein that is not missing or dys-

functional in the patient but which would provide a dif-

ferent therapeutic bene®t. For example, genes encoding

toxic proteins may be introduced into tumour cells. The

Gene Directed Enzyme Prodrug Therapy (GDEPT)

approach is also being developed¹for cancer therapy; in

GDEPT, a gene that encodes a prodrug-activating enzyme

is introduced (e.g., herpes simplex virus thymidine kinase,

which activates ganciclovir by phosphorylation). Genes

that render malignant cells more sensitive to cytotoxic

therapy or which protect normal cells against drug

One of the major limiting factors in the development of

gene therapy strategies is the unavailability of ecient and

selective delivery of DNA to the nucleus of the target cells.

Current delivery vectors (systems for delivery of DNA) are

broadly divided into two types: viral and non-viral.

Approximately 70% of current clinical trials use viral gene

delivery and the majority of these use retroviruses.

Although, at present, viral vectors appear to be more e-

cient than non-viral systems, there are a number of prob-

lems associated with their use, including immunogenicity,

transient duration of gene expression, low capacity for

foreign genes and diculties in generation of suciently

high viral titres.^{5 7}Non-viral vectors fall broadly into

two classes: cationic lipids and cationic polymers (such

as poly-l-Lys (PLL) and polyethyleneimine (PEI)).

Both bind electrostatically to the polyanionic DNA,

provide charge neutralisation and cause condensation of

DNA. Several cationic lipids and lipid±polyamine con-

structs are available for use as transfection agents to

*Corresponding author.: +44-1225-826826; fax: +44-1225-826114;

e-mail: m.d.threadgill@bath.ac.uk

^yPresent address: Centre for Polymer Therapeutics, School of Pharmacy,

University of London, Brunswick Square, London WC1N 1AX, UK.

PII: S0968-0896(00)00113-9

1780

S. W. Garrett et al. / Bioorg. Med. Chem. 8 (2000) 1779±1797

carry DNA across cell membranes in vitro. These include

DOTMA,⁸DOTAP, DOGS, DOSPA, DC-Chol⁹and

CTAP.¹⁰However, there are a number of problems with

lipidic systems. Firstly, complexes of cationic lipids and

DNA tend to aggregate in aqueous solutions, particularly

in the presence of salts or proteins,¹¹thus only small

amounts of DNA can be formulated in this way. After

intravenous injection these hydrophobic particles typi-

cally activate complement;¹²they are oposonised and

subsequently rapidly cleared by the liver and spleen.^13,14

PLL±DNA complexes also activate complement; how-

ever, surface modi®cation of preformed PLL±DNA

Scheme 1. Synthesis of tetramine±MeOPEG constructs 14 and 15. Reagents: (i) propenenitrile, MeOH; (ii) Boc₂O, CH₂Cl₂; (iii) LiAlH₄, Et₂O,

THF; (iv) H₂, Raney Ni, MeOH, NH₃; (v) CF₃CO₂Et, MeOH; (vi) NH₃, MeOH; (vii) chloromethyloxirane, NaOH, H₂O; (viii) 10a, PrⁱOH, Á,

24 h; (ix) 10a, PrⁱOH, Á, 24 h; (x) HCl, CH₂Cl₂.

S. W. Garrett et al. / Bioorg. Med. Chem. 8 (2000) 1779±1797

1781

complexes with PEG considerably reduces complement

activation.¹²Secondly, lipidic systems are unsuitable for

injection into solid tissues such as tumour and muscle

where dispersion of the formulation is thought to be

important;¹³the transfection eciency of these `lipo-

plexes' is low when injected directly into tumours, in

comparison to that of naked DNA.¹⁵Thirdly, lipidic

complexes are relatively unstable.^13,14

active Non-Condensing (PINC) polymers. The aim of

the work presented in this paper was to synthesise a

family of polyamine±poly(ethylene glycol) constructs

which would be capable of forming hydrophilic com-

plexes with DNA, coating the DNA with a sheath of

poly(ethylene glycol) (PEG) chains. PEG is also non-

immunogenic and relatively non-toxic.²³Results of our

study on the binding of the constructs to DNA are also

presented, along with those of our preliminary study on

the eectiveness of two constructs for gene delivery in vivo.

Thus, there has been increasing interest in water-soluble

complexes, as opposed to lipidic systems.^{16 22}These sys-

tems typically consist of a polycation linked to a hydro-

philic polymer. For delivery to solid tissues (muscle and

solid tumours), complete charge neutralisation and con-

densation of DNA does not appear to be a requirement,

as the neutral hydrophilic polymers poly(1-vinylpyrroli-

din-2-one) (PVP) or poly(vinyl alcohol) (PVA)^20,21

appear to be active as gene transfer agents in these

tissues.^{3,20 22}These are the so-called Protective Inter-

Design and Chemical Synthesis of the Constructs

The simplest design of a PEG±polyamine construct

would be to allow a PEG-derived electrophile to react

with the nucleophilic terminal amino groups of a linear

polyamine. N¹,N³-Bis(3-aminopropyl)propane-1,3-di-

amine (`thermine', 6, Scheme 1) was selected as the

Scheme 2. Synthesis of tetramethyltetramine±MeOPEG construct 20 and dimethyltetramine±MeOPEG construct 26. Reagents: (i) Boc₂O, THF;

(ii) Br(CH₂)₃Br, K₂CO₃, DMF; (iii) HCl, CH₂Cl₂; (iv) NaOH, 10a, PrⁱOH; (v) KOBu^t, MeI, THF; (vi) NH₃, MeOH; (vii) phenoxymethyloxirane,

PrⁱOH; (viii) 10b, PrⁱOH.

1782

S. W. Garrett et al. / Bioorg. Med. Chem. 8 (2000) 1779±1797

Scheme 3. Synthesis of ``one ended'' tetramine±MeOPEG constructs 34±37. Reagents: (i) CF₃CO₂Et, EtOH; (ii) Boc₂O, EtOH; (iii) NH₃, MeOH;

(iv) 10a,b, PrⁱOH, Á; (v) HCl, CH₂Cl₂.

initial polyamine, as it has the potential to carry up to four

positive charges at physiological pH to provide good

interaction with the polyanionic DNA. Furthermore,

alkane-a,o-diamines having three or ®ve CH₂units

between the amines are reported²⁴to be considerably

more eective in inducing DNA compaction than were

other alkane-a,o-diamines. Boc protection of the second-

ary amines was necessary to ensure that only the terminal

primary amines reacted with the PEG-derived electro-

phile; this protection should also be readily removable

under conditions which do not destroy the PEG, giving

co-products which are easily separable from the oligo-

meric highly polar target constructs. The required pro-

tected polyamine 4 was accessed through two routes

(Scheme 1). In the ®rst route, treatment of propane-1,3-

diamine 1 with propenenitrile gave the dinitrile 2 quan-

titatively.²⁵The secondary amines were then protected

as their Boc derivatives in 3, in the usual way. McCor-

mick et al.²⁶report that selective reduction of the nitrile

groups of 3 can be achieved using LiAlH₄at 0 ^ꢀC to give

the diBoc-tetramine 4 in 50% yield. In our laboratory,

these conditions gave a poor yield (22%) of the desired

diBoc-tetramine 4, the major product (37%) being the

diBoc-triamine 5. Formation of 5 can be rationalised in

terms of a retro-Michael reaction of the substrate 4 or

an intermediate, initiated by highly basic AlH₄. How-

ever, reduction of the nitriles was achieved by Raney-

Ni-catalysed hydrogenation under forcing conditions in

ammonia-saturated methanol to aord 6. In the second

route, reduction of the nitriles (of 2) was carried out

before selective protection of the secondary amines.

Catalytic hydrogenation under similar conditions gave

thermine 6. Xu et al.²⁷developed conditions for selective

tri¯uoroacetylation of primary amines in the presence of

secondary amines using ethyl tri¯uoroacetate in THF at

low temperature, the selectivity arising from steric fac-

tors. Application of a modi®ed procedure gave the di-

TFA-protected tetramine 7. This material was dicult

to purify; immediate reaction with di-t-butyl dicarbonate

aorded the orthogonally protected tetramine 8 in high

yield. Removal of the tri¯uoroacetyl groups with ammo-

nia gave the desired diBoc-tetramine 4 quantitatively. The

sequence 1!2!6!7!8!4 thus represents the most

ecient route to provide 4 which is now available for

reaction with a suitable PEG-derived electrophile.

The PEG unit for most of the constructs was PEG550

monomethyl ether (MeOPEG550, 9a, a mixture of oli-

gomers of mean MW 550 Da). A speci®c requirement in

converting the alcohol of 9a into a suitable electrophile for

reaction with 4 is that the basicity of all four nitrogen

atoms in the construct should be uncompromised, to per-

mit formation of a tetracation for ecient binding to

DNA under physiological conditions. Oxiranes react

readily with primary amines to form 2-hydroxyalkyl sec-

ondary amines; reaction with secondary amines gives 2-

hydroxyalkyl tertiary amines. We recently used^28,29a,o-

bis(oxiranylmethyl)PEGs as electrophiles in the assem-

bly of cathepsin B-degradable polymers for delivery of

drugs and imaging agents. MeOPEG550 9a was con-

verted to its oxiranylmethyl ether 10a by treatment with

epichlorohydrin; MeOPEG2000 9b was converted^{

^{Experiment performed by Dr. S. E. Matthews, Institute of Macro-

molecular Chemistry, Czech Academy of Sciences, Prague, Czech

Republic. Present address: Inorganic Chemistry Laboratory, Depart-

ment of Chemistry, University of Oxford, South Parks Road, Oxford

OX1 3QR, UK. Compound 10b was a kind gift from Dr. Matthews.

S. W. Garrett et al. / Bioorg. Med. Chem. 8 (2000) 1779±1797

1783

similarly to 10b. In a preliminary experiment, 4 reacted

with two equivalents of 10a in boiling propan-2-ol to

give crude bis-adduct 11. Longer reaction times led to

formation of the tris-adduct 12 (11%) and the tetrakis-

adduct 13 (24%), which were isolable by chromato-

graphy; both were deprotected, giving the constructs 14

and 15, respectively. Clearly, the addition of the bis-

primary amine to the oxirane is not a clean reaction and

the secondary b-hydroxyalkylamine of the intermediate

11 is capable of further nucleophilic addition to oxirane.

to introduce the three Boc groups in 28. Selective

deprotection gave the tri-Boc-thermine 29. The sole

exposed amine in 29 is primary and thus may add twice

to MeOPEG-derived oxiranes. Indeed, treatments of 29

with single equivalents of 10a and 10b gave a separable

mixture of the mono-adducts 30 (21%) and 32 (35%),

respectively, and of the bis-adducts 31 (9%) and 33

(28%), respectively. Reaction with two equivalents of 10a

increased the isolated yield of 31 to 61%. Figure 1 shows

the MALDI-TOF mass spectrum of 33; the peaks corre-

spond to [M+Na]⁺for the PEG oligomers. Deprotection

aorded the adduct salts 34±37.

Our solution to this problem was to arrange the poly-

amine to have secondary amines at the termini, so that

the addition reaction could proceed once only at each

site (Scheme 2). The tetramines 19 and 22 were therefore

designed as appropriate polyamines for reaction with

MeOPEG oxiranylmethyl ethers. The former will give

rise to a PEG±polyamine construct in which all four

amines are methylated, whereas the construct derived

from the latter will have less sterically demanding second-

ary amines in the `inner' positions. Treatment of excess

N¹,N³-dimethylpropane-1,3-diamine 16 with di-t-butyl

dicarbonate³⁰gave the mono-Boc derivative 17. The tet-

ramine was then assembled by alkylation of 17 with 1,3-

dibromopropane, giving 18 in good yield. Deprotection

(giving salt 19) and reaction with 10a cleanly aorded

the desired bis-PEG construct 20 after chromatographic

puri®cation. As tertiary amines, the `inner' amines were

unreactive towards the oxirane and thus did not require

protection/deprotection. In contrast, the ``inner'' (sec-

ondary) amines did require protection in polyamine

derivative 22, prior to reaction with oxiranes. The di-

anion generated from the tri¯uoroacetamide termini of 8

reacted with iodomethane to aord the protected a,o-

dimethyltetramine 21, from which the tri¯uoroacetyl

groups were readily removed. The resulting diprotected

dimethyltetramine 22 was treated with phenoxymethylox-

irane as a model, giving the bis-adduct 24. Using the same

conditions, the required bis-MeOPEG adduct 24 was

prepared and deprotected, giving pure 26 in high yield,

uncontaminated by polyamines bearing more or less

MeOPEG units. Compound 22 was deprotected to pro-

vide 23 as a specimen of this parent polyamine for

comparative DNA-binding studies.

Constructs 11, 14, 15, 20, 26 and 34±37 represent varia-

tions in the number, size and location of the MeOPEG

units and introduction of various numbers of N-Me

groups while retaining the core linear tetramine. Con-

structs 42 and 45 (Scheme 4) are analogous to 34 and

11, respectively, but have a hexamine core. This core, in

which the basic nitrogens are still linked by (CH₂)₃units,

has the potential to exist as a hexacation and thus bind

more strongly to the polyanionic DNA. Hexamines of

this type are commercially unavailable; thus a suitably

tetra-Boc-protected hexamine 40 had to be assembled by

extension from a tetramine. Regioselective cyanoethyla-

tion of 6 and Boc protection of intermediate 38 gave 39.

Hydrogenation of the nitriles aorded the tetra-Boc-hex-

amine 40. Prolonged reaction with one equivalent of 10a

gave a moderate yield of the mono-adduct 41, whereas the

lower reactivity of this larger protected polyamine was

evident in the poor yields of the bis-adduct 43 (12%) and

the tris-adduct 44 (6%) on treatment with two equiva-

lents of the oxirane. Deprotection gave quantitatively the

corresponding mono- and bis-MeOPEG±hexamine con-

structs 42, 45. Thus the set of linear PEG±polyamine

constructs with variation in length and number of PEG

chains, number of cationic sites and sterically demand-

ing N-methylation was complete.

In an alternative design strategy, the PEG unit is

attached to a point in the centre of the polyamine,

rather than at a terminus as in 45. This has the potential

advantage of leaving the terminal primary amines of the

The above constructs carry MeOPEG units at both ends

of the polyamine. To investigate whether enhanced bind-

ing to DNA may be achieved by having one end of the

polyamine sterically unimpeded, a series of constructs was

assembled in which MeOPEG was attached to only one

end of the tetramine. The tetramine for this part of the

study was 6, to minimise possible steric crowding in bind-

ing. This requires the tetramine to be Boc protected at

the two secondary amines and at one of the terminal

primary amines (as in 29, Scheme 3), prior to reaction

with MeOPEG oxiranylmethyl ethers. McCormick et

al.²⁶have prepared 29 in moderate yield by careful

treatment of the diBoc-tetramine 4 with di-t-butyl

dicarbonate. However, it was more ecient to adapt a

one-pot synthesis³¹of tri-Boc-spermine to reaction of 6

with one equivalent of ethyl tri¯uoroacetate at low

temperature, giving the crude mono-tri¯uoroacetamide

27, which was treated in situ with di-t-butyl dicarbonate

Figure 1. MALDI-TOF mass spectrum of protected construct 33.

Peaks correspond to [M+Na]⁺for the PEG oligomers.

1784

S. W. Garrett et al. / Bioorg. Med. Chem. 8 (2000) 1779±1797

polyamine sterically unencumbered. It was considered

essential to preserve molecular symmetry by attaching

the PEG unit to the central nitrogen of the polyamine

and to retain the basicity of this tertiary amine; thus the

target branched polyamine±PEG constructs 53 and 67

(Scheme 5) were based on a linear triamine and a linear

pentamine, respectively.

MeOPEG550 50 was converted to its chloroformate 51

with excess phosgene. Reaction of 51 with the polyamine

derivative 49 was very rapid at 20 ^ꢀC in forming the pro-

tected construct 52; deprotection gave the branched

MeOPEG550 triamine construct as its trihydrochloride

salt 53.

The pentamine construct 67 (Scheme 1) not only has the

potential for formation of a pentacation for increased

electrostatic interaction with DNA but also has a longer

linker between the central tertiary amine and the MeO-

PEG. To assemble this linker, 6-bromohexanoic acid 62

was coupled with the mono-Boc-protected ethane-1,2-

diamine 61.³²The amide 63 then contains a Boc-masked

amine (for later reaction with the PEG-derived electro-

phile) and a bromoalkyl group for alkylation of a suitably

protected pentamine. Protection orthogonal to Boc of all

but the central amine of the pentamine was required. By

the general procedure of Xu et al.,²⁷46 was treated with

The simple triamine 46 was protected at the termini with

Boc, giving 47. Cyanoethylation of the crowded second-

ary amine required forcing conditions to give a good yield

of the nitrile 48, which was reduced to the amine 49. In

these branched constructs, the MeOPEG is not attached

directly to one of the polyamine nitrogens, as it is in 42.

Rather, it is attached through the 3-aminopropyl linker

in 53. This has the bene®t that the nitrogen to which the

MeOPEG unit is attached does not have to be basic, as it

is not required for electrostatic interaction with the DNA;

thus a more reactive MeOPEG electrophile can be used.

Scheme 4. Synthesis of hexamine±MeOPEG constructs 42 and 45. Reagents: (i) propenenitrile, MeOH; (ii) Boc₂O, CH₂Cl₂; (iii) H₂, Raney Ni,

MeOH, NH₃; (iv) 10a (1 equiv), PrⁱOH, Á; (v) HCl, CH₂Cl₂; (vi) 10a (2 equiv), PrⁱOH, Á.

S. W. Garrett et al. / Bioorg. Med. Chem. 8 (2000) 1779±1797

1785

ethyl tri¯uoroacetate at low temperature to give the

crystalline a,o-diamide 54 in good yield. Boc protection

of the central amine was followed by removal of the

terminal tri¯uoroacetyl groups in the usual way,

aording 56. Cyanoethylation (giving 57) and hydro-

genation lead to the mono-protected pentamine 58, car-

rying Boc at the central nitrogen only. The other four

amines were converted to their benzyl carbamates (in 59)

and the Boc group was removed, giving the pentamine

derivative 60, which has only the central secondary

amine exposed. This amine was alkylated eciently with

63 to provide the protected pentamine 64 carrying the

long functionalised side chain at the central nitrogen.

Deprotection of the linker and reaction of the exposed

primary amine with the chloroformate 51 gave the pro-

tected construct 66 in good yield. Hydrogenolysis of the

Cbz groups aorded the target branched MeOPEG550

pentamine construct 67.

Scheme 5. Synthesis of branched triamine±MeOPEG constructs 53 and 67. Reagents: (i) BocON, THF; (ii) propenenitrile, THF; (iii) H₂, Raney

Ni, MeOH, NH₃; (iv) COCl₂, PhMe, CH₂Cl₂; (v) 51, Et₃N, CH₂Cl₂; (vi) HCl, CH₂Cl₂; (vii) (COCl)₂, DMF, CH₂Cl₂; (viii) 61, Et₃N, CH₂Cl₂;

(ix) CF₃CO₂Et, EtOH; (x) Boc₂O, CH₂Cl₂; (xi) NH₃, MeOH; (xii) (BnO₂C)₂O, THF; (xiii) 63, K₂CO₃, DMF; (xiv) H₂, Pearlman's catalyst, MeOH.

1786

S. W. Garrett et al. / Bioorg. Med. Chem. 8 (2000) 1779±1797

DNA Binding

that bind to DNA can displace the intercalator ethidium

from ethidium±DNA complexes. This displacement can

be eected by intercalating compounds and by non-

intercalating agents, such as polycations, which bind as

The DNA-binding anity of the constructs was mea-

sured by an ethidium-displacement assay. Compounds

Figure 2. Graphs of the eects of addition of the linear polyamine±MePEG constructs 14, 20, 34±37, 42 and 45 on the ¯uorescence of a DNA±

ethidium complex. Decrease in ¯uorescence is a measure of the binding of the constructs to DNA.

S. W. Garrett et al. / Bioorg. Med. Chem. 8 (2000) 1779±1797

1787

a result of electrostatic interactions with the polyanionic

DNA. There is correlation between the anity of a

binding agent for DNA and the eciency of displace-

ment of ethidium.³³When ethidium binds to DNA, there

is a large increase in the ¯uorescence of the ethidium; this

¯uorescence is quenched when higher anity com-

pounds displace the ethidium. Simple spectro¯uorimetric

methods can be used to study the interaction of a range

of compounds with DNA; the technique has been

widely described.^{33 37}

Preliminary experiments showed that the ethidium

exclusion assay developed by Gershon et al.³⁴gave

irreproducible results when applied to the MeOPEG±

polyamine constructs. Thus the alternative ethidium

displacement assay³⁵was adopted to measure the bind-

ing of the constructs to DNA. The constructs were

added in small aliquots to a pre-formed ethidium±DNA

complex; decrease in ¯uorescence is a measure of dis-

placement of ethidium and thus of binding of the con-

structs (Fig. 2). The mechanism by which ethidium is

displaced by agents which do not intercalate, such as

polyamines, may be non-competitive and may involve

changes in DNA conformation.³⁶Displacement could

result from polyamine binding in the major groove,

minor groove or along the phosphate backbone;³⁷the

exact mechanism is not known.

Figure 3. Graphs of the eects of addition of the polyamine 23 on the

¯uorescence of a DNA±ethidium complex in the absence ±±^Ð and

presence ^

sure of the binding of the polyamine to DNA.

.... ....

of MeOPEG2000. Decrease in ¯uorescence is a mea-

chains (36 and 37) reduce binding. These eects suggest

that (i) there is steric obstruction to binding caused

either by the PEG chain attached to the polyamine

attempting to bind or the PEG chain of a dierent con-

struct molecule that is already bound, or (ii) the mode

of attachment of the MeOPEG lowers the pK_aof one

nitrogen such that it is not protonated at pH 7.4, the pH

of the assay. To test this latter hypothesis, the binding

of 34 to DNA was also measured at pH 5.5. The results

shown in Figure 4 indicate that 34 displaces ethidium

more eectively at pH 5.5 than it does at pH 7.4. This is

consistent with the amine to which the MeOPEG is

attached having a low pK_a, owing to the b-OH group

which can lower pK_aby ca. 1.3 units.⁴¹

Figure 2 shows the diminution of ¯uorescence of the

ethidium±DNA complex caused by addition of the

constructs. Results are expressed as percentage of the

original ¯uorescence versus mass ratio of construct/

DNA. Only constructs 34, 42, 45 and 66 show any sig-

ni®cant and ratio-dependent interaction with the DNA.

The simplest bis-MeOPEG550±tetramine construct 14

causes no diminution of the ¯uorescence at mass ratios

up to 100. Introduction of four N-Me groups on the

polyamine in construct 20 has no positive eect on the

apparent binding. Similarly, construct 26, which carries

N-Me groups at the termini of the polyamine but has

longer PEG chains, does not appear to interact with

DNA at mass ratios up to 100 (data not shown). The

parent dimethyltetramine 23 (lacking MeOPEG units)

binds strongly to DNA and causes condensation and,

thus, displacement of ethidium at low mass ratios; a mass

ratio of ca. 2.5 causes 50% decrease in ¯uorescence. This

shows that N-methylation per se does not have a

strongly deleterious eect on the binding of polyamines

to DNA. Interactions between PEG and DNA may be

thermodynamically unfavourable;^38,39this may be asso-

ciated with the observed collapse of DNA structure with

increasing concentrations of PEG in aqueous solution.

However, addition of MeOPEG2000 9b (two equiva-

lents per equivalent of 23) does not measurably aect

the binding of 23 to DNA (Fig. 3) and the concentration

of PEG used does not itself induce DNA coil col-

lapse.^38,40Constructs 34±37 have MeOPEG units

attached to only one end of the tetramine and lack N-Me

groups. Of these, the most potent in displacing ethidium

from its DNA complex is the mono-MeOPEG550±tet-

ramine construct 34; even this only achieves a 25%

diminution of ¯uorescence at a mass ratio of 100 (Fig.

2). A second MeOPEG unit (in 35) or longer MeOPEG

Figure 4. Graphs of the eect of pH on the change in ¯uorescence

caused by addition of the polyamine±MePEG construct 34 to

a DNA±ethidium complex. Experiments were performed at pH 7.4

.... ....

. Decrease in ¯uorescence is a measure

Ð±^Ð± and pH 5.5

^

of the binding of the constructs to DNA.

1788

S. W. Garrett et al. / Bioorg. Med. Chem. 8 (2000) 1779±1797

Extending the polyamine to be a hexamine should ame-

liorate this eect, in that the central tetramine should

now be fully protonated at pH 7.4; furthermore, there

should also be some protonation of the terminal amines,

leading to enhanced binding to the DNA. Hence the

mono-MeOPEG550±hexamine construct 42 and the bis-

MeOPEG550±hexamine construct 45 were designed and

evaluated. Construct 42 is relatively ecient at expelling

ethidium from the complex (Fig. 2), indicating that it has

good anity for DNA. The ¯uorescence of the ethidium±

DNA complex is diminished by ca. 50% at a mass ratio of

10 42/DNA, which compares very favourably with the

eect of the analogous tetramine construct 34 (ca. 5%

diminution of ¯uorescence at mass ratio 10). Similarly,

45 appears to bind weakly but signi®cantly at mass ratio

100.

muscle^{45 48}or tumours.¹⁵This activity may be equal or

greater than that produced by lipoplexes,¹⁵despite cell

culture models that predict that lipoplexes are several

orders of magnitude more active. Gene expression from

naked plasmid DNA in vivo after direct injection into

solid tumours can be enhanced by formulation of the

DNA in a hydrophilic polymer solution. This phenom-

enon has been reported after direct injection of a

reporter plasmid into mouse muscle using either PVP or

PVA.^20,21,49There are very little data relating to the

injection in vivo of plasmid DNA formulations in which

the DNA is only partially condensed.

Having no eukaryotic equivalent, the chloramphenicol

acetyltransferase (CAT) gene has become one of the

standard markers used in transfection studies both in

vitro and in vivo. It also bene®ts from a relatively long

half-life (50 h in vivo), making it particularly useful in

animal experiments. Formulations of pCMVCAT (a

plasmid containing the CAT gene) with saline, PVP, 34

or 42 were injected into RIF-1 tumours and the expres-

sion was measured after 48 h (Fig. 5). All three plasmid

formulations produced enhanced expression of CAT (3-

to 4-fold) in comparison to pCMVCAT injected in sal-

ine. There was no statistically signi®cant dierence in

tumour expression between PVP, 34 and 42 formula-

tions. The enhancement of expression of CAT in the

tumours was comparable to that reported by Mumper et

al.,²⁰who used the CAT reporter gene in muscle. The

expression levels obtained following direct injection of

The branched triamine construct 53 does not interact

strongly with DNA, up to mass ratio (construct/DNA)=

100. However, a signi®cant ratio-dependent diminution in

¯uorescence is caused by the pentamine construct 67, with

>35% diminution at mass ratio=100. In 53 and 67, the

MeOPEG unit is attached in such a way that the pK_aof

this central amine should be unaected. Nevertheless,

the electrostatic interaction of the potentially tricationic

polyamine in 53 may not be sucient to hold the MeO-

PEG unit close to the DNA. In 67, several structural

modi®cations may allow binding: (i) it is now a potential

pentacation, which will allow more electrostatic attrac-

tion and hydrogen-bonding with the DNA phosphate

backbone; (ii) the linker is less sterically demanding in

the region which may lie in the DNA groove.

It is interesting to compare these results with those

reported using PEG±poly-l-lysine (PLL±PEG), PEG±

poly(ethyleneimine) (PEI±PEG) and PEG±polyspermine

(PEG±PS) constructs. PLL±PEG has been produced in

two variants, one a block co-polymer involving the N-

terminal a-NH₂¹⁸and the other involving the Lys e-NH₂

groups,¹⁹both using long amine chains (in one case 120

Lys). Both variants condensed DNA to levels close to

those achieved using free poly-l-Lys, irrespective of

whether the Lys e-NH₂polymer had a PEG coverage of

5, 10 or 25 mol%. PEG±PEI and PEG±PS also con-

densed DNA, producing very small particles (12±32 nm).

It is unlikely that the constructs described in this paper

will produce particles, although the ethidium exclusion

assays for 42 and for histone H1 are similar; since histone

H1 produces particles,⁴²it is conceivable that 42 may

also form particles with DNA. A major problem with

PLL±PEG, PEI±PEG and PEG±PS polymers is that, on

addition to DNA, complexes form which are poorly

soluble and aggregate. Thus condensed DNA (particu-

late) delivery systems cannot be produced in suciently

high concentrations for intramuscular or intratumoural

injection.⁴³The PEG±polyamine constructs in the pre-

sent study, including 42, allow formulation of DNA at

2500 mg mL ¹, which is ca. 50Â the maximum con-

centration that can be formulated using poly-l-Lys.⁴⁴

Figure 5. Eect of polymer solution (poly(1-vinylpyrrolidin-2-one),

SG166 or 42) on expression of CAT in RIF-1 tumour tissue. Plasmid

pCMVCAT (50 mg) was injected directly into each tissue in either iso-

tonic saline (20 mL), 5% PVP, 5% 34 or 5% 42 (n=8). [**denotes 99%

signi®cance.]

Although `naked' plasmid DNA has little or no activity

in a typical cell culture transfection experiment, naked

DNA can be active in vivo after direct injection into

S. W. Garrett et al. / Bioorg. Med. Chem. 8 (2000) 1779±1797

1789

plasmid were variable, as reported for mouse skeletal

Experimental

muscle.^48,50Constructs 34 and 42 also enhanced CAT

expression in the tumour tissue, with respect to for-

mulation in saline, but were no more eective than PVP.

These data suggest a generic eect of soluble interacting

agents on gene expression that is not dependent on high

molecular weight. Such an enhancement could be valu-

able for formulation of DNA vaccines; further studies

will probe the mechanism of this enhancement.

General methods

NMR spectra were recorded on samples in CDCl₃, unless

otherwise stated. Mass spectra were obtained by fast atom

bombardment (FAB) in the positive ion mode, unless

otherwise stated. The stationary phase for chromato-

graphy was silica gel. Melting points are uncorrected.

Solutions in organic solvents were dried with MgSO₄. Sol-

vents were evaporated under reduced pressure. The brine

was saturated.

Conclusions

N¹,N³-Bis(2-cyanoethyl)propane-1,3-diamine (2). Prope-

nenitrile (14.3 g, 270 mmol) was added to propane-1,3-

diamine (10.0 g, 130 mmol) in MeOH (15 mL) at 0 ^ꢀC

during 20 min. The mixture was stirred at 0 ^ꢀC for 20 min

and at 20 ^ꢀC for 16 h. Evaporation gave 2 (24.3 g, 100%)

as a colourless liquid: bp_0.3(Kugelrohr) 235 ^ꢀC (lit.²⁵

bp_0.5176 ^ꢀC). Found C, 60.0; H, 9.0, N, 31.2. C₉H₁₆N₄

Versatile, ecient syntheses of protected tetramines,

suitable for coupling to electrophilic MeOPEGs, have

been developed; N-methylation was achieved in high

yield by alkylation of tri¯uoroacetamides. A high yield-

ing synthesis of a protected hexamine has also been

devised. Attachment of MeOPEG (via its oxiranylmethyl

ether) gave the corresponding linear MeOPEG±poly-

amine constructs. This oxirane-addition reaction led to

multiple reaction with some primary amines but chro-

matographic separation of constructs with diering

numbers of MeOPEG units was facile. A novel triamine

and a novel pentamine have also been synthesised and

were linked to MeOPEG550 via the central amine while

retaining its basicity, giving branched constructs.

1

requires C, 59.96; H, 8.94, N, 31.08%; H NMR d 1.49

(2H, s, 2ÂNH), 1.67 (2H, qn, J=6.7 Hz, CH₂CH₂CH₂),

2.53 (4H, t, J=6.7 Hz, CH₂CN), 2.73 (4H, t, J=6.7 Hz,

CH₂CH₂CH₂), 2.91 (4H, t, J=6.7 Hz, CH₂CH₂CN);

¹³C NMR d 18.67, 29.77, 45.08, 47.55, 118.98.

N¹,N³-Bis(2-cyanoethyl)-N¹,N³-bis(1,1-dimethylethoxy-

carbonyl)propane-1,3-diamine (3). Compound 2 (2.50 g,

13.9 mmol) was stirred with Boc₂O (6.06 g, 27.8 mmol)

in CH₂Cl₂(10 mL) at 0 ^ꢀC for 20 min and at 20 ^ꢀC for

5 h. Evaporation gave 3 (5.3 g, 100%) as a colourless oil

(lit.²⁶oil): ¹H NMR d 1.48 (18H, s, 2ÂBu^t), 1.81 (2H, qn,

J=7.3 Hz, CH₂CH₂CH₂), 2.62 (4H, m, CH₂CN), 3.29

(4H, t, J=7.3 Hz, CH₂CH₂CH₂), 3.48 (4H, t, J=6.5 Hz,

CH₂CH₂CN). This material was used without further

puri®cation or characterisation.

The constructs show a range of interactions with DNA,

as demonstrated by ethidium exclusion. The more and

larger the MeOPEG units are, the weaker is the inter-

action of the construct with DNA. Increasing the num-

ber of basic nitrogens in the polyamine enhances DNA

interaction, as demonstrated by the hexamine construct

42. The novel PEG±polyamine constructs all interacted

with DNA to a greater extent than did PVP (data not

shown). However, unlike poly-l-Lys,⁵¹none of the

constructs were able to condense DNA, with the possible

exception of 42, which caused a diminution in ethidium±

DNA complex ¯uorescence of ca. 65% (at mass ratio

100), suggesting a signi®cant conformational change in

the DNA. Poly-l-Lys and many other polycations show

a threshold eect with increasing concentration in the

ethidium-displacement assay. This threshold eect is

consistent with the theory that DNA condensation

induced by multivalent cations occurs when 89±90% of

its negative charge is neutralised.⁵²Interestingly, the

MeOPEG±polyamine constructs 34, 35, 42 and 45 show

a progressive displacement of ethidium with increasing

concentration, which phenomenon is unprecedented in

the literature. Given the nature of the DNA-complex as

determined by the ethidium assays, these novel constructs

may have potential as PINC systems, exploiting Rolland

and Mumper's²²``Interactive Window of Opportunity''.

N¹,N³-Bis(3-aminopropyl)-N¹,N³-bis(1,1-dimethylethoxy-

carbonyl)propane-1,3-diamine (4). Method A. Com-

pound 3 (3.58 g, 9.4 mmol) in dry THF (40 mL) was

added dropwise to LiAlH₄(2.60 g, 69 mmol) in dry Et₂O

(200 mL) under N₂at 0 ^ꢀC during 10 min; the mixture

was stirred at 0 ^ꢀC for 2 h. Aqueous NaOH (1 M) was

added cautiously until eervescence ceased. The super-

natant was decanted and the residue was extracted with

Et₂O (3Â50 mL). Evaporation and chromatography

(CH₂Cl₂:MeOH:35% aq NH₃, 20:10:1) gave 4 (790 mg,

22%) as a colourless oil (lit.²⁶oil): ¹H NMR d ((CD₃)₂SO)

1.39 (18H, s, 2ÂBu^t), 1.46±1.68 (10H, m, 3ÂCH₂CH₂

CH₂+2ÂNH₂), 2.50 (4H, t, J=7.0 Hz, 2ÂCH₂NH₂),

3.10 (4H, t, J=7.0 Hz, 2ÂCH₂NBoc), 3.17 (4H, t,

J=6.8 Hz, 2ÂCH₂NBoc); MS (FAB) m/z 389 (M+H),

189 (M+H 2ÂBoc). Further elution gave 5 (1.15 g, 37%)

as a colourless oil (lit.⁵⁴oil): ¹H NMR d 1.44 (9H, s, Bu^t),

1.46 (9H, s, Bu^t), 1.67 (4H, m, 2ÂCH₂CH₂CH₂), 2.74 (2H,

m, CH₂NH₂), 3.13±3.35 (8H, m, 3ÂCH₂NBoc+NH₂),

3.67 (1H, m, NH); MS (FAB) m/z 332 (M+H), 232

(M+H Boc), 132 (M+H 2ÂBoc).

The successful delivery of pCMVCAT in preliminary

studies in vivo points to the potential utility of these

MeOPEG±polyamine constructs. Further studies are in

progress to determine the nature of the complex

between DNA and these constructs. The results of

fuller studies in vivo on the use of these linear poly-

amine±MeOPEG constructs as gene delivery agents will

be presented elsewhere.⁵³

N¹,N³-Bis(3-aminopropyl)-N¹,N³-bis(1,1-dimethylethoxy-

carbonyl)propane-1,3-diamine (4). Method B. The dini-

trile 3 (2.74 g, 7.2 mmol) in MeOH (10 mL) was saturated

with NH₃at 0 ^ꢀC and was treated with H₂(3000 Torr) in

the presence of W-2 Raney Ni for 72 h. Filtration

1790

S. W. Garrett et al. / Bioorg. Med. Chem. 8 (2000) 1779±1797

(Celite¹), evaporation and chromatography (CH₂Cl₂:

MeOH:35% aq NH₃, 20:10:1) gave 4 (1.93 g, 69%) as

colourless oil as above.

the suspension was ®ltered. The combined ®ltrate and

CH₂Cl₂washings were dried and the solvent and excess

reagent were evaporated to give 10a (10.8 g, 98%) as a col-

1

ourless oil: H NMR ((CD₃)₂SO) d 2.53 (1H, dd, J=5.0,

N¹,N³-Bis(3-aminopropyl)-N¹,N³-bis(1,1-dimethylethoxy-

carbonyl)propane-1,3-diamine (4). Method C. The di-

amide 8 (540 mg, 930 mmol) was heated with 35% aq

NH₃(1 mL) in MeOH (8 mL) at 60 ^ꢀC in a sealed vessel

for 5 h. Evaporation and chromatography (CH₂Cl₂:

MeOH:35% aq NH₃, 20:10:1) gave 4 (360 mg. 100%) as

a colourless oil as above.

2.5 Hz, oxirane 3-H), 2.72 (1H, t, J=5.0 Hz, oxirane 3-H),

3.09 (1H, m, oxirane 2-H), 3.27 (2H, m, oxirane-CH₂), 3.24

(3H, s, Me), 3.41±3.73 (ca. 45H, m, nÂOCH₂CH₂O); MS

(electrospray) m/z 771 (20%), 727 (36%), 683 (56%),

639 (74%), 595 (92%), 551 (100%), 597 (92%), 463

(80%), 419 (58%), 375 (46%) (all M+Na).

N¹,N³-Bis(1,1-dimethylethoxycarbonyl)-N¹,N³-bis(3-(N-

(2 - hydroxy - 3 - (! - methoxyPEG550oxy)propyl)amino)-

propyl)propane-1,3-diamine (11). The diamine 4 (82 mg,

210 mmol) was boiled under re¯ux with the oxirane 10a

(253mg, 420 mmol) in PrⁱOH (3.0 mL) for 24h. Evapora-

tion gave crude 11 (330mg, 100%) as a pale-yellow oil: ¹H

NMR d 1.41 (18 H, s, 2ÂBu^t), 1.62±1.95 (6H, m, 3ÂCH₂

CH₂CH₂), 2.51±2.59 (8H, m, 2ÂCH₂NHCH₂), 3.15±3.25

(8H, m, 4ÂCH₂NBoc), 3.35 (6H, s, 2ÂOMe), 3.41±3.53

(12H, m, 6ÂOCH₂), 3.60±3.78 (ca. 106 H, m, nÂOCH₂),

3.80±3.83 (4H, m, OCH₂+2ÂCHOH); MS ¹³C/¹²C iso-

tope clusters centred at m/z 1490, 1446, 1402, 1358, 1314

(all M+H).

N¹,N³-Bis(3-aminopropyl)propane-1,3-diamine (6). Com-

pound 2 (2.00 g, 11.1mmol) in MeOH (20 mL) was satu-

rated with NH₃for 30min and treated with H₂at 2700

Torr in the presence of W-2 Raney Ni (1.0 g) for 65h.

Filtration (Celite¹), evaporation and distillation (Kugel-

rohr) gave 6 (1.40 g, 67%) as a colourless liquid: bp_0.08

(Kugelrohr) 285 ^ꢀC (lit.²⁵bp_0.0797±100^ꢀC); ¹H NMR

((CD₃)₂SO) d 1.40±1.53 (6H, m, 3 CH₂CH₂CH₂), 1.67

(6H, br, 2ÂNH+2ÂNH₂), 2.47±2.56 (12H, m, 3ÂCH₂

CH₂CH₂); MS m/z 189 (M+H).

N¹,N³-Bis(1,1-dimethylethoxycarbonyl)-N¹,N³-bis(3-tri-

¯uoroacetamidopropyl)propane-1,3-diamine (8). Method

A. EtO₂CCF₃(7.9 g, 56 mmol) was added to 6 (5.1 g,

27 mmol) in MeOH (10 mL) at 0 ^ꢀC during 5 min. The

mixture was stirred at 20 ^ꢀC for 2.5 h. Evaporation gave

crude 7 (10.4 g, 100%) as a colourless oil: ¹H NMR d 1.67

(2H, qn, J=7.0Hz, central CH₂CH₂CH₂), 1.73 (4H, qn,

J=6.1Hz, 2ÂNCH₂CH₂CH₂NCOCF₃), 2.66 (4H, t, J=7

Hz, central CH₂CH₂CH₂), 2.79 (4H, t, J=5.9 Hz, 2Â

NCH₂CH₂CH₂NCOCF₃), 3.44 (4H, t, J=6.1 Hz, 2ÂCH₂

NCOCF₃), 7±10 (4H, br, 4 NH); ¹³C NMR d 27.1, 30.0,

40.2, 48.0, 48.8, 116.2 (q, J_C-F=288Hz, 2ÂCF₃), 157.2 (q,

J_C-F=37 Hz, 2ÂCOCF₃). This material (180mg,

470 mmol) was stirred with Boc₂O for 16 h. Evaporation,

chromatography (EtOAc:hexane, 1:1) and recrystallisa-

tion (EtOAc±hexane) gave 8 (254 mg, 93%) as a white

solid: mp 70±71 ^ꢀC. Found C, 47.7; H, 6.7, N, 9.6.

C₂₃H₃₈F₆N₄O₆requires C, 47.58; H, 6.60, N, 9.65%; ¹H

NMR d 1.47 (18H, s, 2ÂBu^t), 1.74±1.80 (6H, m, 3ÂCH₂

CH₂CH₂), 3.11±3.19 (4H, m, 2ÂCH₂NCOCF₃), 3.21±

3.40 (8H, m, 2ÂCH₂NBocCH₂), 8.29 (2H, br, 2ÂNH);

¹³C NMR d 27.1, 27.7, 28.4, 35.9, 43.1, 44.9, 80.6, 116.0

(q, J_C-F=287 Hz, 2ÂCF₃), 156.8 (q, J_C-F=37 Hz, 2ÂCO

CF₃); MS (FAB +ve ion) m/z 603 (M+Na), 581 (M+H),

481 (M+H Boc), 381 (M+H 2ÂBoc); MS (FAB ve

ion) m/z 579 (M H).

N¹,N³-Bis(1,1-dimethylethoxycarbonyl)-N³-(3-(N-(2-hy-

droxy-3-(! -methoxyPEG550oxy)propyl)amino)propyl)-

N¹-(3-(N,N-bis(2-hydroxy-3-(!-methoxyPEG550oxy)-

propyl)amino)propyl)propane - 1,3 - diamine (12) and N¹,

N³-bis(1,1-dimethylethoxycarbonyl)-N¹,N³-bis(3-(N,N-

bis(2-hydroxy-3-(!-methoxyPEG550oxy)propyl)amino)-

propyl)propane-1,3-diamine (13). The diamine 4 (1.75 g,

4.5 mmol) was stirred at 80 ^ꢀC with the oxirane 10a

(5.4 g, 9.0 mmol) in PrⁱOH (20 mL) for 38 h. Evapora-

tion and chromatography (CH₂Cl₂:MeOH, 6:1) gave 13

(1.40 g, 24%) as a pale-yellow oil: ¹H NMR d 1.45 (18H,

s, 2ÂBu^t), 1.60±1.70 (6H, m, 3ÂCH₂CH₂CH₂), 2.50±2.55

(12H, m, 2ÂN(CH₂)₃), 3.15±3.26 (8H, m, 4ÂCH₂NBoc),

3.38 (12H, s, 4ÂOMe), 3.41±3.49 (6H, m, 3ÂOCH₂),

3.51±3.56 (8H, m, 4ÂOCH₂), 3.56±3.68 (ca. 180H, m,

nÂOCH₂), 3.80±3.84 (4ÂCHOH); MS (electrospray)

¹³C/¹²C isotope clusters centred at m/z 1339, 1318, 1296,

1274, 1252, 1230, 1208 (all [M+H]²⁺). Further elution

gave 12 (560 g, 11%) as a pale-yellow oil: ¹H NMR d 1.47

(18H, s, 2ÂBu^t), 1.60±1.70 (4H, m, 2ÂCH₂CH₂CH₂),

1.95±2.05 (2H, m, CH₂CH₂CH₂), 2.48±2.60 (6H, m,

3ÂNCH₂), 2.95±3.05 (6H, m, 3ÂNCH₂), 3.15±3.25 (8H,

m, 4ÂCH₂NBoc), 3.41 (9H, s, 3ÂOMe), 3.44±3.50 (4H,

m, 2ÂOCH₂), 3.54±3.58 (6H, m, 3ÂOCH₂), 3.58±3.80

(ca. 130H, m, nÂOCH₂), 3.80±3.94 (3H, m, 3ÂCHOH);

MS (electrospray) ¹³C/¹²C isotope clusters centred at m/z

1098, 1076, 1054, 1031, 1009, 988, 966 (all [M+H]²⁺).

N¹,N³-Bis(1,1-dimethylethoxycarbonyl)-N¹,N³-bis(3-tri-

¯uoroacetamidopropyl)propane-1,3-diamine (8). Method

B. The diamine 4 (620 mg, 1.6 mmol) was stirred with

EtO₂CCF₃(1.1 g, 8 mmol) in MeOH (10 mL) for 16h.

Evaporation and chromatography (EtOAc:hexane, 1:1)

gave 8 (890 mg, 96%) as a white solid as above.

N³-(3-(N-(2-Hydroxy-3-(!-methoxyPEG550oxy)propyl)

amino)propyl)-N¹-(3-(N,N-bis(2-hydroxy-3-(!-methoxy-

PEG550oxy)propyl)amino)propyl)propane - 1,3 - diamine

(14). HCl was passed through 12 (2.27 g, 1.1 mmol) in

CH₂Cl₂(40 mL) for 30 min. Evaporation and freeze-

(!-MeOPEG550oxymethyl)oxirane (10a). MeOPEG550

OH 9a (10.0 g, 18mmol) was dried by azeotropic removal

of water with toluene. NaOH (powder, 2.18g, 54mmol),

water (300mL) and chloromethyloxirane (20.35g,

220 mmol) were added and the mixture was stirred vigor-

ously at 60^ꢀC for 3 h. CH₂Cl₂(100 mL) was added and

1

drying gave 14 (2.07 g, 100%) as an o-white glass: H

NMR (D₂O) d 2.13 (6H, m, 3ÂCH₂CH₂CH₂), 3.11±3.27

(18H, m, 9ÂNCH₂), 3.39 (9H, s, 3ÂOMe), 3.53±3.64

(16H, m, 8ÂOCH₂), 3.68±3.88 (ca. 120H, m, nÂOCH₂),

4.14 (3H, 3ÂCHOD).

S. W. Garrett et al. / Bioorg. Med. Chem. 8 (2000) 1779±1797

1791

N¹,N³-Bis(3-(N,N-bis(2-hydroxy-3-(!-methoxyPEG

550oxy)propyl)amino)propyl)propane - 1,3 - diamine (15).

Compound 13 was treated with HCl, as for the synth-

3ÂCH₂CH₂CH₂), 2.40±2.49 (12H, m, 4ÂNMe), 2.57±

2.70 (16H, m, 8ÂNHCH₂), 3.38 (6H, s, 2ÂOMe), 3.49±

3.51 (4H, m, 2ÂOCH₂), 3.54±3.56 (4H, m, 2ÂOCH₂),

3.64±3.66 (ca. 96H, m, nÂOCH₂), 4.00 (2H, m, 2ÂCH

OH); MS ¹³C/¹²C isotope clusters centred at m/z 1390,

1346, 1302, 1258, 1214, 1170, 1126 (all M+H).

1

esis of 14, to give 15 (100%) as an o-white glass: H

NMR (D₂O) d 2.10 (6H, m, 3ÂCH₂CH₂CH₂), 2.97±3.22

(20H, m, 10ÂNCH₂), 3.39 (12H, s, 4ÂOMe), 3.53±3.64

(20H, m, 10ÂOCH₂), 3.66±3.72 (ca. 164H, m, nÂOCH₂),

4.17 (4 H, 4ÂCHOD).

N¹,N³-Bis(1,1-dimethylethoxycarbonyl)-N¹,N³-bis(3-(N-

methyltri¯uoroacetamido)propyl)propane-1,3-diamine (21).

KOBu^tin THF (1.0 M, 2.6 mL, 2.6 mmol) was added

to 8 (712 mg, 1.2 mmol) in dry THF (25 mL) under N₂

and the mixture was stirred for 5 min. Iodomethane

(426 mg, 3.0 mmol) was added and the mixture was stir-

red for 20 min under N₂. The solvent and excess reagent

were evaporated and the residue, in EtOAc, was washed

with water and with brine and was dried. Evaporation of

the solvent gave 21 (730 mg, 98%) as a colourless oil: ¹H

NMR d ((CD₃)₂SO, 22 ^ꢀC) 1.37 (18H, s, 2ÂBu^t), 1.65±

1.77 (6H, m, 3ÂCH₂CH₂CH₂), 2.95 (2H, s, 0.67ÂNMe),

3.09±3.15 (11H, m, 1.33ÂNMe+3.5ÂNCH₂), 3.32±3.44

1,1-Dimethylethyl N-methyl-N-(3-methylaminopropyl)-

carbamate (17). Boc₂O (1.10 g, 5.0 mmol) in dry THF

(10 mL) was added dropwise during 40 min to 16 (1.53 g,

15 mmol) in dry THF (10mL) at 0 ^ꢀC. Stirring continued

at 0 ^ꢀC for 1 h and at 20^ꢀC for 20h. Evaporation and

chromatography (CH₂Cl₂:MeOH:35% aq NH₃, 60:20:1)

1

gave 17 (723mg, 71%) as a colourless oil (lit.³⁰oil): H

NMR d 1.46 (9H, s, Bu^t), 1.69±1.82 (2H, m, CH₂CH₂

CH₂), 2.46 (3H, s, NHCH₃), 2.61 (2H, t, J=7.0Hz,

CH₂NH), 2.85 (3H, s, BocNCH₃), 3.17 (1H, brs, NH), 3.3

(2H, m, CH₂NBoc); MS (FAB) m/z 204.1795 (M+H)

(¹³C¹²C₉H₂₃N₂O₂requires 204.1793); 203.1763 (M+H)

(¹²C₁₀H₂₃N₂O₂requires 203.1760).

(5H, m, 2.5ÂNCH₂); H NMR d ((CD₃)₂SO, 100 ^ꢀC)

1

1.42 (18H, s, 2ÂBu^t), 1.68±1.80 (6H, m, 3ÂCH₂CH₂CH₂),

2.96 (6H, s, 2ÂNMe), 3.10±3.20 (8H, m, 4ÂNCH₂), 3.41

(4H, t, J=7.6 Hz, 2ÂNCH₂); ¹⁹F NMR ((CD₃)₂SO, 22 ^ꢀC)

69.11 (4F, s), 68.17 (2F, s); MS (FAB) m/z 610.3122

(M+H) (¹³C¹²C₂₄H₄₃F₆N₄O₆requires 610.3120),

609.3091 (M+H) (¹²C₂₅H₄₃F₆N₄O₆requires 609.3087).

N¹,N³-Bis(3-(N-(1,1-dimethylethoxycarbonyl)-N-methyl-

amino)propyl)-N¹,N³-dimethylpropane-1,3-diamine (18).

The carbamate 17 (277mg, 1.4 mmol) was stirred with

1,3-dibromopropane (138mg, 680 mmol) and K₂CO₃

(200mg, 1.45mmol) in DMF (4 mL) at 80^ꢀC for 8 h. The

solvent was evaporated and the residue was extracted with

CHCl₃(4Â30mL). Evaporation and chromatography

(CH₂Cl₂:MeOH:35% aq NH₃, 70:10:1) gave 18 (160mg,

52%) as a pale-yellow oil: ¹H NMR d 1.45 (18H, s,

2ÂBu^t), 1.63±1.71 (6H, m, 3ÂCH₂CH₂CH₂), 2.21 (6H, s,

2Âamine NMe), 2.30±2.37 (8H, m, 2ÂCH₂NMeCH₂),

2.85 (6H, s, 2ÂBocNCH₃), 3.22 (4H, t, J=7.0 Hz,

2ÂCH₂NBoc). This material was used without further

characterisation.

N¹,N³-Bis(1,1-dimethylethoxycarbonyl)-N¹,N³-bis(3-me-

thylaminopropyl)propane-1,3-diamine (22). Compound 21

(144mg, 240 mmol) was stirred with 35% aq NH₃(2.0mL)

in MeOH (8 mL) at 55^ꢀC in a sealed vessel for 3.5 h.

Evaporation and chromatography (CH₂Cl₂:MeOH:35%

aq NH₃, 16:8:1) gave 22 (78 mg, 79%) as a colourless oil:

¹H NMR d 1.45 (18H, s, 2ÂBu^t), 1.72±1.74 (6H, m,

3ÂCH₂CH₂CH₂), 2.39 (6H, s, 2ÂNMe), 2.58 (4H, t,

J=7.0Hz, 2ÂCH₂NMe), 2.76 (2H, s, 2ÂNH), 3.12±3.20

(4H, m, 2ÂNCH₂), 3.22±3.28 (4H, m, 2ÂNCH₂); MS

(FAB) m/z 418.3473 (M+H) (¹³C¹²C₂₀H₄₅N₄O₄requires

418.3474), 417.3443 (M+H) (¹²C₂₁H₄₅N₄O₄requires

417.3441), 317 (M+H Boc), 217 (M+H 2ÂBoc).

N¹,N³-Bis(3-methylaminopropyl)propane-1,3-diamine tet-

rahydrochloride (23). Compound 22 was treated with

HCl, as for the synthesis of 14, to give 23 (96%) as a

highly hygroscopic white solid (lit.⁵⁵hygroscopic solid):

¹H NMR (D₂O) d 2.08±2.20 (6H, m, 3ÂCH₂CH₂CH₂),

2.75 (6H, s, 2ÂMe), 3.12±3.24 (12H, m, 6ÂCH₂N); MS

m/z 217 (M+H).

N¹,N³-Bis(3-methylaminopropyl)-N¹,N³-dimethylpropane-

1,3- diamine tetrahydrochloride (19). The bis-carbamate

18 (115 mg, 260 mmol) was stirred with hydrochloric acid

(5M, 5 mL) for 16h. Evaporation and trituration

(CH₂Cl₂) gave 19 (100mg, 100%) as a pale-bu solid: mp

217±220 ^ꢀC; ¹H NMR (D₂O) d 2.15±2.30 (6H, m,

3ÂCH₂CH₂CH₂), 2.75 (6H, s, 2ÂNMe), 2.94 (6H, s,

2ÂNMe), 3.15 (4H, t, J=7.0Hz, 2ÂCH₂N), 3.27±3.35

(8H, m, 4ÂCH₂N); MS (FAB) m/z 321 (M+3 H+2 ³⁷Cl),

317 (M+3 H+³⁷Cl+³⁵Cl), 317 (M+3 H+2 ³⁵Cl), 283

(M+2 H+³⁷Cl), 281 (M+2 H+³⁵Cl), 246.2729 (M+H)

(¹³C¹²C₁₂H₃₃N₄requires 246.2739), 245.2709 (M+H)

(¹²C₁₃H₃₃N₄requires 245.2705).

N¹,N³-Bis(1,1-dimethylethoxycarbonyl)-N¹,N³-bis(3-(N-

(2-hydroxy-3-phenoxypropyl)-N-methylamino)propyl)-

propane - 1,3 - diamine (24). The diamine 22 (45 mg, 110

mmol) was heated at re¯ux with phenoxymethyloxirane

(33 mg, 220 mmol) in PrⁱOH (1.0 mL) for 10 h. Evapora-

tion and preparative layer chromatography (CH₂Cl₂:

MeOH:35% aq NH₃, 140:20:1) gave 24 (20 mg, 26%) as

a colourless oil: ¹H NMR d 1.45 (18H, s, 2ÂBu^t), 1.65±1.80

(6H, m, 3ÂCH₂CH₂CH₂), 2.32 (6H, s, 2ÂNMe), 2.44±

2.60 (8H, m, 2ÂCH₂NCH₂), 3.10±3.35 (10H, m,

4ÂBocNCH₂+ 2ÂOH), 3.97 (4H, m, 2ÂOCH₂), 4.08

(2H, m, 2ÂCHOH), 6.94 (4H, d, J=8.0 Hz, 2ÂPh 2,6-

H₂), 7.25±7.31 (6H, m, 2ÂPh 3,4 5-H₃); MS (FAB) m/z

N¹,N³-Bis(3-(N-(2-hydroxy-3-(!-methoxyPEG550oxy)-

propyl)-N-methylamino)propyl)-N¹,N³-dimethylpropane-

1,3 - diamine tetrahydrochloride (20). The tetramine 19

(83 mg, 210 mmol) was stirred with aq NaOH (5 M,

213 mL, 1.0 mmol) in PrⁱOH (1.0 mL) for 5 min. The oxi-

rane 10a (255mg, 420 mmol) in PrⁱOH (1.0 mL) was added

and the mixture was stirred at 75^ꢀC for 24h. Evaporation

and chromatography (CH₂Cl₂:MeOH:35% aq NH₃, 40:

10:1) gave a pale-yellow oil. Aqueous HCl (9 M, 2.0 mL)

was added. Freeze-drying gave 20 (105 mg, 32%) as an

1

o-white wax: H NMR (free base) d 1.80±1.90 (6H, m,

1792

S. W. Garrett et al. / Bioorg. Med. Chem. 8 (2000) 1779±1797

718.4853 (M+H) (¹³C¹²C₃₈H₆₅N₄O₈requires 718.4836),

717.4811 (M+H) (¹²C₃₉H₆₅N₄O₈requires 717.4802).

1546, 1501, 1457, 1413, 1369 (M+H). Further elution

gave 30 (460mg, 21%) as a colourless oil: ¹H NMR d 1.44

(9H, s, Bu^t), 1.45 (9H, s, Bu^t), 1.46 (9H, s, Bu^t), 1.65±1.78

(6H, m, 3ÂCH₂CH₂CH₂), 2.90±2.98 (4H, m, 2ÂNCH₂),

3.14±3.27 (10H, m, 5ÂNCH₂), 3.38 (3H, s, OMe), 3.54±

3.58 (6H, m) and 3.62±3.66 (ca. 44H, m) (nÂOCH₂CH₂O),

3.82 (1H, m, CHOH); MS (FAB) 1149 (M+H), 1105

(M+H), 1061.7005 (M+H) (¹²C₅₀H₁₀₁N₄O₁₉requires

1061.7060), 1017.6772 (M+H) (¹²C₄₈H₉₇N₄O₁₈requires

1017.6798), 973.6511 (M+H) (¹²C₄₆H₉₃N₄O₁₇requires

973.6536).

N¹,N³-Bis(1,1-dimethylethoxycarbonyl)-N¹,N³-bis(3-(N-

(2-hydroxy-3-(!-MeOPEG2000oxy)propyl)-N-methyl-

amino)propyl)propane-1,3-diamine (25). The diamine 22

(43mg, 103 mmol) was heated at re¯ux with o-MeO-

PEG2000 oxiranylmethyl ether 10b (410mg, 210 mmol) in

PrⁱOH (3.0mL) for 27 h. Evaporation gave 25 (450mg,

1

100%) as a pale-yellow wax: H NMR d 1.45 (18H, s,

2ÂBu^t), 1.66±1.75 (6H, m, 3ÂCH₂CH₂CH₂), 2.25 (6H, s,

2ÂNMe), 2.31±2.45 (8H, m, 2ÂCH₂NCH₂), 3.15±3.20

(8H, m, 4ÂBocNCH₂), 3.38 (6H, s, 2ÂOMe), 3.42±3.60

(12H m) and 3.60±3.77 (ca. 360 H, m) (nÂOCH₂CH₂O),

3.81±3.88 (6H, m, 2ÂOCH₂+2ÂCHOH).

N¹,N³-Bis(1,1-dimethylethoxycarbonyl)-N³-(3-(1,1-di-

methylethoxycarbonylamino)propyl)-N¹-(3-(N-(2-hydroxy-

3-(!-MeOPEG2000oxy)propyl)amino)propyl)propane-1,3-

diamine (32) and N¹,N³-bis(1,1-dimethylethoxycarbonyl)-

N¹,N³-bis(3-(N-(2-hydroxy-3-(!-MeOPEG2000oxy)

propyl)amino)propyl)-N³-(3-(1,1-dimethylethoxy carbonyl-

amino)propyl)propane-1,3-diamine (33). The amine 29

(986 mg, 2.02 mmol) was heated at re¯ux with 10b (4.15 g,

2.02 mmol) in PrⁱOH (21 mL) for 100 h. Evaporation and

chromatography (CH₂Cl₂:MeOH, 7:1) gave 33 (2.53 g,

N¹,N³-Bis(3-(N-(2-hydroxy-3-(!-MeOPEG2000oxy)

propyl)-N-methylamino)propyl)propane-1,3-diamine tetra-

hydrochloride (26). Compound 25 was treated with HCl,

as for the synthesis of 14, to give 26 (100%) as a white

1

wax: H NMR (D₂O) d 2.43 (6H, m, 3ÂCH₂CH₂CH₂),

2.98 (6H, m, 2ÂNMe), 3.31±3.36 (16H, m, 8ÂNHCH₂),

3.38 (6H, s, 2ÂOMe), 3.48±3.83 (ca. 370H, m, nÂOCH₂),

4.33 (2H, m, 2ÂCHOD).

1

28%) as a pale-bu waxy solid: H NMR d 1.44 (9H, s,

Bu^t), 1.45 (9H, s, Bu^t), 1.47 (9H, s, Bu^t), 1.65±1.75 (6H, m,

3ÂCH₂CH₂CH₂), 2.48±2.59 (6H, m, 3ÂNCH₂), 3.08±3.26

(10H, m, 5ÂNCH₂), 3.39 (6H, s, 2ÂOMe), 3.45±3.49 (8H,

m) and 3.55±3.72 (ca. 360 H, m) (nÂOCH₂CH₂O), 3.81

(2H, m, 2ÂCHOH). Further elution gave 32 (1.80 g, 35%)

N¹-(3-Aminopropyl)-N³-(3-(1,1-dimethylethoxycarbonyl-

amino)propyl) - N¹,N³- bis(1,1 - dimethylethoxycarbonyl)-

propane-1,3-diamine (29). EtO₂CCF₃(1.74 g, 12.3 mmol)

in EtOH (50 mL) was added dropwise during 40 min to

6 (2.09 g, 11.1 mmol) in EtOH (50 mL) at 78 ^ꢀC. The

mixture was stirred at 0 ^ꢀC for 30 min. Boc₂O (9.7 g,

44.5 mmol) in EtOH (20 mL) was added dropwise dur-

ing 10 min and the mixture was stirred at 20 ^ꢀC for 16 h.

The solvent was evaporated. The residue, in MeOH

(100 mL), was stirred with 35% aq NH₃(70 mL) in a

sealed vessel for 20 h. Evaporation and chromatography

(CH₂Cl₂:MeOH:35% aq NH₃, 70:10:1!30:10:1) gave 29

(2.76 g, 51%) as a colourless oil (lit.²⁶oil): ¹H NMR d 1.44

(9H, s, Bu^t), 1.46 (9H, s, Bu^t), 1.47 (9H, s, Bu^t), 1.63 (2H,

m, CH₂CH₂NH₂), 1.75 (4H, m, 2ÂBocNCH₂CH₂

CH₂NBoc), 1.97 (2H, brs, NH₂), 2.71 (2H, t, J=6.7Hz,

CH₂NH₂), 3.10±3.28 (10H, m, 5ÂBocNCH₂), 4.85 (0.5H,

brs, 0.5ÂNH), 5.30 (0.5H, brs, 0.5ÂNH); MS (FAB) m/z

490.3678 (M+H) (¹³C¹²C₂₃H₄₉N₄O₆requires 490.3686),

489.3651 (M+H) (¹²C₂₄H₄₉N₄O₆requires 489.3652).

1

as a pale-bu waxy solid: H NMR d 1.44 (9H, s, Bu^t),

1.45 (9H, s, Bu^t), 1.46 (9H, s, Bu^t), 1.66±1.75 (6H, m,

3ÂCH₂CH₂CH₂), 2.85±2.91 (4H, m, 2ÂNCH₂), 3.10±3.25

(10H, m, 5ÂNCH₂), 3.38 (3H, s, OMe), 3.54±3.56 (8H

m) and 3.64±3.70 (ca. 186 H, m) (nÂOCH₂CH₂O), 3.82

(1H, m, CHOH).

N³-(3-Aminopropyl)-N¹-(3-(N-(2-hydroxy-3-(!-MeOPEG

550oxy)propyl)amino)propyl)propane-1,3-diamine tetra-

hydrochloride (34). Compound 30 was treated with HCl,

as for the synthesis of 14, to give 34 (100%) as a white wax:

¹H NMR (D₂O) d 2.12 (6H, m, 3ÂCH₂CH₂CH₂), 3.10±

3.20 (14H, m, 7ÂNCH₂), 3.38 (3H, s, OMe), 3.64±3.72 (ca.

45H, m, nÂOCH₂CH₂O), 4.01 (1H, m, CHOD); MS

¹³C/¹²C isotope clusters centred at m/z 849, 805, 761, 717,

673, 630 (all M+H).

N³-(3-Aminopropyl)-N¹-(3-(N-(2-hydroxy-3-(!-MeOPEG

2000oxy)propyl)amino)propyl)propane-1,3-diamine tetra-

hydrochloride (35). Compound 32 was treated with HCl,

as for the synthesis of 14, to give 35 (100%) as a white wax:

¹H NMR (D₂O) d 2.15 (6H, m, 3ÂCH₂CH₂CH₂), 3.15±

3.28 (14H, m, 7ÂNCH₂), 3.39 (3H, s, OMe), 3.65±3.76 (ca.

185H, m, nÂOCH₂CH₂O), 4.10 (1H, m, CHOD).

N³-(3-Aminopropyl)-N¹-(3-(N-(2-hydroxy-3-(!-MeOPEG

550oxy)propyl)amino)propyl)propane-1,3-diamine tetra-

hydrochloride (36). Compound 31 was treated with HCl,

as for the synthesis of 14, to give 36 (100%) as a pale-

yellow wax: ¹H NMR (D₂O) d 2.14±2.23 (6H, m,

3ÂCH₂CH₂CH₂), 3.10±3.25 (10H, m, 5ÂNCH₂), 3.35±

3.46 (12H, m, 3ÂNCH₂+2ÂOMe), 3.62±3.85 (ca. 100H,

m, nÂOCH₂CH₂O), 4.23 (2H, m, 2ÂCHOD); MS

¹³C/¹²C isotope clusters centred at m/z 1334, 1290, 1246,

1202, 1158, 1114 (all M+H).

N¹,N³- Bis(1,1 - dimethylethoxycarbonyl) - N³- (3 - (1,1 - di-

methylethoxycarbonylamino)propyl)-N¹-(3-(N-(2-hydroxy-

3-(!-MeOPEG550oxy)propyl)amino)propyl)propane-1,3-

diamine (30) and N¹,N³-bis(1,1-dimethylethoxycarbonyl)-

N¹,N³-bis(3-(N-(2-hydroxy-3-(!-MeOPEG550oxy)propyl)

amino)propyl)-N⁰-(3-(1,1-dimethylethoxycarbonylamino)-

propyl)propane-1,3-diamine (31). The amine 29 (959mg,

1.97mmol) was heated at re¯ux with o-MeOPEG550

oxiranylmethyl ether 10a (1.19 g, 1.97mmol) in PrⁱOH

(10 mL) for 48h. Evaporation and chromatography

(CH₂Cl₂:MeOH, 7:1) gave 31 (315mg, 9%) as a colourless

1

oil: H NMR d 1.44 (9H, s, Bu^t), 1.45 (9H, s, Bu^t), 1.46

(9H, s, Bu^t), 1.62±1.74 (6H, m, 3ÂCH₂CH₂CH₂), 2.48±

2.60 (6H, m, 3ÂNCH₂), 3.15±3.24 (10H, m, 5ÂNCH₂),

3.38 (6H, s, 2ÂOMe), 3.50±3.57 (12H, m) and 3.60±3.73

(ca. 90 H, m) (nÂOCH₂CH₂O), 3.83 (2H, m, 2ÂCHOH);

MS (FAB) ¹³C/¹²C ion clusters centred at m/z 1590,

S. W. Garrett et al. / Bioorg. Med. Chem. 8 (2000) 1779±1797

1793

N³-(3-Aminopropyl)-N¹-(3-(N-(2-hydroxy-3-(!-MeOPEG

550oxy)propyl)amino)propyl)propane-1,3-diamine tetra-

hydrochloride (37). Compound 33 was treated with HCl,

as for the synthesis of 14, to give 37 (100%) as a white

wax: ¹H NMR (D₂O) d 1.97±2.13 (6H, m, 3ÂCH₂

CH₂CH₂), 3.07±3.33 (12H, m, 6ÂNCH₂), 3.35±3.46 (4H,

m, 3ÂNCH₂), 3.37 (6H, s, 2ÂOMe), 3.61±3.73 (ca. 370H,

m, nÂOCH₂CH₂O), 4.23 (2H, m, 2ÂCHOD).

N¹-(3-(N-(3-(N-(2-Hydroxy-3-(!-MeOPEG550oxy) pro-

pyl)amino)propyl)amino)propyl) - N³- (3 - (3 - aminopropyl

amino)propyl)propane-1,3-diamine hexahydrochloride (42).

Compound 41 was treated with HCl, as for the synthesis

of 14, to give 42 (86%) as a white wax: ¹H NMR (D₂O)

d 1.93±2.08 (10H, m, 5ÂCH₂CH₂CH₂), 3.01±3.10 (22H,

m, 11ÂNCH₂), 3.24 (3H, s, OMe), 3.44±3.52 (8H, m)

and 3.52±3.60 (ca. 49H, m) (nÂOCH₂CH₂O), 3.73 (1H,

m, CHOD).

N¹,N³-Bis(1,1-dimethylethoxycarbonyl)-N¹,N³-bis(3-(2-

cyanoethylamino)propyl)propane - 1,3 - diamine (38). The

diamine 6 was treated with propenenitrile, as for the

synthesis of 2, to give 38 (3.03g, 98%) as a colourless oil:

¹H NMR d 1.45 (18H, s, 2ÂBu^t), 1.70 (4H, qn, J=6.7Hz,

2ÂCH₂CH₂CH₂), 1.76 (4H, m, BocNCH₂CH₂CH₂N-

Boc+2ÂNH), 2.61 (4H, t, J=6.7 Hz, 2ÂCH₂CN), 2.62

(4H, t, J=6.7 Hz, 2ÂCH₂N), 2.91 (4H, t, J=6.7Hz,

2ÂCH₂N), 3.16±3.43 (8H, m, 4ÂBocNCH₂); MS (FAB)

m/z 496.3704 (M+H) (¹³C¹²C₂₄H₄₇N₆O₄requires

496.3692), 495.3676 (M+H) (¹²C₂₅H₄₇N₆O₄requires

495.3659).

N¹,N³-Bis(1,1-dimethylethoxycarbonyl)-N¹,N³-bis(3-(N-

(1,1-dimethylethoxycarbonyl)-N-(3-(N-(2-hydroxy-3-(!-

MeOPEG550oxy)propyl)amino)propyl)amino)propyl)pro-

pane-1,3-diamine (43) and N¹,N³-bis(1,1-dimethylethoxy-

carbonyl)-N³-(3-(N-(1,1-dimethylethoxycarbonyl)-N-(3-

(N-(2-hydroxy-3-(!-MeOPEG550oxy)propyl)amino)-

propyl)amino)propyl)-N¹-(3-(N-(1,1-dimethylethoxycar-

bonyl) - N - (3 - (N,N - bis(2 - hydroxy - 3 - (! - MeOPEG-

550oxy)propyl)amino)propyl)amino)propyl)propane - 1,3 -

diamine (44). The diamine 40 (845 mg, 1.2 mmol) was

heated at 65 ^ꢀC with 10a (1.45 g, 2.4 mmol) in PrⁱOH

(25 mL) for 44 h. Evaporation and chromatography

(CH₂Cl₂:MeOH, 6:1) gave 44 (167 mg, 6%) as a colour-

N¹,N³-Bis(3-(N-(3-aminopropyl)-N-(1,1-dimethylethoxy-

carbonyl)amino)propyl) - N¹,N³- bis(1,1 - dimethylethoxy-

carbonyl)propane-1,3-diamine (40). The dinitrile 38

(2.92 g, 5.9 mmol) was stirred with Boc₂O (2.64 g,

12 mmol) in CH₂Cl₂(40 mL) at 0 ^ꢀC for 1.5 h and at

20 ^ꢀC for 16 h. Evaporation gave crude 39 (4.1 g, 100%)

as a colourless oil: ¹H NMR d 1.46 (18H, s, 2ÂBu^t), 1.47

(18H, s, 2ÂBu^t), 1.69±1.88 (6H, m, 3ÂCH₂CH₂CH₂),

2.55±2.70 (4H, m, 2ÂCH₂CN), 3.17±3.29 (12H, m,

6ÂCH₂NBoc), 3.48 (4H, t, J=6.9 Hz, 2ÂCH₂CH₂CN).

This material (2.58 g, 3.7 mmol) in MeOH (30 mL) was

saturated with NH₃and was treated with H₂(3500

Torr) in the presence of W-2 Raney Ni (2.0 g) for 65 h.

The suspension was ®ltered (Celite¹) and the solvent was

evaporated from the combined ®ltrate and MeOH wash-

ings. Chromatography (CH₂Cl₂:MeOH:35% aq NH₃, 60:

10:1) gave 40 (2.29g, 88%) as a colourless oil: ¹H NMR d

1.45 (36H, s, 4ÂBu^t), 1.66 (4H, qn, J=6.7 Hz, 2ÂCH₂

CH₂NH₂), 1.71±1.78 (6H, m, 3ÂBocNCH₂CH₂CH₂N

Boc), 1.88 (4H, brs, 2ÂNH₂), 2.70 (4H, t, J=6.7Hz,

2ÂCH₂NH₂), 3.17±3.43 (16H, m, 8ÂCH₂NBoc); MS

(FAB) m/z 704.5381 (M+H) (¹³C¹²C₃₄H₇₁N₆O₈

requires 704.5367), 703.5348 (M+H) (¹²C₃₅H₇₁N₆O₈

requires 703.5333).

1

less oil: H NMR d 1.45 (36H, s, 4ÂBu^t), 1.73 (10H, m,

5ÂCH₂CH₂CH₂), 2.54 (10H, m, 5ÂNCH₂), 3.15±3.19

(16H, m, 8ÂCH₂NBoc), 3.38 (9H, 3ÂOMe), 3.54±3.60

(20H, m, 10ÂOCH₂), 3.60±3.70 (ca. 140H, m, nÂOCH₂

CH₂O), 3.83 (3H, m, 3ÂCHOH); MS (MALDI-TOF) m/z

2685, 2640, 2596, 2551, 2507, 2463, 2419 (all M+H).

Further elution gave 43 (262mg, 12%) as a colourless oil:

¹H NMR d 1.45 (36H, s, 4ÂBu^t), 1.73 (10H, m, 5ÂCH₂

CH₂CH₂), 2.48 (8H, m, 4ÂNCH₂), 3.10±3.20 (16H, m,

8ÂCH₂NBoc), 3.38 (6H, 2ÂOMe), 3.45±3.60 (10H, m,

5ÂOCH₂), 3.61±3.70 (ca. 100H, m, nÂOCH₂CH₂O), 3.85

(2H, m, 2ÂCHOH).

N¹,N³-Bis(3-(N-(3-(N-(2-hydroxy-3-(!-MeOPEG550oxy)

propyl)amino)propyl)amino)propyl)propane - 1,3 - diamine

hexahydrochloride (45). Compound 43 was treated with

HCl, as for the synthesis of 14, to give 45 (100%) as a

white glass: ¹H NMR (D₂O) d 1.95±2.03 (10H, m,

5ÂCH₂CH₂CH₂), 3.04 (24H, ca. t, J=7.6Hz, 12ÂNCH₂),

3.22 (6H, 2ÂOMe), 3.43±3.48 (ca. 14H, m, 7ÂOCH₂),

3.53±3.63 (ca. 120H, m, nÂOCH₂CH₂O), 3.99 (2H, m,

2ÂCHOH).

Bis(3 - (1,1 - dimethylethoxycarbonylamino)propyl)amine

(47). 2-(t-Butoxycarbonyloxyimino)-2-phenylacetonitrile

(BocON) (5.78 g, 23.5 mmol) in THF (50 mL) was added

during 1.5 h to (H₂N(CH₂)₃)₂NH 46 (1.40 g, 10.7mmol) in

THF (20 mL) at 0 ^ꢀC. The mixture was then stirred at

20^ꢀC for 40min. Evaporation and chromatography

(EtOAc:MeOH, 4:1) gave 47 (1.87 g, 54%) as a white

N¹,N³-Bis(1,1-dimethylethoxycarbonyl)-N¹-(3-(N-(1,1-

dimethylethoxycarbonyl) -N-(3- (N -(2-hydroxy -3-(! -

MeOPEG550oxy)propyl)amino)propyl)amino)propyl)-N³-

(3-(N-(3-aminopropyl)-N-(1,1-dimethylethoxycarbonyl)

amino)propyl)propane-1,3-diamine (41). The diamine 40

(537 mg, 760 mmol) was heated at re¯ux with 10a

(465 mg, 750 mmol) in PrⁱOH (10 mL) for 48 h. Eva-

poration and chromatography (CH₂Cl₂:MeOH:35% aq

NH₃, 80:10:1) gave 41 (302 mg, 32%) as a pale-yellow

solid: mp 70±71 ^ꢀC (lit.⁵⁶mp 70±72 ^ꢀC); H NMR d 1.44

1

(18H, s, 2ÂBu^t), 1.64 (4H, qn, J=6.4Hz, 2ÂCH₂CH₂

CH₂), 1.65 (1H, br, NH), 2.65 (4H, t, J=6.4 Hz, 2ÂCH₂

NH), 3.22 (4H, m, 2ÂCH₂NBoc), 5.28 (2H, m, 2Â

NHBoc); MS (FAB) m/z 333.2580 (M+H), (¹³C¹²C₁₅

H₃₄N₃O₄requires 333.2583), 332.23548 (M+H), (¹²C₁₆

H₃₄N₃O₄requires 332.2549).

1

oil: H NMR d 1.45 (36H, s, 4ÂBu^t), 1.72±1.81 (10H,

m, 5ÂCH₂CH₂CH₂), 2.67±2.72 (6H, m, 3ÂNCH₂),

3.14±3.30 (16H, m, 8ÂBocNCH₂), 3.38 (3H, s, OMe),

3.45±3.58 (8H, m) and 3.62±3.70 (ca. 50H, m)

(nÂOCH₂CH₂O+CHOH); MS (FAB) 1320 (M+H),

1276 (M+H), 1232 (M+H), 1188 (M+H), 1144

(M+H).

N,N-Bis(3-(1,1-dimethylethoxycarbonylamino)propyl)-2-

cyanoethylamine (48). The amine 47 (1.11 g, 3.4 mmol)

1794

S. W. Garrett et al. / Bioorg. Med. Chem. 8 (2000) 1779±1797

was heated at re¯ux with propenenitrile (1.5 g, 21 mmol)

in THF (10 mL) for 60 h. Evaporation and chromato-

graphy (EtOAc) gave 48 (1.04 g, 80%) as a pale-yellow

oil (lit.⁵⁷oil): ¹H NMR d 1.44 (18H, s, 2ÂBu^t), 1.64 (4H,

qn, J=6.6 Hz, 2ÂCH₂CH₂CH₂), 2.45 (2H, t, J=6.9Hz,

CH₂CN), 2.48 (4H, t, J=6.7 Hz, 2ÂCH₂NH), 2.76 (2H, t,

J=7.0Hz, CH₂CH₂CN), 3.18 (4H, m, 2ÂCH₂NBoc),

5.12 (2H, m, 2ÂNHBoc); MS (FAB) m/z 386.2853

(M+H) (¹³C¹²C₁₈H₃₇N₄O₄requires 386.2848), 385.2818

(M+H) (¹²C₁₉H₃₇N₄O₄requires 385.2815).

m, 4ÂCH₂N), 3.22 (3H, s, OMe), 3.46±3.48 (2H, m) and

3.54±3.61 (ca. 40H, m) (nÂOCH₂CH₂O), 4.06 (2 H, m,

CH₂O₂CN).

Bis(3-tri¯uoroacetamidopropyl)amine (54). (H₂N(CH₂)₃)₂

NH 46 (15.0 g, 115 mmol) was stirred with EtO₂CCF₃

(33.4 g, 235 mmol) in EtOH (100 mL) for 60 h. The sol-

vent was evaporated. The residue, in EtOAc, was

washed with water and with brine and was dried. Eva-

poration gave 54 (37.0 g, 100%) as a pale-yellow solid:

mp 168±171 ^ꢀC; H NMR d 1.74 (4H, qn, J=6.6 Hz,

1

3-Amino-N,N-bis(3-t-butoxycarbonylaminopropyl)propyl-

amine (49). W-2 Raney Ni (1.0 g) was added to 48

(971 mg, 2.5 mmol) in MeOH (35 mL). NH₃was passed

through the suspension for 30 min at 0 ^ꢀC. The mixture

was treated with H₂(3000 Torr) for 72 h. The suspension

was ®ltered (Celite¹). Evaporation of the solvent from

the combined ®ltrate and MeOH washings gave crude 49

(1.0 g, 99%) as a pale-blue oil, which was used immedi-

ately: ¹H NMR d 1.43 (20H, s+br, 2ÂBu^t+NH₂), 1.63±

1.66 (6H, m, 3ÂCH₂CH₂CH₂), 2.44±2.75 (8H, m,

4ÂCH₂N), 3.12±3.20 (4H, m, 2ÂCH₂NBoc), 5.28 (2H,

br, 2ÂNH); MS m/z 390.3160 (M+H) (¹³C₁C₁₈

H₄₁N₄O₄requires 390.3161), 389.3126 (M+H) (C₁₉H₄₁

N₄O₄requires 389.3128).

2ÂCH₂CH₂CH₂), 2.00 (1H, br, NH), 2.73 (4H, t,

J=6.3 Hz, 2ÂCH₂NH), 3.44 (4H, t, J=6.3 Hz, 2ÂCH₂

NCOCF₃), 8.67 (2H, br, 2ÂNHCOCF₃); ¹³C NMR d

27.8, 39.3, 47.9, 116.1 (q, J_C-F=288 Hz, CF₃), 157.3 (q,

J_C-F=37 Hz, COCF₃); MS (FAB) m/z 325.1192 (M+H)

(¹³C¹²C₉H₁₆F₆N₃O₂requires 325.1180), 324.1162

(M+H) (¹²C₁₀H₁₆F₆N₃O₂requires 324.1146).

1,1-Dimethylethyl N,N-di(3-aminopropyl)carbamate (56).

Compound 54 (270 mg 840 mmol) was stirred with

Boc₂O (182 mg, 840 mmol) in CH₂Cl₂(5 mL) for 2 days.

Evaporation gave crude 55 (360 mg, 100%) as a col-

ourless oil: H NMR d (23 ^ꢀC) 1.47 (9H, s, Bu^t), 1.77

1

(4H, m, 2ÂCH₂CH₂CH₂), 3.25 (8H, m, 4ÂCH₂N), 6.90

ꢀ

1

(1H, br, NH), 8.24 (1H, br, NH); H NMR d ( 40 C)

1.47 (9H, s, Bu^t), 1.61 (2H, m, 2ÂCH₂CH₂CH₂), 1.83 (2H,

qn, J=7.4Hz, CH₂CH₂CH₂), 3.22 (2H, t, J=7.0Hz,

CH₂N), 3.28±3.35 (4H, m, 2ÂCH₂N), 3.37 (2H, td,

J=7.0, 5.4 Hz, CH₂NCOCF₃), 7.55 (1H, m, NH), 8.69

(1H, t, J=5.4 Hz, NH); MS (FAB +ve ion) m/z 446

(M+Na), 424 (M+H); MS (FAB ve ion) m/z 422

(M H). This material 55 (310 mg, 730 mmol) in MeOH

(6 mL) was heated with 35% aq NH₃(3 mL) at 60 ^ꢀC in a

sealed vessel for 5 h. Evaporation and chromatography

(CH₂Cl₂:MeOH:35% aq NH₃, 10:5:1) gave 56 (155 mg,

91%) as a colourless oil (lit.⁵⁸oil): ¹H NMR ((CD₃)₂SO)

d 1.41 (9H, s, Bu^t), 1.75 (4H, tt, J=7.3, 6.8 Hz,

2ÂCH₂CH₂CH₂), 2.76 (4H, t, J=7.3 Hz, 2ÂCH₂NH₂),

3.18 (4H, t, J=6.8 Hz, 2ÂCH₂NBoc), 3.38 (4H, brs,

2ÂNH₂); MS (FAB) m/z 254 (M+Na), 232 (M+H),

132 (M+H Boc).

!-MethoxyPEG 550 chloroformate (51). MeOPEG550

50 (10.0 g, 18 mmol) was stirred with phosgene (CAU-

TION, 20% solution in toluene, 100 mL) in CH₂Cl₂

(150mL) for 48h. This solution was stored until required.

The solvents and excess reagents were evaporated from

required aliquots to give 51 as a colourless oil: ¹H NMR d

3.38 (3H, s, Me), 3.54 (2H, m, CH₂), 3.60±3.70 (ca. 45H,

m, (CH₂)_n), 3.78 (2H, m, CH₂), 4.46 (2H, m, CH₂OCO);

MS m/z 777 (8%) (M+Na), 733 (10%) (M+Na), 689

(12%) (M+Na), 645 (14%) (M+Na), 601 (14%)

(M+Na), 557 (12%) (M+Na), 513 (8%) (M+Na), 755

(7%) (M+H), 711 (8%) (M+H), 667 (10%) (M+H), 623

(12%) (M+H), 579 (13%) (M+H), 535 (12%) (M+H).

N,N-Bis(3-(1,1-dimethylethoxycarbonylamino)propyl)-3-

(! - MeOPEG550oxycarbonylamino)propylamine (52).

The crude amine 49 (94 mg, 240 mmol) was stirred with

o-MeOPEG550 chloroformate 51 (192 mg, 300 mmol)

and Et₃N (50 mg, 500 mmol) in CH₂Cl₂(5 mL) for 16 h.

Evaporation and chromatography (CH₂Cl₂:MeOH, 7:1)

gave 52 (65 mg, 28%) as a colourless oil: ¹H NMR d 1.44

(18H, s, 2ÂBu^t), 1.63±1.66 (6H, m, 3ÂCH₂CH₂CH₂),

2.44±2.50 (6H, m, 3ÂCH₂N), 3.12±3.25 (6H, m,

2ÂCH₂NCO₂R), 3.38 (3H, s, OMe), 3.54±3.56 (2 H, m)

and 3.62±3.70 (ca. 40H, m) (nÂOCH₂CH₂O), 4.20 (2H,

m, CH₂O₂CN), 5.27 (2H, br, 2ÂNHBoc), 5.64 (1H, br,

NH); MS (FAB) m/z 1019 (25%) (M+H), 975 (40%)

(M+H), 931 (46%) (M+H), 887 (55%) (M+H), 843

(57%) (M+H), 799 (50%) (M+H), 755 (44%) (M+H),

711 (30%) (M+H).

1,1-Dimethylethyl N,N-di(3-(2-cyanoethylamino)propyl)-

carbamate (57). The diamine 56 (2.4 g, 10.5 mmol) was

stirred with propenenitrile (1.5 g, 28 mmol) in MeOH

(60 mL) for 40 h. The solvent was evaporated. The resi-

due, in CH₂Cl₂, was washed with water and with brine

and was dried (fraction A). The water was evaporated

from the combined aqueous phases. The residue was

extracted with MeOH (30 mL). Propenenitrile (1.8 g,

34 mmol) was added and the mixture was stirred for 4

days. K₂CO₃(1.5 g, 11 mmol) was added and the mixture

was stirred for 14 h. The solvent was evaporated. The

residue, in CH₂Cl₂, was washed with water and with

brine and was dried. This solution was combined with

fraction A. Evaporation and chromatography (CH₂Cl₂:

MeOH, 4:1!3:1) gave 57 (1.04 g, 30%) as a pale-yellow

oil: ¹H NMR d 1.46 (9H, s, Bu^t), 1.72 (6H, m,

2ÂCH₂CH₂CH₂+2ÂNH), 2.50 (4H, t, J=6.5 Hz,

2ÂCH₂CN), 2.63 (4H, t, J=6.6 Hz, 2ÂCH₂CH₂CN), 2.89

(4H, t, J=6.6 Hz, 2ÂCH₂N), 3.22 (4H, m, 2ÂCH₂NBoc);

MS (FAB) m/z 339.2588 (M+H) (¹³C¹²C₁₆H₃₂N₅O₂

!-MeOPEG550 N-(3-(N,N-di(3-aminopropyl)amino)-

propyl)carbamate trihydrochloride (53). Compound 52

was treated with HCl, as for the synthesis of 14, to give 53

(100%) as a colourless glass: ¹H NMR (D₂O) d 1.75±1.81

(2H, m, CH₂CH₂CH₂), 1.92±2.02 (4H, m, 2ÂCH₂CH₂

CH₂), 2.94 (4H, t, J=7.7 Hz, 2ÂCH₂N), 3.07±3.16 (8H,

S. W. Garrett et al. / Bioorg. Med. Chem. 8 (2000) 1779±1797

1795

requires 339.2589), 338.2566 (M+H) (¹²C₁₇H₃₂N₅O₂

requires 338.2556), 238 (M+H Boc).

(693 mg, 2.1 mmol) and K₂CO₃(285 mg, 2.0 mmol) in

DMF (4.0 mL) at 85 ^ꢀC for 20 h. The solvent was eva-

porated. The residue, in CH₂Cl₂, was washed with water

and with brine and was dried. Evaporation and chro-

matography (EtOAc:MeOH, 6:1!4:1) gave 64 (965 mg,

1,1-Dimethylethyl N,N-di(3-(N-phenylmethoxycarbonyl)-

N - (3 - (phenylmethoxycarbonylamino)propyl)amino)pro-

pyl)carbamate (59). The dinitrile 57 (1.00 g, 3.0 mmol)

in MeOH (20 mL) was saturated with NH₃and was

then treated with H₂(3500 Torr) in the presence of W-2

Raney Ni (1.0 g) for 60 h. The suspension was ®ltered.

The solvent was evaporated from the combined ®ltrate

and MeOH washings to give crude 58, which was stirred

with Cbz₂O (4.23 g, 15 mmol) in THF (30 mL) for 36h.

Evaporation and chromatography (EtOAc/hexane, 1:1

!2:1) gave 59 (1.57 g, 65%) as a colourless oil: ¹H NMR d

1.40 (9H, s, Bu^t), 1.61±1.78 (8H, m, 4ÂCH₂CH₂CH₂),

3.05±3.38 (16H, m, 8ÂCH₂N), 5.09 (8H, s, 4ÂPhCH₂),

5.68 (2H, br, 2ÂNH), 7.33 (20H, m, 4ÂPh-H₅); MS

(FAB) m/z 904 (M+Na), 883.4661 (M+H) (¹³C¹²C₄₈

H₆₄N₅O₁₀requires 883.4687), 882.4646 (M+H) (¹²C₄₉

H₆₄N₅O₁₀requires 882.4653), 782 (M+H Boc).

1

67%) as a colourless oil: H NMR d 1.24±1.42 (4H,

m, 3,4-H₄), 1.45 (9H, s, Bu^t), 1.59±1.68 (10H, m,

5-H₂+4ÂNCH₂CH₂CH₂N), 2.14 (2H, m, 2-H₂),

2.25±2.40 (6H, m, 3ÂCH₂N), 3.15±3.29 (16H, m,

8ÂCH₂NCO), 5.10 (8H, s, 4ÂPhCH₂), 5.74 (1H, m,

NH), 6.43 (1H, m, NH), 7.32 (20H, m, 4ÂPh-H₅);

¹³C NMR d 25.2, 25.6, 25.8, 26.1, 26.4, 27.0, 28.1, 28.4,

28.9, 36.4, 37.7, 38.3, 40.4, 44.4, 45.2, 46.0, 51.2, 51.3,

53.4, 66.5, 67.1, 79.5, 127.8, 128.0, 128.3, 128.5, 136.0,

136.5, 136.7, 156.0, 156.6, 156.9, 173.7; MS (FAB)

12

m/z 1040.5960 (M+H) (¹³C₂C₅₅H₈₀N₇O₁₁requires

1040.5983), 1039.5912 (M+H) (¹³C¹²C₅₆H₈₀N₇O₁₁

requires 1039.5949), 1038.5893 (M+H) (¹²C₅₇H₈₀N₇O₁₁

requires 1038.5915).

N-(2-Aminoethyl)-6-(di(3-(N-phenylmethoxycarbonyl)-N-

(3 - (phenylmethoxycarbonylamino)propyl)amino)propyl)

amino)hexanamide (65). HCl was passed through 64

(860 mg, 830 mmol) in CH₂Cl₂(25 mL) for 30 min. The

solvent and excess reagent were evaporated. The resi-

due, in MeOH (20 mL), was treated with aq NaOH

(5 M, 660 mL, 3.3 mmol). The solvents were evaporated.

The residue, in CH₂Cl₂, was washed with water and

with brine and was dried. Evaporation gave 65 (676 mg,

Di(3-(N-phenylmethoxycarbonyl)-N-(3-(phenylmethoxy-

carbonylamino)propyl)amino)propyl)amine (60). HCl

was passed through 59 (1.33 g, 1.5 mmol) in CH₂Cl₂

(40 mL) for 30 min. The solvent and excess reagent were

evaporated. The residue, in MeOH (10 mL), was treated

with aq NaOH (5 M, 900 mL, 4.5 mmol). The solvents

were evaporated. The residue, in CH₂Cl₂, was washed

with water and with brine and was dried. Evaporation

1

gave 60 (1.07 g, 91%) as a colourless oil: H NMR d

67%) as a colourless oil: H NMR d 1.25±1.33 (4H, m,

1.55±1.77 (8H, m, 4ÂCH₂CH₂CH₂), 2.10 (1H, br, NH),

2.40±2.58 (8H, m, 4ÂCH₂N), 3.06±3.39 (8H, m,

4ÂCH₂NCbz), 5.10 (8H, s, 4ÂPhCH₂), 5.69 (2H, br,

2ÂNH), 7.33 (20H, m, 4ÂPh-H₅); MS (FAB) m/z

783.4168 (M+H) (¹³C¹²C₄₃H₅₆N₅O₈requires 783.4162),

782.4139 (M+H) (¹²C₄₄H₅₆N₅O₈requires 782.4129).

3,4-H₄), 1.61±1.68 (10H, m, 5-H₂+4ÂNCH₂CH₂CH₂N),

2.03±2.16 (4H, m, 2-H₂+CH₂NH₂), 2.25±2.35 (6H, m,

3ÂCH₂N), 2.78 (2H, br, NH₂), 3.15±3.26 (14H, m,

7ÂCH₂NCO), 5.10 (8H, s, 4ÂPhCH₂), 5.75 (1H, m,

NH), 6.27 (1H, m, NH), 7.32 (20H, m, 4ÂPh-H₅); MS

(FAB) m/z 939.5412 (M+H) (¹³C¹²C₅₁H₇₂N₇O₉

requires 939.5425), 938.5383 (M+H) (¹²C₅₂H₇₂N₇O₉

requires 938.5391).

6-Bromo-N-(2-(1,1-dimethylethoxycarbonylamino)ethyl)-

hexanamide (63). 6-Bromohexanoic acid 62 (1.53 g,

7.9 mmol) was stirred with oxalyl chloride (5.0 g, 39mmol)

and DMF (70 mL) in CH₂Cl₂(15 mL) at 0 ^ꢀC for 30min

and at 20^ꢀC for 100 min. The solvent and excess reagent

were evaporated. CH₂Cl₂(5mL) was added and evapo-

rated. The residue was stirred with BocNH(CH₂)₂NH₂

61³²(2.04 g, 12.5mmol) and Et₃N (2.02 g, 10mmol) in

CH₂Cl₂(20 mL) at 0 ^ꢀC for 30min and at 20^ꢀC for

90min. Washing (water, brine), drying, evaporation and

recrystallisation (EtOAc) gave 63 (1.79 g, 68%) as white

6-(Di(3-(N-phenylmethoxycarbonyl)-N-(3-(phenylmethoxy-

carbonylamino)propyl)amino)propyl)amino)-N-(2-(!-Me

OPEG550oxycarbonylamino)ethyl)hexanamide (66). The

amine 65 (408 mg, 440 mmol) was stirred with 51

(320 mg, 500 mmol) and Et₃N (120 mg, 1.2 mmol) in

CH₂Cl₂(6.0 mL) for 3 h. Further 51 (45 mg, 70 mmol) in

CH₂Cl₂(1.0 mL) was added and the mixture was stirred

for 4 h. Evaporation and chromatography (CH₂Cl₂:

MeOH, 20:1) gave 66 (274 mg, 42%) as a colourless oil:

¹H NMR d 1.25±1.38 (4H, m, 3,4-H₄), 1.60±1.70

(10H, m, 5-H₂+4ÂNCH₂CH₂CH₂N), 2.14 (2H, m, 2-

H₂), 2.24±2.38 (6H, m, 3ÂCH₂N), 3.15±3.31 (16H, m,

8ÂCH₂N), 3.38 (3H, s, OMe), 3.53±3.56 (2H, m,

OCH₂), 3.62±3.82 (ca. 36H, m, nÂOCH₂CH₂O), 4.18

(2H, m, CH₂O₂C), 5.10 (8H, s, 4ÂPhCH₂), 5.52 (1H, m,

NH), 5.71 (1H, m, NH), 7.33 (20H, m, 4ÂPh-H₅); MS

(FAB) ¹³C/¹²C isotope clusters centred at m/z 1481

(3%) (M+H), 1437 (4%) (M+H), 1391 (5%) (M+H),

1349 (6%) (M+H), 1305 (6%) (M+H), 1261 (5%)

(M+H), 1217 (4%) (M+H).

ꢀ

1

crystals: mp 104±106 C; H NMR d 1.44 (9H, m, Bu^t),

1.47 (2H, m, 4-H₂), 1.66 (2H, qn, J=7.2 Hz, 3-H₂), 1.87

(2H, qn, J=7.2Hz, 5-H₂), 2.20 (2H, qn, J=7.3 Hz, 2-H₂),

3.27±3.37 (4H, m, NCH₂CH₂N), 3.41 (2H, qn, J=6.8Hz,

6-H₂), 5.04 (1H, br, NH), 6.38 (1H, br, NH); MS (FAB)

m/z 340.1118 (M+H) (¹³C¹²C₁₂H₂₆⁸¹BrN₂O₃requires

340.1139), 339.1097 (M+H) (¹²C₁₃H₂₆⁸¹BrN₂O₃requires

339.1106), 338.1113 (M+H) (¹³C¹²C₁₂H₂₆⁷⁹BrN₂O₃

requires 338.1160), 337.1121 (M+H) (¹²C₁₃H₂₆⁷⁹BrN₂

O₃requires 337.1127).

N-(2-(1,1-Dimethylethoxycarbonylamino)ethyl-6-(di(3-(N-

phenylmethoxycarbonyl)-N-(3-(phenylmethoxycarbonyl-

amino)propyl)amino)propyl)amino)hexanamide (64).

Compound 60 (1.07 g, 1.4 mmol) was stirred with 63

6-(N,N-Di(3-(3-aminopropylamino)propyl)amino)-N-(2-

(! - MeOPEG550oxycarbonylamino)ethyl)hexanamide

(67). The protected construct 66 (75 mg, 50 mmol) was

1796

S. W. Garrett et al. / Bioorg. Med. Chem. 8 (2000) 1779±1797

treated with H₂(3500 Torr) in MeOH (5 mL) in the

presence of Pearlman's catalyst (50 mg) for 40 h. The

mixture was diluted with MeOH (10 mL) and ®ltered

(Celite¹). Aqueous HCl (5M, 2.0 mL, 10mmol) was

added. Evaporation and freeze-drying gave 67 (57 mg,

100%) as a colourless glass: ¹H NMR (D₂O) d 1.60±1.71

(4H, m, 3,4-H₄), 2.06±2.15 (10H, m, 4ÂNCH₂CH₂-

CH₂N+5-H₂), 2.27 (2H, t, J=7.0 Hz, 2-H₂), 3.09±3.30

(22H, m, 11ÂCH₂N), 3.38 (3H, s, OMe), 3.62±3.66 (2H,

m, OCH₂), 3.70±3.75 (ca. 36H, m, nÂOCH₂CH₂O), 4.22

(2H, m, CH₂O₂C); MS (FAB) ¹³C/¹²C isotope clusters

centred at m/z 944 (15%) (M+H), 900 (22%) (M+H),

856 (32%) (M+H), 812 (35%) (M+H), 768 (30%)

(M+H), 724 (30%) (M+H), 680 (22%) (M+H).

tion. On the day of implantation the mice were lightly

anaesthetised using intraperitoneal Hypnorm (mix of

1

fentanyl citrate (0.315 mg mL

)

and ¯uanisone

(10 mg mL ¹), 0.05 mL per mouse) and an injection of

0.05 mL PBS containing 2Â10⁵RIF-1 cells made intra-

dermally midway along the back of the mouse

(approximately 2 cm from the base of the tail). Tumours

were visible after approximately 7 days and were ready for

gene delivery experiments 12±14 days post-implantation.

One day before treatment with DNA, the mice were

weighed and the dimensions of the tumour were recorded.

The average tumour was approximately 6±7 mm in dia-

meter and 250 mg in weight at this stage. The mice were

then randomly assigned to the dierent treatment cages.

The plasmid pCMVCAT was supplied in bulk as a gift

from GeneMedicine Inc. (now Valentis Inc.). Injections

were made of plasmid DNA (50 mg) in an isotonic solu-

tion (20 mL), either alone or with the test polyamine±

PEG construct. After 48 h, the animals were killed and

the excised tumours were weighed, then homogenised

for 30 s in a screw cap sterile tube containing lysis buer

(1.5 mL, potassium phosphate buer (0.1 M, pH 7.8),

0.2% Triton X-100) using an Ultraturax homogeniser.

The tissue suspension was then spun at 12,000Âg for

15 min at 4 ^ꢀC to remove cellular debris. This was fol-

lowed by two freeze±thaw cycles, freezing at 70^ꢀC and

then thawing at room temperature. The protein content of

this supernatant was analysed using a Bio-Rad D_cProtein

assay kit. Supernatant containing 10±100mg of protein

was assayed for the bacterial enzyme chloramphenicol

acetyltransferase type I (CAT), using a commercial

ELISA colourimetric enzyme immunoassay from Boeh-

ringer-Mannheim (Mannheim, Germany). Expression

was normalised with respect to either total tissue weight.

DNA binding studies

Plasmid DNA (pCMVcat) was a generous gift from

GeneMedicine, The Woodlands, Texas, USA. The

experiments were performed in HEPES-buered

saline (HBS), containing 4-(2-hydroxyethyl)piperazine-

1-ethanesulfonic acid (HEPES) (20 mM) and NaCl

(150 mM); the pH was adjusted to 7.4 by addition of aq

NaOH (1.0 M). The experiment at pH 5.5 was per-

formed in MES-buered saline, containing morpholine-

4-ethanesulfonic acid (MES) (20 mM) and NaCl

(150 mM); the pH was adjusted to 5.5 by addition of

aq HCl (1.0 M). Fluorescence was measured on a Perkin

Elmer LS-50B luminescence spectrometer with excita-

tion wavelength 260 nm and emission wavelength

600 nm.

For each assay, aqueous ethidium bromide (3.0 mL,

0.5 mg mL ¹) was added to DNA (6 mg) in HBS (3 mL)

and the mixture was stirred in a ¯uorimeter cuvette until

no change in ¯uorescence occurred (usually 2±3 min).

The test construct in HBS (4.0 mL) was added with con-

tinuous stirring. The new ¯uorescence reading was

taken 60 s later. An additional aliquot of test construct

solution was added. This sequence was repeated until no

further change in ¯uorescence is observed or when the

total volume of construct solution added is 2.5% of the

total volume of the system. The percent relative ¯uor-

escence intensity was determined using

Statistical comparisons were performed using non-para-

metric Kruskal±Wallis and post-hoc Mann±Whitney

tests (Minitab for Windows version 11). These rank

order tests make no assumptions regarding the type of

distribution of values around the mean (e.g., normal,

log-normal, etc.), and are valid for use in comparing two

independent groups that contain the same or dierent

numbers of values. Experimental group medians were

considered signi®cantly dierent from each other or

control group medians if the p value was less than 0.05.

The data are presented in Figure 5 as box plots. The box

represents the data between the lower and upper quar-

tiles, that is, 50% of data, and lines (whiskers) are

drawn to represent the lower and upper extremes. The

median is represented by a horizontal line.

F_r ꢀF_obsF_e=ꢀF₀F_e

ꢀ1

F_ris the relative ¯uorescence, F_obsis the measured

¯uorescence, F_eis the ¯uorescence of ethidium bromide

without DNA and F₀is the initial ¯uorescence of the

ethidium±DNA complex before any construct is added.

Acknowledgements

DNA delivery studies in vivo

Female C3H mice were obtained from Charles River at

6±7 weeks of age and then allowed to acclimatise for a

further week prior to experiments. Before implanting

murine ®brosarcoma RIF-1 cells⁵⁹the fur from the

lower half of the mouse back was removed using clip-

pers. The RIF-1 cells were previously cultured in RPMI

1640 medium supplemented with 15% fetal calf serum,

penicillin and streptomycin, and were not passaged

more than 7 times prior to trypsinisation for implanta-

The authors thank Mr. R. R. Hartell and Mr. D. J.

Wood (University of Bath) for the NMR spectra,

Mr. C. Cryer (University of Bath) for the FAB and

electrospray mass spectra and Mr. M. Domin (Uni-

versity of London) for the MALDI-TOF mass spectra.

SWG thanks the University of Bath for a studentship.

Part of this work was carried out as an Undergraduate

Research Project by ORD as part of the BPharm

degree.

S. W. Garrett et al. / Bioorg. Med. Chem. 8 (2000) 1779±1797

1797

References

30. Saari, W. S.; Schwering, J. E.; Lyle, P. A.; Smith, S. J.;

Engelhardt, E. L. J. Med. Chem. 1990, 33, 97.

31. Blagbrough, I. S.; Geall, A. J. Tetrahedron Lett. 1998, 39,

439.

1. Dachs, G. U.; Dougherty, G. J.; Stratford, I. J.; Chaplin,

D. J. Oncol. Res. 1997, 9, 313.

2. Culver, K. W.; Vickers, T. M.; Lamsam, J. L.; Walling, H.

W.; Seregina, T. Br. Med. Bull. 1995, 51, 192.

3. Mendiratta, S. K.; Quezada, A.; Matar, M.; Wang, J.;

Hebel, H. L.; Long, S.; Nordstrom, J. L.; Pericle, F. Gene

Ther. 1999, 6, 833.

4. Hersch, E. M.; Stopeck, A. T. In Self-assembling Com-

plexes for Gene Therapy: from Laboratory to Clinical Trial;

Kabanov, A. V., Felgner, P. L., Seymour L. W., Eds.; Wiley:

Chichester, 1998; pp 421±436.

32. Kneeland, D. M.; Ariga, K.; Lynch, V. M.; Huang, C.-Y.;

Anslyn, E. V. J. Am. Chem. Soc. 1993, 115, 10042.

33. Cain, B. F.; Baguley, B. C.; Denny, W. A. J. Med. Chem.

1978, 21, 658.

34. Gershon, H.; Ghirlando, R.; Guttman, S. B.; Minsky, A.

Biochemistry 1993, 32, 7143.

35. Tang, M. X.; Szoka, F. C. Gene Ther. 1997, 4, 823.

36. Delcros, J.-G.; Sturkenboom, M. C. J. M.; Basu, H. S.;

Shaer, R. H.; Szollosi, J.; Feuerstein, B. G.; Marton, L. J.

Biochem. J. 1993, 291, 269.

37. Stewart, K. D.; Gray, T. A. J. Phys. Org. Chem. 1992, 5,

461.

38. Vasilevskaya, V. V.; Khoklov, A. R.; Matsuzawa, Y.;

Yoshikawa, K. J. Chem. Phys. 1995, 102, 6595.

39. Minagawa, K.; Matsuzawa, Y.; Yoshikawa, K.; Khoklov,

A. R.; Doi, M. Biopolymers 1994, 34, 555.

40. Lerman, L. S. Proc. Natl. Acad. Sci. USA 1971, 68,

1886.

5. Robbins, P. D.; Tahara, H.; Ghivizzi, S. C. Trends Biotech.

1998, 16, 35.

6. French Anderson, W. Nature 1998, 392, 25.

7. Friedmann, T. Scienti®c American 1997, 276, 80.

8. Felgner, P. L.; Gadek, T. R.; Holm, M.; Roman, R.; Chan,

H. W.; Wenz, M.; Northrop, J. P.; Ringold, G. M.; Danielsen,

M. Proc. Natl. Acad. Sci. USA 1987, 84, 7413.

9. Gao, X.; Huang, L. Biochem. Biophys. Res. Commun. 1991,

272, 280.

10. Cooper, R. G.; Etheridge, C. J.; Stewart, L.; Marshall, J.;

Rudginsky, S.; Cheng, S. H.; Miller, A. D. Chem. Eur. J. 1998,

4, 137.

11. Mahato, R. I.; Rolland, A.; Tomlinson, E. Pharm. Res.

1997, 14, 853.

41. CRC Handbook of Chemistry and Physics, 58th edition.

Weast, R. C., Ed.; CRC Press, Cleveland, Ohio, USA, 1977.

42. Olins, D. E.; Olins, A. L. J. Mol. Biol. 1971, 57, 437.

43. Vinogradov, S. V.; Bronich, T. K.; Kabanov, A. V. Bio-

conj. Chem. 1998, 9, 805.

12. Plank, C.; Mechtler, K.; Szoka, F. C.; Wagner, E. Human

Gene Ther. 1996, 7, 1437.

13. Pouton, C. W.; Seymour, L. W. Adv. Drug Del. Rev. 1998,

34, 3.

44. Wagner, E. In Self-assembling Complexes for Gene Deliv-

ery: from Laboratory to Clinical Trial; Kabanov, A. V.,

Felgner, P. L., Seymour, L. W., Eds.; Wiley: Chichester, 1998;

pp 309±322.

14. Bally, M. B.; Harvie, P.; Wong, F. M. P.; Kong, S.;

Wasan, E. K.; Reimer, D. L. Adv. Drug Del. Rev. 1999, 38,

291.

15. Yang, J.-P.; Huang, L. Gene Ther. 1996, 3, 542.

16. Kabanov, A. V.; Vinogradov, S. V.; Suzdaltseva, Y. G.;

Alakhov, V. Y. Bioconj. Chem. 1995, 6, 639.

17. Toncheva, V.; Wolfert, M. A.; Dash, P. R.; Oupicky, D.;

Ulbrich, K.; Seymour, L. W.; Schacht, E. H. Biochim. Biophys.

Acta 1998, 1380, 354.

18. Wolfert, M. A.; Schacht, E. H.; Toncheva, V.; Ulbrich, K.;

Nazarova, O.; Seymour, L. W. Human Gene Ther. 1996, 7, 2123.

19. Choi, Y. H.; Liu, F.; Kim, J.-S.; Choi, Y. K.; Park, J. S.;

Kim, S. W. J. Controlled Release 1998, 54, 39.

20. Mumper, R. J.; Duguid, J. G.; Anwar, K.; Barron, M. K.;

Nitta, H.; Rolland, A. P. Pharm. Res. 1996, 13, 701.

21. Mumper, R. J.; Wang, J. J.; Klakamp, S. L.; Nitta, H.;

Anwer, K.; Tagliaferri, F.; Rolland, A. P. J. Controlled

Release 1998, 52, 191.

22. Mumper, R. J.; Rolland, A. P. Adv. Drug Del. Rev. 1998,

30, 151.

23. Katre, N. V. Adv. Drug Delivery Rev. 1993, 10, 91.

24. Yoshikawa, Y.; Yoshikawa, Y. FEBS Lett. 1995, 361, 277.

25. Israel, M.; Rosen®eld, J. S.; Modest, E. J. J. Med. Chem.

1964, 7, 710.

26. McCormick, K. D.; Kobayashi, K.; Goldin, S. M.; Reddy,

N. L.; Meinwald, J. Tetrahedron 1993, 49, 11155.

27. Xu, D.; Prasad, K.; Repic, O.; Blacklock, T. J. Tetra-

hedron Lett. 1995, 36, 7357.

28. Matthews, S. E.; Pouton, C. W.; Threadgill, M. D. J.

Controlled Release, in press.

45. Davis, H. L.; Jasmin, B. J. FEBS Lett. 1993, 333, 146.

46. Dowty, M. E.; Williams, P.; Zhang, G.; Hagstrom, J. E.;

Wol, J. A. Proc. Natl. Acad. Sci. USA 1995, 92, 4572.

47. Levy, M. Y.; Barron, L. G.; Meyer, K. B.; Szoka, J. F. C.

Gene Ther. 1996, 3, 201.

48. Wol, J. A.; Malone, R. W.; Williams, P.; Wang, C.;

Acsadi, G.; Jani, A.; Felgner, P. L. Science 1990, 247, 1465.

49. Mumper, R. J.; Barron, M. K.; Anwer, K.; Lessard, R. L.;

Liu, Q.; Nitta, H.; Alila, H.; Rolland, A. Pharm. Res. 1995, 12,

80.

50. Manthorpe, M.; Cornefert-Jensen, F.; Hartikka, J.;

Felgner, J.; Rundell, A.; Margalith, M. Human Gene Ther.

1993, 4, 419.

51. Lucas, P.; Milroy, D. A.; Thomas, B. J.; Moss, S. H.;

Pouton, C. W. J. Drug Targeting 1999, 7, 143.

52. Bloom®eld, V. A. Curr. Opin. Struct. Biol. 1996, 6, 334.

53. Milroy, D. A.; Wood, P. J.; Garrett, S. W.; Threadgill, M.

D.; Pouton, C. W. J. Drug Targeting, submitted.

54. Nakamura, Y. Heterocycles 1988, 27, 1873.

55. Bergeron, R. J.; McManis, J. S.; Liu, C. Z.; Feng, Y.;

Weimar, W. R.; Luchetta, G. R.; Wu, Q. H.; Ortiz-Casio, J.;

Vinson, J. R. Y.; Kramer, D.; Porter, C. J. Med. Chem. 1994,

37, 3464.

56. Marsh, I. R.; Bradley, M. Tetrahedron 1997, 53, 17317.

57. Malabarba, A.; Ciabatti, R.; Kettenring, J.; Scotti, R.;

Candiani, G.; Pallanza, R.; Berti, M.; Goldstein, B. P. J. Med.

Chem. 1992, 35, 4054.

58. Clark, B. P.; Harris, J. R.; Timms, G. H.; Olkowski, J. L.

Tetrahedron Lett. 1995, 36, 3889.

59. Twentyman, P. R.; Brown, J. M.; Gray, J. W.; Franko, A.

J.; Scoles, M. A.; Kallman, R. F. J. Natl. Cancer Inst. 1980,

64, 595.

29. Matthews, S. E.; Pouton, C. W.; Threadgill, M. D. New J.

Chem. 1999, 23, 1087.

Article Doi

Synthesis and characterisation of polyamine-poly(ethylene glycol) constructs for DNA binding and gene delivery

DOI: 10.1016/S0968-0896(00)00113-9

Source and publish data:

Authors:

Article abstract of DOI:10.1016/S0968-0896(00)00113-9

Full text of DOI:10.1016/S0968-0896(00)00113-9

Products guided by the article

R&D Labs maybe for 294864-70-3

Relevant to this article

Hot Product