Synthesis, Characterization and Application of

68Ga-labelled Peptides and Oligonucleotides

by

Irina Velikyan

Thesis for the filosofie licentiate degree

January 2004

UPPSALA UNIVERSITY Institute of Chemistry

Department of Organic Chemistry

2

Contents

Abstract 3 Papers included in the thesis 4 List of abbreviations 5 1. Introduction 6 1.1 Tracer concept 6 1.2 Imaging techniques; Positron Emission Tomography 6 1.3 Nuclear and chemical properties of 68Ga radiometal 7 1.4 Radionuclide production; generator system 8 1.5 Time factor in radiolabelling chemistry 8 1.6 Microwave activation 8 1.7 Bifunctional chelator 9 1.8 Complexation reaction 101.9 Stability of chelate compounds 101.10 Macromolecular tracers 121.11 Specific radioactivity 152. Aims of the study 153. Results and discussion 153.1 Method of obtaining 68Ga 15 3.1.1 68Ge/68Ga generator characterisation 15 3.1.2 Preconcentration and purification of the 68Ge/68Ga generator eluate 173.2 68Ga-labelling of macromolecules 19 3.2.1 Conjugation of DOTA with macromolecules and 68Ga-labelling of the

conjugates 19

3.2.2 68Ga-labelling with non-treated 68Ge/68Ga generator eluate 22 3.2.3 68Ga-labelling using preconcentrated and purified generator eluate 243.3 Microwave activation 253.4 Specific radioactivity 263.5 Purification and characterization of the macromolecular conjugates and their 26 68Ga-labelled counterparts 3.5.1 Purification 26 3.5.2 Chemical characterization 27 3.5.3 Preliminary biological examination of the 68Ga-labelled oligonucleotide 28 conjugates 4. Conclusions 31Acknowledgements 32References 33

3

Abstract

The positron emitting 68Ga radionuclide (T1/2 = 68 min) has the potential of practical interest for clinical PET.

The metallic cation, 68Ga3+, is suitable for complexation reactions with chelators either naked or conjugated with

macromolecules such as peptides and oligonucleotides. Such labeling procedures require pure and concentrated

radiometal preparations, which cannot be sufficiently fulfilled by the presently available 68Ge/68Ga generator

eluate. This thesis presents a method to increase the concentration and purity of 68Ga obtained from a

commercial 68Ge/68Ga generator. DOTATOC (DOTA = 1,4,7,10-tetraazacyclo-dodecane-1,4,7,10-tetraacetic

acid, TOC = D-Phe1-Tyr3–Octreotide) was used as a test molecule for comparing the labeling properties of

different 68Ga preparations. In addition, DOTA-RDG (RGD = Cys2-6; c[CH2CO-Lys(DOTA)-Cys-Arg-

Gly-Asp-Cys-Phe-Cys]-CCX6-NH2) and NODAGATATE (NODAGA = 1,4,7-triazacyclononane-1,4,7-

triacetic, TATE = Tyr3 - Octreotate) were used to prove the concept. The use of the concentrated and purified 68Ga eluate along with microwave activation allowed quantitative 68Ga-labelling of peptide conjugates of ≤1

nanomolar quantities within 10 min. The specific radioactivity of the radiolabelled peptides was improved by a

factor of >100 compared to previously applied techniques using non-treated generator eluate and conventional

heating. A commercial 68Ge/68Ga generator in combination with this method for purification, concentration and

microwave activated labeling resulted in a kit technology for 68Ga-tracer production.

Four 17-mer oligonucleotides modified and functionalised with an hexylamine group in the 3'- or 5'- position

were conjugated with DOTA and labelled with 68Ga using microwave activation. Chemical modification of the

oligonucleotide backbone or sugar moiety did not influence the labelling nor the hybridisation ability of the

oligonucleotides. However, the radioactivity organ biodistribution in rats differed dependent on the

oligonucleotide structure. This indicated that metabolism and non-specific binding were affected by the

backbone and sugar moiety structure.

4

Papers included in the thesis This thesis is based on the following selected papers, which are referred to in the text by their roman numerals. I Velikyan I., Beyer G., Långström B., Microwave supported preparation of 68Ga-bioconjugates

with high specific radioactivity. Bioconjugate Chem., 2003, submitted.

II Velikyan I., Lendvai G., Välilä M., Roivainen A., Yngve U., Bergström M. and Långström B., Microwave accelerated 68Ga-labelling of oligonucleotides. J. Labelled Compd. Radiopharm., 2003, accepted.

III Lendvai G., Velikyan I., Bergström M., Laryea D., Välilä M., Salomäki S., Långström B., Roivainen A., Biodistribution of 68Ga -labelled antisense oligonucleotides targeting human K-ras oncogene in normal rats, Eur. J. Pharm. Science, 2003, submitted.

The title page picture represents a patient PET examination image of midgut carcinoid tumor coming from the intestine with multiple liver metastasis. The tracer, 68Ga-DOTATOC, was synthesised employing microwave activation (paper I).

5

List of abbreviations

β+ Positron DNA Deoxyribonucleic acid DOTA 1,4,7,10-Tetraazacyclododecane-1,4,7,10-tetraacetic acid DOTATOC 4,7,10-Tricarboxymethyl-1,4,7,10-tetraazacyclododecan-1-yl-

acetyl-D-Phe-Cys-Tyr-D-Trp-Lys-Thr-Cys-L-threoninol (DOTA-D-Phe1-Tyr3-Octreotide)

DOTA-RGD Cys2-6; c[CH2CO-Lys(DOTA)-Cys-Arg-Gly-Asp-Cys-Phe-Cys]-CCX6-NH2)

EC Electron capture EDC 1-Ethyl-3(3-dimethylaminopropyl)carbodiimide ESI Electrospray ionisation HPLC High-performance liquid chromatography LC Liquid chromatography MRI Magnetic resonance imaging mRNA Messenger-RNA MS Mass spectrometry NOTA 1,4,7-triazacyclononane-1,4,7-triacetic acid NODAGATATE 1,4,7-Tricarboxymethyl-1,4,7-triazacyclononan-1-yl-acetyl-D-Phe-

Cys-Tyr-D-Trp-Lys-Thr-Cys-L-Thr (NODAGA - Tyr3 - Octreotate) PET Positron emission tomography RNA Ribonucleic acid SPE Solid phase extraction SPECT Single photon emission computed tomography SST Somatostatin SSTR Somatostatin receptor Sulfo-NHS 1-Hydroxy-2,5-dioxo-3-pyrrolidinesulfonic acid SUV Standardized uptake value TATE octreotate TEAA Triethylammonium acetate TFA Trifluoroacetic acid Tyr Tyrosine UV Ultraviolet

6

1. Introduction 1.1 Tracer concept

Tracer concept was described by George de Hevesy in 1923.1 Tracer is defined as a substance which is

introduced into a biological or mechanical system and can be followed through the course of a process. The

tracer amount is low so that it does not disturb the studied process, but provides information on the pattern of

events in the process or on the distribution of the elements involved. Tracers containing radioactive nuclides,

which radiation can be detected, are widely used in medicine and in biomedical research for studies of

physiological processes, in vivo parmacokinetics as well as diagnosis and radiotherapy.

1.2 Imaging techniques: Positron Emission Tomography

Magnetic resonance imaging (MRI), gamma scintigraphy and positron emission tomography (PET) are non-

invasive imaging technologies widely used in radiology.2 MRI exploits the difference in relaxation rates of

water in tissues, translating them into three-dimentional anatomical information. Gamma scintigraphy requires a

radiopharmaceutical containing a nuclide that emits gamma (γ) radiation and a gamma camera or single photon

emission computed tomography (SPECT) camera capable of imaging the patient injected with the gamma

emitting radiopharmaceutical. PET requires a radiopharmaceutical labelled with a positron-emitting

radionuclide (β+) and a PET camera for imaging the patient.3,4 Both SPECT and PET provide information about

structure and physiological function of the body since they employ biologically active tracer molecules. Though,

PET has better spatial resolution of about 4 mm, temproral resolution in the order of seconds and, most

importantly, allows accurate quantification of the images. Positron scan registration is based on the 180º

correlation of the 511 keV photons arising from the annihilation of positrons with electrons and detection by

means of two opposing counters recording only coincident events5 (Figure 1). The registered events are

reconstructed into images representing the spatial distribution of the radioactive source in the body.

Figure 1. (A) A positron particle and an electron annihilate producing two 511 keV gamma photons travelling in opposite directions and (B) registered externally by radiodetectors in the PET camera.

Positron emitting isotope Gamma rays 511keV

Gamma ray

Positron

511keV Electron Gamma ray A

7

Positron emission is exclusively a property of the neutron deficient nuclides. Some positron emitting nuclides

are presented in Table 1.

Table 1. Some positron emitting radionuclides and their decay properties.

Radionuclide Half-life E(β+), keV β+ decay, % 11C 20.3 min 961 100 15O 2.07 min 1732 100 18F 110 min 634 97

66Ga 9.5 h 4153 56 68Ga 68 min 1899 89 124I 4.2 d 2138 25

64Cu 12.8 h 656 19 86Y 14.74 h 3150 34

62Cu 9.76 min 2910 98 52Fe 8.2 h 800 57 110In 69 min 637 62

The choice of a radionuclide depends on: i) the physical half-life that should allow the production and

application of the radioactive tracer, however, it should not have too long half-life not to overexpose the patient

to any unnecessary radiation; ii) the decay mode; iii) chemistry for the tracer preparation; iv) the availability and

cost. Generally, a tracer is chosen that will be selectively taken up by a certain type of tissue, e.g. cancer cells.

PET has been employed to follow response to therapy , image tumor6 7 as well as study neurotransmission and

measure blood flow .

8

5 In particular, 68Ga radiopharmaceuticals have been used for imaging of brain, renal, bone,

blood pool, lung, vascular pool and tumor.9,10

1.3 Nuclear and chemical properties of 68Ga radiometal

Naturally occurring gallium consists of two isotopes 69Ga (60.1% natural abundance) and 71Ga (39.9% natural

abundance). Three radioisotopes can be produced that are useful for incorporation into radiopharmaceuticals.

Two of these 66Ga (T1/2 = 9.45 h) and 68Ga (T1/2 = 68 min) decay by β+ -emission and as such are suitable for

PET imaging, 67Ga (T1/2 = 78 h) decays by γ-emission and is used for SPECT imaging. 68Ga is a generator

produced nuclide and does not require a cyclotron on site. The parent 68Ge is accelerator-produced by the 69Ga(p,2n) reaction.11 68Ge decays with a half-life of 270.8 days solely by orbital electron capture to 68Ga. The

latter decays 89% by the emission of positrons (p → n + β+ + ν) of 1.92 MeV max enegry without additional

gamma contribution and 11% by electron capture (p + e- → n) to stable 68Zn with half-life of 68 min. The long

half-life of the parent nuclide 68Ge (270.8 d) allows a three-year life-span of a generator. The half-life of 68Ga

(T1/2 = 68 min) permits production and application of radiopharmaceuticals.

8

68Ga3+ cation forms stable complexes with many ligands containing oxygen and nitrogen as donor atoms.9,12

Thus, 68Ga is suitable for complexation with chelators, naked or conjugated with peptides or other

macromolecules.

1.4 Radionuclide production; Generator system

Most positron-emitting nuclides are produced in-house in expensive accelerators requiring qualified personnel.

A convenient and less expensive means for radionuclide production is a generator system. Generators consist of

a long-lived parent radionuclide, which decays to a shorter-lived daughter radionuclide.14 The daughter

radionuclide is radiochemically separated from the decaying parent such that the former is obtained in a pure

radionuclide and radiochemical form. Key features of radionuclide generators include low cost, the convenience

of obtaining the desired daughter radionuclide on demand, and availability of the daughter radionuclide in high

specific activity, no-carrier added form. Thus, 80% of SPECT medical examinations using metal radionuclide is

performed with 99mTc due to the fact that it is readily available from the 99Mo/99mTc generator even though the

latter has only one week shelf-life.15 68Ga is a generator obtained radionuclide with strong potential for PET studies. The life span of 68Ge/68Ga

generator is as long as 2-3 years. The major hinder for wider use of 68Ga is its chemical form upon elution from

the generator, low concentration and presence of other metal ions and parent 68Ge in the generator eluate. The

elution of the 68Ga using EDTA-solution14,16,17 or concentrated HCl18,19 requires additional steps for preparation

of 68Ga for radiopharmaceutical production. The metal contaminants that originate mainly from the generator

column material might not only be toxic, but also compete with gallium in complexation reaction. Column

materials proposed so far for 68Ge/68Ga generators are inorganic oxides like aluminium dioxide20,21, titanium

dioxide or tin dioxide22 or selective organic resins comprising, for example, phenolic hydroxyl groups or

pyrogallol.14,18,23-25

1.5 Time factor in radiolabelling chemistry

The radiolabelling reaction time has to be minimised due to the short half-life of the radionuclide. Usually, the

total time including the synthesis, purification and application of a tracer should not exceed three half-lives of

the radionuclide. The synthesis is considered complete when the maximum possible amount of radioactivity is

incorporated into the tracer molecule. This corresponds to the highest radiochemical yield which is a

compromise between the chemical yield and the radionuclide decay.26

1.6 Microwave activation

Microwave activation, providing reaction acceleration, is an attractive tool for radiolabelling chemistry of short-

lived radionuclides. The conventional heat comes from the oil bath and heats up the walls of the vessel first

causing a temperature gradient in the solution. Under microwave irradiation the sample is heated from inside,

9

uniformly at each point resulting in very fast heating. Microwave activation is especially useful for microscale

organic chemistry such as radiolabelling where the sample size is comparable to the penetration depth of the

microwave field.27,28

The technique has been used for labelling of different organic molecules with 131I, 11C, 15O, 18F and 13N.27

The concept of rapid heating as a technique to enhance the volatility of complex fragile species is based on a

kinetic analysis of the competitive decomposition and evaporation processes. The activation energy for

decomposition is lower than that for evaporation. Thus, the macromolecule survives rapid heating at higher

temperature, but decomposes at lower temperature.29

Microwave heating uses the property of some products (liquids and solids) to transform electromagnetic energy

into heat. Those products must have dipole moment or be ionic. Polar solvents are very good (e. g. DMSO,

water, DMF, acetonitrile, NMP, t-butanol), non-polar solvents cannot be used (e. g. hexane, carbon

tetrachloride, toluene), if non-polar solvents have to be used, DMSO or a soluble salt can be added to improve

heating.

1.7 Bifunctional chelator

Bifunctional complexing agents30 which form stable complexes with the radionuclides can be covalently linked

to a carrier molecule. Acyclic bifunctional chelators such as EDTA (ethylenediaminetetraacetic acid), DTPA

(diethylenetriaminepentaacetic acid) or DFO (desferrioxamine) (Figure 2) have been used for labelling of

macromolecules with 111In, 67Ga, 68Ga, 90Y for tumour imaging. Their complexes showed low stability in vivo,

which has been related to the tendency of such anionic complexes to undergo acid- or cation-promoted

dissociation in vivo.31

HOOC N

HOOC

N

COOH

COOH HOOC

N

HOOC N

COOH

N

COOH

COOH

NH2NNHN

NHNCH3

OH

O

O

OH

O

O

OH

O

EDTADTPA

DFO

Figure 2. Some open-chain metal-chelators used for radiopharmaceuticals.

10

The released radioisotope may then be bound by serum proteins, such as transferin or may build up in

radiosensitive organs such as bone/bone marrow or in gastrointestinal mucosa.31 The introduction of

bifunctional macrocyclic complexing agents has led to agents’ slower rates of decomposition.32,33 Ligands such

as NOTA (1,4,7-triazacyclononanetriacetic acid) and DOTA (1,4,7,10-tetraazacyclododecanetetraacetic acid)

have been used for the complexation with 111In and 67/68Ga. 31,33,34 (Figure 3) DOTA and NOTA chelators can be

attached to a carrier molecule via the carboxylate group or the functionalised skeleton carbon. The former

method allows a one step coupling reaction to amines using carbodiimide chemistry.35 DOTA conjugated to a

macromolecule through the one carboxylate pendant arm forms neutral complexes with trivalent metals. Neutral

complexes are more resistant to acid/cation-promoted dissociation over a wide pH range.

O

OH

N N

NN

OH

O

O

OOH

OHO

N

NN

O

O

OH

OH

OH

DOTA NOTA

Figure 3. Examples of macrocyclic chelators used for radiopharmaceuticals.

1.8 Complexation reaction

High stability of tetraazamacrocycles is due to extremely slow dissociation reactions. Typical dissociation

constant values are 105 to 107 times lower than those of open-chain analogues. The hole-size effects influence

both thermodynamics and kinetics of macrocycle complexes. This is because a ligand in its minimum energy

metal-binding conformation will be optimized for a particular size of metal ion and that when other metal ions

are bound, the ligand conformational energy will rise with a resultant decrease in the stability of the complex.

Thus macrocycles are selective for metal ions. For a given ionic size the stability of the complex increases with

increasing charge. Another factor which influences the complexation rate and the stability is the presence of

pendant arms, which help to achieve the full coordination number of the metal ion.36-38 In addition, a pendant

arm can be used for the conjugation to a macromolecule.39

1.9 Stability of the chelate compounds

A bifunctional chelator should meet the following criteria: i) A chelator conjugated to a macromolecule should

bind the radioisotope rapidly and quantitatively; ii) The formed complex should be kinetically stable to cation

release over a pH range of 2-8 and in the presence of the cations like Ca2+, Zn2+ and Mg2+ found in serum;31 iii)

11

Other factors to consider include redox properties, stereochemistry, charge and lipophilicity.40 Neutral

complexes are less sensitive to acid/cation-promoted dissociation compared with anionic complexes. For

example, neutral Ga(III) complex of NOTA has been shown to undergo acid-catalysed dissociation only at low

pH which is unlikely under physiological condition. The Ga(NOTA) complex has such high thermodynamic and

kinetic stability that it remains intact in nitric acid over a period of 6 months.31 The log stability constant (logK)

of the Ga(III)-NOTA was determined as 30.98.41 While the log stability constants of the Ga(III) complex of

DOTA was determined as 21.33.42 This might be expected since the overall charge for Ga(DOTA) is negative

with four carboxylate pendant arms and Ga(III) charge of three. The X-ray crystallography study of NOTA

complex with Ga(III) revealed that the coordination of Ga(III) was octahedral.31 The cavity of NOTA is defined

by the facial triaza plane and an opposite facial plane consisting of three carboxylate oxygen atoms (Figure 4).

The compactness of the triazanonane ring and the steric efficiency of the pendant acetate groups lead to the

formation of complexes of unusually high stability and selectivity for the Ga(III).41

Figure 4. Molecular structure of the gallium complex with NOTA.

DOTA has a larger cavity than NOTA, which results in lower stability of the Ga(III) complex. In the Ga(III)

complex of DOTA-D-Phe-NH2, DOTA adopts a cis-pseudooctahedral geometry with a folded macrocyclic

unit.34,43 Gallium radiopharmaceuticals must be stable enough to avoid trans-chelation of Ga3+ to various iron

binding proteins, particularly transferrin. Transferrin has two binding sites, for which the gallium binding

constants are 20.3 and 19.3.13 In practice, this requirement necessitates the coordination of Ga3+ by polydentate

ligand, typically forming gallium species that are six-coordinate.

1.10 Macromolecular tracers

Peptides: Peptides are important regulators of growth and cellular functions in normal tissue and tumours.

Radiolabelled regulatory peptide analogues are used for in vivo localization of tumors.15,34,44-46 (Figure 5) If

labelled with a positron-emitting radionuclide, they can be used to quantify the radiation dose to a tumour and

critical organs thus allowing dose planning and dose monitoring for successful radiotherapy.

Linker

Chelator

Receptor

Vector

Radionuclide Vector: peptide, antibody Linker: any covalent bond Chelator: open-chain or macrocyclic lRadionu +

igand clide: β or γ emitters

Figure 5. Schematic representation of a vector molecule labeled with a covalentely attached radionuclide.

Somatostatin (Figure 6) analogues such as octreotide, [Tyr3]octreotide and [Tyr3]octreotate (Figure 7), are the

most frequently used of all peptides.47 Design of metal-based diagnostic agents requires knowledge of metal-

ligand complex structure, ligand selectivity, coordination number, lipophilicity, and thermodynamic stability.

Ala Gly Cys Lys Asn Phe Phe Trp

LysThrPheThrSerCys Figure 6. Structural formula of somatostatin, a cyclic disulphide-containing

peptide hormone of 14 amino acids.

Octreotide, an eight amino acid cyclic peptide, has been conjugated with bifunctional chelators such as DTPA,

DFO48,49 and DOTA50 (Figure 8, also see for the chelates Figure 2). The DOTATOC can be labelled with a

variety of metals in the +3 oxidation state such as Ga, In, and Y, and has shown high stability in human serum

and has a high affinity to somatostatin receptors. Studies have shown that DOTATOC labelled with gallium is a

promissing somatostatin analogue developed to-date due to its high tumour to blood ratio.50 DOTATOC has

been labeled with 66Ga, 67Ga and 68Ga.50-52 Replacement of Phe by Tyr (Figure 7) has been shown to increase

affinity of DOTA-Octreotide radioconjugates to SST2-positive tumours both in vitro and in vivo.53

(A) D-Phe-Cys-Phe-D-Trp-Lys-Thr-Cys-Thr(OH)

(B) D-Phe-Cys-Tyr-D-Trp-Lys-Thr-Cys-Thr(OH)

(C) D-Phe-Cys-Tyr-D-Trp-Lys-Thr-Cys-Thr

Figure 7. Structural formulae of

various eight amino acid-

containing analogues: (A)

octreotide, (B) [Tyr3]octreotide,

(C) [Tyr3]octreotate.

12

13

O

OH

N N

NN

OH

O

OOH

O

NN

NHO

HOO

O

O

S

NHN

NN

O

O

ON

OS

NH2

HO

NH

Figure 8. Structural formula of DOTATOC.

Oligonucleotides: An antisense oligonucleotide (Figure 9) is a short, synthetic nucleic acid that manifests the

inhibition effect on the gene expression by selectively hybridizing with its complementary “sense” sequences in

mRNA through Watson-Crick base-pairing54 (Figure 10).

DNA

3´-end

5´-end

Guanine

Cytosine

Adenine

Thymine

O

OPOO

O

N

N

NH

N

O

NH2

ON

N

O

OPOO-

O

NH2

ON

NH

O

O

H3C

OPOO-

O

OPOO-

O

O

Figure 9. An oligonucleotide is a linear polymer built up of monomeric units, the nucleotides. A nucleotide consists of three molecular fragments: sugar, heterocycle, and phosphate that can be chemically modified. The oligonucleotide biological function like replication of DNA, messenger RNA synthesis and protein synthesis is influenced by its three-dimensional structure and its ability to form hydrogen bonding.

OPOO-

O

N

N

N

N

NH2

14

A B

Figure 10. (A)The diagram depicts the flow of genetic information from DNA to the protein. The information contained in genes (DNA) is eventually expressed as the phenotype (protein). (B) Antisense oligonucleotide hybridize to the complementary target mRNA and causes a block of protein translation. Antisense oligonucleotides are of considerable interest for biological studies and particularly molecule-targeted

therapies of cancer. Inhibitory activity of antisense oligonucleotides with phosphodiester, phosphorothioate and

methylphosphonate linkages against viral replication or expression of cellular genes in a number of different

tissue culture systems has been reported.54 Radionuclide labelled antisense oligonucleotides may be used for in

vivo imaging of gene expression. A number of radiolabelling studies has been conducted with gamma emitters,

such as 90Y, 99mTc, 111In and 125I, for in vitro studies and imaging with gamma camera and SPECT.55-60 Methods

for the labelling of oligonucleotides with positron emitting radionuclides, such as 11C, 18F and 76Br, have also

been presented.61-65

A prime requirement of a modified oligonucleotide is that it should be chemically stable and maintain its

hybridisation properties. Phosphodiester oligonucleotides are readily degraded by nuclease enzymes and have a

half-life of less than 30 minutes in the cell.54 A way of achieving metabolically more stable oligonucleotides is

to modify the phosphate backbone. Thus, phosphorothioate antisense oligonucleotides, in which one of the

nonbonding oxygen atoms in each internucleoside phosphate linkage is replaced by a sulfur atom, have the

advantage of greater resistance to degradation by deoxyribonuclease than phosphodiester oligonucleotides.66

The half-life of a phosphorotioate oligonucleotide in serum is 24 hours.67 Another modification is the

introduction of an O-methyl group into the sugar moiety at the 2′-position.68-70

Methylphosphonate, phosphoramidate, morpholino oligonucleotide, mixed-backbone oligonucleotide and

peptide nucleic acid (PNA) have also been studied.71 End-modifications might also increase the stability of the

oligonucleotide from degradation by exonucleases.72 The latter modification also allows introduction of an

amine group enabling conjugation of the oligonucleotide to a chelator73 and further labelling with a metal

radionuclide.

15

1. 11 Specific radioactivity

The concentration of a radioactive material in a sample is described by specific radioactivity and is expressed as

radioactivity per substance amount (Bq/mol). The successful use of a tracer macromolecule requires high

specific activity. The requirement is dictated by biological factors such as limited amount of receptors and

pharmacological side-effects. In particular, due to competition with the labeled peptide for the same receptor,

the presence of unlabelled peptide may have a negative effect on the percent dose uptake of radioactivity.

2. Aims of the study

The aim with the thesis was 68Ga-labelling of macromolecules such as peptides and oligonucleotides. In

particular, the following tasks were addressed:

• To evaluate performance of a commercially available 68Ge/68Ga generator in order to determine the

potential for long term efficient utilization for 68Ga radiopharmaceutical preparation.

• To develop a method for preconcentration and purification of 68Ge/68Ga generator eluate.

• To develop a method for the conjugation of a peptide/oligonucleotide to a bifunctional chelator.

• To develop a fast method for the 68Ga-labelling of a peptide/oligonucleotide with high specific activity.

• To perform chemical characterisation and preliminary biological examination of the bioconjugates and

corresponding 68Ga-labelled counterparts.

• To apply 68Ga-labelled peptides/oligonucleotides for studies of a tumour model in rat, analytical

chemistry and patient diagnostic examinations.

3. Results and discussion

3.1 Method of obtaining 68Ga74

3.1.1 68Ge/68Ga generator characterisation (paper I)

Three units of a commercial generator (Cyclotron Co., Ltd, Obninsk, Russia) were evaluated over a period of up

to 2.5 years. Radionuclide and chemical purities were measured and analyzed on the base of Inductively

Coupled Plasma Atomic Emission Spectroscopy (ICP-AES) analysis, well-counter radioactivity measurements

and energy spectrum analysis. The elution efficiency was defined as the proportion of 68Ga separated during the

elution process, and expressed as percentage of the 68Ge loaded activity at the moment of the measurement. In

the first month the elution efficiency was 78±5% of the theoretical amount and then it decreased slowly and

continuously down to 41±3% after 29 month use. The generator elution efficiency over time was found

16

reproducible (RSD<10%) (Figure 11A).

0

20

40

60

80

100

0 200 400 600 800 1000

Time, [day]

68G

a yi

eld,

[%]

Generator-1 Generator-2 Generator-3

0

200

400

600

800

1000

1 2 3 4 5 6 7Fraction

Elut

ed a

ctiv

ity p

er

frac

tion

68Ga activity given in [MBq]

68Ge breakthrough given in [Bq]

volume 6 mL

BA

Figure 11. (A) 68Ga elution efiiciency for generator-1 during 29 months, generator-2 during 14 months and generator-3 during 3 months . (B) Elution profile of the 68Ge/68Ga generator, one fraction was 1 ml, (fraction 1 = 0.3 ml, fraction 7 = 0.7 ml) giving a total eluted volume of 6 mL. The profile for the 68Ga and the 68Ge breakthrough are identical, the 68Ge breakthrough is ~10-4 %. Fraction 3 (1 mL) contains 60 % of the available 68Ga activity. The 68Ga elution profile and the 68Ge breakthrough are presented in Figure 11B. By fractionating the 68Ga eluate,

approximately 65% of the available activity could be obtained with only 1 mL (the third fraction, Figure 11B).

The 68Ge breakthrough with respect to the eluted 68Ga activity was found to be 0.001 - 0.007% and did not

change during the investigation period. It has been previously shown that 68Ge breakthrough losses of 0.001%

per elution are insignificant compared with 68Ge decay losses, assuming two elutions per day.75

Since the major hinder in complexation chemistry of 68Ga is the presence of competing metal ions in the eluate,

special attention was paid to metal ion analysis and optimization of the elution of 68Ga. The metal ion content of

the 68Ge/68Ga generator eluate depending upon elution frequency is shown in Figure 12

0

1

10

100

1000

0 20 40 60 80 100

Elution period, [h]

Con

cent

ratio

n, [

µg/

L]

Ti

Pb

ZnAl

Pt

BaNi

Cu

Ga

Figure 12. Metal ion content in 6 mL of the generator eluate dependent on the elution time period.

17

The concentration of the metal ions, except for Pb, was constant for a given elution frequency. The generator

requires a daily elution in order to keep the concentration of the contaminant metal ions as low as possible. A

preventive elution 3-4 hours prior to the synthesis is recommended since the interfering metal ion concentration

can be kept at its minimum value. The non-toxic generator matrix material (TiO2) may cause relatively high Ti

concentration (close to 1000 ppm) in the eluate if the generator is not eluted regularly.

Calculations (based on ICP analysis data) of theoretical specific activity and specific activities considering

isotopic dilution, pseudo carriers and concentration of the macromolecule indicated, that the largest contribution

into the specific activity drop came from the pseudo carriers that require high macromolecule concentration. The

consequence is lower 68Ga incorporation (low radiochemical yield) or a drop of the specific radioactivity of the

final 68Ga labeled peptide conjugate.

3.1.2 Preconcentration and purification of the 68Ge/68Ga generator eluate (paper I)

A general disadvantage for all generators is the large 68Ga eluate volume leading consequently to a low 68Ga

concentration. Contamination with the long-lived parent nuclide 68Ge is critical and cationic metal ions

originating from the column material may disturb the labeling procedure due to competition in the chelating

reaction. A method based on anion exchange to purify the eluate and to reduce the volume and HCl

concentration of a generator eluate has been proposed previously.18 Nevertheless, the final volume obtained was

still large (~4 mL). The evaporation to dryness causes activity loss due to the time factor as well as uncontrolled

release of GaCl3 which is relatively volatile. The anion-exchange approach may be a promising way to solve the 68Ge/68Ga generator problems regarding concentration and purification.

The adsorption behaviour of the metal ions from HCl-solutions on the anion exchanger DOWEX AG1-X8 is

well known.76,77 In HCl solution gallium forms strong anionic complexes with Cl¯, the corresponding

[GaCl6]3¯and [GaCl4]¯complexes are strongly adsorbed at the mentioned anion exchange resin from HCl

concentrations > 3 M. While germanium is practically not adsorbed from < 5 M HCl solution.

A number of strong anion exchange resins ( AG 1, Bio-Rad, USA) with different mesh size and crosslinking

percent parameters, as well as commercially available cartridges with quaternary ammonium functional group

were investigated. The resin type, cartridge size, sample pretreatment, column conditioning, sample addition,

column wash, column drying and analyte elution procedures were optimized, showing high reproducibility.

Inductively Coupled Plasma Atomic Emission Spectroscopy (ICP-AES), well-counter radioactivity

measurements and energy spectrum analysis were used as analytical tools in the study.

The pre-concentration protocol utilized the [68GaCl4]- complex formation in 4 M HCl medium. This anionic

complex can be efficiently adsorbed at anion exchange resins from HCl solutions and 68Ga can be efficiently

eluted with a very small volume of water. The results of the optimisation studies concerning the capabilities of

different anion exchange resins are presented in Table 2. The highest 68Ga recovery was obtained using a

column diameter of 4 mm amongst the column diameters of 1.3, 2.3, 4 and 5 mm. The linear flow rate of the

deionised water was 25 cm/min. Among the class of AG1-resins the Bio-Rad AG 1-X8 with fine mesh (200-400

18

mesh) showed the highest retention, recovery and a sharp elution profile.

Table 2. Charactererization of the anion exchange resins and SPE cartridges used in the 68Ga purification and preconcentration study. No Sorbent Counter ion Retention, % Recovery, %

1 AG 1-X8 (100-200 mesh) HO- 81±4 69±3

2 AG 1-X8 (200-400 mesh) HO- 98±2 85±5

3 AG 1-X4 (100-200 mesh) HO- 65±5 80±3

4 AG 1-X2 (200-400 mesh) HO- 45±4 68±2

5 SAX SPEC, Isolute, 50mg Cl- < 4 -

6 SAX SPEC, NTK kemi, 15 mg Cl- < 4 -

7 SAX SPEC, Isolute, 50mg OH- 9 -

8 SAX SPEC, NTK kemi, 15 mg OH- 7 -

9 SAX SPEC, Chromavix, 45 mg HCO3- 99±1 93±2

As seen in Table 2, amongst the tested commercial anionic cartridges only the SEX SPEC Chromavix

cartridge showed high retention and recovery of >90%. This cartridge contains 45 mg of a strong basic

anion exchange resin based on polystyrene-divinylbenzene comprising HCO3- as counter-ions and quaternary

amine functional groups. The adsorption of 68Ga from HCl solution at the anion exchange cartridges

increased rapidly from 0.8% at 0.1M HCl to almost 100 % at 3.8 M HCl under our conditions (4

ml/min, 4 mm column diameter) in agreement with literature data on distribution coefficients76 (D >

105) (Figure 13). Steeply rising portion of the distribution function is due to the formation of the

negatively charged complexes. The maxima would occur at HCl concentration where the negatively

charged complexes become predominant, and the region of decreasing D would correspond to the high

M HCl where the negatively charged complexes are essentially completely formed.

1.E+01

1.E+02

1.E+03

1.E+04

1.E+05

1.E+06

1.E+07

1 2 3 4 5 6

HCL in Mol/L

Dis

trib

utio

n co

efic

ient

D

Figure 13. Distribution coefficient D for the adsorption of 68Ga on the anion-exchange resin of the commercial Cartridge:SEX SPEC, Chromavix, that contains 45 mg of resin. Note: under the practical conditions a value of D = 105 provides a 68Ga retention of ~ 99 % . The distribution coefficients D of 68Ga between the stationary and mobile phase were determined by the column method as weight distribution coefficients, D.

res

liq

liq

res

mm

AAD ∗= where, A res = Ga activity in [Bq] of

the dry resin, A

68

liq = Ga activity of the solution passed through the column in [Bq], m

68

res = amount of dry resin in [g], m liq = amount of the solution in [g].

19

Since the parent nuclide 68Ge is not retained at the resin, the concentrating step is also purification of the 68Ga

from 68Ge breakthrough. The eluate was purified from Al(III) and In(III) as well since the adsorbability of these

ions decreased rapidly with increasing M HCl and became negligible above 3 and 1 M, respectively, in

agreement with literature data78. In the same way the original generator eluate was purified from Ti by 90% as

shown by the ICP-AES analysis data. The elution with deionised water resulted finally in 90±2% of 68Ga

recovery in a total volume of 200 µL only. (Figure 14)

0

10

20

30

40

50

60

70

1 2 3 4 5 6 7Fraction number (one fraction 50 µL)

Elut

ed 68

Ga

activ

ity in

[%]

Figure 14. Elution profile of 68Ga eluted from the SAX SPE Chromavix cardridge with dionized water.

AG 1-X8 (200-400 mesh) resin and Chromafix SAX SPEC cartridges showed comparable results with respect to

retention, recovery and elution profile. However, Chromafix SAX SPEC cartridges are, as a standardized

commercial product, much more suited for further standardization and automation. In total the preconcentration

procedure required 4 to maximum 6 min, that includes the filtration of the original 68Ga eluate in 4 M HCl

through the cartridge, the drying procedure and the elution of the 68Ga in 200 µl pure water.

3.2 68Ga-labelling of macromolecules

3.2.1 Conjugation of DOTA with macromolecules and 68Ga-labelling of the conjugates (papers II, III)

Carboxylic functionality of DOTA chelate (5) can be used to form amide bond with amine functionality of a

macromolecule. Peptides have amine group at the N-terminal and at Lysine amino acid residues. The

oligonucleotides were modified at 5′- or 3′-end with aminohexyl linker in order to introduce amine functionality.

In the thesis are presented four modified antisense oligonucleotides (1 - 4) specific for activated human K-ras

oncogene54: 17-mer phosphodiester oligonucleotide with hexylaminolinker at 5′ end (1); 17-mer phosphodiester

oligonucleotide with hexylaminolinker at 3′ end (2); 17-mer phosphorothioate oligonucleotide with

hexylaminolinker at 5′ (3); and 2′-O-methyl phosphodiester with hexylaminolinker at 5′ (4) (Figure 15).

20

1 X= O-, X ' = H

3 X= S-, X' = H

4 X= O-, X ' = O - CH3

O

O

O Base

PX O

O

P

O

X

X'

OH 2N

P

O

O -O

O

O

O Base

P-O O

OH 2N

OH

X'

2 X= O-, X ' = H

Figure 15. 1 - phosphodiester oligonucleotide with 5´- Hexylamine modification; 2 - phosphodiester oligonucleotide with 3´ - Amino-C7 modification; 3 - phosphorothioate oligonucleotide with 5´- Hexylamine modification; 4 - 2´-O-methyl phosphodiester oligonucleotide with 5´- Hexylamine modification. The antisense and sense sequences used were, respectively, 5′-CTA CGC CAC TAG CTC CA and 5′-TGG AGC TAG TGG CGT AG.

The method originally developed for conjugation of monoclonal antibodies and proteins with N-hydroxy-

sulfosuccinimidyl DOTA35 was adopted and modified for the present studies. The macromolecules (peptides and

oligonucleotides, 6) bearing amine functionality were reacted with the N-hydroxy-sulfosuccinimide ester of

DOTA (7), generated in situ using the water-soluble EDC (1-ethyl-3(3-dimethylaminopropyl)carbodiimide, 8)

as coupling reagent to give DOTA-macromolecule (9) (Scheme 1). The advantage of adding sulfo-NHS (N-

hydroxy-sulfosuccinimide, 10) to EDC (8) reaction is to increase the stability of the active intermediate, which

ultimately reacts with the attacking amine.73 EDC reacts with carboxylate group to form an active ester (O-

acylisourea (11)) leaving group. This reactive complex is subject to rapid hydrolysis in aqueous solutions,

having a rate constant of seconds. This is especially a problem when the target molecule is in low concentration

compared to water, as in the case of macromolecules. Forming a sulfo-NHS ester intermediate from the reaction

of the hydroxyl group on sulfo-NHS with the EDC active-ester complex extends the half-life of the activated

carboxylate to hours. Since the concentration of added sulfo-NHS is usually much greater than the concentration

of target molecule, the reaction preferentially proceeds through the longer-lived intermediate. However, the final

product of this two-step reaction is identical to that obtained using EDC alone: the activated carboxylate reacts

with an amine to give a stable amide linkage.

21

R1 O

OH

CH3 N

N NHN

+CH3

CH3H

Cl-

CH3 NH N N+

CH3CH3

O

R1

OH

Cl-

R1 NHR

ON+ R

H

H

H

NOH

O

O

SO

O

ONa

5

6

7

8

9

10

11

O

OH

N N

NN

OOH

O OH

R1 =

NO

O

O

SO

O

ONa

R1

O

Scheme 1. Reaction scheme of the amide bond formation via sulfo-NHS ester intermediate. R stands for an oligonucleotide modified with hexylamine linker or peptide.

DOTA was conjugated to the peptides with various constitution and length ranging from 8 to 28 amino acid

residues and oligonucleotides with modifications in backbone, sugar moiety and both 3' and 5' ends (Table 3).

The reaction yields were 70-90%.

Table 3. Peptides and oligonucleotides conjugated to DOTA bifunctional chelator.

Number Peptide Residue NH2 group pI MW,

Da Angiotensin II 8 1 6.74 1046.20 Bombesin 14 1 6.85 1619.00 Pancreastatin Fragment 37-52, human 16 1 9.75 1819.00 Neuropeptide Y Fragment 18-36 19 1 10.42 2456.80 Secretin human 27 1 9.45 3039.00 VIP 28 4 9.82 3325.80 Oligonucleotide Phosphodiester (3'- modified) 17 1 - 5249.80 Phosphodiester (5'- modified) 17 1 - 5249.80 Phosphorothioate 17 1 - 5522.13 2′-O –methyl Phosphodiester 17 1 - 5724.00

The conjugates were labeled with 68Ga obtained in 0.1M hydrochloric solution and buffered with sodium acetate

or HEPES to a pH of 4.0-5.0. After addition of the macromolecule to the mixture the reaction was carried out in

a microwave oven (1 min, Scheme 2)79.

22

68Ga3+

Buffer, pH ~ 4.0Microwave activation

O

O -

N N

NN

O -

O

ONH

OO -

R

68Ga3+ in HCl

O

O -

N N

NN

O -

O

ONH

OO -

R

Scheme 2. Reaction scheme for the microwave activated complexation of 68Ga with a macromolecule, where R is a peptide or oligonucleotide – (CH2)6 –. Buffer: NaOAc/HEPES.

Labelling of four oligonucleotide counterparts gave isolated decay corrected radiochemical yields in the range of

30 to 52% (paper II). Radiochemical yields for peptide conjugates were higher than 70%. 68Ga-labelled

Angiotensin II and VIP peptides were used for the characterization of microstructures integrated in miniaturized

systems for chemical analysis, in particular, for the evaluation of peptide adsorption to plastic surface.80,81

3.2.2 68Ga-labelling using non-treated 68Ge/68Ga generator eluate (paper I)

DOTATOC (DOTA-D-Phe1-Tyr3–Octreotide), was used as a test molecule for comparing the labeling

properties of the different 68Ga preparations. Two buffering systems, NaOAc (sodium acetate) and HEPES (N-2-

Hydroxyethylpiperazine-N´-2-ethanesulfonic acid), were used in the study for the optimisation of the labelling

conditions. The complexation reaction of 68Ga with macrocyclic chelator conjugated to the peptide starts already

at room temperature, but the radiochemical yield (RCY) did not exceed 24% and 30% for sodium acetate and

HEPES buffers, respectively (Figure 16A). Conventional heating improved the yield. In the case of the HEPES

buffer, the 68Ga incorporation time was shorter. Microwave activation of the reaction mixture considerably

shortened the reaction time and improved the reproducibility. Nevertheless, a final purification of the labelled

peptide conjugate was still required. An increase of 68Ga concentration was achieved by fractionating of the

generator eluate (Figure 11B). The third fraction of the eluate (1 mL) contained > 60 % of the total available 68Ga activity. The RCY of the reaction at room temperature, conventional heating and microwave activation was

improved when using this fraction (Figure 16B). Under conventional heating condition, quantitative 68Ga

incorporation was achieved within 5 min for 5 nanomols of DOTATOC using the HEPES buffer system. For the

sodium acetate buffer system a factor 3 larger amount of peptide (15 nanomols) and 20 min heating were

required.

23

0

20

40

60

80

100

0 5 10 15 20

Time, [min]

RC

Y, [%

]

B

0

20

40

60

80

0 5 10 15 20Time, [min]

RC

Y, [%

]

A

Figure 16. Time course of 68Ga complexation reaction conducted using the full original 68Ga eluate (6mL) (A) and using only the 1 mL peak fraction of the generator eluate (B) at room temperature (dashed line), conventional heating in a heating block at 95 oC (solid line) and with microwave-activation for 1 min at 90±5 oC (circled) for two different buffer systems: sodium acetate buffer, pH = 4.6, 20 nanomols of DOTATOC ∆ HEPES buffer, pH = 4.2, 20 nanomols of DOTATOC (A) and 5 nanomols of DOTATOC (B)

With microwave-heating the labeling reaction was complete within 1 min, and the incorporation of the activity

was quantitative and the reproducibility of the reaction was improved. No further purification of the labelled

peptide conjugate was required. The amount of peptide needed for a quantitative 68Ga incorporation was 15

nanomols at least when using sodium acetate buffer and 5 nanomols when using HEPES (Figure 17). It is worth

mentioning that the influence of the buffer is more pronounced at lower peptide quantities indicating the

sensitivity of the reaction to potential metal impurities present in the buffers as well as complexing ability of the

sodium acetate itself competing with DOTA when used at high concentration.

0

20

40

60

80

100

120

0 5 10 15 20

DOTATOC, [nanomol]

RC

Y, [%

]

Figure 17. Influence of the buffering system ( sodium acetate, ∆ HEPES) on the radiochemical 68Ga complexation yield for different DOTATOC quantities under 1 min microwave-activation at 90±5 °C. Reaction was conducted using the 1 mL peak fraction of the original generator eluate (see Figure 11B.).

68Ga-DOTATOC of the described preparations was used in 16 patient examinations (see for midgut carcinoid

tumor image the thesis title page picture) and the results were compared with those obtained using 111In-DTPA-

D-Phe1-octreotide. The latter is extensively used in the neuroendocrine tumor diagnosis using planar gamma

camera imaging and SPECT.

The better temporal and spatial resolution of PET allowed detection of the smaller tumours that could not be

24

monitored by SPECT. Moreover, in the case of 68Ga-DOTATOC the whole body dose was < 1 mSv (14-100

MBq injected), in comparison to the examination with 111In-DTPA-D-Phe1-octreotide when the whole body dose

was 20 mSv (175 MBq injected). Another factor to take into consideration is the half-life of 2.3 days for 111In

with consequent unnecessary irradiation of the patient. The comparative study of the imaging of the

somatostatin receptor using 111In-octreotide (SPECT detection) and 68Ga-DOTATOC (PET detection) has also

been performed by Hoffman et al. with the following results.51 68Ga-DOTATOC PET identified 100%, whereas 111In-octreotide planar and SPECT imaging identified only 85% of the total 40 lesions predefined by CT or

MRI51. 68Ga-DOTATOC resulted in high tumour to non-tumour contrast and low kidney accumulation and

yields higher detection rates as compared with 111In-octreotide scintigraphy.

3.2.3 68Ga-labelling using preconcentrated and purified generator eluate (paper I)

Although the 68Ga incorporation was quantitative using only 5 nanomols of DOTATOC, there were still

drawbacks in the approach when using a 1 mL peak fraction of the generator eluate. Firstly, about 40% of the 68Ga activity was wasted and secondly, even smaller peptide amounts might be requested due to cost and

biological properties. Thus a fast technique for the 68Ge/68Ga generator eluate purification and concentration was

introduced (Section 3.1.2).

The influence of the concentration and purification step of the 68Ga eluate in combination with microwave

activation on the labeling reaction is demonstrated in Figure 18 and Table 4. Figure 18 illustrates that the full 68Ga radioactivity eluted from the generator could quantitatively be incorporated into less than 1 nanomol of a

peptide conjugate. Labeling tests with equimolar quantities (68Ga : bioconjugate = 1 : 1) using 6 pmols

DOTATOC resulted in a maximum activity incorporation of 11 % and the specific activity of the labelled

bioconjugate only by factor of 10 less than theoretical one. In this combination the original full 68Ga eluate was

purified and concentrated as described in the section 3.1.2. and the labeling was performed during 1 min

microwave activation.

RCY vs [DOTATOC]

0

20

40

60

80

100

120

0,001 0,01 0,1 1 10

DOTATOC, [nanomol]

RC

Y, [%

]

Figure 18. Influence of the DOTATOC amount on the radiochemical yield of the 68Ga complexation reaction using the full available 68Ga activity in 200 µl volume after the preconcentration and purification step in HEPES buffer system. solid line : 1 min microwave-activation at 90±5 °C dashed line: 5 min conventional heating at 95 °C

25

Table 4 presents details concerning age of the generator, variation of the amount of peptide and variation of the

ligand itself. Independent on age of the generator, overviewing a period of 2.5 years, stable and reproducible 68Ga incorporation of > 99 % into 0.3 – 1 nmols of the peptide conjugate was obtained. It is worth to note that

even with the lower 68Ga amount and concentration with the time course (2.5 years) it is still possible to get a

quantitative incorporation with one nanomol of the bioconjugate.

Table 4. Labeling conditions and practical quality parameters of the 68Ga-labeled peptide conjugates obtained from preconcentrated and purified 68Ga preparations (200-220 µL) under microwave-activation. Due to quantitative radionuclide incorporation no further purification of the labelled product is needed. Generator age,

month Peptide conjugate Peptide amount, nanomol

Activity incorporation

[ % ]

Starting full 68Ga activity from

generator, [MBq ]

Specific activity after synthesis [MBq/nmol ]

1 DOTATOC 1 99.0 1289.0 997.7

1 DOTATOC 0.5 99.9 1286.7 2003.0

1 DOTATOC 0.3 98.8 1251.3 3307.0

14 DOTATOC 1 99.9 357.0 308.7

14 DOTATOC 0.5 99.9 337.2 537.2

14 DOTATOC 0.3 99.9 329.4 843.7

29 DOTATOC 1 99.9 78.8 61.0

1 NODAGATATE 1 99.9 1275.9 1011.1

1 NODAGATATE 0.5 99.9 1110.9 1758.4

14 NODAGATATE 1 98.7 357.4 281.3

14 NODAGATATE 0.5 96.8 323.2 493.8

29 NODAGATATE 2 98.0 77,6 30.5

1 DOTA-RGD 1 99.9 1253.0 976.0

1 DOTA-RGD 0.5 99.1 1213.9 1869.9

1 DOTA-RGD 0.25 96.5 1256.7 3756.4

14 DOTA-RGD 0.5 99.9 362.4 570.4

14 DOTA-RGD 0.25 96.2 369.8 953.0

29 DOTA-RGD 1 98.4 72.2 56.1

3.3 Microwave activation (papers I-III)

In the study, the macromolecules, their conjugates and 69,71Ga-labelled counterparts were exposed to microwaves

and then analysed by LC-ESI-MS to confirm their stability. Compared to synthesis with conventional heating

microwave application shortened the synthesis time considerably and improved the radiochemical yield. Due to

the shortened reaction time the amount of radioactive material and the product specific activity was increased by

21%. But microwave activation not only reduced chemical reaction time, but also reduced side reactions,

increased yield, and improved reproducibility (Figure 19).

0.6 A 68Ga-DOTATOC

0.3

0.0

0.4

B 68Ga-DOTATOC Impurities

0.2

0.0

0 5 10

Minutes

3.4 Specific radioactivity

In the present work the specific radioactivity was increased by a fac

applied technique using the non-treated generator eluate and to lite

specific radioactivity of 68Ga with respect to macromolecule mass

equimolar quantities (68Ga : bioconjugate = 1 : 1) using 6 pmols DOT

incorporation of 11 % and the specific activity of the labelled bioco

theoretical one.

3.5 Purification and characterization of the macromolecular

68Ga-labelled counterparts

3.5.1 Purification (papers I-III)

Purification of radiolabeled compounds is usually performed using s

(LC) or solid phase extraction (SPE). The purification step causes loss

as well as losses on the purification systems. To speed up this step solid

radioactivity losses on the cartridge is 20-50% since the strong orga

recovery. Another complication is the eluent used for the product el

Figure 19. HPLC-radiochromatogram of a 68Ga-DOTATOC: (A) 5 nanomols of DOTATOC, 1 min microwave activation at 90±5 oC; (B) 5 nanomols of DOTATOC, 5 minconventional heating at 95 oC.

26

tor of 100-150 compared to previously

rature data50. The maximum theoretical

is 100 GBq/nmol. Labeling tests with

ATOC resulted in a maximum activity

njugate only by factor of 10 less than

conjugates and their

emi-preparative liquid chromatography

es of radioactivity due to the time factor

SPE cartridges are usually used, but the

nic solvents cannot be used for better

ution since most of the eluents are not

compatible with the biological systems under investigation. Thus either evaporation of the product solvent or

change of the solvent using SPE cartridges is needed. This additional step decreases the radiochemical yield

further. In the present study, due to the developed method for quantitative radioactivity incorporation, the

omission of the purification step was possible (Figure 20). Moreover, the labelling product was obtained in

HEPES buffer which is compatible with biological systems.

-0 .1

0

0 .1

0 .2

0 .3

0 .4

0 .5

0 .6

0 2 4 6 8 1 0

re te n tio n tim e in [m in ]

radi

atio

n si

gnal

6 8G a D O T A T O C

-0 .1

0

0 .1

0 .2

0 .3

0 .4

0 .5

0 .6

0 2 4 6 8 1 0

re te n tio n tim e in [m in ]

radi

atio

n si

gnal

-0 .1

0

0 .1

0 .2

0 .3

0 .4

0 .5

0 .6

0 2 4 6 8 1 0

re te n tio n tim e in [m in ]

radi

atio

n si

gnal

6 8G a D O T A T O C

3.5.2 Chemical characterization

Chemical characterization and analysis prior to application are necessa

amount of the radiolabelled compound. The analysis should be performed

of activity. For macromolecular bioconjugates, appropriate means of c

performance liquid chromatography (HPLC) and mass spectrometry (MS

the addition of authentic reference substance to the labeled tracer and coel

series with a radioactivity and a UV detector. This method is convenient

synthesis. The HPLC analysis developed in this study is accomplished

control (QC) of the radiopharmaceutical prior to patient injection (Figu

was synthesized under the same conditions as its radioactive counterpart

The identity of the compounds was confirmed by LC-ESI-MS. In additi

assessed by performing the labelling reaction with both conjugated an

product was detected in the reaction with unconjugated macromolecules,

to the chelator.

The stability of radiolabelled oligonucleotides in water and 50% ethano

analysis of aliquots taken from the labelling reaction mixture every 15 m

were detected in the stability study. The radiochemical purity of 68Ga-DO

ethanol was >95% for more than four hours. This time corresponds to 3-

time required for biological experiments.

Figure 20. HPLC-radiochromatogram of a 68Ga-DOTATOC preparation (0.5 nanomols of DOTATOC, 1 min microwave activation at 90±5 oC, total reaction volume 220 µL). The68Ga incorporation yield is 99.9% and the specific activity of the labeled peptide conjugate was 2 GBq/nmol.

27

ry to ensure the identity, purity and

within short time to minimize the loss

haracterization generally include high

). The most commonly used method is

ution on a HPLC column connected in

and can easily be performed for each

within 10 min allowing fast quality

re 20). Authentic reference substance

, but using the stable 69,71Ga isotopes.

on, the position of the 68Ga-label was

d unconjugated macromolecules. No

indicating that the label was attached

l was monitored by radio HPLC with

in. No additional radio-HPLC signals

TA-oligonucleotides in water and 50%

4 physical half-lives of 68Ga and is the

28

3.5.3 Preliminary biological examination of the 68Ga-labelled oligonucleotide conjugates (papers II, III) The impact of the modifications and labelling on the oligonucleotide probe performance was investigated conducting: i) specific hybridisation of antisense oligonucleotide to a complementary 17-mer phosphodiester sense oligonucleotide in solution; ii) activity organ distribution in rats; iii) whole body autoradiography of rats. i) Hybridization: The hybridization products were analysed using polyacrylamide gel electrophoresis in a cell-

free system and the results were visualised by ethidium bromide staining and autoradiography. The results of the

concentration dependent hybridisation of the phosphorothioate counterpart are shown in Figure 21.

Autoradiography

Hybrid

30 bp20 bp

10 bpSense

1 2 3 4 5 6 7 8 9 10 11 12

PAGE

HybridAntisense

1 2 3 4 5 6 7 8 9 10 11

1 2 3 4 5 6 7 8 9

AntisenseHybrid

AntisenseHybrid

A B

C

Figure 21. Concentration dependent hybridisation of 17-mer antisense phosphorothioate oligonucleotide (20 pmol in 1 µL) to the complementary 17-mer sense phosphodiester oligonucleotide in solution. The antisense:sense concentration ratios in the lanes are as follows: 1) 1:1/60, 2) 1:1/30, 3) 1:1/15, 4) 1:1/5, 5) 1:1/2, 6) 1:1, 7) 1:2, 8) 1:3, 9) 1:4 and the references are sense oligonucleotide (lane 10), antisense oligonucleotide (lane 11) and molecular weight marker (lane 12). A) PAGE picture of the concentration dependent hybridization study; B) Autoradiography of the polyacrylamide gel (A); C) The scaled up region of interest.

Nine samples with a constant concentration of the radiolabelled antisense oligonucleotide and a gradually

increasing concentration of the sense oligonucleotide were analysed (Figure 21, lanes 1-9). As a reference the

sense oligonucleotide (lane 10), the antisense oligonucleotide (lane 11) and molecular weight marker (lane 12)

samples were loaded onto the polyacrylamide gel. The gradual increase of the intensity of the hybrid bands, the

absence of the free radioactive bands of antisense at the higher concentrations of the sense oligonucleotide and

29

the absence of free sense bands at the lower concentration of the sense oligonucleotide serve as an indication of

the concentration dependent hybridisation. All four oligonucleotide counterparts were able to hybridise to the

complementary sense oligonucleotide.

ii) Organ distribution of 68Ga-labelled phosphodiester, phosphorothioate, and 2'-O-methyl phosphodiester

oligonucleotides in normal rats: Since the oligonucleotides studied do not have any biological target in rats,

their tissue distribution reflects their non-specific interactions and elimination. The measurement of the organ

radioactivity at 20, 60 and 120 min time points after i.v. administration of labelled phosphodiester showed the

highest values in the liver followed by the urinary bladder, bone marrow and spleen.

0,005,00

10,00

15,0020,0025,0030,00

35,0040,0045,00

PD PS

2'-O

-Me

PD PS

2'-O

-Me

PD PS

2'-O

-Me

PD PS

2'-O

-Me

PD PS

2'-O

-Me

Kidney Liver Urinarybladder

Bonemarrow

Spleen

SUV

20 min60 min120 min

Figure 22. The five organs of highest radioactivity uptake after injection (20, 60 and 120 min time points) of 68Ga-labelled phosphodiester (PD), phosphorothioate (PS) and 2'-O-methyl phosphodiester (2'-OMe) oligonucleotides.

Uptake in the kidney was predominant in the phosphorothioate and 2'-O-methyl phosphodiester distribution

patterns. Bone marrow, kidney, liver, spleen and urinary bladder were among the five organs with the highest

SUV values in each oligonucleotide distribution pattern (Figure 22). The distribution pattern in almost all tissues

seemed to vary with the nature of the oligonucleotide backbone.

iii) Whole-body autoradiography: As complement to quantitative organ distribution, whole-body

autoradiography provides qualitative images without missing organs or more detailed information about the

30

tissues. The highest uptake of radioactivity of phosphodiester oligonucleotide was observed in the liver and

urinary bladder (Figure 23a). With radioactivity decreasing in magnitude, kidney and spleen were the next

tissues, followed by the lung, and finally skin and bone. The highest accumulation of phosphorothioate

oligonucleotide in the kidney cortex and relative high phosphorothioate uptake in the parotid gland were

observed (Figure 23b). 2'-O-methyl phosphodiester showed similar distribution pattern to phosphorothioate, but

the uptake in the kidney cortex was noticeably lower (Figure 23c).

a)

Urinary bladder

c)

LiverLung Kidney

Urinary bladder

Parotid glandKidney Spleen Lung

Heart

Figure 23. Ex vivo autoradiography obtained 20 min after the injection of 68Ga-labelled phosphodiester (a), phosphorothioate (b), 2'-O-methyl phosphodiester (c) antisense oligonucleotides.

b)

Liver

The 68Ga-labelled oligonucleotides were used to study biodistribution and biokinetics in vivo in athymic rats

each bearing a tumor of A549 cells, containing K-ras point mutation in codon 12, and a tumor of BxPC-3 cells,

containing wild-type K-ras.82 The injected 68Ga-oligonucleotides revealed high quality PET images, allowing

quantification of the biokinetics in major organs and in tumours.

31

4 Conclusions • A method for DOTA conjugation and 68Ga labeling of olignucleotides/peptides has been developed.

• The modifications of the oligonucleotides such as the introduction of hexylaminolinker either at the 3′-

or 5′- end, substitution of non-bridging oxygens by sulphur or introduction of O-methyl group at sugar

2′ position did not influence the conjugation or radiolabelling or the hybridisation ability of the

oligonucleotides. The 68Ga-oligonucleotide counterparts showed different radioactivity organ

distribution reflecting their various metabolism and non-specific binding in rats.

• A method for concentration and purification of 68Ge/68Ga generator eluate was developed.

• The introduction of 68Ga purification and concentration step in combination with the microwave

activation makes it possible to use the commercial 1850 MBq generator over a period of more than 2

years for patient studies. The proposed technique is suited for automation and the development of a

preparetion kit is in progress.

• Microwave activation was found to be efficient to accelerate and improve the complexation reaction of 68Ga with bifunctional chelators, DOTA and NOTA, conjugated to peptides and oligonucleotides.

• The introduction of 68Ga purification and concentration step in combination with the microwave

activation allowed omission of the purification step in the preparation of 68Ga labelled peptide

conjugates and improved the specific radioactivity of the product by a factor of 100-150.

32

Acknowledgements I would like to thank all people who have helped me and contributed to this thesis. First of all I express my sincere gratitude to my supervisor Professor Bengt Långström for accepting me as a PhD student, for his constant support, guidance and enthusiasm, for providing excellent working facilities and conditions. Special thanks to Professor Gerd Beyer, Cyclotron Unit, Geneva University Hospital, Geneva Switzerland, for constructive, educational discussions and fruitful collaboration. Many thanks to Dr. Johan Sandell for thorough discussion on the thesis and practical, instructive comments. My colleagues at the preclinical laboratory for creating a friendly and stimulating atmosphere. Especially, Professor Mats Bergström for always being ready to answer practical and theoretical questions on any PET topics. Elisabeth Bergström for constant support and provision with laboratory stuff. Helena Wilking for constant help with language correction, enjoyable talks and discussions and also for the remarks on the thesis. Gabor Lendvai for enjoyable and fruitful collaboration. Past and present members of the group BLå. In particular, Ulrika Yngve for the first introduction to the lab equipment and 68Ga-labeling procedure. Koichi Kato, Hisashi Doi and Bert Windhorst for educative chemistry talks and instructive remarks on the “oligo” manuscript. Martin Laven for always being ready to help and the interesting and fruitful collaboration. Obaidur Rahman for interesting and useful talks on chemistry subject and for useful comments on the thesis. Jonas Eriksson for nice talks and constant help. Julien Barletta and Olexiy Itsenko for chats and introduction to the “classical” organic chemistry lab. Linda Samuelsson for encouraging me speaking Swedish. Mimmi Lidholm for the fruitful and enjoyable collaboration, for always positive and optimistic attitude. Jonas Eriksson, Tommy Lindell, Johan Ulin for the work on the “Gallea” automated system. Eva Pylvännen, Paula Delking, Anders Nilsson, Peter Hjelm and Tommy Lindell for invaluable secretarial and technical assistance. Everybody at Uppsala Imanet AB and the Department of Organic Chemistry for the support, help and for making me feel always welcomed. Professor Helmut Maecke, Institute of Nuclear Medicine, Division of Radiological Chemistry, Basel, Switzerland, for collaboration and for giving me an opportunity to present the results of my research to the PET audience at 4th and 6th European COST Action B12 meetings (Dresden-2002, Pisa-2003). Dr. Hege Karlsen, Amersham Health, Olso, Norway, Maria Välilä and Dr. Anne Roivainen, Turku PET centre, Finland for collaboration. The Swedish Council is acknowledged for its support by grant K3464. My Swedish, Hungarian, Russian, Assyrian, Ukrainian, English, Armenian, Sudanian friends for the support, help and fun outside the PET world. I am a lucky person having you in my life. My beloved husband Misha and my beloved daughter Sona, without your love, endless patience, understanding and constant support it would not be possible to accomplish this thesis.

33

References (1) Hevesy, G. Biochem. J. 1923, 17, 439-445. (2) Weissleder, R. Nature Reviews Cancer 2001, 2, 11-18. (3) Raichle, M. E. In Advances in chemistry series: Washington, 1981. (4) Lundqvist, H.; Lubberink, M.; Tolmachev, V. European Journal of Physics 1998, 19, 537. (5) Ter-Pogossian, M. M. In Principles of Nuclear Medicine; Wagner, H. N., Szabo, Z., Buchanan, J. W.,

Ed.; W.B.Saunders: Philadelphia: W.B.Saunders: Philadelphia, 1995, pp 342-346. (6) Eary, J. F.; Krohn, K. A. European Journal of Nuclear Medicine and Molecular Imaging 2000, 27, 1737-

1739. (7) Price, P. Trends Mol Med 2001, 7, 442-446. (8) Långström, B.; Hartvig, P.; D., N., Ed.; Mercel Dekker Inc.: New York, USA, 1992. (9) Green, M. A.; Welch, M. J. Nucl. Med. Biol. 1989, 16, 435-448. (10) Anderson, C. J., Welch, M.J. Chem. Rev. 1999, 99, 2219-2234. (11) Lambrecht, R. M.; Sajjad, M. Radiochimica Acta 1988, 43, 171-179. (12) Pearson, R. G. JACS 1963, 85, 3533-3539. (13) Harris, W. R.; Messori, L. Coordin Chem Rev 2002, 228, 237-262. (14) Rösch, F.; Knapp, F. F. In Handbook of nuclear and radiochemistry; Vertes, A., Ed.; Kluwer Publishers,

2002, pp 1-36. (15) Heppeler, A., Froidevaux, S., Eberle, A. N., Maecke, H.R. Current Medicinal Chemistry 2000, 971-994. (16) Yano, Y.; Anger, H. O. J. Nucl. Med. 1964, 5, 485-488. (17) Carlton, J. E.; Hayes, R. L. Int. J Appl Radiat Isot 1971, 22, 44-45. (18) Schuhmacher, J.; Maier-Borst, W. Int. J. Appl. Radiat. Isot. 1981, 32, 31-36. (19) Ambe, S. Appl Radiat Isotopes 1988, 39, 49-51. (20) Kopecky, P.; Mudrova, B.; Svoboda, K. International Journal of Applied Radiation and Isotopes 1973,

24, 73-80. (21) Kopecky, P.; Murdova, B. Int. J Appl Radiat Isot 1974, 25, 263-268. (22) Loc'h, C.; Maziere, B.; Comar, D. J. Nuc. Med. 1980, 21, 171-173. (23) Neirinckx, R. D.; Davis, M. A. Generator for gallium-68 and compositions obtained therefrom., USA,

Massachusetts Institute of Technology (Cambridge, MA), 1981. (24) Nakayama, M.; Haratake, M.; Koiso, T.; Ishibashi, O.; Harada, K.; Nakayama, H.; Sugii, A.; Yahara, S.;

Arano, Y. Anal. Chim. Acta 2002, 453, 135-141. (25) Neirinckx, R.; Davis, M. J Nucl Med 1980, 21, 81-83. (26) Långström, B.; Bergson, G. Radiochem. Radioanal. Letters 1980, 43, 47-54. (27) Elander, N.; Jones, J. R.; Lu, S. Y.; Stone-Elander, S. Chem. Soc. Rev. 2000, 29, 239-249. (28) Ohlsson, T.; Vengtsson, N. E.; Risman, P. O. J. Microwave Power 1974, 9, 129. (29) Beuhler, R. J.; Flanigan, E.; Green, L. J.; Friedman, L. Journal of the American Chemical Society 1974,

96, 3990-3999. (30) Moi, M. K.; Meares, C. f.; McCall, J. M.; Cole, W. C.; De Nardo, S. J. Anal. Biochem. 1985, 148, 249-

253. (31) Broan, C. J., Cox, J.P.L., Craig, A.S., Kataky, R., Parker, D., Harrison, A., Randall, A.M., Ferguson, G. J.

Chem. Soc. Perkin Trans. 2 1991, 87-99. (32) Morphy, J. R.; Parker, D.; Alexander, R.; Bains, A.; Carne, A. F.; Eaton, M. A. W.; Harrison, A.;

Millican, A.; Phipps, A.; Rhind, S. K.; Titmas, R.; Weatherby, D. J. Chem. Soc., Chem Commun. 1988, 156-158. (33) Craig, A. S.; Parker, H. A.; Bailey, N. R. J.Chem.Soc., Chem. Commun. 1989, 1793-1794. (34) Eisenwiener, K. P.; Prata, M. I.; Buschmann, I.; Zhang, H. W.; Santos, A. C.; Wenger, S.; Reubi, J. C.;

Macke, H. R. Bioconjug Chem 2002, 13, 530-541. (35) Lewis, M. R., Raubitschek, A., Shively, J.E. Bioconjugate Chem. 1994, 5, 565-576. (36) Keire, D. A.; Jang, Y. H.; Li, L.; Dasgupta, S.; Goddard, W. A.; Shively, J. E. Inorg Chem 2001, 40,

4310-4318. (37) Kumar, K.; Chang, C. A.; Francesconi, L. C.; Dischino, D. D.; Malley, M. F.; Gougoutas, J. Z.; Tweedle,

M. F. Inorg Chem 1994, 33, 3567-3575. (38) Alexander, V. Chem Rev 1995, 95, 273-342. (39) Kaden, T. A. Chimia 2000, 54, 574-578. (40) Reichert, D. E.; Lewis, J. S.; Andreson, C. J. Coordin Chem Rev 1999, 184, 3-66. (41) Clarke, E. T.; Martell, A. E. Inorg Chim Acta 1991, 181, 273-280. (42) Clarke, T. E., Martell, A.E. Inorganic Chimica Acta 1991, 190, 37-46.

34

(43) Heppeler, A. S. F., H.R.Mäcke, E.Jermann, M.Behe, P.Powell, M.Hennig Chem. Eur. J. 1999, 5, 1974-1981.

(44) Liu, S.; Edwards, D. S. Chem Rev 1999, 99, 2235-2268. (45) Thakur, M. L. Nucl Med Commun 1995, 16, 724-732. (46) Behr, T. M.; Behe, M.; Becker, W. Q J Nucl Med 1999, 43, 268-280. (47) Breeman, W. A.; de Jong, M.; Kwekkeboom, D. J.; Valkema, R.; Bakker, W. H.; Kooij, P. P.; Visser, T.

J.; Krenning, E. P. Eur J Nucl Med 2001, 28, 1421-1429. (48) Smith-Jones, P. M.; Stolz, B.; Burns, C.; Albert, R.; Reist, H. W.; Fridrich, R.; Mäcke, H. M. J. Nucl.

Med. 1994, 35, 317-325. (49) Stolz, B.; smith-Jones, P. M.; Albert, R.; Reist, H.; Mäcke, H.; Bruns, C. Horm. metab. Res 1994, 26,

453-459. (50) Henze, M.; Schuhmacher, J.; Hipp, P.; Kowalski, J.; Becker, D. W.; Doll, J.; Mäcke, H. R.; Hofmann, M.;

Debus, J.; Haberkorn, U. J. Nucl. Med. 2001, 42, 1053-1056. (51) Hofmann, M.; Maecke, H.; Borner, R.; Weckesser, E.; Schoffski, P.; Oei, L.; Schumacher, J.; Henze, M.;

Heppeler, A.; Meyer, J.; Knapp, H. Eur J Nucl Med 2001, 28, 1751-1757. (52) Ugur, O.; Kothari, P. J.; Finn, R. D.; Zanzonico, P.; Ruan, S.; Guenther, I.; Maecke, H. R.; Larson, S. M.

Nucl Med Biol 2002, 29, 147-157. (53) Froidevaux, S.; Eberle, A. N. Biopolymers 2002, 66, 161-183. (54) Milligan, J. F.; Matteucci, M. D.; Martin, J. C. J. Med. Chem. 1993, 36, 1923-1937. (55) Hnatowich, D. J. The Quarterly Journal Of Nuclear Medicine: Official Publication Of The Italian

Association Of Nuclear Medicine (AIMN) [And] The International Association Of Radiopharmacology (IAR) 1996, 40, 202-208.

(56) Dewanjee, M. K.; Haider, N.; J., N. J. Nucl. Cardiol. 1999, 6, 345-356. (57) Hnatowich, D. J. J. Nucl. Med. 1999, 40, 693-703. (58) Kobori, N.; Imahori, Y.; Mineura, K.; Ueda, S.; Fujii, R. Neuroreport 1999, 10, 1271-1274. (59) Urbain, J. L. C. J. Nucl. Med. 1999, 40, 498-504. (60) Watanabe, N.; Sawai, H.; Endo, K.; Shinozuka, K.; Ozaki, H.; Tanada, S.; Murata, H.; Sasaki, Y. Nucl

Med Biol 1999, 26, 239-243. (61) Hedberg, E.; Långström, B. Acta Chem. Scand. 1997, 51, 1236-1240. (62) Hedberg, E.; Långström, B. Acta Chem. Scand. 1998, 52, 1034-1039. (63) Pan, D.; Gamdhir, S.; Toyokuni, T.; Iyer, M.; Acharya, N.; Phelps, M. E.; Barrio, J. R. Bioorg. Med.

Chem. Lett 1998, 8, 1317-1320. (64) Yngve, U.; Hedberg, E.; Lövqvist, A.; Tolmachev, V.; Långström, B. Acta Chem. Scand. 1999, 53, 508-

512. (65) Kühnast, B.; Dollé, F.; Terrazino, S.; Rousseau, B.; Loc'h, F. V.; Hinnen, F.; Doignon, I.; David, C.;

Crouzel, C.; Tavitian, B. Bioconjugate Chem. 2000, 11, 627-636. (66) Ho, P. T.; Parkinson, D. R. Semin Oncol 1997, 24, 187-202. (67) Agrawal, S. Trends Biotechnol. 1996, 14, 376-387. (68) Boiziau, C.; Larrouy, B.; Sproat, B. S.; Toulme, J. J. Nucleic Acids Res 1995, 23, 64-71. (69) Larrouy, B.; Boiziau, C.; Sproat, B.; Toulme, J. J. Nucleic Acids Res 1995, 23, 3434-3440. (70) Zhang, R.; Lu, Z.; Zhao, H.; Zhang, X.; Diasio, R. B.; Habus, I.; Jiang, Z.; Iyer, R. P.; Yu, D.; Agrawal,

S. Biochem Pharmacol 1995, 50, 545-556. (71) Younes, C. K.; Boisgard, R.; Tavitian, B. Curr Pharm Design 2002, 8, 1451-1466. (72) Crooke, S. T. Annu. Rev. Pharmacol. Toxicol. 1992, 32, 329-376. (73) Hermanson, G. T. Bioconjugate Techniques; 1 ed., 1996. (74) Velikyan I.; Beyer G.; Långström B. Method of obtaining 68Ga, filed patent application, PZ0333, 2003. (75) Ehrhardt, G.; Welch, M. J Nucl Med 1978, 19, 925-929. (76) Kraus, K. A.; Nelson, F. In Conference on peaceful uses of atomic energy: United Nations, 1956; Vol. 7,

pp 113-125. (77) Kraus, K. A.; Nelson, F. In Ion exchange and chromatography in analytical chemistry; 195, A. S. T. P.,

Ed.; American society for testing materials, 1958, pp 27-57. (78) Nelson, F.; Murase, T.; Kraus, K. A. Journal of Chromatography A 1964, 13, 503-535. (79) Velikyan I.; Långström B. Microwave method, filed patent application, PZ0334, 2003. (80) Laven M.; Velikyan I.; Djodjic M.; Ljung J.; Berglund O.; Markides K.; Långström B.; Wallenborg, S.

Manuscript in preparartion 2003. (81) Laven M.; Wallenborg, S.; Velikyan I.; Bergström S.; Djodjic M.; Ljung J.; Berglund O.; Edenwall N.;

Markides K.; B., L. Manuscript in preparartion 2003. (82) Roivainen, A.; Tolvanen, T.; Salomäki, S.; Lendvai, G.; Velikyan, I.; Numminen, P.; Hoffren, M.-L.; Välilä,

M.; Sipilä, H.; Bergström, M.; Härkönen, P.; Lönnberg, H.; Långström, B. J Nucl Med 2003, accepted.

35

Paper I

36

Microwave supported preparation of 68Ga-bioconjugates with high specific radioactivity

Running Title: 68Ga - Peptides

I. Velikyan†,‡, G. J. Beyer§, B. Långström†,‡,*

†Department of Organic Chemistry, Institute of Chemistry, BMC, Uppsala University, Box 599, SE-751 24 Uppsala, Sweden, ‡Uppsala Imanet AB, PO Box 967, SE-751 09 Uppsala, Sweden, §Cyclotron

Unit, Geneva University Hospital, Geneva Switzerland

For correspondence contact: Bengt Långström Uppsala Imanet, PO Box 967, SE-751 09 Uppsala Phone: +46 18 666 900 Fax: +46 18 666 819 E-mail: Bengt.Langstrom@Uppsala.Imanet.se

37

Abstract The generator produced positron-emitting 68Ga (T1/2 = 68 min) is of potential interest for clinical PET. 68Ga as a metallic cation is suitable for complexation reactions with chelators, naked or conjugated with peptides or other macromolecules. Large 68Ga generator eluate volumes, metal traces from the generator column material or from added reagents ligands, however, disturb a fast, reliable and quantitative labeling procedure. In this paper we describe a simple technique, based on anion exchange, aiming firstly, to increase the 68Ga concentration, secondly to purify it from competing impurities and thirdly to obtain a fast and quantitative 68Ga labeled peptide conjugate, that can be applied without further purification in humans. Within 5 min one can obtain from the original 6 mL generator eluate a 200 µL 68Ga preparation (volume reduction by a factor 30) that is suitable for direct and quantitative labeling of peptide conjugates. DOTATOC (DOTA-D-Phe1-Tyr3–Octreotide), (DOTA = 1,4,7,10-tetraazacyclo-dodecane-1,4,7,10-tetraacetic acid) was used as a test tracer for comparing the labeling properties of the different 68Ga preparations. In combination with microwave activation peptide conjugates of 0.5 – 1 nanomolar quantities could be labeled within 10 min with the full 68Ga activity of a generator. Further purification of the 68Ga labeled peptide conjugate was no longer required since the nuclide incorporation was almost quantitative. The specific radioactivity (with respect to the peptide) was improved by a factor ~100 compared to the previously applied techniques using the original generator eluate. The commercial 68Ge/68Ga generator from Obninsk in combination with this system for purification and concentration with an integrated microwave-activated labeling technology resulted in a kit –like technology for 68Ga-tracer production. The first automated prototype using this technology is being tested. Key words: 68Ga-radiopharmaceuticals, peptide conjugate, 68Ge/68Ga generator, microwave activation Introduction The positron-emitting 68Ga (T1/2 = 68 min) radionuclide is of great practical interest for clinical PET. Especially the new approaches in radionuclide therapy based on radio-labeled peptides call for an individual dosimetric PET imaging protocol. The 68Ga is a generator produced nuclide and does not require cyclotron on site. The long half-life of the parent nuclide 68Ge (270.8 d) provides in principle a long life-span generator. The 68Ga decays by 89% through positron emission of 1.92 MeV max energy, without additional gamma contribution. The short half-life of 68Ga (T1/2 = 68 min) permits application of suitable 68Ga activities while maintaining an acceptable radiation dose to the patient. The only stable chemical form in solution is the 68Ga3+ cation, that forms stable complexes with many ligands containing oxygen and nitrogen as donor atoms. Thus, 68Ga is suitable for complexation with chelators, naked or conjugated with peptides or other macromolecules. Many 68Ga radiopharmaceuticals have been used so far for imaging of brain, renal, bone, blood pool, lung, vascular pool and tumor (1). Some of the restrictions for wider use of 68Ga are the chemical form in which it is obtained from the generator and contamination of the generator eluate with traces of the generator column material. Column materials proposed so far for 68Ge/68Ga generators are inorganic oxides like aluminium dioxide, titanium dioxide or tin dioxide or selective organic resins comprising phenolic hydroxyl groups or pyrogallol (2-5). Elution of the 68Ga is performed using EDTA-solution (5), relative high concentrated HCl (3) or 0.1 M HCl (Cyclotron Co., Ltd, Obninsk, Russia). There are advantages with the latter generator: the 68Ge is loaded onto a titanium dioxide column, that is a non-toxic material; and, more importantly, 68Ga is eluted in the cationic form with 0.1M HCl, allowing universal application for different classical radiopharmaceutical preparations. A general disadvantage related to all generators is the large primary 68Ga eluate volume leading consequently to a low 68Ga concentration. Contamination with the long-lived parent nuclide 68Ge (breakthrough) is generally a critical value and cationic metal ions (mainly from the column material) disturb the labeling procedure due to competition in the complexation reaction. A method based on anion exchange to purify the eluate and to reduce the volume and HCl concentration of a generator eluate has been proposed previously (3). Nevertheless, the final volume obtained was still large (~4 mL). Evaporation to dryness causes activity loss due to the time factor as well as uncontrolled release losses since the GaCl3 is relatively volatile. The anion-exchange approach, however, seems to be a promising way to solve both concentration and purification problems of 68Ge/68Ga generator eluate more generally. The adsorption behaviour of the elements from HCl-solutions on the anion exchanger DOWEX AG1-X8 is well known (6). In HCl solution gallium forms strong anionic complexes with Cl¯ and the corresponding [GaCl6]3¯and [GaCl4]¯complexes are strongly adsorbed on the mentioned anion exchange resin from HCl concentrations > 3 M. In contrast, germanium is practically not adsorbed from < 5 M HCl solution; only from HCl > 6 M is it found that Ge follows Ga. The factor that is most crucial for radiolabeling procedures with short-lived radionuclides is the time needed for a certain process. Microwave activation providing the acceleration is an attractive tool. This technique has been used for labelling different organic molecules with [131I], [11C], [15O], [18F] and [13N] (7). Microwave activation not only shortens the chemical reaction time, but also reduces side reactions, increases the yield and improves reproducibility.(7) Microwave activation is especially useful for microscale organic chemistry like radiolabelling where the size of the sample is comparable to the penetration depth of the microwave field.(7; 8) Microwave-activation has already shown its potential in

38

speeding up 68Ga complexation with DOTA bifunctional chelator coupled to oligonucleotides (9). In the present paper we wish to describe a simple technological procedure based on anion exchange reaction, that allows purification of 68Ga obtained from commercial generators and to reduce the volume of the preparation in one step thus meeting the requirement for direct highly efficient labeling of modern bioconjugates under microwave-activation conditions. Special attention has been paid to reasonable high speed, reliability and simplicity, quantitative product yield and purity. DOTA-D-Phe1-Tyr3 – Octreotide (DOTATOC, where DOTA is the chelator 1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetic acid) was used as a test molecule to study and optimize the labeling procedure.

Materials and methods 68Ga and 68Ge/68Ga-Generator characterization 68Ga (T1/2 = 68 min, β+ = 89% and EC=11%) was available from a 68Ge/68Ga-generator-system (Cyclotron Co., Ltd, Obninsk, Russia) where the 68Ge (T1/2 = 270.8 d) was attached to a column of an inorganic matrix based on titanium dioxide. The nominal 68Ge activity loaded onto the generator column was 1850 MBq (50 mCi). The specified shelf-live of the generator was 2-3 years. The 68Ga was eluted with 6 mL of 0.1 M hydrochloric acid. The elution profile was determined by fractionating and measuring the 68Ga activity in each successive fraction of the eluate. The 68Ga elution yield was expressed as the ratio between the obtained 68Ga and the 68Ge activity (in equilibrium with the 68Ga daughter nuclide) on the generator column at the time of elution. Aliquots of the generator eluates were counted using an ionisation chamber and a NaI(Tl) scintillation detector immediately and after 24-48 h in order to determine the 68Ge breakthrough in the eluate. The 68Ge contamination was expressed as % 68Ge in the 68Ga eluate in each sample. The 68Ga generator eluates were analyzed for the content of trace metals by inductively-coupled plasma atomic emission spectrometry (ICP-AES) using a Spectroflame P instrument (Spectro, Germany). The instrument was calibrated towards standards from Spectrascan (Reference material AB, Ulricehamn, Sweden). The detection limit of the applied Inductively Coupled Plasma Atomic Emission Spectroscopy (ICP-AES) analysis was approximately 0.0001%. Purification and concentration of the 68Ga eluate using anion exchange resins Several strong basic anion exchange resins were investigated for optimization [Dowex AG 1-X8 (100-200 mesh), AG 1-X8 (200-400 mesh), AG 1-X4 (100-200 mesh) and AG 1-X2 (200-400 mesh), see Table1]. Three commercially available strong anion exchange cartridges were tested as well: Isolute (UK) (50mg, 1mL, Cl¯-form); NTK kemi (USA), (15 mg, 3 mL, Cl¯ -form,) and Chromafix 30-PS-HCO3, Macharey-Nagel (Germany). The resins were suspended in 1 M HCl for 1 h, washed several times with deionized water, suspended in 1 M NH3 for 10 min, washed with water and resuspended in 1 M HCl. After this procedure the resins were centrifuged and kept under 5M HCl for the further use. Small plastic tubes of different sizes were used as small columns. They were plugged with glass wool or polyethylene filter elements and packed with the wet resin. Just before use the packed columns were conditioned and activated washing successively with 5 M HCl, 1 M HCl, H2O and again 5 M HCl. The commercial cartridges were preconditioned in the same way. The 68Ge/68Ga-generator was eluted according to the manufacturers protocol with 6 mL 100 mM HCl solution. 5 mL of 30% HCl was added to the 6 mL of the generator eluate giving finally a HCl concentration of 4.0 M. The resulting 11 mL solution in total was passed through a pretreated anion exchange column at a flow rate of 4 mL/min (linear flow speed 25 cm/min) at room temperature. The columns were washed with 2 mL 4 M HCl and then dried by sucking filtered air through the column, in order to eliminate excess 4 M HCl. The 68Ga was then eluted with small fractions of deionized water (50-200µl) at a flow rate of 0.5 mL/min. The elution profile of the small purification columns and the elution yield of 68Ga were determined by measuring the 68Ga activity in the 50 µL fractions of the eluate and normalizing to the end of the elution of the main generator column. The elution yield is expressed as % of eluted 68Ga from the total 68Ga activity present at the generator column at the beginning of elution. The 68Ge contamination was determined by re-counting the fractions 48h after collection. Distribution coefficients The distribution coefficients D of 68Ga between the stationary and mobile phase were determined by the column method as weight distribution coefficients, D.

res

liq

liq

res

mm

AAD ∗=

with A res = 68Ga activity in [Bq] of the dry resin,

A liq = 68Ga activity of the solution passed through the column in [Bq], m res = amount of dry resin in [g], m liq = amount of the solution in [g].

39

The HCl concentration of the 68Ga solution was altered in the range between 0.1 M and 4.3 M by addition of corresponding amounts of 30% HCl for this systematic study. Aliquot samples of the 68Ga solutions (starting and eluate) were measured with a well-type scintillation counter under standardized geometry conditions. D rises from 105 to 2·106 for [HCl] = 3.5 – 4.3 M. 68Ga retention on the cartridge of ~99 % is assured under these conditions. MW-activation Microwave activations were performed with a SmithCreatorTM oven (Personal Chemistry AB, Uppsala, Sweden) with monomodal radiation. Labelling of the peptides with the original 68Ga – generator eluate The pH of the original 68Ge/68Ga-generator eluate (6 mL, containing full 68Ga activity or the 1 mL peak fraction that contained 60-65% of the available 68Ga activity) was adjusted to pH ~ 4.6 by adding sodium acetate (Aldrich, in solid form) to give finally a 0.4 M solution with regard to acetate or to pH ~ 4.2 by adding HEPES (4-(2-Hydroxyethyl) piperazine-1-ethanesulfonic acid, Sigma) to give finally a 1.0 M solution with regard to HEPES. Then 0.1-20 nanomols in 1-20 µL of 1 mM and 1-5 µL of 0.1 mM DOTATOC solution (in water) were added. The reaction mixture was transferred into a Pyrex glass vial for microwave activation and exposed for 1 min to 90±5 °C. Alternatively, the mixture was heated to 95 °C in a conventional heating block for up to 20 min. Quality control Analytical liquid chromatography (LC) was performed using a HPLC system from Beckman (Fullerton, CA, USA) consisting of a 126 pump, a 166 UV detector and a radiation detector coupled in series. Data acquisition and handling was performed using the Beckman System Gold Nouveau Chromatography Software Package. The column used was a Vydac RP 300 Å HPLC column (Vydac, USA) with the dimensions 150 mm × 4.6 mm, 5 µm particle size. We applied gradient elution with the following parameters: A = 10 mM TFA; B = 70% acetonitrile (MeCN), 30% H2O, 10mM TFA with UV-detection at 220 nm; flow was 1.2 mL/min; 0-2 min isocratic 20% B, 20-90% B linear gradient 8 min, 90-20% B linear gradient 2 min. Retention time (tR) for the radio-HPLC signals was 6.56±0.02, 5.66±0.03 and 6.66±0.02 min for DOTATOC, DOTA-RGD and NODAGATATE, respectively. In addition, electrospray ionization mass spectrometry (ESI-MS) was perfirmed using a Fisons Platform (Micromass, Manchester, UK) with positive mode scanning and detecting [M+2H]2+ and [M+3H]3+ species. DOTATOC was detected at m/z =711.26, 69,71Ga-DOTATOC at m/z = 746.0, DOTA-RGD at m/z = 549.1 for [M+2H]2+ and 823.44 for [M+3H]3+ and 69,71Ga-DOTA-RGD at m/z = 857 for [M+2H]2+ and 571 for [M+3H]3+. The 69,71Ga- conjugates synthesised under indentical to labeling conditions were used for the identification of the radio-HPLC chromatogram signals. Labelling of the DOTA-conjugates with the preconcentrated and purified 68Ga eluate: 1. Using sodium acetate buffer 5 µL of 10 M NaOH solution were added to the 200 µL of the 68Ga preconcentrated eluate containing 92±4% of the initially available activity. Then the mixture was added to 16 mg of solid sodium acetate buffering the solution to pH ~ 4.2. Then 0.1 – 10 nanomoles of an aqueous solution of DOTATOC were added (1-10 µL of a 1 mM solution for the range of 1-10 nanomols or 1-5 µL of 0.1 mM solution for the range of 0.1 – 0.5 nanomoles of the peptide conjugate, respectively). The reaction mixture was transferred to a Pyrex glass vial with an insert to accommodate the small volume (200±20 µL) for microwave activation. The heating time in a conventional heating block at 95 °C was up to 20 min and in the microwave furnace 1 min at 90±5 °C. The obtained product was analyzed applying the identical protocol as described above. 2. Using HEPES buffer The procedure was identical to the sodium acetate protocol, however instead of 16 mg solid sodium acetate we used 72 mg of powdered HEPES to obtain identical pH conditions. The labeling was performed for DOTATOC and DOTA-RGD (RGD is Cys2-6; c[CH2CO-Lys(DOTA)-Cys-Arg-Gly-Asp-Cys-Phe-Cys]-CCX6-NH2)(10) conjugates. Labelling of NOTA- conjugate with preconcentrated and purified 68Ga eluate NODAGATATE (NODAGA - Tyr3 - Octreotate)(11), where the bifunctional chelator is 1,4,7-triazacyclononane-1,4,7-triacetic acid, was labeled with the preconcentrated and purified 68Ga eluate according to the following standardized protocol: 3 µL of 10 M NaOH solution were added to the 200 µL of the 68Ga preconcentrated eluate containing 92±4% of the initially available 68Ga activity. Then the mixture was transferred to a vial containing 72 mg HEPES powder buffering the solution to 3 < pH < 3.5. Then 0.2-5 nanomols of the NODAGATATE were added in 1-5 µL of 1 mM aqueous solution or 2-5 µL of 0.1 mM aqueous solution of the conjugate. The resulting 200±20 µL reaction mixture was then transferred to a Pyrex glass vial with an insert to accommodate the small volume for microwave activation. The heating time in the microwave furnace was 1 min at 95±5 °C. The obtained prod uct was analyzed applying the protocol described above.

40

Results and discussion Elution characteristics The generator elution yield over time revealed reproducible performance of the generators (Figure 1). In the first month the elution yield was 78±2% of the theoretical amount and then it decreased slowly and continuously down to 41% after 29

month use. The 68Ge elution profile and the 68Ge breakthrough are presented in Figure 2. By fractionating the 68Ga eluate one can obtain approximately 65% of the available activity with only 1 mL (the third fraction, Figure 2). The 68Ge breakthrough with respect to the eluted 68Ga activity was found to be 0.001 - 0.007% and the 68Ga/68Ge ratio did not change during the investigation period. It should be mentioned that 68Ge breakthrough losses of 0.001% per elution are insignificant compared with 68Ge decay loss, assuming two elutions per day.

Metal content The metal ion content of the 68Ge/68Ga generator eluate depending upon elution frequency is shown in Figure 3. The concentration of the metal ions, except for Pb, was constant for a given elution frequency. The generator requires daily elution in order to keep the concentration of the contaminant metal ions as low as possible. A preventive elution four hours prior to the synthesis is recommended since the interfering metal ion concentration can be kept as its minimum value. The generator matrix material (TiO2) may cause relatively high non-toxic Ti concentration (close to 1000 ppm) in the eluate if the generator is not eluted regularly. The metals act as pseudo carriers requiring, consequently, higher macromolecule conjugate concentrations. The consequence is lower 68Ga incorporation (low labelling yield) or a drop in the specific radioactivity of the final 68Ga labeled peptide conjugate. The proposed concentration and purification step will reduce the metal content as shown below and increase the specific radioactivity of the final product. Labeling reaction The complexation reaction of 68Ga with a macrocyclic chelator conjugated to the peptide starts already at room temperature, but the radiochemical yield (RCY) did not exceed 24% and 30% for sodium acetate and HEPES buffers, respectively (Figure 4A). Conventional heating improved the situation but the yield and reaction time were still unsatisfactory. In the case of the HEPES buffer, the 68Ga incorporation time was shorter. Microwave activation of the reaction mixture considerably shortened the reaction time and improved reproducibility. Nevertheless, a final purification step was still required for obtaining a high quality labelled peptide conjugate (Figure 4A). It is expected that a smaller volume of the reaction mixture will improve both, RCY and reaction time. In a first phase, the reduction of the reaction mixture volume was achieved by fractionation of the generator eluate (Figure 2). The third fraction of the eluate (1 mL) contained > 60 % of the total available 68Ga activity. Using this 68Ga peak fraction the RCY for the reaction at room temperature, conventional heating as well as for microwave activation was improved (Figure 4B). Under conventional heating conditions, quantitative 68Ga incorporation was achieved within 5 min for 5 nanomols of DOTATOC using the HEPES buffer system. For the sodium acetate buffer system a factor 3 larger amount of peptide (15 nanomols) and 20 min heating were required. With microwave-heating the labeling reaction was complete within 1 min, the incorporation of the activity was quantitative and the reproducibility of the reaction was improved. No further purification of the labelled peptide conjugate was required. From this study we learnt that the amount of peptide needed for quantitative 68Ga incorporation was 15 nanomols at least when using sodium acetate buffer and 5 nmol in the case of HEPES (Figure 5). It is worth noting that the difference in the buffer influence is more pronounced at lower peptide quantities indicating the sensitivity of the reaction to potential metal impurities present in the buffers as well as complexing ability of the sodium acetate itself competing with DOTA when used at high concentration. Although the 68Ga incorporation was quantitative using only 5 nanomols of DOTATOC, there were still drawbacks in this approach. Firstly, about 40% of the 68Ga activity is wasted and the secondly, even smaller peptide amounts may be required due to cost and biological properties. Thus a fast technique for the 68Ge/68Ga generator eluate purification and concentration was introduced. Concentration and purification step for 68Ga Our pre-concentration protocol utilized [68GaCl4]- complex formation in 4 M HCl medium. This anionic complex can be efficiently adsorbed on anion exchange resins from HCl solutions and is efficiently eluted with a very small volume of water. The results of the optimisation studies concerning the capabilities of different anion exchange resins are presented in Table 1. The adsorption of 68Ga from HCl solution on anion exchange cartridges increased rapidly from 0.8% at 0.1M HCl to almost 100 % at 3.8 M HCl under our conditions (4 ml/min, 4 mm column diameter) in agreement with literature data on distribution coefficients(6) (D > 105). The best 68Ga recovery was obtained using a column diameter of 4 mm. The linear flow rate of the deionised water was 25 cm/min. Among the class of AG1-resins, Bio-Rad AG 1-X8 with fine mesh (200-400 mesh) showed the best retention and recovery characterstics and a sharp elution profile. The best agent for eluting the 68Ga from the small anionic cartridges was deionised water: the HCl content at the resin was more than sufficient to avoid hydrolysis. To use sodium acetate solution directly as eluting agent is not recommended since the recovery yield of 68Ga is

41

very poor (> 30 % losses) as most likely the anionic Ga-acetate complex is adsorbed to the resin. 50% EtOH as eluting agent could be an alternative and shows nearly the same recovery properties as pure water alone. Since the parent nuclide 68Ge is not retained on the anion resin, the concentration step can also be seen as a purification of 68Ga from 68Ge breakthrough. In the same way the original generator eluant was purified from Ti by 90% as shown by the ICP-AES analysis data. As seen in Table 1, amongst the tested commercial anionic cartridges only the SEX SPEC Chromavix cartridge showed very satisfactory results. This cartridge contains 45 mg of a strong basic anion exchange resin based on polystyrene-divinylbenzene with HCO3

- as the counter-ion and quaternary amine functional groups. Using this cartridges we obtained 99.1 % retention at a flow rate of 2 ml/min and still 99 % at a flow rate of 5 ml/min. Although the lower flow rate assured higher and stable retention values, the decay losses due to time were important. Thus, it is generally more beneficial to apply a higher flow rate and accept 1-3 % worse retention. Elution with deionised water resulted finally in 90±2% of 68Ga recovery in a total volume of 200 µL only. Another 150 µl water volume would provide only 3 % increase of the 68Ga recovery (Figure 2, insert). AG 1-X8 (200-400 mesh) resin and Chromafix SAX SPEC cartridges showed comparable results with respect to retention, recovery and elution profile (Table 1). However, Chromafix SAX SPEC cartridges which are a standardized commercial product are much more suited for further standardization and automation. In total the preconcentration procedure required 4 to maximum 6 min, which includes filtration of the original 68Ga eluate in 4 M HCl through the cartridge, the drying procedure and the elution of the 68Ga in 200 µl pure water. Combined protocol: preconcentration and microwave supported labelling: The influence of the purification step of the 68Ga eluate in combination with microwave activation is demonstrated in Figure 6 and Table 2. Figure 6 illustrates that one can incorporate the full 68Ga radioactivity eluted from the generator into less than 1 nanomol of a peptide conjugate. In this combination the original full 68Ga eluate is purified and concentrated as described and the labeling is performed during 1 min microwave activation. Table 2 contains more details concerning the age of the generator, variation of the amount of peptide and variation of the ligand itself. Labeling tests with equimolar quantities (68Ga : bioconjugate = 1 : 1) using 6 pmols DOTATOC resulted in a maximum activity incorporation of 11 % and the specific activity of the labelled bioconjugate only a factor 10 less than the theoretical value. Independent of the age of the generator and over a period of 2.5 years, we obtained stable and reproducible 68Ga incorporation of > 99 % into 0.3 – 1 nmols of the peptide conjugate. It should be stressed that even with the lower 68Ga amount and concentration with time (2.5 years) it is still possible to obtain quantitative incorporation with one nanomol of the bioconjugate. Quality control The HPLC conditions were selected so that the QC of the radiopharmaceutical prior injection could be completed within 10 min. The retention time (tR) was 6.56±0.02, 5.66±0.03 and 6.66±0.02 min for DOTATOC, DOTA-RGD and NODAGATOC, respectively (Figure 7). Conclusions

The introduction of the purification and concentration step in combination with the microwave activation may have very positive impact on the further utilization of 68Ga in nuclear medicine, generally. The proposed technique allows to eliminate an additional purification step in the preparation of 68Ga labelled peptide conjugates and improves the specific radioactivity of the product by a factor 100. The over-all process can be performed in 10 – 12 min starting from the end of the original generator elution. The HPLC quality control (another 10 minutes) can be performed prior to injection in humans. Due to the excellent 68Ga activity economy, one can assume using the commercial 1850 MBq generator over a period of more than 2 years for patient studies. The proposed technique is suited for automation and the development of a corresponding KIT-like technology is in progress. Acknowledgement The authors wish to thank sincerely Prof. Helmut Maecke (Institute of Nuclear Medicine, Division of Radiological Chemistry, University Hospital Base, Basel, Switzerland) for providing us with DOTATOC and NODAGATATE and Dr. Hege Karlsen (Dept. of Synthetic Chemistry, Amersham Health, Olso, Norway) for the provision of DOTA-RGD. Jean Pettersson, Department of Analytical Chemistry, Uppsala University is gratefully acknowledged for performing the ICP-analysis. The Swedish Council is acknowledged for its support by grant K3464.

42

References (1) Green, M.A. and Welch, M.J. (1989) Gallium radiopharmaceutical chemistry. Nucl. Med. Biol. 16, 435-448. (2) Neirinckx, R.D. and Davis, M.A. (1981) Generator for gallium-68 and compositions obtained therefrom. US-A-4264468, United States Patent, USA. (3) Schuhmacher, J. and Maier-Borst, W. (1981) A new 68Ge/68Ga radioisotope generator system for production of 68Ga in dilute HCl. Int. J. Appl. Radiat. Isot. 32, 31-36. (4) Nakayama, M., Haratake, M., Koiso, T., Ishibashi, O., Harada, K., Nakayama, H., Sugii, A., Yahara, S. and Arano, Y. (2002) Separation of Ga-68 from Ge-68 using a macroporous organic polymer containing N-methylglucamine groups. Anal. Chim. Acta 453, 135-141. (5) Rösch, F. and Knapp, F.F. (2002) Radionuclide generators. Handbook of nuclear and radiochemistry (A. Vertes, Ed.) pp. 1-36, Kluwer Publishers. (6) Kraus, K.A. and Nelson, F., 1956. Anion exchange studies of the fission products, Conference on peaceful uses of atomic energy, United Nations, pp. 113-125. (7) Elander, N., Jones, J.R., Lu, S.Y. and Stone-Elander, S. (2000) Microwave-enhanced radiochemistry. Chem. Soc. Rev. 29, 239-249. (8) Ohlsson, T., Vengtsson, N.E. and Risman, P.O. (1974) The frequency and temperature dependence of dielectric food data as determined by a cavity pertubation technique. J. Microwave Power 9, 129. (9) Velikyan, I., Lendvai, G., Välilä, M., Roivainen, A., Yngve, U., Bergström, M. and Långström, B. (2003) Microwave accelerated 68Ga-labelling of oligonucleotides. J. Labelled Compd. Radiopharm., in press. (10) Karlsen, H., (2003), Personal communication. (11) Maecke, H., (2003), Personal communication.

43

0

20

40

60

80

100

0 200 400 600 800 1000

Time, [day]

68G

a yi

eld,

[%]

Generator-1 Generator-2 Generator-3

Figure 1. 68Ga elution efiiciency for generator-1 during 29 months, generator-2 during 14 months and generator-3 during 3 months .

0

2 0 0

4 0 0

6 0 0

8 0 0

1 0 0 0

1 2 3 4 5 6 7

frac tion

elut

ed a

ctiv

ity p

er fr

actio

n in

[MB

q] fo

r 68

Ga

and

[Bq]

for

68G

e

68G a activ ity g iven in [M B q]

68G e break through g iven in [B q]

0

10

20

30

40

50

60

70

1 2 3 4 5 6 7Fraction num ber (one fraction 50 µ L)

reco

vere

d 68

Ga

activ

it [%

]

vo lum e 200 µ L

0

2 0 0

4 0 0

6 0 0

8 0 0

1 0 0 0

1 2 3 4 5 6 7

frac tion

elut

ed a

ctiv

ity p

er fr

actio

n in

[MB

q] fo

r 68

Ga

and

[Bq]

for

68G

e

68G a activ ity g iven in [M B q]

68G e break through g iven in [B q]

0

10

20

30

40

50

60

70

1 2 3 4 5 6 7Fraction num ber (one fraction 50 µ L)

reco

vere

d 68

Ga

activ

it [%

]

vo lum e 200 µ L

v o lu m e 6 m L

Figure 2. Elution profile of the 68Ge/68Ga generator where one fraction was 1 ml, (fraction 1 = 0.3 ml, fraction 7 = 0.7 ml) giving a total eluted volume of 6 mL. The profiles for the 68Ga and the 68Ge breakthrough are identical, the 68Ge breakthrough is ~10-4 %. Fraction 3 (1 mL) contains >60 % of the available 68Ga activity. The Insert shows the elution profile of 68Ga eluted from the small SAX SPE Chromavix cartridge with small volumes of deionized water. More than 90 % of the initial 68Ga activity was obtained in 200 µL solution.

44

0

1

10

100

1000

0 20 40 60 80 100

Elution period, h

Con

cent

ratio

n, µ

g/L

TiPb

ZnAl Pt

BaNi

Cu

Ga

Figure 3. Metal ion content in 6 mL of the generator eluate dependent on the elution time period.

0

20

40

60

80

100

0 5 10 15 20

Time, min

RC

Y, %

B

0

20

40

60

80

0 5 10 15 20Time, min

RC

Y, %

A

Figure 4. Time course of 68Ga complexation reaction conducted using the full original 68Ga eluate (6mL) (A) and using only the 1 mL peak fraction of the generator eluate (B) at room temperature (dashed line), conventional heating in a heating block at 95 oC(solid line) and with microwave-activation for 1 min at 90±5 oC (circled) for two different buffer systems: sodium acetate buffer, pH = 4.6, 20 nanomols of DOTATOC ∆ HEPES buffer, pH = 4.2, 20 nanomols of DOTATOC (A) and 5 nanomols of DOTATOC (B)

45

0

20

40

60

80

100

120

0 5 10 15 20

DOTATOC, [nanomol]

RC

Y, [%

]

0

20

40

60

80

100

120

0.01 0.1 1 10

[DOTATOC], nanomol

RC

Y, %

Figure 6. Influence of the DOTATOC amount on the radiochemical yield of the 68Ga complexation reaction using the full available 68Ga activity in 200 µl volume after the preconcentration and purification step in HEPES buffer system. solid line : 1 min microwave-activation at 90±5 °C dashed line: 5 min conventional heating at 95 °C

Figure 5. Influence of the buffering system ( sodium acetate, ∆ HEPES) on the radiochemical 68Ga complexation yield for different DOTATOC quantities under 1 min microwave-activation at 90±5 °C. The reaction was conducted using the 1 mL peak fraction of the original generator eluate (see Figure 2.).

46

able 1. Charactererization of the anion exchange resins and cartridges used in the 68Ga purification and preconcentration

Counter ion Retention, % Recovery, %

-0.1

0

0.1

0.2

0.3

0.4

0.5

0.6

0 2 4 6 8 10

retention time in [min]

radi

atio

n si

gnal

68Ga DOTATOC

-0.1

0

0.1

0.2

0.3

0.4

0.5

0.6

0 2 4 6 8 10

retention time in [min]

radi

atio

n si

gnal

-0.1

0

0.1

0.2

0.3

0.4

0.5

0.6

0 2 4 6 8 10

retention time in [min]

radi

atio

n si

gnal

68Ga DOTATOCFigure 7. HPLC-radiochromatogram of a 68Ga-DOTATOC preparation (0.5 nanomols of DOTATOC, 1 min microwave activation at 90±5 oC, total reaction volume 220 µL). The 68Ga incorporation yield is 99.9% and the specific activity of the labeled peptide conjugate was 2 GBq/nmol.

Tstudy. With a retention of > 98 % and a recovery yield of better than 90 % the commercial anion cartridge SAX SPEC, Chromavix, showed the most promising results.

No Sorbent

1 AG 1-X8 (100-200 mesh) HO- 81±4 69±3

2 AG 1-X8 (200-400 mesh) HO- 98±2 85±5

3 AG 1-X4 (100-200 mesh) HO- 65±5 80±3

4 AG 1-X2 (200-400 mesh) HO- 45±4 68±2

5 SAX SPEC, Isolute, 50mg Cl- < 4 -

6 SAX SPEC, NTK kemi, 15 mg Cl- < 4 -

7 SAX SPEC, Isolute, 50mg OH- 9 -

8 SAX SPEC, NTK kemi, 15 mg OH- 7 -

9 SAX SPEC, Chromavix, 45 mg HCO3- 99±1 93±2

47

able 2. Labeling conditions and practical quality parameters of the 68Ga-labeled peptide conjugates obtained from

Peptide conjugate Activity

tion Starting full Ga Specific activity

Tpreconcentrated and purified 68Ga preparations under microwave-activation. Due to quantitative radionuclide incorporation, no further purification of the labelled product is needed.

Generator age, Peptide 68

month amount, nanomol

incorpora [ % ]

activity from generator, [MBq ]

after synthesis [MBq/nmol ]

1 DOTATOC 1 99.0 1289.0 997.7

1 DOTATOC 0.5 99.9 1286.7 2003.0

1 DOTATOC 0.3 98.8 1251.3 3307.0

14 DOTATOC 1 99.9 357.0 308.7

14 DOTATOC 0.5 99.9 337.2 537.2

14 DOTATOC 0.3 99.9 329.4 843.7

29 DOTATOC 1 99.9 78.8 61.0

1 NODAGATATE 1 99.9 1275.9 1011.1

1 NODAGATATE 0.5 99.9 1110.9 1758.4

14 NODAGATATE 1 98.7 357.4 281.3

14 NODAGATATE 0.5 96.8 323.2 493.8

29 NODAGATATE 2 98.0 77,6 30.5

1 DOTA-RGD 1 99.9 1253.0 976.0

1 DOTA-RGD 0.5 99.1 1213.9 1869.9

1 DOTA-RGD 0.25 96.5 1256.7 3756.4

14 DOTA-RGD 0.5 99.9 362.4 570.4

14 DOTA-RGD 0.25 96.2 369.8 953.0

29 DOTA-RGD 1 98.4 72.2 56.1

48

Paper II

49

Short Title 68

1,2 2 3 3 1,2

M. Bergström and B. Långström

1

om@Uppsala.Imanet.se

Microwave Accelerated 68Ga-Labelling of Oligonucleotides

: Ga - Labelled DOTA - Oligonucleotides

I. Velikyan , G. Lendvai , M. Välilä , A. Roivainen , U. Yngve , 2 1,2

Department of Organic Chemistry, Institute of Chemistry, Uppsala University, Box

531, SE-751 21 Uppsala, Sweden, 2Uppsala Imanet, 3PO Box 967, SE-751 09 Uppsala, Sweden, Fax: +46 18 666 819, Turku PET Centre,

PO Box 52, FIN-20521, Turku, Finland.

For correspondence contact: Bengt Långström Uppsala Imanet, PO Box 967, SE-751 09 Uppsala Phone: +46 18 666 900 Fax: +46 18 666 819 E-mail: Bengt.Langstr

50

UMMARY

es are extensively used for characterisation of gene expression in vitro and have now been studied as

g

3 were studied.

RDS: 68Ga; radiolabelled antisense oligonucleotides; DOTA; microwave activation

ides can inhibit gene expression.1 It makes them of considerable interest for biological studies and

fication

23-25 68

131 11 15 18

e for

S Oinhi

ligonucleotidbitors of gene expression in vivo in various diseases. Labelled antisense oligonucleotides are therefore of potential

interest for possible in vivo imaging of gene expression, considering the biology of tumours and applications in designinnovel molecule-targeted therapies. In the present work a method of microwave accelerated 68Ga-labelling of oligonucleotides and analysis of the resulting tracers are described. Four modified and functionalised 17-mer oligonucleotides with an hexylamine group in the '- or 5'- positionThe oligonucleotides were conjugated to the bifunctional chelator, 1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetic acid (DOTA ), and then labelled with 68Ga (T1/2=68 min) using microwave activation. The isolated decay-corrected radiochemical yields ranged from 30 to 52%. Labelled products were stable in water and ethanol for more than 4 h. The impact of the labelling procedure on the oligonucleotide probes was investigated using hybridization to a complementary 17-mer sense oligonucleotide in solution. Chemical modification did not influence either the labelling or hybridisation ability of the oligonucleotides. The radiolabelled oligonucleotides will be used for the further in vitro and in vivo biology studies.

EY WOK NTRODUCTION I

Ap

ntisense oligonucleotarticularly molecule-targeted therapies of cancer. Labelled antisense oligonucleotides may be used for in vivo imaging of

gene expression using appropriate radionuclide and oligomer base sequence.2, 3 Labelling studies have been conducted withgamma emitters, such as [90Y], [99mTc], [111In] and [125I]4-8 as well as with positron emitting radionuclides, such as [11C], [18F] and [76Br]9-13 The reported methods modify the oligonucleotide by addition of labelled or radiolabel chelating groups. Compared to radiohalogenetion, radiometal ion complexation reactions have simpler chemistry and allow tracer productionkits. Labelling with positron emitting radionuclides offers a distinct advantage over gamma emitting radioisotopes since Positron Emission Tomography (PET) has a greater spatial resolution and allows not only imaging of biological processes but also biodistribution quantification. Generator available radiometals are preferable to costly cyclotron produced ones. Toimprove the stability and maintenance of the hybridisation properties of nucleic acids, a number of modified oligonucleotides like phosphorothioate14, 2′- O-methyl15-17, methylphosphonate, phosphoramidate, morpholino oligonucleotide, mixed-backbone oligonucleotide and peptide nucleic acid (PNA) has been studied. 3 End-modiincreases the stability of the oligonucleotide to degradation by exonucleases18 and allows introduction of an amine groupfor the further conjugation of the oligonucleotide to a chelator 19. The resulted conjugate can then be labelled with a metalradionuclide. 68Ga meets the major requirements for a candidate in the labelling of oligonucleotides. It forms stable complexes with non-

20-22cyclic and macrocyclic bifunctional chelators containing nitrogen and oxygen donors. Ga has favourable detectioncharacteristics decaying 89% by positron emission. In addition, β+

max energy of 1.9 MeV provides good resolution. 68Ga is obtained from 68Ge (T1/2=270.8 d) in a generator-system with about 1.5 years life span. Its 68 min half-life is sufficient to follow certain biochemical processes. Another factor that is crucial for radiochemistry and tracer production is the labelling synthesis time. Microwave activation providing the acceleration is an attractive tool. The technique has been used for labelling with [ I], [ C], [ O], [ F] and [13N].26 Microwave activation seems to be useful for microscale organic chemistry like radiolabelling where the size of the sample is comparable to the penetration depth of the microwave field.26, 27 The purpose of the present study was to develop a rapid oligonucleotide labelling method that would also give high radiochemical yields (RCY). Our study has shown that the microwave celac erated complexation method is applicablthe 68Ga labelling of oligonucleotides and peptides. In this paper we present 68Ga labelling and product characterization of the following four modified antisense oligonucleotides (1 - 4) specific for activated human K-ras oncogene: 17-mer phosphodiester oligonucleotide with hexylaminolinker at 5′ end (1); 17-mer phosphodiester oligonucleotide with hexylaminolinker at 3′ end (2); 17-mer phosphorothioate oligonucleotide with hexylaminolinker at 5′ (3); and 2′-O-mphosphodiester with hexylaminolinker at 5′ (

ethyl 4).

RESULTS AND DISCUSSION

he structural formulas of the modTfu

ifications introduced into the oligonucleotides are shown in Figure 1. The amine ng nctionality introduced at the 3′- or 5′-end was used to covalently link the oligonucleotide to bifunctional complexi

agents28 which can form kinetically stable complexes with the radionuclides of interest. The chelator DOTA (5) has been

51

e shown to form stable complexes with Ga(III).29 Conjugation protocols developed for antibody labelling30 were shown to bapplicable for DNA conjugation and radiolabelling.31, 32 The method originally developed for conjugation of monoclonal antibodies and proteins with N-hydroxy-sulfosuccinimidyl DOTA33, 34 was applied in this work for the oligonucleotide conjugation with 6 (Scheme 1, Table 1). The conjugate obtained was then used for the labelling with 68Ga. As in the case of 76Br labelling of oligonucleotides35, where the labelling procedure worked for oligonucleotides of different lengths (30, 20, 12 and 6-mer) equally well, the method used in the present work was shown to be applicable independent of the modification of the oligonucleotides. The oligonucleotides were reacted with the N-hydroxy-sulfosuccinimide estDOTA (

er of 6), generated in situ using water-soluble 1-ethyl-3(3-dimethylaminopropyl) carbodiimide (EDC) as the couplin

reagent, to give DOTA-oligonucleotide (g

7) (Scheme 1, Table 1). The advantage of adding N-hydroxy-sulfosuccinimide (Sulfo-NHS) to the EDC reaction is to increase the stability of the active intermediate, which ultimately reacts with the attacking amine.19 The half-life of the intermediate activated carboxylate is crucial in our case where the target oligonucleotide molecule is at low concentration compared to water.19 Purification of the conjugation product wperformed taking particular care to ensure absence of free DOTA, since the latter would compete in the labelling wDOTA-oligonucleotide and decrease the radiochemical yield. The maximum amount of

as ith

the labelling reaction was in the pg range. The action to

-

68Ga available from the generator foroligonucleotide amount was in the µg range. In order to provide the concentration necessary for the labelling retake place, the volume of the 68Ga generator eluate solution was minimised. The 68Ga elution profile showed that over 50% of the maximum possible activity was obtained in the second fraction and the highest relative 68Ge breakthrough was in the first fraction of the eluate. Accordingly, the first part of the eluate was discarded and the next 1 mL of 0.1M hydrochloric solution of 68Ga(III) was collected and buffered with sodium acetate to a pH of 5.5 to avoid precipitation of gallium in theform of gallium trihydroxide.36 After addition of the oligonucleotide to the latter mixture, the reaction was carried out in a microwave oven (1 min at 100W, Scheme 2). Labelling of all four oligonucleotide counterparts gave isolated decay-corrected radiochemical yields in the range of 30 to 52% (Table 1). The specific activity of the antisense 68Ga-DOTAoligonucleotides (8) obtained ranged from 0.1-1.5 MBq/nmol. The position of the 68Ga-label was assessed by performing the labelling reaction with both conjugated and unconjugated

s

nd

edure using conventional heating (radiochemical yield of 19%) the microwave application

d

s (8

oligonucleotides. No product was detected in the reaction with unconjugated oligonucleotide, indicating that the label wamost likely attached to the chelator. The radiochemical yield was found to improve with increasing concentration of oligonucleotide. However, increasing the concentration of the oligonucleotide may render purification steps difficult adecrease the specific activity. Compared to the synthesis procin the present work shortened the synthesis time considerably and improved the radiochemical yield. Due to the shortened reaction time, the amount of radioactive material and the product specific activity was increased by 21%. Furthermore, microwave activation not only reduces the chemical reaction time, but also reduces side reactions, increases the yield, animproves reproducibility.26 In the present case, the analytical radiochemical yield was increased 2 fold compared to our previous results with conventional heating. The stability of radiolabelled oligonucleotide ) in water and 50% ethanol was monitored by radio HPLC with analysis of

aliquots taken from the labelling reaction mixture every 15 min. 68Ga was used as the internal reference. The ratio of the areas of the free 68Ga and labelled oligonucleotides radiosignals remained constant indicating that the loss of activity wasonly due to the decay of the radionuclide. Moreover, no additional radiosignals were detected in the stability study. The radiochemical purity of 68Ga-DOTA-oligonucleotides (8) in water and 50% ethanol was >95% for more than four hours. This time corresponds to 3-4 physical half-lives of 68Ga and is the time required for biological experiments and diagnosticalinvestigations. The impact of th

e labelling procedure on the oligonucleotide probes was investigated using their specific hybridisation to a

e

four

EXPERIMENTAL

odium acetate (99.995%, Aldrich) and doubly distilled HCl (Aldrich) were used in the labelling. 1-ethyl-3-(3-7,10-

complementary 17-mer phosphodiester sense oligonucleotide in solution. The results were analysed using polyacrylamide gel electrophoresis in a cell-free system and the results were visualised by ethidium bromide staining and autoradiography. The results of the concentration dependent hybridisation of the phosphorothioate counterpart are shown in the Figure 2. Nine samples with a constant concentration of the radiolabelled antisense oligonucleotide and a gradually increasing concentration of the sense oligonucleotide were analysed (Figure 2, lanes 1-9). As a reference the sense oligonucleotid(lane 10), the antisense oligonucleotide (lane 11) and molecular weight marker (lane 12) samples were loaded onto the polyacrylamide gel. The gradual increase of the intensity of the hybrid bands, the absence of the free radioactive bands of antisense at the higher concentrations of the sense oligonucleotide and the absence of free sense bands at the lower concentration of the sense oligonucleotide serve as an indication of the concentration dependent hybridisation. All counterparts were able to hybridise to the complementary sense oligonucleotide.

Sdimethylaminopropyl)carbodiimide (EDC) (Sigma), N-Hydroxysulfosuccinimide (Sulfo-NHS) (Sigma) and 1,4,

52

tetraazacyclododecane-1,4,7,10-tetraacetic acid (DOTA) (Macrocyclics, USA) were used for the conjugation. 17-mer phosphodiester (1) and phosphorothioate (3) oligonucleotides of antisense strand sequence with a hexylamine linker in5′-position, 17-mer phosphodiester oligonucleotide (

the 2) of antisense strand sequence with a hexylamine linker in the 3′-

position as well as 17-mer phosphodiester oligonucleotide of sense strand sequence were purchased from ScandinavianGene Synthesis AB (Köping, Sweden). 2′-O-methyl phosphodiester oligonucleotide (

4) was synthesised at Turku PET

centre, Turku, Finland and was used as received. The antisense and sense sequences used were, respectively, 5′-CTA CCAC TAG CTC CA and 5′-TGG AGC TAG TGG CGT AG. Deionised water (18.2 MΩ) was used in all reactions. The purchased chemicals were used without further purification. Thin-layer chromatography (TLC) was performed on PEI-Ce

GC

llulose F plates (Merck, Germany) with 0.4 M NaH2PO4, pH =

tions were performed using a Beckman (Fullerton, CA, USA) System (a 126

= 100% acetonitrile (MeCN),

0-15 isocratic; λ = 280nm ter,

ed

89% and by EC 11%) was available from Ge (T1/2 =

nal Chemistry AB, Uppsala, Sweden) with

onjugation of DOTA to oligonucleotides: DOTA (32 mg, 66 µmol) and Sulfo-NHS (14 mg, 65 µmol) in H2O (250 µl)

3.5 as the running buffer. Ultraviolet (UV) absorbance was visualised using short- and long-wave length ultraviolet light. Radioactivity on TLC-plates was measured by storage-phosphorous autoradiography using the Molecular Dynamics Phosphor Imager® (Sunnyvale, CA, USA). Analytical liquid chromatography (LC) separapump, a 166 UV detector and a radiodetector coupled in series). Data collection was performed using the Beckman System Gold Nouveau Chromatography Software Package. The columns used were; I: Vydac RP 300 Å high performance liquid chromatography (HPLC) column (Vydac, USA) 150 × 4.6 mm ID, 5 µm; II: Fast Desalting HR 10/10 fast protein liquid chromatography (FPLC) gel filtration column (Pharmacia Biotech, Uppsala, Sweden); System A (I): flow 1.5 mL/min, a = 20 mM triethylammonium acetate buffer (TEAA); blinear gradient 0-10% b 2-4 min, 10-30% b 4-9 min, 30-50 % b 9-15 min; λ = 254nm System B (II): flow 1.5 mL/min, a = H2O; b = phosphate buffered saline (PBS), 100%bElectrospray ionisation mass spectroscopy (ESI-MS) was performed using a Fisons Platform (Micromass, ManchesU.K.). Data were acquired in the negative ionisation mode by scanning from m/z 500 to m/z 1100. Samples were dissolvin 2.5 mM imidazole, 2.5 mM piperidine and aqueous 50% isopropyl alcohol.37 For purification procedures C18 SupelcleanLC-18 solid phase extraction (SPE) tubes 500 mg supplied by Supelco (Bellefonte, PA, USA) and NAP 5 columns (Amersham Pharmacia Biotech AB, Uppsala, Sweden) were used. 68Ga generation and recovery: 68Ga (T1/2 = 68 min, decay by β+ 68

270.8 d) in a generator-system (Cyclotron C., Obninsk, Russia) where 68Ge is attached to an ion-exchange column and the 68Ga is eluted in 0.1 M hydrochloric acid. 68Ge activity is 1850 MBq (50 mCi). 68Ga yield in 5 ml of 0.1M HCl is 50%. Breakthrough of 68Ge is < 0.01%. The shelf life of the generator is 2-3 years. Microwave activations were performed with a MicroWell 10 oven (Persomonomodal radiation. Cwere added to EDC (13 mg, 68 µmol) in H2O (250 µl) and stirred on ice for 30 min then warmed to room temperature togive DOTA-sulfo-NHS. A 100 fold excess of DOTA-NHS (6) solution was added dropwise to an oligonucleotide in 1M carbonate buffer (pH 9) and then cooled on ice. The mixture was left at room temperature for 10 hours. The reaction mixture was first purified by gel filtration on NAP 5 columns and 100 µL of 1M TEAA (Triethylammonium acetate Bwas added to 1 mL of the product eluate (H

uffer)

acuum

;

Ga – Labelling of oligonucleotides: Sodium acetate was added to the eluate from the Ge/ Ga-generator (36 mg to 1 mL)

2O). The mixture was then applied to a C-18 SPE column (Supelco), the column was washed with 50 mM TEAA (5 mL), 50 mM TEAA containing 5% acetonitrile (3 mL) and the DOTA-oligonucleotide was eluted with water:acetonitrile 50:50 (1 mL). The water-acetonitrile fraction was dried using a vcentrifuge. The products were analysed using electrospray ionisation mass spectrometry. Analysis in negative mode after direct infusion resulted in the following data: 1. DOTA-phosphodiester: MS (ESI-) m/z: 662.27 [M-8H]8-; 756.36 [M-7H]7-

882.91 [M-6H]6-. Reconstitution of the data gave M = 5303.71; 2. DOTA-phosphorothioate: MS (ESI-) m/z: 656.58 [M-8H]9-; 738.56 [M-7H]8-. Reconstitution of the data gave M = 5917.35; 3. DOTA-2′-O-methyl phosphodiester: MS (ESI-) m/z: 674.02 [M-6H]9-; 770.19 [M-8H]8-; 885.00 [M-7H]7-. Reconstitution of the data gave M = 6148.84

68 6868

to give a pH of approximately 5.5 and the mixture was vortexed well yielding the gallium acetate complex.36 Then DOTA-oligonucleotide (7) (10-100 nmol) was added and the mixture was transferred into a Pyrex glass vial for microwave activation for 1 min at 100 W. The reaction mixture was cooled to room temperature then 1 mL of 150 mM TEAA inwas added. The mixture was applied to a C-18 SPE-column (Supelco), which was then washed with 50 mM TEAA (1 mL),50mM TEAA containing 5% acetonitrile (1mL) and the product (

H2O

8) was eluted with ethanol: water 50:50 (1 mL) or water:acetonitrile 50:50 (1 mL). The reaction mixture was analysed by HPLC using Vydac RP and Fast Desalting HR 10/10 FPLC gel filtration columns.

ynthesis of 2′-O-methyl phosphodiester oligonucleotide: 2′-O-methyl phosphodiester oligonucleotide was synthesized on

om

San ABI synthesizer (392 DNA/RNA synthesizer; Applied Biosystems, CA) using 2′ methyl 5′ dimethoxytrityl cyanoethyl phosphoramidite RNA monomers (Glen Research, VI). After the synthesis the oligonucleotide was kept at 55 °C in concentrated ammonia overnight to cleave the oligonucleotide from the solid support and release protecting groups fr

53

chleicher & Schuell, 00,

s performed using HPLC with a reversed phase column (LiChroCART 250 x 10 mm,

er)

ybridisation in solution: The Ga-labelled antisense oligonucleotides (8

the bases. Subsequently the ammonia was evaporated and the residue was dissolved in water. The oligonucleotide was isolated from the solid support by filtration using 0.2 µm size filter (SGermany) and purified from the protecting groups using HPLC with an anionic exchange column (SynChropak AX3250 x 4.6 mm; Eichrom Technologies, Inc, Il,). HPLC conditions: A/B gradient, where solution A was KH2PO4 (0.05 M, pH = 5.6) in formamide:water (50:50, v/v). Solution B was solution A + (NH4)2SO4 (0.6 M). Gradient elution was done byincreasing the concentration of B. Desalting of the oligonucleotide waHypersil ODS 5 µm; Merck KGaA, Darmstadt, Germany). HPLC conditions: A/B gradient, where solution A was water and B was water:acetonitrile (50:50, v/v). The collected separated fraction was evaporated and dissolved in water. The product was characterized with a mass spectrometer (PE Sciex API 365 Triple Quadropole LC/ESI-MS/MS; Perkin-Elm

68H ) were obtained in 1 mL of water:acetonitrile

ge

(20

aturing PAGE gel, run at 200 V for 1 hour and 20 min. A 10 base

e

ONCLUSION

he 17-mer oligonucleotides were labelled with 68Ga radionuclide using microwave activation to speed up the the

the . All

CKNOWLEDGMENTS

his work was financially supported by Swedish Research Council, grant K 5107 and the Swedish Cancer Society.

EFERENCES

. Agrawal S, Kandimalla ER. Curr Cancer Drug Targets 2001; 1: 197-209.

harmaceutical Design 2002; 8: 1451-1466. ogy 2003; 30: 207-214.

6. , Haider N, J. N. J Nucl Cardiol 1999; 6: 345-356. 99; 10: 1271-1274.

Doignon I, David C, Crouzel C, Tavitian B.

10. 1997; 51: 1236-1240.

a Chem Scand 1999; 53: 508-512. 8: 1317-

(50:50) solvent. The concentration of the total oligonucleotide mass (labelled and unlabelled) was determined from standard calibration curves of the UV absorption at 254 nm. After evaporation of the solvent using a vacuum centrifu(Labconco CentriVap, USA), the 68Ga-antisense was dissolved in 1 x TES buffer (50 mM TRIS pH 8.0, 50 mM NaCl, 1 mM EDTA). The hybridisation samples were prepared as follows: a gradually increased concentration of the sense phosphodiester oligonucleotide (from 0.33 pmol to 80 pmol) was added to constant concentration of 68Ga-antisense pmol in 1 µL) and the total volume was adjusted to 10 µL with 1 x TES. As reference solutions 40 pmol of 68Ga-antisense oligonucleotide and sense oligonucleotide were used. All hybridisation mixtures were kept on ice and then at 40 oC for 10 min for hybridisation followed by gel electrophoresis. Gel electrophoresis was performed using a 20 % non-denpair (bp) DNA Step Ladder (Promega, Madison, WI) was used as a molecular weight marker. After electrophoresis the gel was stained using ethidium bromide to visualize the DNA and photographed under UV-light. Subsequently, the gel was exposed to a phosphor imaging plate for 12 hours, scanned using a Phosphorimager SITM device and analyzed using ImagQuant 5.1 software (Molecular Dynamics Inc, Sunnyvale, CA, USA). C Tcomplexation of the metal with the DOTA bifunctional chelator. The modifications of the oligonucleotides likeintroduction of hexylaminolinker either at the 3′- or 5′- end, substitution of non-bridging oxygens by sulphur or introduction of O-methyl group at sugar 2′ position did not influence the conjugation and radiolabelling result orhybridisation ability of the oligonucleotides. Labelled products were stable in water and ethanol for more than 4 hoursfour oligonucleotide counterparts retained their ability to hybridise in solution. The 68Ga-labelled oligonucleotides presented in this work will be used for the further in vitro and in vivo biological experiments. A T

R 12. Tavitian B. Q J Nucl Med 2000; 44: 236-255. 3. Younes CK, Boisgard R, Tavitian B. Current P4. Liu C-b, Liu G-z, Liu N, Zhang Yu-m, He J, Rusckowski M, Hnatowich DJ. Nucl Med Biol5. Watanabe N, Sawai H, Endo K, Shinozuka K, Ozaki H, Tanada S, Murata H, Sasaki Y. Nucl Med Biology 1999;

26: 239-243. Dewanjee MK

7. Kobori N, Imahori Y, Mineura K, Ueda S, Fujii R. Neuroreport 198. Stalteri MA, Mather SJ. Nucl Med Commun 2001; 22: 1171-1179. 9. Kühnast B, Dollé F, Terrazino S, Rousseau B, Loc'h FV, Hinnen F,

Bioconjugate Chem 2000; 11: 627-636. Hedberg E, Långström B. Acta Chem Scand

11. Hedberg E, Långström B. Acta Chem Scand 1998; 52: 1034-1039. 12. Yngve U, Hedberg E, Lövqvist A, Tolmachev V, Långström B. Act13. Pan D, Gamdhir S, Toyokuni T, Iyer M, Acharya N, Phelps ME, Barrio JR. Bioorg Med Chem Lett 1999;

54

14. C, Parkinson DR. Semin Oncol 1997; 24: 187-202. ng Z, Iyer RP, Yu D, Agrawal S. Biochem Pharmacol

16. Sproat B, Toulme JJ. Nucleic Acids Res 1995; 23: 3434-3440.

: 169-181. Jr SLW. NuclMedBiol 1991; 18: 813-816.

325.

24. s CF, de Lima JJ. J Inorg Biochem 2000; 79: 359-363.

9.

; 148: 249-253. 74-1981.

njugate Chem 1994; 5: 565-576. hem ASAP published on Web

35. da M, Lu L, Eriksson B, Watanabe Y, Bergström M, Långström B. Eur J Pharm

36. P. Analytical chemistry of gallium, Ann Arbor: Ann Arbor Science Publihers. Moscow,

37. , Griffey RH. Rapid Commun in Mass Spectrom 1995; 9: 97-102.

1320. Ho PT

15. Zhang R, Lu Z, Zhao H, Zhang X, Diasio RB, Habus I, Jia1995; 50: 545-556. Larrouy B, Boiziau C,

17. Boiziau C, Larrouy B, Sproat BS, Toulme JJ. Nucleic Acids Res 1995; 23: 64-71. 18. Crooke ST. Annu Rev Pharmacol Toxicol 1992; 32: 329-376. 19. Hermanson GT. Bioconjugate Techniques, 1996. 20. Hnatowich DJ. Int J appl Radiat Isotopes 1977; 2821. Otsuka FL, Welch MJ, Kibourne MR, Dence CS, Dilley WG, 22. Smith-Jones PM, Stolz B, Burns C, Albert R, Reist HW, Fridrich R, Mäcke HM. J Nucl Med 1994; 35: 317-23. Broan CJ, Cox, J.P.L., Craig, A.S., Kataky, R., Parker, D., Harrison, A., Randall, A.M., Ferguson, G. J Chem Soc

Perkin Trans 2 1991; 87-99. Prata MI, Santos AC, Geralde

25. Craig AS, Parker HA, Bailey NR. J Chem Soc, Chem Commun 1989; 1792. 26. Elander N, Jones JR, Lu S, Stone-Elander S. Chem Soc Rev 2000; 29: 239-2427. Ohlsson T, Vengtsson NE, Risman PO. J Microwave Power 1974; 9: 129. 28. Moi MK, Meares Cf, McCall JM, Cole WC, De Nardo SJ. Anal Biochem 198529. Heppeler ASF, H.R.Mäcke, E.Jermann, M.Behe, P.Powell, M.Hennig. Chem Eur J 1999; 5: 1930. Hnatowich DJ, Layne WW, Childs RL. Science 1983; 220: 613-615. 31. Chu BC, Orgel LE. Proc Natl Acad Sci USA 1985; 82: 963-967. 32. Dewanjee MK. Diag Oncol 1993; 3: 189-208. 33. Lewis MR, Raubitschek, A., Shively, J.E. Bioco34. Lewis MR, Kao, J.Y., Anderson, A.-L.J., Shively, J.E., Raubitschek, A. Bioconjug C

00/00/0000 2001; page est: 4.9. Wu F, Yngve U, Hedberg E, HonSci 2000; 10: 179-186. Dymov AM, Savostin A1968. Greig M

55

1 X= O-, X' = H

3 X= S-, X' = H

4 X= O-, X' = O - CH3

O

O

O Base

PX O

O

P

O

X

X'

OH2N

P

O

O-O

O

O

O Base

P-O O

OH2N

OH

X'

2 X= O-, X' = H

Figure 1. 1 - phosphodiester oligonucleotide with 5´- Hexylamine modification; 2 - phosphodiester

oligonucleotide with 3´ - Amino-C7 modification; 3 - phosphorothioate oligonucleotide with 5´- Hexylamine

modification; 4 - 2´-O-methyl phosphodiester oligonucleotide with 5´- Hexylamine modification

56

NN

N N

COOH

HOOC

HOOC

COOH

NN

N N

COOH

HOOC

HOOC

C NH(CH2)6RO

5

7

NN

N N

COOH

HOOC

HOOC

C O NS O-O

O

O

O

O

2) R(CH2)6NH2

6

1) EDC

NHO

SO

OONa

O

O

Scheme 1. Conjugation of DOTA (5) to the oligonucleotide bearing the hexylamine linker. R stands for four

oligonucleotide counterparts (see Table 1 and Scheme 1).

Table 1. Isolated decay corrected radiochemical yield (RCY) of the 68Ga labelling of oligonucleotides (1 - 4) complementary to human K-ras oncogene.

R RCY (%)

Phosphodiester

1 5′-CTA CGC CAC TAG CTC CA 50

2 CTA CGC CAC TAG CTC CA-3′ 42

Phosphorothioate

3 5′-CTA CGC CAC TAG CTC CA 30

2′-O –methyl Phosphodiester

4 5′-CTA CGC CAC TAG CTC CA 52

57

N

N N

N

COOH

C

O

NH(CH2)6 R

68Ga(III)

COOH

CH3COONa, pH=5.5

68Ga(III) in 0.1 M HCl

N

N N

N 1 min, 100WC

O

NH(CH2)6 RHOOC HOOC

HOOCHOOC

7 8 Scheme 2.

68Ga labelling of DOTA – oligonucleotides 7. For R see Table1 and Scheme 1.

Autoradiography

Hybrid

30 bp20 bp

10 bpSense

1 2 3 4 5 6 7 8 9 10 11 12

PAGE

HybridAntisense

1 2 3 4 5 6 7 8 9 10 11

1 2 3 4 5 6 7 8 9

AntisenseHybrid

AntisenseHybrid

A B

C

Figure 2. Concentration dependent hybridisation of 17-mer antisense phosphorothioate oligonucleotide (20 pmol in 1 µL)

to the complementary 17-mer sense phosphodiester oligonucleotide in solution. The antisense:sense concentration ratios in

the lanes are as follows: 1) 1:1/60, 2) 1:1/30, 3) 1:1/15, 4) 1:1/5, 5) 1:1/2, 6) 1:1, 7) 1:2, 8) 1:3, 9) 1:4 and the references are

sense oligonucleotide (lane 10), antisense oligonucleotide (lane 11) and molecular weight marker (lane 12). A) PAGE

picture of the concentration dependent hybridization study; B) Autoradiography of the polyacrylamide gel (A); C) The

scaled up region of interest.

58

Paper III

59

Biodistribution of 68Ga-labelled phosphodiester, phosphorothioate, and 2'-O-methyl

phosphodiester oligonucleotides in normal rats

Gabor Lendvaia, b, Irina Velikyana, c, Mats Bergströma, d, Daniel Laryeaa, Maria Väliläe, f, Satu

Salomäkie, f, Bengt Långströma, c, Anne Roivainena, e.

aUppsala Research Imaging Solutions AB, S-751 85 Uppsala, Sweden bDepartment of Medical Sciences, Uppsala University Hospital, S-751 85 Uppsala, Sweden cDepartment of Organic Chemistry, Institute of Chemistry, Uppsala University, S-751 21 Uppsala,

Sweden dDepartment of Pharmaceutical Biosciences, Uppsala Biomedical Centre, S-751 24 Uppsala, Sweden eTurku PET Centre, Turku University Central Hospital, FIN-20521 Turku, Finland fDepartment of Chemistry, University of Turku, FIN-20014 Turku, Finland

Address for correspondence and reprint requests:

Mats Bergström Uppsala Research Imaging Solutions AB S-751 85 Uppsala Sweden Tel: 46-18-471 5374 Fax: 46-18-471 5370

E-mail: Mats.Bergstrom@pet.uu.se

60

Abstract Antisense oligonucleotides may hybridise with high selectivity to mRNA sequences allowing monitoring of gene expression or inhibition of the manifestation of altered genes inducing diseases. As part of the development of positron emission tomography methods, 17 mer antisense phosphodiester, phosphorothioate and 2'-O-methyl phosphodiester oligonucleotides specific for point mutationally activated human K-ras oncogene were labelled with 68Ga radionuclide using a chelator coupled to the probe. Hybridisation in solution and non-denaturing PAGE with a subsequent exposure of the gels was performed to verify the hybridisation properties after labelling. The biodistribution was studied in male Sprague-Dawley rats by injecting 2 MBq of 68Ga-oligonucleotides via the tail vein and measuring the organ radioactivity concentration after 20, 60, 120 min or using whole-body autoradiography with 10 MBq 68Ga-oligonucleotide and 20 min incubation time. Control experiments were performed with 68GaCl3 and 68Ga-chelator complex. The results revealed that labelling did not change the hybridisation abilities. The biodistribution pattern depended on the nature of oligonucleotide backbone. Bone marrow, kidney, liver, spleen and urinary bladder were the five organs of highest uptake with each oligonucleotide. Phosphodiester accumulated highly in the liver, whereas high kidney uptake dominated the phosphorothioate and 2'-O-methyl phosphodiester patterns. This study supplies a base for the further development of 68Ga-labelled oligonucleotides as pharmacokinetic tools and a potential future use for in vivo imaging of gene expression. Keywords: Antisense oligonucleotides; Hybridisation in solution; Distribution in vivo; Whole-body autoradiography; Positron emission tomography; 68Ga

1. Introduction In today’s molecular medicine, the hypothesis is that many diseases result from altered patterns of gene expression, which converts cells into the “sick” phenotypes. For example, alterations in the cellular genes, which directly or indirectly control cell growth and differentiation, are generally considered the main cause of cancer. Members of the ras gene family, H-ras, K-ras and N-ras, are thought to be involved in normal cell growth and maturation. However, a point mutation induced by carcinogens or environmental factors causing an amino acid alteration at one of the three critical positions in the protein results in conversion to a form that is involved in the formation of tumours. Approximately 10-20 % of all human tumours have a mutation in one of the ras oncogenes; however over 90 % of pancreatic adenocarcinomas, about 50 % of adenocarcinomas of colon, lung adenocarcinomas, thyroid carcinomas, and a large fraction of haematological malignancies have been found to be associated with point mutationally activated ras oncogenes (Bos, 1989). The growing number of identified genes behind various human diseases has promoted the idea of using this information for treatment by “knocking down” their expressions. Antisense oligonucleotides may be such a molecule that could reach this goal on the basis of arresting the mRNA in the cell and thereby preventing it from being translated into a protein. These short, synthetic nucleic acids manifest the inhibition effect on the gene expression by hybridizing with their complementary “sense” sequences in mRNA through Watson-Crick base-pairing. The therapeutic utility of antisense nucleic acids was already suggested by Stephenson following the first encouraging experiments (Zamecnik, Stephenson, 1978) which entailed the testing of several antisense oligonucleotides in different in vitro model systems. However, the most intensive research concerning the in vivo applications has been conducted in the past decade. Successful “knock-down” of gene expression has been demonstrated in cell cultures (Monia et al., 1992; Standifer et al., 1995; Hanecak et al., 1996; Lai et al., 1997), and in small animal experiments (Gray et al., 1993; Nesterova, Cho-Chung, 1995; Bilsky et al., 1996; Edsall et al., 1996; Lai et al., 1996). A number of human trials have started with the aim of inhibiting disease-related genes in viral infections (Crooke et al., 1994; Amado et al. 1999), inflammatory diseases (Hartmann et al., 1997), and cancer (Akhtar, Agrawal, 1997; Ho, Parkinson, 1997, Cowsert, 1997). There is at least one antisense drug available, Vitravene (Marwick, 1998), and others are in clinical trials (Opalinska, Gewirtz, 2002). For effective inhibition, the antisense oligonucleotides should be specific and accessible to the target, stable in vivo, and possess minimal non-specific interactions. The fact that natural oligonucleotides with their phosphodiester backbone proved unstable to serum and cellular nucleases (Sands et al., 1994) led to the search for analogues with improved stability. The phosphorothioate oligonucleotide, which is resistant to metabolism (Crooke, et al., 1996; Agrawal, Iyer, 1997; Agrawal, et al., 1997a; Henry et al., 1999), is the most studied analogue. Methylphosphonate, phosphoramidate, 2'-O-methyl RNA, morpholino oligonucleotide, mixed-backbone oligonucleotide and peptide nucleic acid (PNA) are other modified alternatives, which have been tested (Younes et al., 2002) to find good candidates. Obviously, any chemical modification of the oligonucleotide could be expected to have effect on the sensitivity to nucleases, elimination, membrane passage, protein binding etc. and thereby causes major alterations in the pharmacokinetics, essential for its biological activity. A method allowing pharmacokinetic measurements of oligonucleotides in vivo would be of value to characterise

61

one important aspect of antisense oligonucleotides having a therapeutic potential (Stein, Cheng, 1993; Tavitian et al., 1998). Positron emission tomography (PET) is a most advanced technology allowing to image biological processes (e.g. molecular interactions) in vivo and to obtain knowledge about the fate of drugs and substances pharmacokinetics, pharmacodynamics, and metabolism (Langstrom, Bergstrom, 1995; Jones, 1996; Langstrom et al., 1999; Paans, Vaalburg, 2000; Eckelman, 2002). Therefore, it is logical to suggest this technology as a tool in the development of antisense therapy with respect to selective accumulation in target organs or for the assessment of drug concentration in organs where side effects may be induced. Additionally, PET might be able to offer a non-invasive diagnostic tool to analyse gene expression, converting in vitro (situ) hybridisation to in vivo hybridisation. To be able to record antisense oligonucleotides in vivo with PET, methods were developed by which different oligonucleotides and other analogues may be labelled using 18F, 76Br, 125I

and 68Ga as radionuclides (Kuhnast et al., 2000;

Yngve et al., 1999; Velikyan, Langstrom, 2002). In our previous publication, we examined 76Br-labelled antisense oligonucleotides of different lengths (Wu et al., 2000). Following up and extending the used preclinical methods, we have in the present study in vitro and ex vivo investigated

68Ga-labelled 17 mer phosphodiester, phosphorothioate and 2'-O-

methyl phosphodiester antisense oligonucleotides specific for point mutationally activated K-ras oncogene to assess the effect of labelling on the hybridisation properties and to obtain information about biodistribution in rats. We chose the K-ras oncogene as a model gene towards in vivo imaging because of its significance in cancer biology and oncology. 2. Methods 2.1. Oligonucleotides Phosphodiester and phosphorothioate oligodeoxynucleotides were purchased from Scandinavian Gene Synthesis AB, Köping, Sweden. 2'-O-methyl phosphodiester oligoribonucleotide was synthesised and purified at the Department of Organic Chemistry, University of Turku, Turku, Finland. Antisense oligonucleotides of 17 mer with a hexylamine linker in the 5'-position, targeting codon 12 point mutation, were designed based on the sequence of human K-ras oncogene exon 1 (Kita et al., 1999). According to the GenBank search (Blast Sequence Similarity Searching; NCBI), the selected sequence did not have any complementary sequence in normal rat tissue. Oligonucleotides corresponding to mutation specific sense sequence were utilised as controls. The antisense and sense sequences used for this work were 5'-CTACGCCACTAGCTCCA-3' and 5'-TGGAGCTAGTGGCGTAG-3', respectively. 2.2. 68Ga-labelling of oligonucleotides Commercially available chemicals were used without further purification: Sodium acetate (99.995%, Aldrich) and doubly distilled HCl (Aldrich) for the labelling chemistry; 1-Ethyl-3-(3-dimethylaminopropyl)carbodiimide (EDAC) (Sigma), N-Hydroxysulfosuccinimide (Sulfo-NHS) (Sigma) and 1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetic acid (DOTA) (Macrocyclics, USA) for the conjugation chemistry. Deionised water (18.2MΩ) was used in all reactions. DOTA (32 mg, 66 µmol) and Sulfo-NHS (14 mg, 65 µmol) in H2O (250 µl) were added to EDAC (13 mg, 68 µmol) in H2O (250 µl) and kept on ice for 30 min, then warmed to room temperature to yield DOTA-sulfo-NHS (Hermanson, 1996; Yngve, 2001). DOTA-NHS-solution in 100-fold excess compared to oligonucleotide concentration was added to the oligonucleotide in carbonate buffer 1 M, pH 9 and cooled on ice. The mixture was left at room temperature for 10 h (Yngve, 2001; Lewis et al., 1994). The reaction mixture was first purified by gel filtration through NAP 5 columns (Sephadex G-25; Amersham Pharmacia Biotech AB, Uppsala, Sweden). Then 100 µl of 1M Triethylammonium acetate (TEAA) was added to 1 ml of the product eluate (H2O) and the mixture was applied to a C-18 SPE column C18 (Supelclean LC-18 solid phase extraction (SPE) tubes; Supelco, Bellefonte, PA). The column was washed with 50 mM TEAA (5 ml), then with 50 mM TEAA containing 5% acetonitrile (3 ml), and the DOTA-oligonucleotide was eluted with water:acetonitrile 50:50 (1 ml). The water:acetonitrile fraction was evaporated, redissolved in water and stored in refrigerator until use. 68Ga (T1/2 = 68 min) was obtained from a 68Ge (T1/2 = 270.8 d) generator-system (Cyclotron C, Obninsk, Russia) by elution of the anion-exchange column wtih 0.1 M hydrochloric acid. Sodium acetate was added to the eluate from the 68Ge/68Ga-generator (36 mg to 1 ml) to give a pH of approximately 5.5 (Dymov, Savostin, 1970). DOTA-oligonucleotide (10-100 nmol) was then added and the mixture incubated at +100 ºC for 10 min. The reaction mixture was cooled to room temperature, and 1 ml of 150 mM TEAA was added. The mixture was applied to C-18 SPE-column. The column was washed first with 50 mM TEAA (1 ml), then with 50 mM TEAA containing 5% acetonitrile (1 ml), and the product was eluted with ethanol:water 50:50 (1 ml). The ethanol was evaporated from the 68Ga-labelled oligonucleotide using a vacuum centrifugal concentrator (Labconco, USA). Analytical liquid chromatography (LC) separations were performed with a Beckman (Fullerton, CA, USA) System consisting of a 126 pump, a 166 UV detector and a radiodetector coupled in series. The used columns were: (I) Vydac RP 300 Å HPLC column (Vydac, USA) 150 x 4.6 mm ID, 5 µm, and (II) Fast Desalting HR 10/10 fast protein liquid chromatography (FPLC) gel filtration column (Pharmacia Biotech, Uppsala, Sweden). System A (I): flow 1.5 mL/min, a = 20 mM TEAA; b = 100 % acetonitrile (MeCN), linear gradient 0-10 % b 2-4 min, 10-30 % b 4-9 min, 30-50 % b 9-15 min; λ = 254 nm. System B (II): flow 1.5 mL/min, a = H2O; b = phosphate buffered saline (PBS),

62

100 % b 0-15 isocratic; λ = 280 nm. For data collection, the Beckman System Gold Nouveau Chromatography Software Package was applied. 2.3. In vitro study: hybridisation in solution The 68Ga-labelled antisense oligonucleotides (phosphodiester, phosphorothioate and 2'-O-methyl phosphodiester) were obtained in 50% ethanol. After evaporation of the solvent using a vacuum centrifugal concentrator (Labconco, USA), the 68Ga-antisense was dissolved in 1xTES (50 mM TRIS, pH 8.0, 50 mM NaCl, 1 mM EDTA). The hybridisation mixture was prepared as follows: a gradually increased amount of sense oligonucleotide (from 0.33 pmol to 80 pmol) was added to a constant concentration of 68Ga-antisense (20 pmol in 1 µl). The total volume was adjusted to 10 µl with 1xTES. As reference solutions, 40 pmol/µl of 68Ga-antisense and unlabelled sense oligonucleotides were used. All mixtures were kept on ice until commencement of hybridisation at 40 oC for 10 min and subsequent gel electrophoresis. The gel electrophoresis was performed with 20 % non-denaturing PAGE gel and run at 200 V for 80 min. A 10 base-pair DNA Step Ladder (Promega, Madison, WI, USA) was applied as a molecular weight marker. After electrophoresis, the gel was stained using ethidium bromide to visualise the DNA and was photographed under UV-light. Subsequently, the gel was exposed to a phosphor imaging plate for 12 h, scanned with a Phosphorimager SITM and analysed using Image Quant 5.1 software (Molecular Dynamics Inc, Sunnyvale, CA, USA). 2.4. Ex vivo studies 2.4.1. Animals Male Sprague-Dawley rats (age 8-12 weeks, weight approx. 360 g) were used. The animals permission was granted by the local Research Animal Ethics Committee C 241/98. The animals were kept at a constant temperature (20 oC) and humidity (50 %) in a 12 h light-dark cycle, and given free access to laboratory animal food and water. Two rats were used for each time point and the experiments were repeated three times. 2.4.2. Organ distribution The 68Ga-antisense oligonucleotides were injected into the tail vein of the rats as a bolus of 2 MBq/rat with 700 µl saline as the vehicle. The animals were not anaesthetized. In separate experiments, the same amount of 68GaCl3 or DOTA-conjugated 68Ga (pH adjusted to 6.0 with 1 M NaOH) was given. The injected amount versus the body weight of the rats (expressed as mean ± S.D.) was the following: phosphodiester 35.5 ± 43.0 µg/kg; phosphorothioate 33.9 ± 20.7 µg/kg; 2'-O-methyl phosphodiester 51.8 ± 26.6 µg/kg; 68GaCl3 4.39 ± 0.085*10-6 µg/kg; 68Ga-DOTA 0.62 ± 0.41 µg/kg. After 20, 60 or 120 min, the rats were sedated with CO2-O2 mixture and samples of blood, heart, lung, liver, kidney, pancreas, spleen, intestine, adrenal gland, urinary bladder, testis, brain, muscle, bone, bone marrow, skin, and parotid gland were taken, weighed, and measured in a calibrated well-counter for 30 s. The radioactivity concentration was expressed as SUV (Standardized Uptake Values).SUV=(organ activity/organ weight)/(total given radioactivity/rat body weight) 2.4.3. Whole-body autoradiography The animals were injected via the tail vein with 10 MBq of radiolabelled antisense oligonucleotides with 700 µl saline as the vehicle. The injected amount versus the body weight of the rats (expressed as mean ± S.D.) was the following: phosphodiester 41.5 ± 2.3 µg/kg; phosphorothioate 106.5 ± 17.5 µg/kg; 2'-O-methyl phosphodiester 85.7 ± 4.9 µg/kg. After 20 min, the animals were sacrificed with CO2-O2 mixture and embedded in carboxymethyl cellulose (CMC) in a mould and frozen in liquid hexane cooled by dry ice (−70 oC) for 20 min to accomplish rapid freezing. The resulting CMC blocks were mounted in a cryomicrotome (PMV Cryo-microtome, Stockholm, Sweden) where 50 µm thick sections were cut and adhered to tape (d’Argy et al., 1984). The sections were heat-dried for a few minutes and then exposed to phosphor imaging plates for 15 h. The scanning operation, image display and analysis were performed using the software mentioned under 2.3. In a separate experiment, 68GaCl3 or DOTA-conjugated 68Ga (pH adjusted to 6.0 with 1 M NaOH) were given at the same amount and the animals were killed 1 h after the injection. The injected amount versus the body weight of the rats (expressed as mean ± S.D.) was with 68GaCl3 3.82 ± 0.077*10-6 µg/kg, and with 68Ga-DOTA 4.58 ± 0.18 µg/kg. 3. Results 3.1. Hybridisation The stained PAGE gels showed that the various 68Ga-labelled antisense oligonucleotides − phosphodiester, phosphorothioate and 2'-O-methyl phosphodiester − retained their potency and were able to hybridise. The duplex formation between labelled antisense and the complementary sense oligonucleotides was proportional to the increase of sense oligonucleotide (Fig. 1/a, b, c). The antisense and sense oligonucleotide bands appearing at concentrations when they were in excess corresponded to the reference oligonucleotide bands. The autoradiogram of the gels were in accordance with the Polaroid picture, since the radioactivity of 68Ga-labelled antisense oligonucleotide decreased and reappeared in the

63

hybrid showing an increase as more and more sense became available. The line of hot spots of the labelled antisense correlated with the reference. Since the hybrid became double-stranded nucleic acid similar to molecular weight marker, its band appeared at 17 bp heights on the gel, whereas the single-stranded antisense and sense used for control did not emerge at the same heights because of their folded structure. 3.2. Organ distribution The measurement of the organ raioactivity after i.v. administration of labelled phosphodiester showed the highest values in the liver (average SUV of all three time points is 6.5), followed by the urinary bladder, bone marrow and spleen (Fig. 2). Uptake in the kidney was predominant in the phosphorothioate and 2'-O-methyl phosphodiester distribution patterns with an average SUV of 30 and 12, respectively (Fig. 3-4). Bone marrow, kidney, liver, spleen and urinary bladder were among the five organs with the highest SUV values in each oligonucleotide distribution pattern (Fig. 5). The SUV values of the rest of the organs were below 1.0. The lowest uptake was observed in the brain with an SUV of 0.03. On the average, phosphorothioate oligonucleotide had higher SUV values in the tissue samples compared to that of phosphodiester (3.4 fold) and 2'-O-methyl phosphodiester (2.7 fold). As references, 68GaCl3 and DOTA-conjugated 68Ga distribution studies were performed. The 68GaCl3 was retained very prominently in all organs; the SUV ranged between 3.7 and 0.5 for most of the tissues; in the brain, however, it was 0.09 (data not shown). The SUVs of bone marrow and lung placed second and third after blood measured 3.5 and 1.7, respectively. In contrast, 68Ga-DOTA did not accumulate extensively in the organs; in 14 out of 17 samples, no SUV value exceeded 0.37. However, the two highest radioactive concentrations were observed in the bone marrow and in the liver with an average SUV of 5.8 and 3.9, respectively, followed by the urinary bladder and kidney (Fig. 6). 3.3. Whole-body autoradiography The highest uptake of radioactivity of phosphodiester oligonucleotide was observed in the liver and urinary bladder (Fig. 7/a). With radioactivity decreasing in magnitude, kidney and spleen were the next tissues, followed by the lung, and finally skin and bone. For radioactive phosphorothioate oligonucleotide, labelling of the kidney cortex was predominant together with the urinary bladder. The radioactive distribution pattern in the liver was uneven – a high uptake in some areas but low in other areas. The low magnitude of uptake in the liver corresponded almost with that in the parotid gland, the lung and spleen. Uptake in skin was lowest, though higher than the background radioactivity (Fig. 7/b). 2'-O-methyl phosphodiester showed similar distribution pattern to phosphorothioate, but the uptake in the kidney cortex was noticeably less (Fig. 7/c). One hour after the 68GaCl3 injection, the most pronounced uptake was observed in the liver and spleen. Radioactivity in the lung was also considerable. Uptake in the kidney, bone marrow, lymph nodes and vessels, and skin was much lower but well above the background. In contrast, the DOTA-conjugated 68Ga was slightly taken up by most of the organs. The highest uptake was seen in the urinary bladder and some parts of the liver and lung (small pockets of radioactivity distributed all over in these organs). According to decreasing magnitude of radioactivity, the next tissue was kidney cortex. Uptake in the liver and lung, spleen, skin and bone marrow was slightly above the background. The experiment also revealed that the contents in the stomach and some parts of the intestine contained radioactivity to a high extent. 4. Discussion The purpose of the present work was to acquire knowledge of the distribution in normal tissues of radionuclide-labelled oligonucleotides having different modifications as the first step in the strategy aimed to develop methods for quantitative measurement of antisense oligonucleotides in vivo in humans using PET. For the possible utilisation of these methods either in antisense therapy or for imaging of gene expression, it is important to obtain information about the quantitative accumulation in the target organs versus non-targeted tissues, as well as information about the metabolism, elimination profile and rate. Besides, the oligonucleotides have been demonstrated to be toxic (Sarmiento et al., 1994; Agrawal et al., 1997b). In order to understand the possible organ specific toxicity it might be important to know the concentration reached in the organ. Furthermore, the chemical modification made in the backbone for better stability, or even the labelling method, might result in different tissue distribution and kinetic properties. For example, the radiolabelled antisense oligonucleotides might alter in properties from the unlabelled ones used as drug. Before the PET imaging method can be applied in humans, it is essential to further evaluate the behaviour of the labelled oligonucleotide analogues in a better perspective. In our previous works (Yngve et al., 1999; Wu et al., 2000), 76Br-labelled antisense oligonucleotides were used, since 76Br has a rather long (16.2 h) half-life that makes this radionuclide suitable for long-lasting applications. However, it is not advantageous to administer to humans because it decays only by 55 % via positrons and additionally generate high-energy gamma rays. The

68Ga radionuclide, (decaying by 89 % via positrons and by 11 % via EC, half-life 68 min), is an

alternative and its half-life might be enough to follow biological processes for 3-4 hours (Lubberink et al., 2002).

64

Furthermore, using a 68Ge/68Ga generator instead of a cyclotron, it is produced more easily and using a metal chelator, the labelling procedure is simpler; in addition, 68Ga-labelled oligonucleotides are more stable compounds (Roivainen et al., submitted December 2002; Velikyan et al., submission is in progress). In connection with labelling, the experiments revealed that 68Ga labelling using DOTA as a chelator did not alter the hybridisation abilities of either oligonucleotides, but the amount hybridised was concentration dependent. When performing similar hybridisation in solution experiments with 76Br-labelled 30 mer phosphodiester and phosphorothioate oligonucleotides, only the phosphodiester retained its hybridisation capacity (Yngve, 2001). The previous negative results might be methodological artefacts of separating double-stranded nucleic acid from single-stranded one. The distribution pattern in almost all tissues seemed to vary with the nature of the oligonucleotide backbone. Since the oligonucleotides studied do not have any biological target in rats, their tissue distribution reflects their non-specific interactions and elimination. Tavitian et al. examined in vivo the fate of 18F-labelled oligonucleotides of the same backbone modifications used in the present study in baboons. Five minutes after injection, radioactivity uptake was found to be high in the heart, liver and kidney for all three oligonucleotides, only the presence of phosphodiester in the urinary bladder and the presence of phosphorothioate in the spleen were different. One hour after i.v. administration, phosphodiester was found in the bladder, and phosphorothioate showed high amounts of radioactivity in the liver and kidney, while 2'-O-methyl RNA appeared to be intermediate between those two, showing high bladder and noticeable kidney parenchyma radioactivity (Tavitian et al., 1998). Our results are in agreement with these observations except the intense presence of phosphodiester in the liver. Sands et al. compared the biodistribution of 3H-labelled phosphodiester and phosphorothioate oligonucleotides in mice, but their results differ from our phosphodiester distribution data (Sands et al., 1994). Their results showed that tissues did not take up phosphodiester prominently: 1 min after injection very little intact oligonucleotide was found in any organ; however, its relative initial concentrations were, in a decreasing order of radioactivity, dominated by kidney, blood, heart, liver, lung and spleen. The discrepancies in the data might be explained by the difference in metabolism of the three dissimilar species used, since Sands et al. reported very high radioactivity in the spleen, most likely because of the aggregation of metabolites. Nevertheless, phosphorothioate oligonucleotide exhibited similar rapid accumulation in all tissues, especially in the kidney showing an increased uptake over time. Organ to blood ratios at 2 h were in their study 5:1 for all organs, 84:1 for the kidney, and 20:1 for the liver, while the corresponding ratios in our data were 7:1 for all organs, 84.3:1 for the kidney, and 10.8:1 for the liver. Raynaud et al. reported similar ratios in mice (Raynaud et al, 1997). In the previous study, the three highest uptakes were also observed in the kidney, liver and spleen with 20 mer phosphorothioate oligonucleotide (Wu et al., 2000). Others have also published high kidney and liver accumulation of phosphorothioate oligonucleotides in rats (Cosum et al., 1993; Zhang et al., 1995; Agrawal, Iyer, 1997; Philips et al., 1997; Zhao et al., 1998). Moreover, the organ pattern of uptake seems to be similar across species (mouse, rat, monkey), which provides a level of confidence that preclinical animal models can be predictive towards exposure in the clinic (Geary et al., 1997, 2001). As complement to quantitative distribution, whole-body autoradiography as a technique provides qualitative images without missing organs or more detailed information about the tissues. A prominent accumulation of phosphorothioate oligonucleotide in the kidney cortex was observed, which has been reported to be due to significant uptake in the proximal tubule and the glomerular capsule epithelia (Oberbauer et al., 1995; Buttler et al., 1997). The whole-body autoradiography revealed relative high phosphorothioate uptake in the parotid gland, which could have been easily missed. Philips et al. found the salivary gland among the high accumulating organs in rats, as well (Philips et al., 1997). The relatively high uptake of bone marrow with phosphorothioate oligonucleotide has also been reported (Cossum et al., 1993; Buttler et al., 1997; Zhao et al., 1998), and it seems to be related to megakaryocytes, myeloid and macrophage cells, rather than to lymphocytes. This might be associated with the toxicological effect of antisense oligonucleotides (Sarmiento et al., 1994), and even with the fact that gallium resembles iron (Koizumi et al., 1990). For the presentation of radioactivity concentration, the SUV nomenclature commonly applied in the PET field was used instead of % ID/g. The SUV denotes tissue radioactivity in relation to an average body concentration, so that the dependence of animal weight can be avoided. In addition, the concentration of the compound itself in an organ, expressed in mg/kg, can easily be calculated as the total administered amount in mg/kg multiplied by the organ’s SUV value. For the study, the K-ras oncogene was chosen as the model gene because of its significance in cancer biology and oncology. The ras oncogene mutations are found in a variety of human tumour types but not in normal tissue, thus making ras mRNAs suitable targets for following the progression of malignancy using a non-invasive imaging technique with radiolabelled antisense oligonucleotide tracers. Before this might be performed, however, further tests are required, mostly to investigate the in vivo behaviour of these oligonucleotides and to find out whether they have specific affinity to tumours (Roivainen et al., submitted December 2002). 5. Conclusions The antisense phosphodiester, phosphorothioate and 2'-O-methyl phosphodiester oligonucleotides specific for point mutationally activated human K-ras retained their hybridisation abilities after 68Ga labelling and showed different organ distribution reflecting their various metabolism and non-specific binding in rats. However, the high bone marrow uptake as well as the function of 68Ga in the organism remains to be clarified.

65

Acknowledgements This work was supported by a grant from Nordic Cancer Union, and by European Union COST Action B12. References Agrawal, S., Iyer, R.P., 1997. Perspectives in antisense therapeutics. Pharmacol. Ther. 76, 151-160. Agrawal, S., Jiang, Z., Zhao, Q., Shaw, D., Cai, Q., Roskey, A., Channavajjala, L., Saxinger, C., Zhang, R., 1997a. Mixed-backbone oligonucleotides as second generation antisense oligonucleotides: in vitro and in vivo studies. Proc. Natl. Acad. Sci. 94, 2620-2625. Agrawal, S., Zhao, Q., Jiang, Z., Oliver, C., Giles, H., Heath, J., Serota, D., 1997b. Toxicologic effects of an oligodeoxynucleotide phosphorothioate and its analogs following intravenous administration in rats. Antisense Nucleic Acid Drug. Dev. 7, 575-584. Akhtar, S., Agrawal, S., 1997. In vivo studies with antisense oligonucleotides. Trends Pharmacol. Sci. 18, 12-18. Amado, R.G., Mitsuyasu, R.T., Zack, J.A., 1999. Gene therapy for the treatment of AIDS: animal models and human clinical experience. Front. Biosci. 4, 468-475. Bilsky, E.J., Bernstein, R.N., Hruby, V.J., Rothman, R.B., Lai, J., Porreca, F., 1996. Characterization of antinociception to opioid receptor selective agonists after antisense oligodeoxynucleotide-mediated "knock-down" of opioid receptor in vivo. J. Pharmacol. Exp. Ther. 277, 491-501. Bos J.L., 1989. ras oncogenes in human cancer: a review. Cancer Res. 49, 4682-4689. Butler, M., Stecker, K., Bennett, C.F., 1997. Cellular distribution of phosphorothioate oligodeoxynucleotides in normal rodent tissues. Lab. Invest. 77, 379-388. Cossum, P.A., Sasmor, H., Dellinger, D., Truong, L., Cummins, L., Owens, S.R., Markham, P.M., Shea, J.P., Crooke, S., 1993. Disposition of the 14C-labeled phosphorothioate oligonucleotide ISIS 2105 after intravenous administration to rats. J. Pharmacol. Exp. Ther. 267, 1181-1190. Cowsert, L.M., 1997. In vitro and in vivo activity of antisense inhibitors of ras: potential for clinical development. Anticancer Drug Des. 12, 359-371. Crooke, S.T., Graham, M.J., Zuckerman, J.E., Brooks, D., Conklin, B.S., Cummins, L.L., Greig, M.J., Guinosso, C.J., Kornbrust, D., Manoharan, M., Sasmor, H.M., Schleich, T., Tivel, K.L., Griffey, R.H., 1996. Pharmacokinetic properties of several novel oligonucleotide analogs in mice. J. Pharmacol. Exp. Ther. 277, 923-937. Crooke, S.T., Grillone, L.R., Tendolkar, A., Garrett, A., Fratkin, M.J., Leeds, J., Barr, W.H., 1994. A pharmacokinetic evaluation of 14C-labeled afovirsen sodium in patients with genital warts. Clin. Pharmacol. Ther. 56, 641-646. d'Argy, R., Ullberg, S., Stalnacke, C.G., Langstrom, B., 1984. Whole-body autoradiography using 11C with double-tracer applications. Int. J. Appl. Radiat. Isot. 35, 129-134. Dymov, A.M., Savostin, A.P., 1970. Analytical Chemistry of Gallium. Ann Arbor Science Publishers. Distributed by Keter Inc., New York, N.Y. Eckelman, W.C., 2002. Accelerating drug discovery and development through in vivo imaging. Nucl. Med. Biol. 29, 777-782.

66

Edsall, S.A., Knapp, R.J., Vanderah, T.W., Roeske, W.R., Consroe, P., Yamamura, H.I., 1996. Antisense oligodeoxynucleotide treatment to the brain cannabinoid receptor inhibits antinociception. Neuroreport. 7, 593-596. Geary, R.S., Yu, R.Z., Levin, A.A., 2001. Pharmacokinetics of phosphorothioate antisense oligodeoxynucleotides. Curr. Opin. Investig. Drugs. 2, 562-573. Geary, R.S., Leeds, J.M., Henry, S.P., Monteith, D.K., Levin, A.A., 1997. Antisense oligonucleotide inhibitors for the treatment of cancer: 1. Pharmacokinetic properties of phosphorothioate oligodeoxynucleotides. Anticancer Drug Des. 12, 383-393. Gray, G.D., Hernandez, O.M., Hebel, D., Root, M., Pow-Sang, J.M., Wickstrom, E., 1993. Antisense DNA inhibition of tumor growth induced by c-Ha-ras oncogene in nude mice. Cancer Res. 53, 577-580. Hanecak, R., Brown-Driver, V., Fox, M.C., Azad, R.F., Furusako, S., Nozaki, C., Ford, C., Sasmor, H., Anderson, K.P., 1996. Antisense oligonucleotide inhibition of hepatitis C virus gene expression in transformed hepatocytes. J. Virol. 70, 5203-5212. Hartmann, G., Bidlingmaier, M., Eigler, A., Hacker, U., Endres, S., 1997. Cytokines and therapeutic oligonucleotides. Cytokines Cell. Mol. Ther. 3, 247-256. Henry, S.P., Templin, M.V., Gillett, N., Rojko, J., Levin, A.A., 1999. Correlation of toxicity and pharmacokinetic properties of a phosphorothioate oligonucleotide designed to inhibit ICAM-1. Toxicol. Pathol. 27, 95-100. Hermanson, G.T., 1996. Bioconjugate Techniques. Academic Press, San Diego. Ho, P.T., Parkinson, D.R., 1997. Antisense oligonucleotides as therapeutics for malignant diseases. Semin. Oncol. 24, 187-202. Jones, T., 1996. The imaging science of positron emission tomography. Eur. J. Nucl. Med. 23, 807-813. Kita, K., Saito, S., Morioka, C.Y., Watanabe, A., 1999. Growth inhibition of human pancreatic cancer cell lines by anti-sense oligonucleotides specific to mutated K-ras genes. Int. J. Cancer. 80, 553-558. Koizumi, K., Uchiyama, G., Araki, T., Hihara, T., Ogata, H., Monzawa, S., Kachi, K., Onishi, H., Oba, H., Toyama, K., 1990. Visualization of the bone/bone marrow of lower extremities in Ga-67 whole-body images. Ann. Nucl. Med. 4, 15-18. Kuhnast, B., Dolle, F., Terrazzino, S., Rousseau, B., Loc'h, C., Vaufrey, F., Hinnen, F., Doignon, I., Pillon, F., David, C., Crouzel, C., Tavitian, B., 2000. General method to label antisense oligonucleotides with radioactive halogens for pharmacological and imaging studies. Bioconjug. Chem. 11, 627-636. Lai, J., Crook, T.J., Payne, A., Lynch, R.M., Porreca, F., 1997. Antisense targeting of delta opioid receptors in NG 108-15 cells: direct correlation between oligodeoxynucleotide uptake and receptor density. J. Pharmacol. Exp. Ther. 281, 589-596. Lai, J., Riedl, M., Stone, L.S., Arvidsson, U., Bilsky, E.J., Wilcox, G.L., Elde, R., Porreca, F., 1996. Immunofluorescence analysis of antisense oligodeoxynucleotide-mediated 'knock-down' of the mouse delta opioid receptor in vitro and in vivo. Neurosci. Lett. 213, 205-208. Langstrom, B., Kihlberg, T., Bergstrom, M., Antoni, G., Bjorkman, M., Forngren, B.H., Forngren, T., Hartvig, P., Markides, K., Yngve, U., Ogren, M.., 1999. Compounds labelled with short-lived beta(+)-emitting radionuclides and some applications in life sciences. The importance of time as a parameter. Acta Chem. Scand. 53, 651-669. Langstrom, B., Bergstrom, M., 1995. Is PET a tool for drug evaluation? In: Comar, D. (Ed.), PET for drug development and evaluation. Developments in nuclear medicine volume 26. Kluwer Academic Press, Dordrecht, pp. 37-51. Lewis, M.R., Raubitschek, A., Shively, J.E., 1994. A facile, water-soluble method for modification of proteins with DOTA. Use of elevated temperature and optimized pH to achieve high specific activity and high chelate stability in radiolabeled immunoconjugates. Bioconjug. Chem. 5, 565-576. Lubberink, M., Schneider, H., Bergstrom, M., Lundqvist, H., 2002. Quantitative imaging and correction for cascade gamma

67

radiation of 76Br with 2D and 3D PET. Phys. Med. Biol. 47, 3519-3534. Marwick, C., 1998. First "antisense" drug will treat CMV retinitis. J. Am. Med. Assoc. 280, 871. Monia, B.P., Johnston, J.F., Ecker, D.J., Zounes, M.A., Lima, W.F., Freier, S.M., 1992. Selective inhibition of mutant Ha-ras mRNA expression by antisense oligonucleotides. J. Biol. Chem. 267, 19954-19962. Nesterova, M., Cho-Chung, Y.S., 1995. A single-injection protein kinase A-directed antisense treatment to inhibit tumour growth. Nat. Med. 1, 528-533. Oberbauer, R., Schreiner, G.F., Meyer, T.W., 1995. Renal uptake of an 18-mer phosphorothioate oligonucleotide. Kidney Int. 48, 1226-1232. Opalinska, J.B., Gewirtz, A.M., 2002. Nucleic-acid therapeutics: basic principles and recent applications. Nat. Rev. Drug Discov. 1, 503-514. Paans, A.M., Vaalburg, W., 2000. Positron emission tomography in drug development and drug evaluation. Curr. Pharm. Des. 6, 1583-1591. Phillips, J.A., Craig, S.J., Bayley, D., Christian, R.A., Geary, R., Nicklin, P.L., 1997. Pharmacokinetics, metabolism, and elimination of a 20-mer phosphorothioate oligodeoxynucleotide (CGP 69846A) after intravenous and subcutaneous administration. Biochem. Pharmacol. 54, 657-668. Raynaud, F.I., Orr, R.M., Goddard, P.M., Lacey, H.A., Lancashire, H., Judson, I.R., Beck, T., Bryan, B., Cotter, F.E., 1997. Pharmacokinetics of G3139, a phosphorothioate oligodeoxynucleotide antisense to bcl-2, after intravenous administration or continuous subcutaneous infusion to mice. J. Pharmacol. Exp. Ther. 281, 420-427. Sands, H., Gorey-Feret, L.J., Cocuzza, A.J., Hobbs, F.W., Chidester, D., Trainor, G.L., 1994. Biodistribution and metabolism of internally 3H-labeled oligonucleotides. I. Comparison of a phosphodiester and a phosphorothioate. Mol. Pharmacol. 45, 932-943. Sarmiento, U.M., Perez, J.R., Becker, J.M., Narayanan, R., 1994. In vivo toxicological effects of rel A antisense phosphorothioates in CD-1 mice. Antisense Res. Dev. 4, 99-107. Standifer, K.M., Jenab, S., Su, W., Chien, C.C., Pan, Y.X., Inturrisi, C.E., Pasternak, G.W., 1995. Antisense oligodeoxynucleotides to the cloned delta receptor DOR-1: uptake, stability, and regulation of gene expression. J. Neurochem. 65, 1981-1987. Stein, C.A., Cheng, Y.C., 1993. Antisense oligonucleotides as therapeutic agents--is the bullet really magical? Science. 261, 1004-1012. Tavitian, B., Terrazzino, S., Kuhnast, B., Marzabal, S., Stettler, O., Dolle, F., Deverre, J.R., Jobert, A., Hinnen, F., Bendriem, B., Crouzel, C., Di Giamberardino, L., 1998. In vivo imaging of oligonucleotides with positron emission tomography. Nat. Med. 4, 467-471. Velikyan, I., Langstrom, B., 2002. 68Ga-Labelling of Oligonucleotides. European Union COST Action B12, Dresden. Wu, F., Yngve, U., Hedberg, E., Honda, M., Lu, L., Eriksson, B., Watanabe, Y., Bergstrom, M., Langstrom, B., 2000. Distribution of (76)Br-labeled antisense oligonucleotides of different length determined ex vivo in rats. Eur. J. Pharm. Sci. 10, 179-186. Yngve, U., 2001. Labelling of various macromolecules using positron emitting 76Br and 68Ga. Synthesis and Characterisation. Acta Universitatis Upsaliensis, Comprehensive summaries of Uppsala Dissertations from the Faculty of Science and Technology 605. Uppsala. ISBN 91-554-4939-5. Yngve, U., Hedberg, E., Lovquist, A., Tolmachev, V., Langstrom, B., 1999. Synthesis of N-succinimidyl 4-(76Br)bromobenzoate and its use in conjugation labelling of macromolecules. Acta Chem. Scand. 53, 508-512. Younes, C.K., Boisgard, R., Tavitian, B., 2002. Labelled oligonucleotides as radiopharmaceuticals: pitfalls, problems and

68

perspectives. Curr. Pharm. Des. 8, 1451-1466. Zamecnik, P.C., Stephenson, M.L., 1978. Inhibition of Rous sarcoma virus replication and cell transformation by a specific oligodeoxynucleotide. Proc. Natl. Acad. Sci. 75, 280-284. Zhang, R., Diasio, R.B., Lu, Z., Liu, T., Jiang, Z., Galbraith, W.M., Agrawal, S., 1995. Pharmacokinetics and tissue distribution in rats of an oligodeoxynucleotide phosphorothioate (GEM 91) developed as a therapeutic agent for human immunodeficiency virus type-1. Biochem. Pharmacol. 49, 929-939. Zhao, Q., Zhou, R., Temsamani, J., Zhang, Z., Roskey, A., Agrawal, S., 1998. Cellular distribution of phosphorothioate oligonucleotide following intravenous administration in mice. Antisense Nucleic Acid Drug Dev. 8, 451-458.

69

Figure 1. Photographed gels (left) and autoradiograms of the exposed gels (right) after hybridisation in solution of 17 mer 68Ga-labelled phosphodiester (a), phosphorothioate (b), and 2'-O-methyl phosphodiester (c) antisense oligonucleotides to 17 mer phosphodiester sense oligonucleotide. Applied antisense:sense concentration ratios in the lanes: 1) 1:1/60, 2) 1:1/30, 3) 1:1/15, 4) 1:1/5, 5) 1:1/2, 6) 1:1, 7) 1:2, 8) 1:3 and 9) 1:4. Lane 10) sense oligonucleotide (unlabelled), 11) 68Ga-labelled antisense oligonucleotide and 12) molecular weight marker (10 bp ladder).

70

0123456789

1011

Bloo

d

Hea

rtPa

ncre

asIn

test

ine

Test

isAd

rena

l gla

nd

Lung

Sple

enKi

dney

Urin

ary

blad

der

Live

r

Brai

nM

uscl

e

Skin

Bone

Bone

mar

row

Paro

tid g

land

SUV

20 min

1 hour

2 hours

Figure 2. Organ concentrations of radioactivity (presented as SUV, mean±S.D.) obtained 20, 60 and 120 min after the injections of 68Ga-phosphodiester oligodeoxynucleotide.

0

5

10

15

20

25

30

35

40

Bloo

dH

eart

Panc

reas

Inte

stin

eTe

stis

Adre

nal g

land

Lung

Sple

enKi

dney

Urin

ary

blad

der

Live

rBr

ain

Mus

cle

Skin

Bone

Bone

mar

row

Paro

tid g

land

SUV

20 min1 hour2 hours

Fig. 3. Organ concentrations of radioactivity (presented as SUV, mean±S.D.) obtained 20, 60 and 120 min after the injections of 68Ga-phosphorothioate oligodeoxynucleotide.

71

02468

1012141618

Bloo

d

Hea

rtPa

ncre

asIn

test

ine

Test

isAd

rena

l gla

nd

Lung

Sple

enKi

dney

Urin

ary

blad

der

Live

rBr

ain

Mus

cle

Skin

Bone

Bone

mar

row

Paro

tid g

land

SUV

20 min

1 hour

2 hours

Figure 4. Organ concentrations of radioactivity (presented as SUV, mean±S.D.) obtained 20, 60 and 120 min after the injections of 68Ga-2'-O-methyl phosphodiester oligoribonucleotide.

0

5

10

15

20

25

30

35

Kidney Liver Urinarybladder

Bonemarrow

Spleen

SUV

Phosphodiester

Phosphorothioate

2'-O-MethylPhosphodiester

Figure 5. The five organs of highest radioactivity uptake (presented as average SUV of the three time points) after injection of 68Ga-labelled phosphodiester, phosphorothioate and 2'-O-methyl phosphodiester oligonucleotides.

72

02468

1012141618

Bloo

d

Hea

rtPa

ncre

asIn

test

ine

Test

isAd

rena

l gla

nd

Lung

Sple

enKi

dney

Urin

ary

blad

der

Live

r

Brai

nM

uscl

e

Skin

Bone

Bone

mar

row

Paro

tid g

land

SUV

20 min

1 hour

2 hours

Figure 6. Organ concentrations of radioactivity (presented as SUV, mean±S.D.) obtained 20, 60 and 120 min after the injections of 68Ga-DOTA.

a)

Urinary bladder

c)

Liver Lung Kidney

Urinary bladder

Parotid gland Kidney Spleen Lung

Heart

Figure 7. Ex vivo autoradiography obtained 20 min after the injection of 68Ga-labelled phosphodiester (a), phosphorothioate (b), 2'-O-methyl phosphodiester (c) antisense oligonucleotides.

b)

Liver