Qin Guangming, Zhang Yongxue, Hnatowich D.J., An Rui, Gao Zairong, Cao Wei, Cao Guoxiang
Guoxiang Dept. of Nuclear Medicine, Union Hospital, Tonji Medical College, Huazhong University of Science and Technology. Wuhan, 430022 P.R. of China This paper was awarded the Masahiro Iio Young Scientist Award, during the World Federation of Nuclear M

Correspondence:

Guangming Qin 1277 Jiefangdadao Avenue Department of Nuclear Medicine Union Hospital, Wuhan, 430022 P.R. of China E Mail: bruceqin72@hotmail.com

Cita/Reference:
Guangming Qin. Technetium-99m Labeled Antisense Oligonucleotide-Noninvasive Tumor Imaging in Mice. Alasbimn Journal 5(19): January 2003. Article N° AJ19-1.

Abstract

Single-stranded RNA and DNA oligonucleotides may be useful as radiopharmaceuticals for antisense and other in vivo applications if convenient methods for stably attaching radionuclides such as ^99mTc can be developed. The c-myc oncogene works in cooperation with other oncogenes in a variety of malignant tumors. The concentration of c-myc messenger RNA increases rapidly 30 to 50 fold during DNA synthesis, thus making it a suitable target for following the progression of malignancy by noninvasive imaging with radiolabeled antisense oligonucleotide probes.

Introduction

Specific diagnosis of malignant tumors in an early stage is still a problem to be solved. Though X ray CT, MRI, Ultrasonagraphy and conventional nuclide imaging are broadly introduced in clinic, none of them can give a specific diagnose in early stages, especially to mammary and lung cancers. Nowadays, this difficulty might be overcome by means of antisense imaging. With the development of modern molecular biology, malignant tumors are recognized as the results of oncogene activation and tumor suppressor gene depression. In cancer, oncogenes were amplified so that multiple copies of mRNAs and hybridization sites would be available for binding and retention of radiolabled antisense probes for noninvasive imaging with the gamma camera ⁽¹⁾.

Antisense oligonucleotides are synthetic single-strand deoxyribonuclic acids(DNAs) or oligoribonuclic acids(RNAs) designed to have a base sequence complementary to that of the targeted gene or mRNAs. These short base sequences are supposed to bind in a base-specific manner with the targeted gene or mRNAs by some antisense mechanisms, thus to interfere with the process of gene transcription or translation. As being restricted by various roadblocks such as nuclear membrane crossing, genotoxicity suspicion and unclear binding rules of Hoogsteen triplet formation, antisense strategies usually choose to interfere with the process of translation. Several theories were promoted to interpret the mechanism of translation arrest. It was originally considered that when hybridized with the sense region of the targeted mRNA, the antisense DNA might hinder the protein synthesis by physically blocking the mRNA migration through the ribosome. It is now known that the ribosomal complex can unwind the DNA/RNA duplex and might permit unhindered translation. So in many and possibly most cases, the mechanism at work is more likely to be mRNA degradation by RNase H enzymes that recognize the antisense DNA/RNA duplex. Whatever mechanism acts is most likely concerned by antisense therapy industries. For antisense imaging studies in which scintigraphy is performed after administration of antisense DNAs radiolabeled with imaging radionuclide, an increased, detectable amount of mRNA transcript together with specific retain of antisense probes in the cytoplasm are of the factors most concerned.

In application of antisense imaging, antisense DNAs must survive in plasma and in the cytosol long enough to locate and bind to their target. The main source of in vivo instability of native phosphodiester DNAs, especially single-strand DNAs, is enzyme degradation. Exo- and endonucleases (which attack DNA from its end and interior regions, respectively) are ubiquitous and are responsible for the rapid in vivo degradation of single-strand phosphodiester DNAs. For this reason, native phosphodiester DNAs are almost considered unsuitable and useless in antisense imaging. Few investigations were reported in antisense imaging using antisense phosphodiester DNAs. But in 1994, with Indium-111-DTPA-antisense phosphodiester probes, Dewanjee and his colleagues did make successful imaging in a mammary tumor-bearing mouse model. It seems that more investigations are necessary before a final validation of the phosphodiester DNAs is made. In this study, a c-myc mRNA antisense oligonucleotide (phosphodiester) was radiolabeled with 99mTc via the bifunctional chelator S-Acetyl-NHS-MAG3, with permission by Dr Donald J. Hnatowich, who along with Dr P. Winnard et al provided this method ⁽²⁾. The labeled antisense probe was injected into a mammary tumor-bearing mouse model to which the gamma imaging was then performed.

Materials and Methods

The 15-base, single stranded antisense oligonucleotides complementary to the translation beginning sites of c-myc oncogene messenger RNA and it's sense sequence were selected. The antisense oligonucleoptides (Sangon Biotech, CA) had a base sequence of 5'-amine-AAC,GTT,GAG,GGG,CAT-3'. The molecular weight was about 4.6 kDa. The melting temperature in physiological saline was calculated to be 46ºC. The DNAs were purchased unpurified and were used without further purification. They were generally handled under sterile conditions; all solutions were sterilized by terminal filtration through a 0.22µm filter, and sterile pipette tips were used. All other pipette tips and tubes were autoclaved prior to use.

Technetium-99m-pertechnetate was obtained from a ⁹⁹Mo-^99mTc nuclide generator (Atomic Energy Institute, China). The S-Acetyl-NHS-MAG3 was a gift from Dr. Donald J. Hnatowich (Massachusetts Medical Center, USA).

Oligonucleotides Conjugation and Labeling
Oligonucleotides conjugation and labeling procedure was performed according to the method of P. Winnard et al (2).

A solution of single stranded amine-derivatized DNA (165-990µg) was prepared and heated to 55ºC for 10 min and
immediately plunged into ice water to dissociate any DNA duplexes. The DMF solution of NHS-MAG₃ was then added to the stirred DNA solution. The solution was incubated at room temperature for 15-20 min in the dark. After
purification, fractions off the P4 column (BioRad, Melville, NY) were collected and the absorbance at 260 nm of each measured (Gene Quant RNA/DNA Calculator, Pharmacia Biotech). The peak fractions were lyophilized and stored at -70ºC for periods from several days to more than four months before labeling.

To a sterile test tube containing the MAG3-DNA(about 10-100µg,10-100µl) was added sufficient 99mTc-pertechnetate solution(10-50µl) to provide about 50-100µCi/µg of DNA. To this was added the tartrate solution and stannous ion solution. After 15min at room temperature, the labeled DNA was purified on a 0.7×20cm gel filtration column of Sephadex G-25 using sterile 0.25 M ammonium acetate, pH 5.2, or saline, as eluant. Radioactivity and absorbency at 260nm were used to identify and quantitate peak fractions.

Control labeling was performed in which the native, unconjugated DNA was subjected to the identical labeling procedure to assess the extent of nonspecific labeling.

Sep-Pak Studies
Radiochemical purity was estimated by Sep-Pak C-18 (Waters, Milford, MA) column chromatography. A column was conditioned with 10ml of absolute ethanol followed by 10ml of 1mM HCl. After the sample was loaded, the first elution, with 10ml of 1mM HCl, removed 99mTc-pertechnetate and 99mTc-tartrate. The second elution, with 10ml of ethanol/saline (1:1), removed the labeled DNA. Radiolabeled colloids remained on the column.

In addition to measuring radiochemical purity, Sep-Pak studies were also used to establish whether the conjugation and labeling procedure had diminished the ability of the labeled DNA to hybridize to its complement. The radiolabeled MAG3-DNA was analyzed as above by Sep-Pak column chromatography before and after the addition of the complementary DNA.

Serum Incubation Studies
Labeled MAG3-DNA was incubated at a concentration of 10µg/ml in fresh human serum at 37ºC. Samples were periodically removed over 24hr for Sep-Pak analysis.

Preparation of Animal Model
Ehrlich carcinoma mice (bearing mammary adenocarcinoma in the ascites) were obtained from the Department of Pharmacology, Tongji Medical College, Huazhong University of Science and Technology. The ascites was suctioned into a sterile syringe and diluted to a concentration of 5-10×10⁶ tumor cells/ml. A colony of KM mice (15-20g) were inoculated with 1×10⁶ tumor cells in the right thigh, and the tumors were allowed to grow for 6-7 days to a size of 1.0-1.5 cm in diameter.

Biodistribution Studies
A total of 40 mice(two groups:20 sense phosphodiester and 20 antisense phosphodiester) were injected intravenously in the tail vein with 50 µCi of 99mTc-labeled oncogene probes (sense and antisense). The animals were killed by cervical dislocation at different time points from 45min to 24hr. The tumor, blood, heart, liver, spleen, lung, kidney, intestinal bowel, muscle and bone were harvested and weighed. The tracer distribution in the viscera and tumor was determined with a gamma counter in the 130-150 keV window to include the 140 keV of the ^99mTc radionuclide. The mean and s.d. values, relative radioactivity ratios of tumor-to-blood and tumor-to-muscle, percent of injected dose/gram were calculated for each tumor bearing mouse model.

Tumor Imaging Studies
A total of 8 mice (two groups: 4 sense phosphodiester and 4 antisense phosphodiester) were injected intravenously in the tail vein with 1-2 mCi of ^99mTc-labeled oncogene probes, immobilized with ketamine hydrochloride and imaged at 30min, 1h, 2h, 4h, 6h and 24hr with a gamma camera fitted with a medium-energy pin-hole collimator (Sopha DSX SPECT, France). Images were acquired for a present time of 30 to 1000s using a 256×256 matrix with a 20% energy window set at 140 keV.

In a test of specificity, the unlabeled, unmodified antisense oligonucleotide was administered to block base-specific binding sites of mRNA 2 hr before the administration of ^99mTc labeled antisense oligonucleotides. To eliminate the interference of free ^99mTc, mice were also imaged after the same dose of ^99mTc-pertechnetate was administered.

Results

DNA Conjugation and Labeling
In this study, the amine-derivatized DNAs were readily conjugated with the NHS derivative of S-Acetyl-NHS-MAG₃. Figure 1 presents UV absorption chromatograms obtained by reversed phase Sep-Pak C-18 analysis of unmodified oligonucleotide (left), MAG₃-conjugated oligonucleotide before purification (right). After conjugation with MAG₃, the presence of unmodified oligonucleotide was reduced to a minor constituent (right). The peak absorbance shift demonstrates that the conjugation was largely complete.

Discussion

Most radionuclides used in nuclear medicine are metals. In the field of diagnostic imaging, the radiolabel of choice is almost always 99mTc because of it's relatively short half-life, pure ã emission, 140keV ideal energy for imaging and allowance for daily access. One straightforward approach to radiolabel DNAs with metallic radionuclides is first to derive the antisense DNA on either of its ends with a primary amine, possibly attached by a suitable linker to minimize steric hindrance. The amine then may be conjugated with various metal bifunctional chelators, such as the anhydrides of diethylenetriamine pentaacetic acid (DTPA) or the N-hydroxysuccinimide esters of SHNH and mercaptoacetyltriglycine (MAG₃). DTPA has previously been used for radiolabeling antibodies with 99mTc but was abandoned when the instability of the label was judged to be unacceptably high. As an alternative, SHNH can form stable labeling with DNAs. But Hnatowich et al. reported that DNAs labeled in this manner showed nonspecific serum protein binding, which would interfere with the biodistribution and thus affect the imaging results ⁽³⁾.
MAG₃ seems to show better characteristic in labeling DNAs with 99mTc. In this study, a mean of 80.11%(s.d=2.96%, n=4) labeling efficiency was obtained. The radiolabeled antisense probe still kept the affinity to bind to its complement, as shown in Figure 2. After 30min incubation in fresh human serum at 37¨C, the peak radioactivity shift into two peaks. It seemed that protein binding had occurred. But since 1hr thereafter, the peak radioactivity regressed to its primary site, which most likely suggested that the labeled DNAs had dislocated from the serum proteins. This short-time serum protein binding seemed to be beneficial in protecting the labeled DNAs from being attacked by nuclease.

The extraordinary properties of DNA and RNA suggest that there is potential for the use of these oligonucleotides as radiopharmaceuticals. Despite many uncertainties concerning mechanisms, synthetic single-strand antisense deoxyribonucleic acids are now in clinical trials for the chemotherapy of several cancers, including follicular lymphoma and acute myelogenous leukemia ^(4-7). The question considered here is whether antisense DNAs will also be important to future nuclear medicine imaging. For successful antisense imaging, at least the following criteria must be fulfilled: 1, sufficient oncogene mRNA products in cytoplasm; 2, antisense probes of suitable base sequences; 3, efficient methods for antisense probes labeling; 4, good permeability and in vivo stability ^(8,9).

The c-myc oncogene works in cooperation with other oncogenes in a variety of cancers. It is an early response gene whose expression is involved in the signal transduction pathways leading to cellular proliferation ^(1,10-15). The concentration of the c-myc mRNA increases rapidly 30- to 50-fold during DNA synthesis, and the half-life of c-myc mRNA is sufficiently long, which makes it a suitable target for antisense imaging ⁽¹⁶⁾. This level and time of mRNA retention in cytoplasm may be sufficient for in vivo hybridization, retention of labeled probe and noninvasive imaging. As reported by Dewanjee and his colleagues in their imaging studies, the specific binding to isolated mRNAs was about 20-fold higher for the labeled antisense compared to the sense DNAs. The nude mice obtained about 10%-12% of the radioactivity on injected antisense DNAs and only about 1% on sense DNAs at 1hr postadministration. Preferential accumulation was revealed in tumored mice with respectable tumor/muscle ratios of 20 in antisense probes and of 0.95 in sense probes as early as 30min⁽¹⁷⁾.

In contrast to single-strand DNAs, single-strand RNAs form extensive secondary structures. The secondary structures of mRNA may be required for stability and possibly for recognition by those proteins regulating translation. These structures have implications for chemotherapy, since antisense DNAs show lower affinities for duplex regions. The affinity of antisense DNA for duplex, compared to singlet, regions of mRNA has been reported to be 105-106-fold lower. Therefore, antisense strategies usually seek to target only single-strand regions of the mRNA. Since knowledge of the molecular structure of mRNA is limited, one common approach is to target either the initiation codon (AUG) and adjacent sequences or the untranslated sequences on either the 3' or 5' end ⁽¹⁸⁾. The c-myc mRNA antisense oligonucleotide chosen in this study was designed to target the initiation site of translation, in the hope that these regions will be accessible. This selection was proved to be act by our encouraging results similar to those of Dewanjee's. Maybe because of it's relatively high molecule or difference in radiolabeling, the tracer showed much slower pharmacokinetics in accumulation to the target and clearance from background tissues in our study than in Dewanjee et al's.

The tracer distribution reached an equilibrium value after 45 minutes post administration for both antisense and sense probes and did not change for viscera, connective tissues and tumor. The tumor could be imaged with the antisense probe as early as 45min and reached the highest tumor-to-muscle ratio of 5.9 at 4hr. This allows diagnosis of malignant tumors using short-time radionuclides as 99mTc. Besides the tumor and those highly accumulated organs as the liver and kidneys, the cellular elements of blood, mainly the nucleated cells (neutrophiles, lymphocytes and reticulocytes), retain the probe as a result of internalization and hybridization with low copy of mRNAs. Freshly formed platelets also retained a small fraction of the probes. This resulted in prolonged background clearance that the most encouraging results were not achieved until 20hr later when the background activity was diminished sufficiently. Considering the high uptake of tracers in the liver, intestinal bowels and kidneys, the 99mTc-MAG₃ c-myc probe seemed unsuitable for detecting tumors in these regions. From these points of view, more investigations should be taken to improve the target/nontarget ratio and to accelerate the clearance of background activity.

Conclusion

S-Acetyl-NHS-MAG₃ is an efficient bifunctional chelator for 99mTc labeling with oligonucleotides. Using 99mTc labeled antisense probes, malignant tumors with over-expressed oncogene mRNAs may be specifically imaged at an early stage.

References

1	Martin D. Abeloff. Clinical Oncology. Second Edition. 77-118. back to
2	P. Winnard, Chang F,Rusckowski M,et al. Preparation and Use of NHS-MAG3 for Technetium-99m Labeling of DNA. Nuclear Medicine & Biology 1997; 24: 425-432. back to
3	D.J. Hnatowich, P. Winnard Jr., F. Virzi, et al. Technetium-99m labeling of DNA oligonucleotides. J Nucl Med 1995; 36:2306-2314. back to
4	Ma DD. Synthetic oligonucleotides as therapeutics: the coming of age. Biotechnol Annu Rev 2000; 5:155-96. (192 ref.) back to
5	Cotter FE. Antisense oligonucleotides for haematological malignancies. Haematologica 1999 Jun; 84:19-22(10 ref.). back to
6	Green DW, Roh H, Pippin J, et al. Antisense oligonucleotides: an evolving technology for the modulation of gene expression in human disease. J Am Coll Surg 2000 Jul; 191(1): 93-105(79ref.). back to
7	Hnatowich DJ. Antisense and nuclear medicine. J Nucl Med 1999; 40:693-703. back to
8	Lebedeva I, Benimetskaya L, Stein CA, et al. Cellular delivery of antisense oligonucleotides. Eur J Pharm Biopharm 2000 Jul; 50(1): 101-19 (255 ref.). back to
9	David Piwnica-Worms. Making Sense Out of Anti-Sense: Challenges of Imaging Gene Translation with Radiolabeled Oliginucleotides. J Nucl Med 1994; 35:1064-1066. back to
10	10 Persons DL. Quantitation of HER-2/neu and c-myc gene. Med Pathol 1997; 10:720-7. back to
11	Salqia R. Molecular abnormalities in lung cancer. J Clin Oncol 1998; 16:1207-17. back to
12	Seth A, Gupta S, Davis RJ. Cell cycle regulation of the c-myc transcriptional activation domain. Mol Cell Biol 1993; 13:4125-36. back to
13	Gu W, Checova K,Tassi V, et al. Opposite regulation of gene transcription and cell proliferation by c-myc and MAX. Proc Natl Acad Sci USA 1993; 90:2935-39. back to
14	Johnson JM, Carroll WL. c-myc hypermutation in Burkitt's lymphoma. Leuk Lymphoma 1992; 8:431-9. back to
15	Dang CV. C-myc target genes involved in cell growth, apoptosis and metabolism. Mol Cell Biol 1999; 19:1. back to
16	Park M, Vande Woude GF. Oncogene: genes associated with neuplastic diseases. In: Seriver CR, Beaudet AL, Sylws. Valle D, et al. The metabolic basis of inherited diseases. New York. Mc Graw Hill 1989:251-277. back to
17	Dewanjee MK, Abdol KG,Kapadvanjwala M,et al. Noninvasive Imaging of c-myc Oncogene Messenger RNA with Indium-111-Antisense Probes in a Mammary Tumor-Bearing Mouse Model. J Nucl Med 1994; 35: 1054-1063. back to
18	Scherr M, Rossi J.J., Sczakiel G., et al. RNA accessibility prediction: a theoretical approach is consistent with experimental studies in cell extracts. Nucleic Acids Res 2000 Jul 1; 28(13): 2455-61. back to