A Comparison of in Vitro and in Vivo Stability in Mice of Two

Aug 27, 2003 - Jiang He, Guozheng Liu, Surong Zhang, Jean-Luc Vanderheyden, Ning Liu, Changbin Liu,. Yumin Zhang, Suresh Gupta, Mary Rusckowski, ...
0 downloads 0 Views 71KB Size
1018

Bioconjugate Chem. 2003, 14, 1018−1023

A Comparison of in Vitro and in Vivo Stability in Mice of Two Morpholino Duplexes Differing in Chain Length Jiang He, Guozheng Liu, Surong Zhang, Jean-Luc Vanderheyden, Ning Liu, Changbin Liu, Yumin Zhang, Suresh Gupta, Mary Rusckowski, and Donald J. Hnatowich* Division of Nuclear Medicine, Department of Radiology, University of Massachusetts Medical School, 55 Lake Avenue North, Worcester, Massachusetts 01655. Received June 19, 2003

The stability of hybridized duplexes is an important criterion for any radiopharmaceutical application of DNAs or their analogues such as phosphorodiamidate morpholinos (MORFs). Objective: The stabilities in vitro and in mice of the duplex between MORF and its complement (cMORF) were investigated for two different chain lengths, a 15-mer MORF compared to the identical MORF but elongated to a 25-mer. Methods: The hybridization characteristics of the 15-mer MORF with its complementary 15-mer and that of the 25-mer with its complementary 25-mer MORF were measured using surface plasmon resonance (SPR) analysis. For radiolabeling with 99mTc, the 15- and 25-mer MORF, both with a primary amine via a 10-member linker on the 3′ equivalent end, were conjugated with NHS-MAG3. The 15- and 25-mer cMORFs were conjugated via their amines to carbodiimidazole treated poly(methyl vinyl ether-alt-maleic acid) (PA) such that about 50 cMORFs were attached to each polymer molecule in both cases (estimated MWs about 300 and 450 kDa, respectively). After hybridization in vitro, both the PA-cMORF15-99mTc-MORF15 and PA-cMORF25-99mTc-MORF25 homoduplexes were evaluated by size exclusion HPLC in saline, after incubation in 37 °C serum and in urine obtained 30 min post IV administration to normal mice. Biodistributions were obtained up to 18 h post administration. Results: By SPR, the affinity constants for the homoduplexes were both about 109 M-1 with the 25/25 only about 25% higher than the 15/15. However, the affinity constants for the 15/25 and 25/15 heteroduplexes showed a surprisingly 13-fold difference. By HPLC analysis, all duplexes were stable in saline; however, analysis of serum incubates and urine containing PAcMORF15-99mTc-MORF15 showed an immediate and pronounced low molecular weight peak that was identified by a shift assay to be 99mTc-MORF15. The comparable peak in both fluids was much less pronounced in the case of PA-cMORF25-99mTc-MORF25. Whole body radioactivity levels also fell much more rapidly in mice receiving the 15-mer conjugate (65 vs 30% eliminated at 18 h) and biodistribution results showed higher kidney levels for the 15-mer conjugate. Results with the PA-cMORF25-99mTcMORF15 heteroduplex were more similar to that obtained with the 15-mer homoduplex than the 25-mer homoduplex. Conclusion: Despite what is reported to be high hybridization affinities, both the homoduplex and heteroduplexes prepared with 99mTc-MORF15 were found to be unstable in serum and in vivo toward dissociation to free 99mTc-MORF15. By contrast, homoduplex prepared with 99m Tc-MORF25 showed higher stability. These differences in hybridization stability may be important considerations in radiopharmaceutical design.

INTRODUCTION

This laboratory has been investigating the potential of DNA and DNA-like oligomers for radiopharmaceutical use. We earlier examined both the phosphodiester and phosphorothioate DNAs (1, 2). Subsequently, we examined peptide nucleic acids (PNAs)1 as the first commercially available DNA analogue (3). Recently, we have begun examining phosphorodiamidate morpholinos (MORFs), another commercially available DNA analogue (4). Apart from pharmacokinetic behavior, one important criteria of DNA and other oligomers for radiopharmaceutical use is hybridization stability. Although melting temperatures and affinity constants provide a measure of stability (5-7), oligomers under consideration for in vivo use should be studied in 37 °C serum and, preferably, in animals. * To whom correspondence should be addressed: Prof. Donald J. Hnatowich, Division of Nuclear Medicine, Department of Radiology, University of Massachusetts Medical School, 55 Lake Avenue North, Worcester, MA 01655. Phone: 1-508-856-4256. Fax: 1-508-856-4572. E-mail: [email protected].

One unique feature of oligomers for radiopharmaceutical design is the ability to alter base sequence and chain length. In preliminary studies from this laboratory, MORFs labeled with 99mTc showed significant differences in biodistribution depending upon chain length (and therefore base sequence) from 15- to 25-mer (8). This present investigation was conducted to further establish the influence of base sequence and chain length, now on hybridization stability. A 15-mer MORF (MORF15) and a 25-mer MORF (MORF25) were separately conjugated to a polymer (PA) and hybridized in vitro with complementary MORF15 (cMORF15) and complementary 1 Abbreviations: MORF, phosphorodiamidate morpholino; SPR, surface plasmon resonance; PA, poly(methyl vinyl etheralt-maleic acid); PNA, peptide nucleic acid; NHS-MAG3, Nhydroxysuccinimidyl S-acetylmercaptoacetyltriglycine; DMF, dimethylformamide; NMP, N-methyl-2-pyrrodinone; DIEA, diisopropylethylamine; RUs, resonance units; EDTA, ethylenediaminetetraacetic acid; HEPES, N-[2-hydroxyethyl]piperazineN′-[2-ethanesulfonic acid]; ka, association rate constant; kd, dissociation rate constant; KA, affinity constant; 99mTc, 99mtechnetium.

10.1021/bc0341019 CCC: $25.00 © 2003 American Chemical Society Published on Web 08/27/2003

Stability of Mice Morpholino Duplexes

MORF25 (cMORF25) respectively, each radiolabeled with 99m Tc via MAG3. The stability of each duplex was then investigated in vitro and in vivo in mice. MATERIALS AND METHODS

All MORFs and cMORFs whether biotinylated or amine modified on the 3′ equivalent end were purchased purified (Gene-Tools, Corvallis, OR) and were used as received. The MORF15 base sequence was 5′-equivalentTGT-ACG-TCA-CAA-CTA-linker-primary amine (or biotin) and that of MORF25 was T-GGT-GGT-GGGMORF15. The base sequences of the cMORFs were complementary with the linker and primary amine also on the 3′-equivalent end. The molecular masses ranged from 5000 to 9000 Da. The MORFs were therefore identical to those used by us previously (4, 8). NHydroxysuccinimidyl S-acetylmercaptoacetyltriglycine (NHS-MAG3) was synthesized in house according to published procedures, and the structure was confirmed by elemental analysis, proton NMR, and mass spectroscopy (9). Normal male CD-1 mice were purchased with a mean body weight of 20-30 g (Charles River laboratories, Wilmington, MA). The serum was from normal male CD-1 mice. Poly(methyl vinyl ether-alt-maleic acid) (PA, average MW ca. 45 000), anhydrous dimethylformamide (DMF), N-methyl-2-pyrrodinone (NMP), diisopropylethylamine (DIEA), and 1,1′-carbonyldiimidazole were purchased (Sigma-Aldrich, St. Louis, MO). All other chemicals were reagent grade and were used without purification. A radionuclide generator (Bristol-Myers Squibb Medical Imaging, Inc., North Billerica, MA) provided the 99m Tc-pertechnetate. Synthesis and Radiolabeling of MAG3-Conjugated MORF. A solution of MORF15 or MORF25 was prepared at a concentration of 2 mg/mL in 0.1 M HEPES buffer pH 8.0. A 50 mg/mL solution of NHS-MAG3 in anhydrous DMF was then added dropwise to the stirred MORF solution until a MAG3 to MORF molar ratio of 20:1 was reached. After incubation at room temperature overnight, the conjugated oligomer was purified on a 0.7 × 20 cm P4 (Bio-Gel, Bio-Rad, Melville, NY) gel filtration column eluted with 0.25 M ammonium acetate buffer, pH 5.2. The UV absorbency at 265 nm of fractions off the column was measured (U-2000, Hitachi Instruments, Danbury, CT). Concentration of conjugated oligomer was estimated with respect to MORF at this frequency using a molar absorbency value of 158 000 for MORF15 and 258 350 for MORF25 reported by the manufacturer. Samples were stored as such at -20 °C for further use. Generally, 7 µg of the MAG3-conjugated MORF was labeled on each occasion. To 25 µL of the MAG3-MORF solution in acetate buffer was added 0.5 mg of sodium tartrate from a fresh 50 mg/mL solution in pH 9.3 buffer (0.5 M ammonium bicarbonate, 0.25 M ammonium acetate, 0.18 M ammonium hydroxide), followed by 0.5-3 mCi of 99mTc-pertechnetate. Finally, 4 µg of stannous chloride dihydrate from a fresh 1 mg/mL solution in 10 mM HCl was quickly added with agitation. The final pH was about 7.6. After heating at 100 °C for 20 min, the entire labeled MAG3-MORF15/25 preparation was purified on a 0.7 × 20 cm P4 column using 50 mM phosphate buffered-saline pH 7.2 as eluant. Labeling efficiency was calculated as the percentage of radioactivity eluting as 99m Tc-MORF compared to the total radioactivity applied to the column. Each P4-purified preparation of radiolabeled MAG3MORF was analyzed by size exclusion HPLC using a single 1 × 30 cm Superose 12 column (Amersham

Bioconjugate Chem., Vol. 14, No. 5, 2003 1019

Pharmacia Biotech, Piscataway, NJ) with 50 mM phosphate buffer pH 7.2 as eluant. In-line UV absorbency at 265 nm and radioactivity were used to identify and quantitate peak fractions. Recovery of radioactivity was routinely determined. Synthesis of PA-cMORF. The polymer PA was previously used in this laboratory as a platform for the construction of PNA polymers (10) and MORF polymers (4). The polymer (10 mg) was dissolved in 1.0 mL of the aprotic solvent N-methyl pyrrolidinone (NMP) and to this was added 20.4 mg of 1,1′-carbonyldiimidazole and 3.0 µL of diisopropylethylamine (DIEA). The mixture was incubated at room temperature for 2 h. To 500 µL of a 2.0 mg/mL solution of cMORF15 or cMORF25 in NMP, a designated amount of the activated PA mixture and the equivalent moles of DIEA were added to reach a 100:1 molar ratio of cMORF to PA. The mixture was incubated overnight at room temperature. An aliquot of the solution prior to purification was used to estimate the average number of cMORF groups bonded to each PA molecule by SE HPLC with UV detection at 265 nm. Since PA does not absorb appreciably at 265 nm, the peak areas of free, nonconjugated cMORF and of PAcoupled cMORF15/25 were compared. As an alternative method of estimating the average number of groups per molecule, radiolabeled MORF15 and MORF25 were used to measure the groups per molecule of PA-cMORF15 and PA-cMORF25, respectively. Each was added at tracer concentrations to an aliquot of its polymer and the radioactivity on free, nonconjugated cMORF compared to that on PA-coupled cMORF was determined. The PA-cMORF conjugates were purified by open column gel filtration chromatography on 1 × 30 cm Sephadex G100 column (Amersham Pharmacia Biotech, Piscataway, NJ) using water as eluant. The concentration of PA-cMORFs in the recovered fractions with respect to cMORFs was estimated by UV absorbency using the above molar absorbency values. Surface Plasmon Resonance Measurements. Recent improvements in instrumentation for biomolecular interaction analysis by surface plasmon resonance (SPR) has made it easier to measure rates of association and dissociation directly. Thus, SPR may be used to measure the rate of binding of an analyte on a surface to which its partner has been immobilized (11). SPR is an optical phenomenon that arises when light illuminates thin conducting films under specific conditions. SPR permits the generation of sensorgrams in which the refractive index changes due to this binding are measured in resonance units (RUs) (12). In this investigation, surface plasmon resonance was used to estimate the association rate constants for duplex formation between immobilized cMORFs and MORFs. The analysis was performed on a BIAcore-2000 (BIAcore, Piscataway, NJ) instrument operating at room temperature. Biotinylated cMORF15 and biotinylated cMORF25 were each added to a streptavidin-dextran-coated sensor chip only until a response of about 200 RUs was reached. The absence of mass transfer effects was confirmed by running separately one concentration of free MORF at three different flow rates (10, 30, and 75 µL/min) and demonstrating identical RU responses in all cases. Solutions of free MORF were prepared at six concentrations (0.0-5.0 µM) in the same running buffer (10 mM HEPES, 150 mM sodium chloride, 3.4 mM Na2-EDTA, 0.005% P20, pH 7.4). Dissociation was followed for 25 min and the chip surface was regenerated by injection of 100 mM HCl. In-line reference subtraction was performed at each concentration to correct for bulk refractive index changes. The resulting

1020 Bioconjugate Chem., Vol. 14, No. 5, 2003

sensorgrams were analyzed using instrument software (BIAevaluation 3.0, BIAcore) by assuming 1:1 Langmuir interactions. Stability of Radiolabeled Polymer in Mouse Serum. The polymer conjugates of PA-cMORF15 and PAcMORF25 were usually radiolabeled by adding trace amount of 99mTc-MORF15 or 99mTc-MORF25 respectively to prepare the homoduplexes followed by incubation at room temperature for 30 min. Occasionally, PA-cMORF25 was radiolabeled in this manner with 99mTc-MORF15 to form the heteroduplex. When necessary to provide a radiolabeled polymer of at least 95% radiochemical purity, the preparation was purified by ultrafiltration (Microcon 100 kDa MWCO, Millipore Corporation, Bedford, MA). Quality control by size exclusion HPLC was routinely performed. Repeat measurements by HPLC were then used to assess the stability of the radiolabeled polymer conjugate following incubation in mouse serum at 37 °C for 1 h. Serum aliquots were removed and analyzed by HPLC using 50 mM sodium phosphate buffer pH 7.2 as eluant. Recoveries with respect to radioactivity were routinely measured. Urine Analysis. Normal CD-1 male mice received by tail vein 0.1 mL of saline containing about 20-100 µCi of both radiolabeled homoduplex polymers carrying 2-4 µg of cMORF. The mice were then sedated by halothane inhalation and urine was collected at 0.5 h postinjection. Urine radioactivity and whole body radioactivity after urine removal were both measured in a dose calibrator. A shift assay was performed on the urine samples by HPLC analysis before and after the addition of PAcMORF to establish whether the 99mTc in urine was still bound to MORF. Recoveries were routinely measured. Biodistribution of Radiolabeled Polymer in Normal Mice. In groups of four, normal CD-1 male mice each received by tail vein 0.1 mL of saline containing about 20-100 µCi of each of the four PA-cMORF-99mTc-MORF in each case carrying 2-4 µg of cMORF. Use of a large polymer such as PA was important to this measurement because its slow clearance provided sufficient time for dissociation of the duplex to become evident. Whole body radioactivity was monitored over 18 h by repeatedly placing individual mice in a dose calibrator. Thereafter, the mice were sedated by halothane inhalation and sacrificed by cervical dislocation. Blood was removed by cardiac puncture, the mice were then dissected, and selected organs were removed and rinsed in saline and weighed before counting in a NaI (Tl) well counter along with blood samples and an aliquot of the injectate. The biodistributions are reported as the percentage of the injected dose per gram (%ID/g) of tissue corrected for background radioactivity, physical decay during counting, and retention of radioactivity in the tail. RESULTS

PA-cMORF15/25 Conjugates. The HPLC chromatographic profiles of both PA-cMORF15 and PA-cMORF25 after conjugation but before purification all showed by both radioactivity and 265 nm UV detection two roughly equal peaks, one at 14 min due to PA-cMORF and a second at 26 or 28 min due to free cMORF15 or cMORF25, respectively. In both cases, an average of 40-50 groups of cMORF per PA molecule was calculated from the ratio of UV peak areas (at this frequency, the absorbency of PA is negligible). The HPLC profiles of radioactivity were used to determine the number of cMORF on PA molecule in solution that are accessible to radiolabeled MORF. The 99m Tc-MORF15 or 99mTc-MORF25 added at tracer con-

He et al. Table 1. Kinetics Constants of Single-Stranded MORFs Hybridization Obtained from BIAcore a MORFs

ka (1/Ms)

kd (1/s)

KA (1/M)

25-25 15-15 15-25 25-15

1.01(0.02) × 105 3.80(0.23) × 105 2.08(0.07) × 105 1.73(0.06) × 104

9.25(0.29) × 10-5 4.30(0.20) × 10-4 2.19(0.14) × 10-4 2.52(0.26) × 10-4

1.10(0.05) × 109 8.87(0.81) × 108 9.50(0.92) × 108 6.92(1.00) × 107

a 25-25: MORF25 binding to immobilized biotinylated cMORF25. 15-15: MORF15 binding to immobilized biotinylated cMORF15. 15-25: MORF15 binding to immobilized biotinylated cMORF25. 25-15: MORF25 binding to immobilized biotinylated cMORF15. (standard deviation in parentheses). N ) 4.

Figure 1. Representative sensorgrams obtained from BIAcore for MORF25 binding to immobilized biotinylated cMORF25(25-25) and MORF15 binding to immobilized bintinylated cMORF15 (15-15). The decreased kd of MORF25 is clearly visible in a normalized overlay sensorgrams obtained for dissociation phase(boxed inlay).

centrations to unpurified PA-cMORF15 or PA-cMORF25 respectively distributes between free cMORFs and cMORF accessible on PA polymer according to their relative concentration. In this way, the average number of cMORFs accessible to hybridization in solution was estimated in both cases to be about 40 ( 5 in good agreement with the UV results to suggest that all cMORFs on both PAs are accessible to free MORF in solution. The final molecular masses were calculated to be 300 and 450 kDa, respectively. Surface Plasmon Resonance. The association rate constants (ka), the dissociation rate constants (kd), and the affinity constants (KA ) ka/kd) determined using SPR analysis are summarized in Table 1. Between the two homoduplexes, the 25-mer MORF showed about a 4-fold lower association rate, but a dissociation rate about a 5-fold lower such that its affinity constant was about 25% higher than that of the 15-mer MORF. A surprising result of these measurements on the heteroduplexes was the large difference in association rate constants depending upon which MORF was immobilized and which was free. Thus, while the dissociation rate constants were about the same, the association rate constant for MORF15 binding to immobilized MORF25 was about 12-fold higher than the reverse. For this reason, the affinity constants between the heterduplexes showed a large difference with hybridization of MORF25 to immobilized MORF15 showing the lowest value among all four combinations. Figure 1 presents typical sensorgrams for the hybridization of MORF25 to immobilized cMORF25 and MORF15 to immobilized cMORF15 obtained at room temperature showing association during injection followed by dissociation. The lower dissociation constant for the MORF15 homoduplex relative to the MORF25 het-

Stability of Mice Morpholino Duplexes

Bioconjugate Chem., Vol. 14, No. 5, 2003 1021

Figure 2. Size exclusion HPLC radiochromatograms of PA-cMORF15-99mTcMORF15 (A), PA-cMORF25-99mTcMORF25 (B), and PA-cMORF25-99mTcMORF15 (C) in saline (top row) and after 1 h incubation in 37 °C serum (bottom row).

Figure 3. Size exclusion HPLC radiochromatograms of urine obtained 30 min post administration to mice of PA-cMORF15-99mTcMORF15 before (A) and after (B) the addition of PA-cMORF15.

eroduplex is clearly visible over the 25 min dissociation period. Stability of Radiolabeled Polymer Conjugate in Mouse Serum. Figure 2 presents HPLC radiochromatograms of radiolabeled PA-cMORF15 (panel A) and PAcMORF25 (panel B) homoduplexes and radiolabeled PAcMORF25-99mTc-MORF15 heteroduplex (panel C) in saline (top row) and after 1 h in 37 °C serum incubation (bottom row). Before incubation, all three samples show only the polymer peak at 14 min, but only the PAcMORF25 homoduplex continues to show only this peak after incubation. By contrast, both the PA-cMORF15 homoduplex and PA-cMORF25 heteroduplexes in serum show a prominent second peak at 26-28 min. In both cases, this latter peak has been shown to be due to free labeled MORF15 by a shift assay in which this peak disappears following addition of PA-cMORF15. The figure therefore presents evidence for rapid hybridization instability only in 37 °C serum and only in the case of MORF15, either as the homoduplex or the heteroduplex. Analysis of Urine from Mice Administered Radiolableded Polymer. Figure 3 (panel A) presents a HPLC radiochromatrogram of urine pooled from three mice at 0.5 h post administration of PA-cMORF1599m Tc-MORF15. The radiochromatogram shows one prominent peak at 28 min suggestive of free 99mTc-MORF15. That this peak is indeed due to free MORF is shown by the shift in the radiochromatographic profile (panel B) after the addition of PA-cMORF15. The addition has resulted in a complete shift in the radioactivity profile

Figure 4. Whole body radioactivity levels vs time in normal mice receiving PA-cMORF15-99mTcMORF15 (0, 15/15), PAcMORF25-99mTc-MORF25 (b, 25/25) or PA-cMORF25-99mTcMORF15 (2, 25/15).

to higher molecular weight (earlier retention time) due to polymer hybridization. That the urine from mice receiving PA-cMORF25-99mTc-MORF25 contained much less radioactivity (15%ID) is further evidence for increased instability of the latter complex. A shift assay was not attempted on urine from mice receiving PA-cMORF25-99mTc-MORF25 because of the lower radioactivity level. Whole Body Radioactivity Retention. The whole body radioactivity over 18 h post administration of the three radiolabeled polymers is shown in Figure 4. The average whole body radioactivity fell much more rapidly (only 35 vs 70% remaining at 18 h) following administration of 99mTc-MORF15 whether on a homoduplex or heteroduplex. Since all three polymer have molecular masses beyond the 60 kDa cutoff for glomerular filtration, the rapid clearance must be the result of in vivo instability. Furthermore, urine radioactivity obtained from mice receiving PA-cMORF25-99mTc-MORF15 and PA-cMORF1599m Tc-MORF15 was essentially identical and much higher than that from animals receiving PA-cMORF25-99mTcMORF25 (20-30 vs 2-5% ID). This evidence shows that an important fraction of 99mTc-MORF15, whether hybridized in a homoduplex or heteroduplex, rapidly dissociated in vivo. Biodistribution in Normal Mice. Table 2 presents the biodistribution at 18 h in percent injected dosage per gram for the three radiolabeled polymers. Both polymers radiolabeled with 99mTc-MORF15 exhibited significantly lower liver and spleen level but higher kidney level than that of the PA-cMORF25 homoduplex. Despite being

1022 Bioconjugate Chem., Vol. 14, No. 5, 2003

He et al.

Table 2. Biodistribution in Normal Mice at 18 h Post Administration of Polymera organ

25/25

25/15

15/15

liver heart kidney lung spleen muscle blood whole body

20.5(4.24) 0.53(0.18) 1.47(0.39) 0.44(0.10) 16.5(6.83) 0.08(0.07) 0.15(0.12) 69.0(0.82)

12.3(2.13) 0.15(0.02) 2.62(1.51) 0.19(0.05) 5.61(0.71) 0.03(0.01) 0.02(0.0) 29.7(2.06)

17.9(5.33) 0.44(0.20) 3.67(1.17) 0.71(0.13) 7.26(1.81) 0.07(0.01) 0.05(0.02) 34.0(6.53)

a PA-cMORF15-99mTc-MORF15 (15/15), PA-cMORF25-99mTcMORF25 (25/25) and PA-cMORF25-99mTc-MORF15 (25/15). %ID/g (standard deviation in parenthesis). N ) 5.

hybridized to the same polymer, the biodistribution of PA-cMORF25-99mTc-MORF15 is obviously more similar to that of PA-cMORF15-99mTc-MORF15 than PA-cMORF25-99mTc-MORF25.

24 h. However, after only 1 h in 37 °C serum, the homoduplexes and the heteroduplex prepared with 99mTc-cMORF15 showed by HPLC evidence for free 99m Tc-cMORF15. The 99mTc-cMORF15 was also shown to be the most prominent radiolabeled species present in urine of mice receiving the radiolabeled polymers. Furthemore, whole body retention of radioactivity was much lower in animals administered polymers labeled with 99m Tc-MORF15. Evidence for free 99mTc-MORF25 was in all cases much less prominent. In conclusion, hybridization stability in the case of morpholinos was shown to be dependent upon chain length when subjected to 37 °C serum environments both in vitro and in vivo. If this phenomenon can be shown to be general and not restricted to the particular base sequence selected for this investigation, then hybridization stability may be an important criterion in the selection of oligomers as potential radiopharmaceuticals. ACKNOWLEDGMENT

DISCUSSION

The use of oligomers such as DNAs, PNAs, and MORFs in the design of radiopharmaceuticals would benefit from an improved understanding of their in vitro properties and their behavior in vivo. That oligomers may exist in a large number of distinct chemical forms without apparent impairment of their hybridization properties is an advantage not enjoyed by many biologicals and will explain the large variety of available oligomers. Just as phosphorothioate DNAs display properties unique from that of the phosphodiester DNAs, PNAs, and MORFs display their own unique properties. It may therefore be possible eventually to select among different oligomers for those with particular properties suited to a particular application. This may be especially true for oligomers because of the ability to vary not only the backbone supporting the nitrogenous bases but also both the chain length and base sequence (13, 14). We have recently noted that the pharmacokinetics of MORFs when radiolabeled with 99mTc show important differences depending upon chain sequences and base lengths (8). The objective of this present investigation was to extend these observations by examining the influence of chain length on the hybridization stabilities in vivo of two MORFs. Using surface plasmon resonance, both the association and dissociation rate constant, and therefore the affinity constants for all four combinations of MORFs were measured. The affinity constants for both homoduplexes averaged about 109 M-1 and therefore are respectable for PNA duplexes (15) or DNA duplexes (5) of comparable lengths. As shown in Table 1, the 25-mer homoduplex displayed the highest affinity of the four combinations while one of the heteroduplexes, 25-mer MORF on immobilized 15-mer cMORF showed the lowest affinity by about a factor of 13. Surprisingly, the other heteroduplex, 15-mer MORF on immobilized 25-mer cMORF showed affinities comparable to that of the homoduplexes. Possibly the difference may be due to steric hindrance exhibited when hybridizing the 25-mer MORF to the small 15-mer, that may be less severe when the 15-mer is hybridized to the immobilized 25-mer. While affinity constants may reflect in vivo stabilities, they do not necessarily predict the kinetics of dissociation. The evidence of this investigation is definitive in showing instabilities toward hybridization dissociation in the case of the 15-mer MORF in 37 °C serum and in vivo. Incubations in saline at room temperature for the homoduplexes and the heteroduplex of this investigation showed by HPLC no evidence of dissociation over at least

Financial support for this investigation was provided in part by the National Institutes of Health (CA 79507). The authors are grateful to Dr. Kara Herlihy (BIAcore, Inc., Piscataway, NJ) for helpful suggestions and Mr. Bhavesh Patel (Theseus Imaging, Boston, MA) for valuable assistance with the BIAcore measurements. NOTE ADDED AFTER ASAP POSTING

Changes to the captions of Figures 3 and 4 were made to the version posted August 27, 2003. The corrected version was reposted August 29, 2003. LITERATURE CITED (1) Hnatowich, D. J., Winnard, P., Jr., Virzi, F., Fogarasi, M., Sano, T., Smith, C. L., Cantor, C. R., and Rusckowski, M. (1995) Labeling deoxyribonucleic acid oligonnucleotides with 99mTc. J. Nucl. Med. 36, 2306-2314. (2) Hnatowich, D. J., Mardirossian, G., Fogarasi, M., Sano, T., Smith, C. L., Cantor, C. R., Rusckowski, M., and Winnard, P., Jr. (1996) Comparative properties of a technetium-99mlabeled single-stranded natural DNA and a phosphorothioate derivative in vitro and in vivo in mice. Pharm. Exp. Ther. 276, 326-334. (3) Mardirossian, G., Lei, K., Rusckowski, M., Chang, F., Qu, T., Egholm, M., and Hnatowich, D. J. (1997) In vivo hybridization of technetium-99m-labeled peptide nucleic acid (PNA). J. Nucl. Med. 38, 907-913. (4) Mang’era, K., Liu, G., Wang, Y., Zhang, Y., Liu, N., Gupta, S., Rusckowski, M., Hnatowich D. J. (2001) Initial investigations of 99mTc-labeled morpholinos for radiopharmaceutical applications. Eur. J. Nucl. Med. 28, 1682-1689. (5) Niemeyer, C. M., Burger, W., and Hoedemakers, R. M. (1998) Hybridization characteristics of biomolecular adaptors, covalent DNA-streptavidin conjugates. Bioconj. Chem. 9, 168-175. (6) Hashem, G. M., Pham, L., Vaughan, M. R., and Gray, D. M. (1998) Hybrid oligomer duplexes formed with phosphorothioate DNAs: CD spectra and melting temperatures of S-DNA. RNA hybrids are sequence-dependent but consistent with similar heteronomous conformations. Biochemistry 37, 61-72. (7) Jaroszewski, J. W., Clausen, V., Cohen, J. S., and Dahl, O. (1996) NMR investigations of duplex stability of phosphorothioate and phosphorodithioate DNA analogues modified in both strands. Nucleic Acids Res. 24, 829-834. (8) Liu, G., Zhang, S., He, J., Liu, N., Gupta, S., Rusckowski, M., and Hnatowich, D. J. (2002). The influence of chain length and base sequence on the pharmacokinetic behavior of 99mTc-morpholinos in mice. Quart. J. Nucl. Med. 46, 233243. (9) Winnard, P., Jr., Chang, F., Rusckowski, M., Mardirossian, G., and Hnatowich, D. J. (1997) Preparation and use of NHS-

Stability of Mice Morpholino Duplexes MAG3 for technetium-99m labeling of DNA. Nuclear Med. Biol. 24, 425-432. (10) Wang, Y., Chang, F., Zhang, Y., Liu, N., Liu, G., Gupta, S., Rusckowski, M., and Hnatowich, D. J. (2001) Pretargetingwith amplification using polymeric peptide nucleic acid (PNA). Bioconj. Chem. 12, 807-816. (11) Jonsson, U., Fagerstam, L., Ivarsson, B., Johnsson, B., Karlsson, R., Lundh, K., Lofas, S., Persson, B., Roos, H., Ronnberg, I., Sjolander, S., Stenberg, E., Stahlberg, R., Urbaniczky, C., Ostlin, H., and Malmqvist, M. (1991) Realtime biospecific interaction analysis using surface plasmon resonance and a sensor chip technology. Biotechniques 11, 620-627.

Bioconjugate Chem., Vol. 14, No. 5, 2003 1023 (12) Malmqvist, M. (1993) Biospecific interaction analysis using biosensor technology. Nature 36, 186-187. (13) Braasch, D. A., and Corey, D. R. (2002) Novel antisense and peptide nucleic acid strategies for controlling gene expression. Biochemistry 41, 4503-4510. (14) Hnatowich, D. J. (2000) Antisense and Nuclear Medicine. Where are we now? Cancer Biother. Radiopharm. 15, 447-457. (15) Jensen, K. K., Orum, H., Nielsen, P. E., and Norden, B. (1997) Kinetics for hybridization of peptide nucleic acids (PNA) with DNA and RNA studied with the BIAcore technique. Biochemistry 36, 5072-5077.

BC0341019