Download Technetium-99m Radiolabeling: Oligonucleotide Binding and Label Stability Study and more Lecture notes Engineering in PDF only on Docsity!
are used, either to block transcription of genes within the
nucleus of cells or to block translation of messenger RNA
within the cytoplasm (1,2). Accordingly, if radiolabeled, these oligonucleotides may usefully carry radioactivity to
targeted cells or tissues. This and other applications of
these reagents as radiopharmaceuticals, however, will re
quire that methods be developed for radiolabeling them
with diagnostic and, possibly, therapeutic radionuclides.
Methods for radiolabeling oligonucleotides with beta-emit
@ ting radionuclides such as 3H, and 32P are well estab
lished (3—6);however, the use of oligonucleotides radiola
beled with imageable radionuclides is in its infancy.
Methods have been reported for labeling oligonucleotides
with ‘@I(6,7) and, as such, these methods should be useful
with ‘@Iand ‘@‘Iradionuclides with imaging properties.
Depending on the pharmacokinetic properties of an oligo
nucleotide of interest, it is likely that the imaging radionu
clide of choice will often be 99mTc Recently, a method for
radiolabeling DNA with @Tcwas described which relies
upon a derivative of DTPA attached to the oligonucleotide
to form chelates with reduced @‘@‘Tc(8). DTPA has previ
ously been used for radiolabeling antibodies with @“Tcbut
was abandoned when the instability of the label was judged
to be unacceptably high (9). Accordingly, this study was
performed to investigate an alternative approach to label
oligonucleotides stably with @Tc.A hydrazino nicoti
namide (SHNH) moiety has been shown to form stable
complexes with @Tcwhen conjugated to antibodies (10).
In this investigation, an amine-derivatized DNA was con
jugated with SHNH and the properties of the label evalu
ated. To provide a useful comparison, the same DNA was
alsoconjugatedwithDTPA,as hasbeendescribedbyoth
ers (2,11), for radiolabeling with “Inusing procedures
routine for antibody labeling (12).
In this investigation, 22-base, single-stranded DNASwere
selected because their small size (about 8 kDa) was ex
pected to facilitate rapid whole-body clearance after admin
istration. Moreover, a 22-base oligonucleotide is large
enough to virtually exclude any possibility of an accidental
match within the genomic DNA and its transcripts (1).
DNAS used were derivatized on one end with a biotin
moiety to provide a useful means of establishing labeling
efficiency and label stability. Using this feature, radiola
beled DNAS may be easily distinguished from unbound or
@ngIe-strandedRNA and DNA oligonucleotides may be useful
as radiopharmaceuticais for antisense and other in vivo appli
cations ifconvenient methods for stably attaching radionuclides such as @‘Tccan be developed. Methods: To radiolabel DNA with @‘1c,we have used the hydrazino nicotinamkle (SHNH)
rnolety devebped elsewhere. The diethylenetriaminepenta
acetic acid (DWA) chelate was used to label DNAwith 111lnfor comparison. Complementary 22-base, single-stranded oligonu cleotides were obtained, each wfth a pnmary amine attached to
either the 3' or 5' end and witha biotinmoietyon the opposite
end. The DNA was conjugated with SHNH by a N-hydroxy succinimide derivative and with DTPA by the cyclic anhydride.
Results: Reversed-phase HPLCanalysis showed that essen
tially complete conjugation was achieved in both cases. The
purifiedSHNH-DNAwas radiolabeledwith @rcby transche
lation from glucoheptonate at labeling effiCiencieSof up to 60%
and DTPA-DNA @4th111Inacetate at up to 100% efficiency.
After labeling, the ability of the DNAs to bind to streptavidin
through the biotinmoieties and to hybridIzewiththeir compie
mentary DNA in saline was retained for both radiolabels as
determined by size-exclusion HPLC analysis. HPLC radiochro matograms of serum incubates showed a shift of @‘@‘Tc,but not
1111n,to a high molecular weight, strongly suggesting serum
protein binding in the fomier case only. Low-molecular wsight degradation products were seen with ‘11ln,but not with @Tc and may be related to the use of phosphodiester-linked oligo nucleotides. As a further measure of label stability, the DNAs were bound to strepta@Adin-conjugated magnetic beads and incubated in fresh 37°Chuman serum. Less than 4% of @Fc
and 14% of 1111nwas lost in 24 hr. Conclusion: Amino-modi
fled,single-stranded DNAcan be stably radiolabeledwith @1c
by the SHNHmoietywithoutloss of function.
Key Words oligonucleotides;technetium-99m;radiolabeling
J NuciMed1995362306-
he extraordinary properties of DNA and RNA suggest
that there is potential for the use of these oligonucleotides
as radiopharmaceuticals. For example, there is current in
terest in antisense applications in which oligonucleotides
ReceivedAug.4,1994;revisionacceptedDec.29,1994.
For correspondence or reprints contact: D.J. HnatOWich,PhD, Department of NudearMedicine,Universftyof MassachUsettsMed@alCent&,Worcester,MA
2306 The Journal of Nuclear Medicine•Vol.36 •No. 12 •December 1995
Technetium-99m Labeling of DNA
Oligonucleotides
D.J. Hnatowich, P. Winnard Jr., F. Virzi, M. Fogarasi, T. Sano, C.L. Smith, CR. Cantor and M. Rusckowski
Department of Nuclear Medicine, University of Massachusetts Medical Center, Worcester, Massachusetts, and Center for Advanced Biotechnology and Departments of Biomedical Engineering Phartnacolo@jiand Biology, Boston University, Boston, Massachusetts
0
@ 0 00 BrnN @ oxo,_@
FiGURE1. Structureof A and B DNA
chains used in this investigation.
dissociated label by size-exclusion analysis before and after
the addition of streptavidin (13).
In this study, we describe the results of conjugating 5ev
eral single-stranded DNAS with SHNH and DTPA fol
lowed by labeling with @Tcand ‘‘‘In,respectively. The
stability of the label was determined in serum, and biodis tribution studies were performed in normal mice.
MATERIALS AND METhODS Four 22-base, single-stranded DNA sequences were purchased
(Operon Technologies,Alameda,CA) for this investigation.The
base sequences were 5'-biotinTA ATA CGA CFC ACF ATA GGG AGamine-3' (A-chain) and 5'-amineGG TAC AGG TCf
CAC TGT ATG ACbiotin(C-chain)and their complements(B-
and D-chains, respectively). Figure 1 presents the structure of the
A- or B-chain (i.e., with the biotin moiety on the 5' end). As
shown, the biotin moiety was attached directly through a 15- member amide-polyether linker to the terminal phosphate, while the amine was normally attached to the terminal phosphate group through a 6-member methylene carbon spacer. The molecular weight of each chain was about 8.2 kDa. The melting and anneal
ing temperatures in physiologicalsaline were calculated to be
57—60°Cand 32—41°C,respectively, for both pairs (14). The DNAS were purchased unpurified and were used without further purifi cation. They were generally handled under sterile conditions; all solutions were sterilized by terminal filtration through a 0.22-sm filter, and sterile pipette tips were used. All other pipette tips and tubes were autoclaved prior to use. Streptavidin was purchased and used without further purifica tion. The DNAs were stored dry as received, at refrigerator tern peratures and were dissolved at a concentration of 1—4mg/ml in sterile water when needed. After solubilization, 20—500p.g DNA
were added to sterile plasticvialswhichwere immediatelyfrozen
for storage. Technetium-99m-pertechnetate was obtained from a @Mo@@mTcradionuclide generator and ‘‘‘Inwas purchased as the chloride complex. Streptavidin-conjugated superparamagnetic polystyrene beads (Dynabeads M-280, Dynal, A.S., Lake Success, NY) were stored at refrigerator temperatures as recommended by the manufacturer. The N-hydroxysuccinimide (NHS) derivative of the hydrazino nicotinamide moiety was a gift from Dr. M. Abrams,
(JohnsonMattheyInc., West Chester, PA). The cyclicanhydride
of DTPA was synthesized as previously described (15).
Oligonucleotide Conjugation
In this investigation,all four DNAchainswere derivatizedwith
either SHNH or DTPA. In the former case, the NHS-derivative of SHNH moiety was dissolved in dry dimethylformamide to a con centration of 8.6 mg/ml (10). For conjugation, the DNA solution
was diluted with a sterile bicarbonate buffer so that the final
concentrations were 2.0 mg/ml DNA, 1 M NaCl, 0.25 M NaHCO3, 1.0 mM EDTA at pH 8.3—9.0.The EDTA was added to complex cations such as calcium. The DNA solutions were incubated for 30
mmat 45°Cjustprior to SHNHadditionto dissociateanysecond
ary complexes between the primary amines and the phosphate
backbone.Between0.2 to 1 mgof DNAwasconjugatedat a 1—
molarexcessof SHNHto DNAbyaddingthe necessaryvolumeof
dimethylformarnidesolutiondropwiseto the DNA solutionwhile
vortexing.The solutionwas incubatedat room temperature for 1
hr. Followingincubation,the conjugatedoligonucleotidewas pu
rifled on a 0.7 x 20 cm gel filtration column of P-4 using a sterile 0.25 M ammonium acetate, 1.0 mM DTPA buffer at pH 5.2 as eluant. The DTPA was added to complex excess stannous ion and help prevent radiocolloid formation. Fractions (0.4 ml) were col lected and the absorbance (260 nm) of each measured. Oligonu
cleotideconcentrationswere estimatedusingan extinctioncoeffi
cient determined in this laboratory of 30 @.d/@gfor a 0.1% solution
measuredat 260nm.The absorbanceof SHNHat thiswavelength
and under these conditions was found to be insignificant. The fractions collected contained DNA at concentrations in the 0.5—1.
mg/mirange.
ConjugationwithDTPAwasachievedwiththe cyclicanhydride
as previously described for proteins (12). Approximately 1 mg of the oligonucleotide was dissolved in 10 ml 0.4 M HEPES buffer, pH 8.5, and this solution was added rapidly to a round-bottom test tube containing 100 mg of the thy anhydride while vortexing. Thus,
the DTPA:DNAmolar ratio was 2110:1.The solutionwas incu
bated at room temperature for 30 mm. The conjugatedoligonu
cleotides were originally purified on a gel-filtration column of P- using 10% ethanol as an eluant, but the inability to remove an
unidentifiedradiochemicalcontaminantusingthis columnneces
sitatedthe use of an alternativepurificationmethod.The reaction
Labeling DNA Oligonucleotides with Technetium-99m •Hnatov@chat al. 2307
T c@,o 3.ri,N@
L0X0@ 120
0 0 B.s•N 0'
NH,
IC) c\
z
Cl)
@ 0.
RESULTS
Oligonucleotide Conjugation
In this study, the amine-derivatized DNAS were conju
gated with an NHS derivative of SHNH and with the cyclic
anhydride of DTPA. Figure 2 presents UV absorption chro
matograms obtained by reversed-phase HPLC of unmodi
fled A-chain (panel 1), SHNH-conjugated A-chain prior to
purification (panel 2), and DTPA-conjugated A-chain after
purification (panel 3). The unmodified A-chain is clearly
resolved into one major peak at a retention time of about
12 mm and several minor peaks.
Conjugation with SHNH was accomplished at molar ra
tios of SHNFI:DNA of between 1—25:1.No differences in
the ability of the conjugated oligonucleotide to hybridize or
to accept a @Tclabel were observed at these molar ratios
(data not presented). Accordingly, all subsequent conjuga
tions were performed at a 1:1 molar ratio. After conjugation with either SHNH or DTPA, the pres
ence of unmodified A-chain was reduced to a minor con
stituent (Fig. 2). The peak absorbance shifted to 21 mm in
the SHNH case and 3 mm in the DTPA case. The unmod
ified DNA peak is essentially absent, demonstrating that
conjugation was largely complete.
Because DTPA-coupled oligonucleotides purified on a
P-2 column contained unacceptably high levels of uniden
rifled radiochemical contaminant(s) after radiolabeling; pu
rification was ultimately achieved on an anion exchange
column. Figure 3 shows the UV absorbance profiles of
fractions off this column. The high acidity (pH 2.7) of the initial eluant apparently results in complete protonation of
free DTPA which then elutes from the column early. The
DNA, however, still possesses a high negative charge
through its phosphate groups and is retained. As the ionic
strength is increased, however, the increasing sodium ion
concentration apparently neutralizes this charge and per
mits the DNA to elute. The three major peaks (eluting at
20, 100 and 150 ml in the figure) were analyzed by gel
TIME (mm.)
FiGURE2. UV absorptionchromatogramsobtained by re
versed-phase HPLC analysis of unmodified A-chain (trace 1), of SHNH-conjugated A-chaln prior to purification (trace 2) and of DTPA-conjugated A-chaln prior to puiiflcation(trace 3).
washed three times with the washing buffer. The serum, the com bined washes and the beads were then counted in a NaI(Tl) well counter.
AnimalBiodistñbutionShidues Biodistributions of both @“Tc-and ‘DIn-labeledD-chain were determined in normal CD-i male mice. Each animal received by
tail vein administration0.1 ml of saline containingeither 7.5 @g
@ (7.3 MCi)“In-or 10 (7 MCi)99mTc..labeledDNA. Animals
were killed by spinal dislocation 2.5 hr postadministration. Sam ples of organs were rinsed in cold saline and were counted along
with a blood sampleand an aliquot of the injectatein a NaI(Tl)
wellcounter.The biodistributionswerereported as the percentage
of administered radioactivity per gram of tissue.
vol. (ml) 0— 50— 144 145—
1% HAc, pH 2. 17@NH 4Ac. pH 5. 1% HAc, pH 2.
0.
0.
@ 0.3^ z
a 0
c,i
FiGUREa uv absorptionprofileof free
tions obtained by anion exchange chroma tography of DTPA-conjugatedC-chsin. Ofthe three major peaks (at 20, 100 and 150 ml), only the latter was shown to contain DNk
@@ :: r\ @: /
@@ :@ @JJ@@/
@ 0.0 --/@.
0 20 40 60 80 100 120 140 160 150 VOLUME (ML)
Labeling DNA Oligonucleotides with Technetium-99m •Hnatowich et al. 2309
FIGURE 4. Radiochromatograms ob
talned by size-exclusionHPLCanalysis of
99mTc@labeiedC-chain (leftcolumn)and 111ln labeled C-chain (right column) in saline (trace 1), in saline containing biotin-saturated streptavidin (trace 2), in saline containing streptavidin (trace 3) and in saline containing complementary D-chain bound to streptavi din (trace 4).
electrophoresis. Only the latter peak contained DNA (see
below). The recovery of DNA in this purification was 16%.
Oligonucleotide Labeling
@ When 100 or more of the SHNH-conjugated DNAS at
a concentration of at least 250 @tWmIwere radiolabeled with
up to 1 mCi @°@Tc,labeling efficiencies ranged between
30%—60%,as determined by Sephadex 0-50 chromatogra
phy. Attempts to label smaller amounts of DNA or at lower
concentrations resulted in reduced efficiencies. These ob
servations were independent of the DNA chain. As con trols, the unmodified DNAS were radiolabeled in the iden tical fashion. Labeling efficiencies were less than 5% in the case of each 99mTc control. When the @“Tc-labeledoligo nucleotides were analyzed by size-exclusion HPLC, a single
peak was always observed. Recoveries occasionally ap
proached 100%, although typical recoveries were 81% ±
5% (s.d., n = 4). That recoveries could be increased by increasing the salinity of the eluant suggests that this reten
tion was due, in part, to ionic interactions of the charged
DNAS with the column support.
Typically, specific activities of 50 @Ci4tgDNA were
achieved when glucoheptonate was replaced with tricine. Properties of the labeled DNAS such as biotin binding,
hybridization, protein binding in serum or serum stability
appeared to be unchanged with respect to DNA labeled via
glucoheptonate.
Labeling efficiencies of the DTPA-conjugated and
DEAE-purified oligonucleotides approached 100% under
most conditions of DNA concentration and at specific ac
0 U)
tivities of 60 @Ci/.tg.The unconjugated C-chain under
identical conditions retained only 2% of the “Inlabel.
Figure 4 presents size-exclusion HPLC radiochromato
grams of C-chain radiolabeled with 99mTc(left column) and
“In(rightcolumn)before(trace1) andafter(trace3) the
addition of excess streptavidin. Both labeled DNAS elute in
a single peak which largely shifts to a higher molecular
weight (i.e., smaller elution volumes) with the addition of
the streptavidin. Similar shifts to high molecular weight
were observed with the addition of the streptavidin-D-chain
(complementary) construct (trace 4). Also in Figure 4 are
radiochromatograms showing no shift with the addition of
biotin-saturated streptavidin (trace 2). Recoveries in these
HPLC analyses were always 80%—90%.Because
[ @TcJpertechnetateisretained,thegoodrecoveriesdem
onstrate that oxidation of the label to pertechnetate was not
an important mode of instability.
Elecfro@ Shidies
The first four lanes in the electrophoretogram of Figure
5 contain unmodified DNA chains A through D, respec
tively, stained with ethidium bromide. Whereas chains C
and D show only one band at the position expected for a
22-base DNA, that of chains A and B show two distinct
bands, one co-migrating with chains C and D plus an addi
tional band with an apparently lower molecular weight.
Several other minor bands corresponding to higher molec
ular weight have occasionally been seen as well. Lanes 5—
contain aliquots of three peak fractions (at 20, 100 and 150
ml, respectively) from the DEAE anion exchange purifica
99m 111 In
ELUTION VOLUME
2310 The Journal of NuclearMedicine•Vol.36 •No. 12 •December 1995
@@TcOrgan 111ln
% lD/g s.d. % big s.d.
Resultspresented as percent injecteddose per gram (% big) with
s.d.ofthemean.Thelevelsof @1cinalltissuesaresignificanflyhigher
than 111ln(p < 0.001, Student's unpairedt-test).
useful for in vivo imaging applications. Recently, a method
of radiolabeling oligonucleotides has been reported in
which a derivative of DTPA is covalently attached to an
amine group on a modified single-stranded DNA (2,21).
The derivitization of an amine-containing DNA with DTPA
and with another polyaminopolycarboxylate (ethylenedia
minetetraacetic acid) was earlier reported and used for
binding stable iron ions (11). DTPA and its derivatives
have been useful for labeling proteins such as antibodies
with 1111nand several other radionuclides (19), but they
have been less successful for labeling with @Tcbecause of
poor label stability (9). Since label stability was a major concern of this investigation, the SHNH moiety was used as
an alternative to DTPA for labeling DNA with 99mTc.The
SHNH moiety was developed for labeling antibodies with @Tc(10); this laboratoryhas demonstratedthat, when bound to antibodies by the SHNH moiety, @Tcdisplays
acceptable stability in vitro and in vivo (16). Accordingly,
the SHNH moiety was considered herein for DNA labeling.
As shown in Figure 2, the primaiy amino groups attached
to the oligonucleotides of this study were readily conju
gated with the NHS derivative of SHNH. Following conju
gation, the UV peak of the unmodified DNA virtually
disappeared on reversed-phase HPLC analysis of the con
jugated oligonucleotide. In its place appeared another peak
at greater retention time. The results with DTPA are iden
tical except that, in this case, the conjugated DNA eluted
earlier than the unmodified DNA. Whereas in past studies
from this laboratoiy of antibody conjugation with SHNH,
the presence of high molecular weight protein aggregates
was a concern (16), no evidence of DNA aggregation was
observed in this investigation. The concern in the use of the
cyclic anhydride of DTPA is not aggregation but cross linkingthroughthe twoanhydridegroups(19).As shownin
Figure 5, however, DNA dimers or oligomers, which would
100@
0111 In 75
99m
-@.%. Te
z 5Q@ 0
TABLE I
Biodistribution Results in Normal Mice 2.5 Hours Posthtravenous Administration of Technetium-99m and
Indium-i11-Labeled C-chains
Liver0.390.952.50.58Heart0.030.010.540.12KidneysI
.30.215.61.4Lung0.080.030.870.23Stomach0.210.392.50.67Spleen0.10.030.970.27Mus
(n=6)0.
(n=6)0.
TIME (hours)
FIGURE7. Percentagedissociationof @Tcand 1111nfromIa
baled D-chain bound to magnetic beads duting 37°Cincubation in fresh human serum for up to 24 hr. Errorbars represent I s.d. of the mean (n = 5 at all data points).
Forthe 1111n-labeledC-chain,althoughbindingto serum
proteins was not observed (Fig. 6), a second peak corre
sponding to lower molecular weights was seen after 1 hr in
serum and which became the prominent peak at 24 hr.
Because 1111nlabeled to proteins by DTPA has been shown
to be stable during serum incubations (19), the position of
this peak strongly suggests degradation, possibly of the
DNA phosphodiester-backboneby nucleases.
MimaI BlodistñbutionStud@s
The biodistribution at 2.5 hr postadministration to nor
mal mice are presented in Table 1 as the percentage of
injected @Tcand 1111n.Important differences between
labels are readily apparent. The levels of @“Tcin all tissues
were significantly higher than 1111n(p < 0.001, Student's
unpaired t-test). A partial explanation for these differences
most probably is related to the differences in serum protein
binding for the two injectates; as demonstrated above (Fig.
6), only the @Tc-DNAbinds to serum proteins. This
would raise activity levels in blood and, by virtue of the
blood content of tissues, raise activity levels in tissues as
well. From the percentage of organ weight due to blood
(20) and from the known blood activity, it is possible to
estimate that less than 15% of the activity levels in tissue of
Table 1 (other than heart and spleen) were due to the blood
pool. The differences between labels may therefore be a
consequence of other factors such as instability of the @1In
DNA to degradation in serum.
DISCUSSiON
The goal of this study was to develop a method of
labeling single-stranded DNAS with @“@Tcwhich would be
2312 The Journal of Nuclear Medicine •Vol. 36 •No. 12 •December 1995
be readily apparent in this analysis, are not evident. Finally,
there is little likelihood that treatment with either NHS or
cyclic anhydride would derivatize functional groups of the
bases themselves (11,22).
The SHNH-DNAs were radiolabeled with @mTcfollow
ing an approach essentially identical to that used routinely
for the labeling of SHNH-conjugated antibodies with this
radionuclide (10, 16). No attempt was made to maximize
labeling efficiency, yet typical values were 40%—60%with
@‘Tc.Resultsimprovedwiththe substitutionof tricinefor glucoheptonate as transchelator (17). Labeling at room
temperature proceeded more rapidly and the specific activ
ities achievable appeared to be much higher.
In this study, the determination of labeling efficiency and
label stability was aided by the use of oligonucleotides
derivatized with biotin. The peak @‘@Tcand “Inactivity in
the radiochromatographic profiles obtained by size-exclu
sion HPLC (Fig. 4) and the migration pattern on gel elec
trophoresis (Fig. 5) both showed a distinct shift toward a
higher molecular weight in the presence of streptavidin.
The fact that a similar shift was not evident when biotin
saturated streptavidin was added conclusively demon
strated that the label was on the oligonucleotide. The com
parable shift observed when the biotin-saturated,
streptavidin-bound complementary chain was added is fur
ther evidence. As shown in Figure 7, the biotin moiety was
also useful in demonstrating serum stability with the
streptavidin-conjugated beads.
That the shifts discussed above were not complete in any
case is probably an indication that the unpurified oligonu
cleotide preparations contained nonbiotinylated DNA
chains. For instance, the oligonucleotides used in this study
may have been contaminated with a variety of DNA species
with different chain lengths or missing the biotin moiety.
Polyacrylamide electrophoresis studies (Fig. 5) have clearly shown the presence of lower molecular weight contaminat ing DNA in preparations of A- and B-chains.
The stability of a label in serum is among the most
important factors for an agent under consideration for in
vivouse. Becauseof the highstabilityof @“@Tcon antibod ies facilitated by the SHNH moiety (16 ), it may not be surprising that @Tcshowed only 4% dissociation from
DNA when incubated on beads in serum (Fig. 7). More
meaningful, perhaps, is the stability in serum of the label
when incubated in solution. The rapid serum protein bind
ing of @Tc-DNMobserved in this study, however, inter
fered with the determination of stability in that fashion.
Nevertheless, some measure of serum stability is evident in
that the 99mTclabel was bound, at least in part, to serum
proteins by DNA rather than in some “free―chemical form
of 99mTc This is clear from the large percentage of label
which could be prevented from binding to proteins by the
addition of calf thymus DNA and which could be displaced
by streptavidin-conjugated beads.
It is instructiveto speculate on the mechanismof 99mTc..
DNA binding to serum proteins. It is tempting to suggest
that the binding may be related to the lipophilicity of the
6-member methylene linker by which the amine is conju
gated to DNA. An identical DNA chain but with a hydrox
yl-modified 3-carbon linker, however, was also used in this
research. Even though this linker is considerably less li
pophilic, the serum binding properties of the @“@Tc-labeled
oligonucleotide remained unchanged (data not presented).
The possibility also exists that the binding may be related to
the base sequences selected for this investigation. The same
DNA chain, however, showed no tendency towards serum
protein binding when labeled with “In(Fig. 4). Finally, an
alternative explanation exists because the SHNH moiety
does not, in itself, satisfy the chelation requirements of
reduced 99mTc Since the complex must be “capped―with
glucoheptonate or other species (10), it is possible that serum proteins may participate in this process with the result that the label would be seen to bind to serum pro teins. The nature of the binding will need to be established.
Although serum protein binding was not observed for the
“In-labeledC-chain, evidence of degradation upon serum
incubation was observed only in this case (Fig. 6). Degra
dation by nucleases of oligonucleotides with unprotected
phosphodiester backbones, such as that used in this re
search, is a common observation (1 ). That the @mTc@la@
beled C-chain shows no evidence of labeled degradation
products during serum incubation may be a consequence of
the serum protein binding.
CONCLUSION
Oligonucleotides derivatized with a primaiy amine can
be conjugated with SHNH and radiolabeled stably with
99mTc The stability of the label is comparable to that
observed by us for “Inradiolabeled to the same oligonu
cleotides by DTPA. One interesting observation from this
work is that @Tc-DNA,when labeled in this fashion, binds
to serum proteins. The binding is probably related to the
SHNH moieties since similar binding was not observed for
“In-DNA.A consequence of this binding was higher
99mTcblood levels in vivo, but another consequence may be increased stability to nucleases in serum since significant
degradation products were observed in this work only for
“In.Furtherinvestigationsshouldestablishwhetherse
rum binding of @“Tc-DNMwill interfere with targeting in
vivo.
ACKNOWLEDGEMENTS
The authorsthank Dr. MichaelAbrams,JohnsonMattheyInc.,
for providing the NHS-SHNH used in this investigation and Ms. Eleni Millona and Ms. Sophia Kim for their assistance in the
completionof the manuscript.Thisworkwassupportedin part by
DE-FGO2-93ER61656 from the U.S. Department of Energy and by CA59785 from the National Cancer Institute.
REFERENCES
,. UhlmannE,PeymanA.Antisenseoligonucleotides:anewtherapeuticprin
ciple. Chem Rev 1990;90:543—584.
- DewanjeeMK.Radiolabeledantisenseprobes:diagnosisand therapy.Diag Oncoll993;3:189—208.
Labeling DNA Oligonucleotides with Technetium-99m •Hnatowich at al. 2313
