Imaging Genetic Information - Nanotechnology for Biomedical Imaging and Diagnostics - page 367

13.2.1 dna as a Biomarker

While chromosomal DNA is the primary source of genetic information, it is only
present in two copies, which would require an extremely sensitive imaging agent
with a very low background signal. In a typical mammalian cell, a single DNA copy
of a DNA sequence would correspond to about 1 pm concentration (10 −12 m). Some
sequences of DNA exist as multiple copies such as the DNA coding for rRNA, for
which 300-400 copies are dispersed over five human chromosomes. There are also
genetic diseases that have multiple sequence repeats that can reach thousands of
bases, such as Friedreich's ataxia, fragile X syndrome, and Huntington's disease
[20], which would also greatly increase the target sequence concentration.
Aside from concentration, DNA is not a particularly good target for imaging
because it exists almost exclusively in a duplex form and is therefore inaccessible for
base pairing by simple single-strand antisense agents. DNA can be recognized in its
duplex form, however, by triplex-forming ODNs (TFOs) [21, 22], peptide nucleic
acids (PNAs) [23, 24], minor groove binding polyamides [25], or engineered zinc
finger proteins [26, 27]. The TFOs require polypurine tract sequences, which need to
be quite long to be unique (18 or so nucleotides for humans) and are not as unique.
On the other hand, polyamides and zinc finger proteins can recognize heterogeneous
DNA sequences by hydrogen bonding interactions in the minor groove and can be
made to recognize quite long sequences. A strategy for detecting triplet repeat DNA
by hairpin PNAs has recently been described [28], and a self-assembling catalytic
zinc finger system based on β-lactamase has been developed that greatly amplifies
the signal from the binding of two zinc fingers to a single DNA target sequence [29].
Another problem with genomic DNA as a target is that nontranscribed DNA is tightly
packaged into chromatin, and binding of a probe to the nucleosome-bound sequences
(150 mer stretches of DNA) would be greatly inhibited. Further complicating mat-
ters, chromosomal DNA is sequestered in the nucleus and protected by a double
nuclear membrane that must be traversed by the imaging agent (Fig. 13.3). Transport
across the nuclear membrane can be facilitated by attachment of nuclear localization
peptide sequences (NLS), which have been used to deliver nucleic acids into the cell
nucleus through the nuclear pore [30]. The most serious concern in using chromo-
somal DNA as a target is that agents that bind to the DNA might also induce DNA
mutations, deletions, and rearrangements that could lead to cancer, as has been
observed for triplex-forming oligonucleotides [31, 32]. Overall, the low copy number,
duplex nature of the DNA, and potential for mutating the DNA do not make chromo-
somal DNA itself an attractive target for genetic imaging agents.
mitochondrial DNA, unlike chromosomal DNA, is present in a much higher copy
number in a cell, with about 5 copies per mitochondrion, and 1000 mitochondria/cell,
bringing the total to about 5000 copies/cell or 5 nm in concentration. mitochondrial
DNA, however, is a circular DNA that has only 16,532 bp compared to the 2 billion
base pairs of chromosomal DNA and only encodes 13 proteins, 22 transfer RNAs
(tRNAs), and 2 rRNAs [33]. As such, there is much less in the way of disease
information to be imaged, though more than 300 mutations have been identified that
are associated with disease symptoms. mitochondrial DNA is not very accessible, as
it is sequestered in the mitochondrion by two membrane barriers (Fig. 13.3). Signal