Everyone knows about genes. Genes
are information. They meticulously describe every component that makes up and
runs the tiny cells that form our bodies. The long path between the
single-cell fertilized egg and the billions of coordinated cells that make up
a human being is basically a sequential story of this genetic information
being read, translated and used to incrementally reach the next small step. As
cells divide, they use this gene-encoded information to change their
structure, function and interaction. Thus when the time comes to form the nose
and mouth of a developing baby; inherited information is accessed, subtle
changes are wrought, and a grandfather’s smile beams back at him from his
grandson’s face.
A salient point is that every
single cell contains the same genes, inherited from the fertilized egg – it
is how, when (and how much) each of the individual genes is used that makes
all the difference. This process is called gene expression. Theoretically, if
we had a complete and exact understanding of the timing and level of the
expression of every gene during development we would be a long way towards a
real understanding of the overall process.
Until very recently, obtaining such
a complete and exact understanding was the stuff of science fiction.
Determining the expression of a single gene has been a meticulous process and
there are thousands of genes simultaneously expressed in most cells. However,
recent advances brought about by an unlikely collision between genetics and
“Silicon Valley” high tech have paved the way for the necessary
simultaneous and accurate expression analysis to become a reality. So called
“gene chips”, developed through the same basic technology used to make
computer chips, are revolutionizing biomedicine and will soon become standard
tools in almost every aspect of medical diagnostics and practice.
Gene chips “work” based on a
basic aspect of the genetic molecules - each gene sequence is unique and
defines a complementary physical structure that is also unique. In other
words, the section of DNA that defines the gene for the blood protein,
hemoglobin, is a three dimensional structure. One aspect of this structure is
that it consists of two complementary components and these will only recognize
and bind to their own complementary partner. Therefore, even in a complex
mixture of thousands of individual gene molecules, only the complementary
components will recognize each other and bind together.
Gene chips are made by precisely
positioning multiple copies of chemically-synthesized DNA pieces corresponding
to known gene sequences onto the surface of the chip substrate. The resulting
chip is a matrix of microscopic, tightly defined sections each containing
sequence corresponding to an individual gene. Current manufacturing
methodology allows for precisely creating literally thousands of such
individual regions on the chip surface aligned in a known pattern. Thus a gene
chip can represent the entire genetic information of an organism, spread out
like a tiny checkerboard. During use, a solution containing the DNA or message
RNA from the target cell is flooded across the chip surface and the DNA/RNA
sequences in the cellular sample are allowed to interact with the chip
sequences under carefully controlled and optimized conditions. When
complementary sequences interact, they tightly bind. Therefore if gene
sequences for the hemoglobin gene are present in the cellular sample, they
will uniquely bind only to the tiny square of the chip “checkerboard”
containing the hemoglobin sequence. This is explained graphically in figure
1.
Figure 1. The gene chip
The key to monitoring this binding
is that the cellular DNA/RNA sample is first labeled with a fluorescent dye.
The chip can then be “read” by ultra-sensitive fluorescent detectors that
simply quantify the level of the dye-DNA present at each tiny, gene-specific
spot on the chip. This level provides an accurate measure of the amount of
each gene sequence present in the original sample. Thus in a single run, the
presence and level of every gene can be accurately determined.
The greatest challenge in applying
chip analysis to embryonic samples involves the fact that the amounts of DNA/RNA present in these samples is
minute. In standard medical analyses, samples are created from volumes of
blood or large biopsies containing thousands of cells. Embryonic samples of
interest consist of only a few or perhaps single cells. Our laboratory has
pioneered techniques through which the minute amount of DNA/RNA present in a
single cell (for instance an egg or single embryonic cell) can be amplified to
produce a representative sample that is suitable for use with gene chips and
other protocols.
Using gene expression chips, a
single experiment can provide solid information on the expression of thousands
of genes. This expression profile is essentially a “snapshot” of what is
going on at that time in the cell. While this information is itself of great
interest, the real power of such experiments emerges when comparisons can be
made between different cells, time points, treatments, and underlying
conditions. For instance, in reproductive biology we would like to know how
the pattern of expressed genes changes as the early embryo divides and
developmental changes are made. Knowing which genes are critical to normal
early developmental processes will be critical to understanding these
processes during assisted reproductive protocols. Of potentially even greater
interest will be the ability to compare gene expression patterns between
normal “good quality” embryos and those manifesting developmental
problems. These comparisons will no doubt reveal genes that are either
directly involved in abnormal development or genes that are markers of such
development. Such marker genes could potentially be used in diagnostic
scenarios to identify embryos with normal or abnormal development. Such
expression analysis could also potentially identify embryos that harbor
genetic disease mutations or other abnormalities.
One of the most difficult
components of chip analysis concerns dealing with the enormous amount of data
generated. As stated, for every experiment, the level of thousands of genes is
available. However, the majority of these may not play a critical role in the
process of interest. Sifting through this information for critical components
and important variation is facilitated by a variety of software programs that
organize and analyze chip data. Figure 2 shows part of a typical analysis
output from such a program called GenMapp. In this case, the gene expression
patterns of two groups of oocytes were compared. Genes that are expressed at a
higher level in one group in relation to the other are indicated in red while
those expressed at a lower level are indicated in blue. This section
represents an analysis of just a few dozen genes out of over 9,000 analyzed.
In this comparison, the gene for the oocyte protein known as CENPE which is
involved with the regulation of chromosome behavior was expressed at a lower
level in the compared group.
Figure 2. Gene expression
comparison between 2 groups of human embryos
This type of difference could
indicate an underlying deficit in important reproductive processes such as
proper chromosome segregation. This is exactly the type of information that
will be critical to developing a specific understanding of such
processes.
Detailed expression analysis with
human reproductive material is just beginning. This work will no doubt be one
of the most important developments in the history of assisted reproduction
research. Information from this analysis will provide an exciting new window
and perspective on the basic processes of reproduction. Hopefully, in just a
few years, it will begin to have real, practical impact on the improvement of
clinical treatment options available to infertile patients.
