Shapiro seminar and directed evolution

Joel Duff (
Sun, 27 Sep 1998 20:46:03 -0700


There was some recent discussion of Shapiro's views only days after which I
found out that Shapiro was to speak here at Southern Illinois University
under the title: "Evolutionary Theory at the End of the 20th Century"

Here are some of the notes that I took for anyone interested. Its a bit
jumbled but years of writing on the computer has left me nearly unable to
read my own writing.

Summary of talk:

Shapiro began by tracing some of the significant developments in science
which included Darwin, Rediscovery of Mendel, Discovery of DNA,
McClintock's mobile elements, Watson and Crick, and Operon theory. The
main point was that biology much like physics was atomistic in its thinking
(and from the rest of the talk, he thinks it still is and we need to move
on in the next century) in the beginning of the century. Not until operon
theory did we have a sense of the complexity of gene regulation and thus do
not see things as "one trait, one gene."

The fundamental change though is that we see cells as having the ability in
themselves to have control over their own change. The cell has the
machinery to rearrange its own genome. This then takes things out of the
realm of stochastic events - everything being a pure accident. This is
where his pseudo-directed evolution come into play later.

His two points at the top were:
1) Genomes function as integrated systems not as isolated objects
2) We now know because of molecular studies the potential is in cells for
potentially genome wide change (not local atomistic changes but genome

At this point he made a quick hit on engineers speaking about what the word
"complex" entailed. He said (wish I could quote but years of computer
typing have rendered me nearly helpless writing by hand) complexity does
not mean something with a of objects but organization, hierarchy, and
complex specific interactions between those objects even if each object is
very simple.

He then proceeded for about 20 minutes to give some examples of specific
complex interactions and genome architectures. The evolution of our
understanding of the LacZ operon (fascination really), different uses of
the same genome in butterflys and caterpillars

In a continuing analogy throughout the talk he compares genomes with
computers emphasizing several times that information theory and computer
systems analysis will be the future of biology in this area.

The genome is an information storage organ:
1) Coding for proteins, RNA - genome must be dynamically accessible at
nearly all times
Further there are many potential programs within a similar genome all
running at the same time often using overlapping portions of programs
controlled only by various types of regulation.
2) Must have the capacity for replication, distribution into progeny
3) Have error/damage ability
4) Must be able to reprogram when necessary
this may happen in even a normal life cycle and he gave the example of the
immune system which has a means of directed purposeful high rates of
mutation on a particular DNA segment with leads to the creation of nearly
infinite combinations of antibodies.
The crux though is that this may be necessary in responses to CRISIS!!

I think this was the root of the seminar. He built up all the examples of
regulation and amazing complexity and then asks: if the cell can regulate
itself in all these manners would in NOT be surprising if cells could
regulate the ability to create new genetic combinations by increasing
mutation. These would include adaptive mutation, genome shock
(McClintock) and interspecific hybridization. More on this central point
at the end.

Continuing he develops the primary mechanism for genome wide rearrangement
(and thus new genetic potential)
Genome System Architecture and Natural Genetic Engineering in Evolution
(Slide title)
1) Genome is hierarchically organized into systems composed of codons for
many different aspects of gene expression, genome maintenance and genetic
change. The codons, in general, can be considered the basic units.
2) Repetitive DNA elements are basic codons for controlling gene expression
and genome maintenance. For example the -10, -35 sequences in bacteria,
TATTA boxes etc are really repetitive units found all over the genome
3) Each species has its own functions '"system architecture" which is
determined largely by repetitive rather than protein coding DNA,
consequently repetitive DNA is more taxonomically specific than single-copy
4) Cells have neutral genetic engineering capabilities necessary for rapid
reorganization, especially of repetitive DNA.

He gave an example of the protein kinase domain which is a series of highly
conserved amino acids found in a large number of genes. This domain and
corresponding DNA sequence can be considered a repetitive sequence and thus
is the building block by which it can be spit and attached to other ends.
At this point he gave many examples of how cells have the ability to cut
and splice DNA (it was a truly impressive list of capabilities!).

Shapiro then contrasts what he called "unit" thinking and "system" thinking
which was quite interesting followed by examples of flagella and HOX genes.

Again emphasized that it is the repetitive elements (also a long plea not
to use the word JUNK DNA which I agree with) that give the structure to the
genome. Repetitive domains are the tag or address by which the programs
are read from the genome.
Much much more and the types of repetitive DNA and their various functions
and distributions.
Roles of repetative DNA
- transcription factor binding sites
- RNA processsing (splicing, polyA attenuation)
- Replication origins
- Centromeric repeats
- Telomeres and subelomeric repeats
- Heterochromatin/satellite RNA effects on transcription
- Amplification of triple repeats
- rDNA spacer repeats needed for sex chromosome pairing

Several catchy examples of the importance of repetitive DNA for evolution:
1) Sibling species such as Drosophila, apparently two sibling species have
been examined and they barely differ from one another at the sequence level
in terms of coding sequence but their repetitive DNA is totally different,
it is as if their genes have simply been scrambled and the program is being
read differently even though it may be the same genes.
2) He reiterated the immune system example as directed mutation to modify
3) The champion in genome rearrangement though was the example of
Oxytricha nova which is a truly bizarre protist. It has two nuclei one of
which is usually inert. I need to look up the paper (Trends in Genetics
1992 vol 8 no 12) to read about this but the gist was that the chromosomes
don't have centromeres instead the genome when undergoing sexual
reproduction (notably under times of severe stress) cut themselves into
100s of thousands of small segments that migrate into the new cell and them
reconstruct themselves once again. A truly amazing feet showing that small
pieces of DNA can be cut up at repetitive DNA sites and reorganized and
then put back together again not necessarily in the exact gene order but
with the programs still able to run.
4) Lastly he gave an example of recent evidence that shows that
transposable elements do not place themselves randomly in the genome but
can be targeted toward specific gene families etc..

Essentially Shapiro is proposing that organisms have systems that in
themselves cause the advent of new genetic variation rather than waiting
for a x-ray from the sky above to do the dirty work.

More in a wrap up post later,