Pharyngula

Evo-devo is not the whole of biology

Sometimes a plan just comes together beautifully. I’m flying off to London tomorrow, and on the day I get back to Morris, I’m supposed to lead a class discussion on the final chapters of this book we’ve been reading, Endless Forms Most Beautiful. I will at that point have a skull full of jet-lagged, exhausted mush, and I just know it’s going to be a painful struggle. Now into my lap falls a wonderful gift.

There was a review in the NY Review of Books that said wonderful things about Carroll’s work, and in particular about the revolutionary nature of evo-devo. This prompted Jason Hodin, an evo-devo researcher himself (whose work I’ve mentioned before) to write a rebuttal and send it off to NYRB…which they chose not to publish. So he sent it to me, with permission to post it.

(If Pharyngula is going to be second choice to the NY Review of Books, I’m not going to complain.)

Anyway, I’m almost as guilty as Carroll of hawking the wares of the evo-devo bandwagon and traveling roadshow, so this is a welcome balancing corrective. The complete text is below the fold.

REVOLVING EVOLUTION

To the editors:

Ziff and Rosenfeld ("Evolving Evolution" May 11,
2006) set the bar pretty high for themselves. In the first line of their NYR
article they claim that the past 20 years have not only yielded "major
changes in our understanding of evolution," but indeed that these
purported changes have gone "virtually unnoticed." By the second
paragraph, we learn that the new field of study responsible for these major new
insights is Evo Devo, which is shorthand for the melding of the fields of
evolutionary and developmental biology.

The cover picture of the May 11 NYR issue is of a Drosophila
melanogaster
"fruit" fly embryo.
Since I spent nearly half of the past 20 years studying Evo Devo in this very
same critter, I wondered to myself: did I miss something? Was there some earth
shattering evolutionary insight that we have had concerning this noble insect
that managed to evade my notice? Was I too deep in the dense trees of my Ph.D.
research to see the forest of ground breaking discovery that surrounded me the
whole time?

Ziff and Rosenfeld (henceforth "Z&R") sprint
through the history of 20th century molecular genetics and reach the
following conclusion: all evidence pointed to the fact that organisms are
different due to each "having evolved its own unique set of genes over
millions of years." At first glance, this synopsis seems reasonable. There
must be a genetic basis for the differences between organisms. But it seems
that special emphasis is to be placed on the words "set of genes."
According to this supposedly dominant world-view, organisms are different
because they have different "sets of genes."

But wait a minute. Wasn’t it Gregor Mendel, the founder of
the field of genetics, who showed us that genes existed in different states?
There were round and wrinkled peas, and the differences between these two types
were not due to different "sets of genes," but instead to different versions (called alleles) of the very same genes! In fact,
the field of modern population genetics is much like the field of early 20th
century population genetics in supposing that differences among populations and
even species is due to unique combinations of alleles.

But perhaps what Z&R meant was that this basic
understanding of genetics was all forgotten until about 20 years ago. Was the
middle of the last century the Dark Age of biology, where we forgot our past,
and insisted that evolution must be the result of new "sets of genes"
appearing somehow in different organisms? And was it the sequencing of diverse
genomes in the past 20 years that has led to a biological Enlightenment?

Unfortunately, the historical record does not uphold this
just-so story of modern biology. For example, the evolutionary biologist Emil
Zuckerkandl used the relatively crude data available to him in 1975 to conclude
"that new structural genes must nearly consistently derive from
preexisting structural genes, and that new functions can be evolved only on the
basis of old proteins." Two years later, François Jacob wrote an influential
article in Science entitled "Evolution and Tinkering" in which he
stated:

"The appearance of new molecular
structures during much of biological evolution must…have rested on alteration
of preexisting ones. This is exemplified by the finding that large segments of
genetic information, that is of DNA, turn out to be homologous [evolutionarily
related], not only in the same organism, but also…among those that are
[evolutionarily] distant."

Z&R’s claims for the important discoveries attributable to
the past 20 years of research go further still: "surprising, too, is the
evidence that all animals, from worms to humans, probably descend from one or a
few primitive bacteria." Darwin would be shocked to hear that additional,
modern evidence supporting his fundamental principle of common descent would be
seen 150 years later as "surprising."

Straw men are around every corner in Z&R’s article. For
example, "the presence of a body plan in the genome, whether of a fly, a
whale, or a human, was unexpected by embryologists." One wonders where
these misinformed embryologists thought the body plan resided: in the cosmos
perhaps? If, instead, the authors are here referring to embryologists in a
pre-genetic age, their statement is true, but rather meaningless.

I don’t mean to denigrate my field of Evo Devo, nor do I
intend to suggest that no critical insights have come from it. Perhaps the most
important contribution of the field is methodological. For years, evolutionary
biologists have gone along their merry way, largely ignoring the rapidly
expanding fields of molecular biology, cell biology and development, with their
increasingly impenetrable jargon. For their part, researchers in the latter,
more "reductionist" fields planned and interpreted their experiments
in the absence of an explicit evolutionary perspective. As Z&R correctly
point out, the division between evolution and development traces back to the
NeoDarwinian synthesis in the early 20th century, where
developmental biologists were not at the table.

But this was an unnatural separation; Darwin himself stated
in Origin that developmental biology was
"one of the most important subjects in the whole round of history."
When Stephen Jay Gould published Ontogeny and Phylogeny in 1977, alongside other less visible but equally
insightful contributions by David Wake and Père Alberch, the two fields were
long overdue for reconciliation. Still, it took a few decades before the two
disciplines started to truly become reunited. That process came on the heels of
the Hox gene discoveries, so touted by Z&R and their modern Evo Devo hero,
Sean Carroll.

The finding of similarities in Hox gene display patterns in
fly and mouse embryos so impressed Carroll and other developmental biologists
that the "Hox code" seemed a possible Rosetta stone of evolution. It
was even claimed in the early 1990′s that this Hox pattern defined what it
meant to be animal! Putting aside the extreme provincialism inherent in this
focus on "animalness" (as Hox genes are absent from complex multicellular
creatures like fungi, algae and plants, without whom animals would not exist),
it turns out to be false that Hox genes are even necessary to produce an
animal. Sea urchin larvae, for example, which have bilateral symmetry, complex
organs, a full gut, a nervous system, muscles, and all other definitively
animal features, are formed in the absence of the canonical Hox gene
expression.

Still, the similarity in Hox gene displays in many animals
is notable. But here, again, Z&R follow Carroll’s lead in vastly
overemphasizing the importance and function of these genes in embryos. For
example, we are told by Z&R that "the production of Hox genes divides
the [fly] embryo into a series of segments," but this is not accurate. The
segmentation genes do this job, while Hox genes act later to confer identity on
the already defined segments. Interestingly, we are shown some of the data
supporting this idea in Figure 2 of Z&R’s article, but are not given the
full story.

Overstated claims for the importance of Hox genes do not
stop there. We are told that "the giraffe’s longer [neck] vertebrae may
have developed because of mutations in the Hox genes controlling the size of
our vertebrae." This intriguing hypothesis, unfortunately, is completely devoid
of any supporting evidence.

A nagging question remains: what have the similarity
in Hox gene displays shown us? Evolution is all about diversity, but the Hox
discovery points to sameness. Caroll’s 2005 book Endless Forms Most
Beautiful,
so lauded by Z&R, can be
described as one great genuflection over this very issue, and the conclusion is
that changes in gene regulation must underlie diversity. Nevertheless, as
Z&R note, this finding was not at all a novel discovery of the last 20
years. It has been the dominant paradigm of molecular genetics since François
Jacob and Jacques Monod’s pioneering work of the 1950′s and 60′s. Or, as Jacob
wrote in 1977: "What makes one vertebrate different from another is a
change in time of expression and in the relative amounts of gene
products." This sounds to me rather indistinguishable from what Z&R
call "Sean Carroll’s view [that] what creates diversity is the pattern in
which genes are turned ‘on’ and ‘off’."

Was this all a misreading of Carroll by Ziff & Rosenfeld
(neither of whom are biologists)? Is Carroll much more circumspect than his
reviewers? Was the nuance and subtlety of Carroll’s Endless Forms simply missed by Z&R? Not exactly. First of all,
as Carroll explains in the "Sources" section at the end of his book,
he "elected not to name every individual associated with every work
pertinent to the story" in the interest of his "broad audience."
In fact, he names almost no sources in the text, and it is reasonable for the
uninformed reader to assume that a lot of the ideas are Carroll originals. Here
for example, is the context in which Carroll describes his move to studying
‘fruit’ flies after completing his immunology Ph.D. in 1983:

The common perception [among
biologists]…was that the rules of development and physiology differed
enormously between mammals and bugs or worms.

This is simply not so. Examining any book on developmental
biology or animal physiology dating, say, from 1950-1985, will yield ample
evidence that biologists have long appreciated the fundamental developmental
and physiological commonalities among all animals. In fact, this understanding
even predated Darwin. Furthermore, insects in general, and Drosophila flies specifically, had been recognized as important
study organisms in both development biology and evolutionary biology for
decades preceding Carroll’s postdoctoral career choice. Despite the
implication, Carroll was walking a very well trodden path.

Carroll concludes Endless Forms thusly (my emphasis): "Virtually everything I
have described in chapters 3-10 has been discovered in the past 20 years…[and
has] forced biologists to rethink completely their picture of how
forms evolve." It is this kind of hyperbole that led Z&R to vastly
overstate the impact that Evo Devo in general, and Carroll in particular, has
had on evolutionary biology. Such hyperbole might increase book sales and keep
the grant money flowing in the short term, but it ultimately makes all of us
practitioners of Evo Devo look pretty silly.

In my opinion, one major contribution of the field of Evo
Devo has been in the synthesis itself, and this is still incomplete. If we are
to come to any subtle understanding of the diversity of life we must consider
all aspects of biology, from paleontology to genetics to development, cell and
molecular biology, field ecology and physiology, ALL in the context of
evolutionary biology. Perhaps the important insight of Evo Devo is merely a
reminder that this has been true all along. Some of us just forgot it for a
while.

Jason Hodin

Hopkins Marine Station

Pacific Grove, CA

To presage my classroom strategy on Wednesday the 18th of October: now discuss. Here, I’ll also mention that the ‘extreme provincialism inherent in this
focus on "animalness"’ is also discussed on Evolgen, and in this comment by Larry Moran. It’s not to suggest that evo-devo isn’t important, but that it isn’t the whole universe, and it did not come blasting out of nowhere.