Ontogeny & Phylogeny V

Xuhua Xia xia at cc.umanitoba.ca
Sat May 21 17:11:12 EST 1994


==================================================================
Human ontogeny is simply a matter of enzymatic transformation of
groceries.

--------- Dobzhansky (I forgot the original wording, and might have
          cited wrong.)
===================================================================
                      Ontogeny & Phylogeny:
                    We need molecular biology

Let me first share with you some of my experience accumulated while
attending a neurobiology symposium called Altschul Symposium held
in University of Saskaschewan at Saskatoon. Most basic studies
presented are descriptive, i.e., documenting what gene is expressed
where and when, and what would happen when a particular gene fails
to express where and when it normally should. 

I suspect that most of modern-day biologists would be offended when
their research is labelled as descriptive. If this is indeed the
case, let me hurriedly add that a good description of a biological
pattern almost always leads to new insight. For example, one French
biologist showed in the symposium that a homeotic mutation in the
mouse led to several extra skeletal elements which had not
mammalian equivelents, but had counterparts in reptiles. No one
would doubt the evolutionary significance of this description.

In this and the following 3 or 4 postings I will talk about a few
powerful techniques in molecular biology that can be used to
describe ontogeny in relation to the study of ontogeny and
phylogeny. In consideration of those readers with a computer or
engineering background, I will limit one method per posting, and
only the conceptual aspect of the method and its utility will be
illustrated. This posting will focus on subtraction hybridization.

Let us digress a bit to talk about a cDNA library. A cDNA library
represents a set of mRNA species present in a particular body part
(or a particular cell) at a particular developmental stage. Such a
library can be written as:

     L = {gene1, gene2, gene3, ......, geneN}

Suppose we now make a cDNA library from a gorilla embryo at one
embryonic stage, and another cDNA library from a chipanzee embryo
at the same stage. Now we have

     L(gorilla) = {Ggene1, Ggene2, ......, GgeneN1}; and
     L(chimp)   = {Cgene1, Cgene2, ......, CgeneN2}.

Subtraction hybridization is a technique that tells us what gene is
present in L(gorillar) but absent in L(chimp), and vice versa. In
addition, available data also allow us to have an approximate
answer to the intersection of the two sets. The following data
matrix is then generated:
=================================================================
Genes         Species at equivalent embryonic stage
        ---------------------------------------------------------
          gorilla   chimp     human     .    .    .    .

gene1     1         1         1
gene2     1         1         0
gene3     1         0         1
gene4     0         1         1
.         .         .         .
.         .         .         .
.         .         .         .
geneN     1         1         0
=================================================================

Such a data set will tell us the difference between gorilla and
chimp, between chimp and human and between human and gorilla. It is
most likely that there will be no difference among them during
early ontogeny, but then, at embryonic stage ES5, say, a gene or a
gene group sets gorilla aside from chimp and human. Then, at ES9,
say, another gene or gene group arises in the chimp and sets chimp
and human apart. Isn't this nice!

Once we get a series of cDNA libraries fron gorilla, chimp and
human at different ontogenetic stages, further insights can arise.
For example, if the chimp library at embryonic week 5 is most
similar to the human library at embryonic week 7, then we can infer
that the hypothetical "neoteny gene" responsible for human
evolution must have already come into operation before embryonic
week 7 in human ontogeny. The table below suggest that this gene
might be expressed at embryonic stage 3. If we knock out genes
expressed at embryonic stage 3 one by one, then we may find one
particular gene, when knocked out, cause accelerated development of
the human embryo...... We can then knock out such genes in cows to
reduce their lengthy misery of pregnancy. (I think I can write a
good science fiction based on this.)

=================================================================
Embryonic                Embryonic weeks
stage     -------------------------------------------------------
          gorilla   chimp     human     .    .    .    .

1         1         1         1
2         2         2         2
3         3         3         3
4         4         4         5
5         5         5         7
6         6         6         10
.         .         .         .
.         .         .         .
.         .         .         .
N         .         .         .
=================================================================

(to be continued)
-- 
====================================================================
Xuhua Xia
University of Manitoba
xia at ccu.umanitoba.ca



More information about the Mol-evol mailing list