Subject: Criticism of Molecular Evolution SUMMARY: I defend molecular evolution against th
From: L.A. Moran
Subject: Criticism of Molecular Evolution
Organization: UTCS Public Access
From: lamoran@gpu.utcs.utoronto.ca (L.A. Moran)
Message-ID: 1992Feb9.230116.28211@gpu.utcs.utoronto.ca
Newsgroups: talk.origins
SUMMARY: I defend molecular evolution against the criticisms of Schwabe
and Warr as posted by Chuck Maier.
Chuck Maier (cmaier@uiuc.edu) writes in support of "world class" scientist
Dr. Kouznatsov. Readers will recall that Kouznetsov made some silly
statements about molecular evolution. Before addressing Chuck's concerns
I should point out that the molecular data provides excellent and independent
support for our ideas about evolution. However, the data is not perfect, there
are a few cases which do not seem to fit our notions about evolution. (The
emphasis here is on "few"; maybe half a dozen anomalies in several thousand
comparisons.) In many cases the unusual results have led to further insights
about evolution and its mechansims and most of the so-called anomalies can
be easily explained. Readers of talk.origins need to keep in mind that we
are dealing with real science here, and science that is technologically
complex (ie. cloning and sequencing genes). Sometimes the data is less than
perfect. Chuck says,
"I would like to enter some corroborating evidence to what
Dr. Kouznetsov is saying about the selective presentation of
molecular data. From the evolutionary literature a paper by:
Schwabe, Christian, and Warr G., Perspectives in Biology and
Medicine, 27, 3 spring 1984 pp465... "A Polyphyletic View:
The genetic Potential Hypothesis"
Schwabe discusses protein evolution theory:
"Proteins are thought to evolve mainly through gene duplications
and RANDOM PROCESSES of mutation ... . The Neo-Darwininan argument
is that by [PURE] CHANCE such mutations ( that by definition give
rise to GREATER COMPLEXITY of an organism) [quotes in original] will
be adaptively advantageous."
I couldn't find this paper but I see no problems so far. The only quibble
is that Schwabe and Warr should have pointed out that not all adaptive
mutations give rise to greater complexity. Also there are mutations that
are neutral, but Chuck continues,
"and on the reality of neutral mutations he says:
"However, we need a very much better measure of the rate of neutral
mutations (if such truly occur) [quotes in original] in structural
proteins before we can ... make a reasonable test. ...[T]his
neutralist theory makes many ASSUMPTIONS that are in principle hard
to justify."
This quotation suggests that Schwabe and Warr haven't done their homework.
There is abundant evidence for neutral mutations in structural proteins.
I have no idea what kind of assumptions Schwabe and Warr are referring to.
As examples of neutral mutations I draw your attention to human polymorphisms.
In the case of hemoglobin genes there are more than 200 variants in the
population and most of these have no known effect. In fact, based on our
knowledge of hemoglobin structure, we do not expect these variants to be
anything but neutral. Those who would argue against neutralism have to
explain how all 200 of these different genes can confer selective advantage
on the individuals that carry them.
The human glucose-6-phosphate dehydrogenase (G6PD) gene provides us with
an even more extreme example. There are more than 300 variants in the human
population. (This is the most polymorphic human gene that I know of.)
All of the variant proteins seem to be functionally equivalent (with a few
special exceptions). They represent neutral mutations by any definition that
matters.
Chuck continues,
"and on all the ad hoc explanations evolutionists use to explain
anomalous molecular sequence data: he lists general stories
evolutionists use:
" 3. Genes can mutate or remain stable, migrate laterally from
species to species, spread through a population by MECHANISMS
WHOSE OPERATION IS NOT FULLY UNDERSTOOD, evolve cooordinately,
splice, stay silent and exist as pseudogenes .
4. Ad hoc arguments CAN BE INVENTED (such as insect vectors
or viruses) that transport a gene into places where no
monophyletic logic could otherwise explain its presence.
This liberal spread of rules,..., does not just sound facetious,
but ... robs monophyletic molecular evolution of its vulnerability
to disproof, and thereby of its ... status of a scientific theory."
It's hard to understand exactly what Schwabe and Warr are getting at here.
There are clear examples of genes that are spliced and pseudogenes are
quite common in organisms with complex genomes. We also know about genes
that are transferred laterally between species BY MECHANISMS THAT ARE
WELL UNDERSTOOD. I'm not aware of any arguments that postulate transferal
of a gene by some mysterious mechanism nor am I aware of any data that
would require such an explanation. Thus I must conclude that Schwabe
and Warr don't know what they are talking about when they make the statement
in the last paragraph. (I considered the possibility that Chuck, or other
creationists are misquoting Schwabe and Warr but these statements are
consistent with what Schwabe writes elsewhere.)
I cannot agree with Chuck's characterization of evolutionist explanations
as "ad hoc". Chuck, are you aware of some of these explanations? Do you
really consider that splicing and pseudogenes are examples of "ad hoc"
explanantions? If so, then I suggest that you look over an introductory
biochemistry or molecular biology book. We will also encounter multigene
families in this series of postings. They are not something that evolutionists
make up.
Chuck continues,
"Schwabe presents evidence in this paper of molecules not fitting the
monophyletic pattern as he says:
"We will show below that branching patterns are not correct as far
as relaxin sequences are concerned [his work]. There are other
inconsistencies involving the insulins of the hystricomorphs
[cited work], as well as goose and hen lysozymes [cited work]
that are not readily compatible with a monophyletic scheme of
vertebrate evolution. It is *NEARLY IMPOSSIBLE TO JUDGE* how well
other sequenced proteins fit into the scheme because too many
liberties with hypothetical deletions, gene duplication, and other
genetic conveniences have been taken in the determination of
sequence relatedness."
I will address each of the specific examples (relaxins, lysozymes, and
insulins) in other postings. Schwabe and Warr's criticisms of the rest of
the voluminous data on molecular evolution seems piqued. There are lots and
lots of examples of phylogenetic trees where the alignments are straight-
forward and gene duplications or other "genetic conveniences" (whatever those
are) are not a problem. True, there are some where the data is not as good as
we would like, but that does not justify maligning ALL of the work.
It turns out that Christian Schwabe has uncovered a genuine problem in his
own work on relaxin sequences. He seems to believe that this invalidates all
other work. (Schwabe is a member of the Dept. of Biochemistry and Cell
Biology, Medical University of South Carolina, Charleston SC: I don't know
what his religion is or whether it influences his science.)
SUMMARY: The data on rates of evolution of relaxins does not conform to
our ideas about molecular evolution.
Relaxin is a polypeptide hormone which plays a role in birth and pregnancy
in mammals. The hormone is produced in the corpus luteum during pregnancy
and it causes a softening of the connective tissue of the cervix and pubic
ligaments. This leads to widening of the birth canal.
The structure of the hormone is similar to that of insulin and to insulin-
like growth factors. These three proteins share about 30% sequence similarity
which suggests that they are derived from a single ancestral gene that
duplicated twice in the lines leading to mammals. All of these proteins are
synthesized initially as single polypeptides of about 150 amino acids but
they are processed by cleavage to give an A chain and a B chain that are
joined by covalent bonds. For example,
B chain C chain A chain
preproinsulin ---------------------------------------------------
^ ^ ^
| | |
cut at these sites
all members of ------------ A chain this is the active
the insulin family | | form of the hormone
have this structure ------------- B chain which is secreted
Since the C chain is discarded following processing it does not contribute
to the activity of the hormones (insulin, relaxin, insulin-like growth
factors). The amino acid sequence of this part of the protein is not
conserved (very much) between species. Even the size (length) of this region
can vary considerably. Variability in the C region is often quoted as an
excellent example of neutral mutations.
Chuck Maier is quoting from a paper by Schwabe and Warr in order to
demonstrate that scientists have misinterpreted the data on the evolution
of genes and proteins,
"I won't attempt to put in the table, but he (Schwabe) makes these
remarks on his relaxin data:
"Thus, since pig and rat relaxins are as different from each other
(55%) as is shark relaxin from either pig or rat relaxin, all must
therefore have a common point of divergence, either at the time of
the mammalian radiation (70,000,000 years ago) or at the time of
shark divergence from the branch that would later give rise to
mammalia (about 700,000,000 years ago). It is clear that both
interpretations are unacceptable to the paleontologists."
He notes other problems as well with the relaxin data in other
organisms."
Relaxin is usually considered to be one of the LEAST conserved proteins
known. In most cases mammalian proteins are at least 80% similar and in
the case of highly conserved proteins they can be identical. This means
that rat and pig proteins, for example, differ by 0-20% in most cases.
(Recall that most mammals shared a common ancestor less than one hundred
million years ago.)
As Schwabe points out there are several examples of relaxin sequences that
differ by more than 50% even though the species from which these relaxins
are derived are closely related. The most reasonable explanation is that
there is little constraint (by natural selection) on most of the sequence.
Such an explanation could account for the apparent anomaly that is quoted
above. Note that we are dealing with an unusually small protein (about 50
amino acids) so that single amino acid differences translate to 2%.
However, more recent data from Schwabe's lab (1) is more difficult to explain.
He has shown that the relaxin sequences from two closely related species of
whale difffer by 6% - this is further evidence that relaxin genes are not
highly conserved (one would expect the sequences of other proteins to be
identical). But, the sequence of one of the whale relaxins is 96% identical
to the pig relaxin! Such similarity was not expected. (Schwabe's group
deliberately set out to find relaxins that were related to the pig hormone.
They do not state how many species they examined before finding ones that
showed similarity.)
The opening sentence of the abstract of the Schwabe et al. paper says it
all,
"The tendency toward extremely high variablity among relaxins from
purportedly closely related species has come to an abrupt end with
the discovery of quasi-porcine relaxin in the minke whale and the
Bryde's whale."
Note the subtile shift in explanation. Whereas before Schwabe was concerned
about the large differences he now seems to accept that high variability
is a good explanation. The problem now is that whales are TOO closely
related to pigs! [The other relaxins that have been sequenced are human and
rat. The fossil, morphological, and sequence data confirms that whales
SHOULD be more closely related to pigs than to rats or primates so that
is not part of the anomaly. The only anomaly is why the pig and whale
sequences are not more different. It is not clear to me that Schwabe et al.
understand this point.]
In Schwabe et al. (1) paper he reveals his confusion about mammalian evolution
in general and he closes by saying,
"The model of molecular evolution can not accomodate such
exceptions and one must consider that recent proposals
concerning the multiplicity of life's origins and the mechanism
of polymerization of repetitive primordial DNA oligomers as
described by Ohno may be a viable alternative and that
similarities and differences in primary structures of proteins
may in fact be a relic of primordial chemistry."
This is silly. By all criteria relaxin genes are relatively recent inventions
and are derived from genes that are ancestral to other insulin like hormones.
They are not relics of primordial chemistry. I don't know why the rate of
evolution of relaxin genes in whales and pigs has slowed relative to the
lines leading to humans and rats but I don't think that this particular
little anomaly deserves much attention.
(1) Schwabe, C. et al. (1989) Cretacean Relaxin.
J. Biol. Chem. 264, 940-943.
Laurence A. Moran (Larry)
Dept. of Biochemistry lamoran@gpu.utcs.utoronto.ca
Medical Sciences Blgd. (416) 978-2704
University of Toronto FAX (416) 978-8548
Toronto M5S 1A8, CANADA
From: L.A. Moran
Subject: Evolution of lysozyme genes
SUMMARY: In spite of what Christian Schwabe says, there are no problems
in explaining the molecular evolution of lysozymes.
Chuck Maier is quoting from a paper by Schwabe and Warr in order to
demonstrate that scientists have misinterpreted the data on the evolution
of genes and proteins,
"and more on the explanations of anomalous sequence data:
"It is instructive to look at additional examples of purportedly
anomalous protein evolution and note that the explanations
permissible under the molecular clock theories cover a range of ad
hoc explanations apparently limited *ONLY BY THE IMAGINATION* .
... the FAILURE OF THE CHACHALACA AND GOOSE LYSOZYMES TO FIT THE
ACCEPTED SCHEME OF EVOLUTION FOR THESE SPECIES IS POSTULATED TO
RESULT FROM AN INCORRECT NONMOLECUALAR PHYLOGENY IN THE CASE OF
THE CHACHALACA, AND FROM THE POSSIBility that an anologous, rather
than homologous, lysozyme has been studied in the case of the goose.
We can see no possiblity of a definitive test of these explanations."
[Lysozyme is an enzyme that digests the cell walls of bacteria.]
I don't know what a CHACHALACA is, can anyone help me out? I haven't come
across it in my brief readings of lysozyme papers. As for the goose lysozyme
the postulated explanation was that birds contain two lysozyme genes (say
A and B) and that the goose lysozyme A is being compared to chicken lysozyme
B. (ie. analogous rather than homologous genes are being compared). In the
jargon of biochemistry the chicken and goose A genes would be "orthologous"
while the chicken B gene and the goose A gene would be "paralogous".
A gene duplication occurred in an extinct ancestor to both chickens and
geese and the A and B genes evolved separately from that point on. Because
the gene duplication predates the divergence of chickens and geese then
the A and B genes will be more different from each other than either the
bird A genes or the bird B genes. This is a small gene family, according
to the explanation.
chicken goose chicken goose
A gene A gene B gene B gene
| | | |
| | | |
| | | |
| | | |
------------ ------------
| |
| |
---------------------------
|
ancestral gene
How could we decide if this is the correct explanation? How about looking
for the other genes in birds in order to test the hypothesis? This is the
way science is supposed to work; you formulate a hypothesis (the genes are
parologous) and you test it. Why do Schwabe and Warr say, "We can see no
possiblity of a definitive test of these explanations."
The "goose-type" lysozyme gene in chickens was cloned and sequenced last year.
This proves that the earlier explanation was correct. The correctness of the
explanation should not have been a surprise to Schwabe and Warr because
even in 1984, when they wrote their paper, it was known that gene families
were common in vertebrates. (Are you beginning to get the impression that
Schwabe and Warr are not experts in this field? Do you think that that
explains why they are not widely known in the field?)
[The data on the "goose-type" lysozyme gene is not published yet. You can
see it for yourself by retrieving the sequence from GenBank, the DNA
sequence database. To do this send an Email message with no subject header
to "retrieve@genbank.bio.net". The letter should contain only the accession
number of the DNA sequence "X61001" (no quotation marks). In a few minutes
the sequence will appear in your mailbox.]
Chuck Maier continues to quote Schwabe and Warr in an attempt to discredit
molecular evolution,
"Supposedly polyploidy and gene duplication/mutation have given rise
to all functional complexity of life. Is there any evidence for this?
Schwabe says:
"...the major conclusion to which we wish to draw attention is that
these findings strongly suggest that many of the genes purportedly
produced by gene duplication have been present very early in the
development of life.
In fact, we can ask if they were not present so early that we must
question whether any gene has come about by duplication or whether
all have been there **FROM THE BEGINNING***, as a potential for
species development."
What nonsense! Insects and plants have a single globin gene while mammals
have several genes. Not only is it logical to conclude that there has
been a gene duplication in the mammalian lines but there is very good
evidence that such duplications have occurred. Furthermore, there are
many examples of additional recent duplications of globin genes in living
humans - this demonstrates clearly that such events are common.
We have similar strong data for lysozyme and insulin genes. Both of these
gene families show recent duplications during mammalian evolution. In the
case of the lysozyme genes the multiple duplications occur in ruminant
lines where lysozyme is expressed in the stomach and is used to digest
the bacteria that degrade cellulose in the foregut. Presumably multiple
copies of these lysozyme genes are needed in order to make more enzyme. In
other mammals lysozyme is almost exclusively a defensive anti-bacterial
enzyme which is made in macrophages in the blood or in tears (and sometimes
in milk).
The sequence data for lysozyme and related enzymes indicates that it is
derived from a gene in an organism that was ancestral to mammals. Duplications
of this ancestral gene gave rise to two other genes; lactalbumin (lactose
formation) and calcium binding lysozyme (found in milk) as well as
conventional lysozyme. This is how evolution works, existing genes are
modified and new genes are created by duplicating old ones.
Laurence A. Moran (Larry)
Dept. of Biochemistry lamoran@gpu.utcs.utoronto.ca
Medical Sciences Blgd. (416) 978-2704
University of Toronto FAX (416) 978-8548
E-Mail Fredric L. Rice / The Skeptic Tank