Papilio joanae, mtDNA and species differences

[Felix Sperling]

Thanks Patrick Foley, for bringing up the questions, and Jeff Oliver,
for the crisp answer to one aspect of them.
 
To rephrase Jeff's response, sometimes (and it seems true far more
often than not) female hybrids between butterfly species are more
likely to be inviable or sterile than male hybrids. This pattern was
observed a long time ago and enshrined as "Haldane's Rule". Since
mtDNA, as far as we know for butterflies, is only inherited from the
mother, and since we know that female hybrids are often
disproportionately eliminated, mtDNA should be less likely to move
between species than any other marker. That seems to be true in the
majority of cases that I am aware of, but there are also clear
exceptions.
 
As for the rate of evolution of mtDNA, that statement can only be
made with reference to particular nuclear genes. In general, mtDNA
does seem to evolve much faster than nuclear protein coding genes. A
published example is a paper by Reed and Sperling (1999) in Molecular
Biology and Evolution. On the other hand, based on the pattern in
other insects, I'll bet that mtDNA is much slower than microsatellite
markers or even some of the spacer regions between nuclear genes.
In general, I would say that mtDNA is about as useful a marker as
they come for genetic investigations at the species boundary. But it
can't be expected to give the whole story, in part because it has a
different inheritance pattern (maternal) from the rest of the genome.
 
Speaking of genes that might be good markers of species boundaries, a
particularly interesting class of genes may be the X-linked genes
(=genes on the X, or sex, chromosomes, also known as W chromosomes in
lepidoptera). Based on a literature survey of well-studied cases in
Lepidoptera, I (1994, Canadian Entomologist) and D. Pashley Prowell
(in a book chapter published 1998 by Oxford U. Press) have tried to
draw attention to the fact that about half of all genes coding for
differences between closely related species seem to be controlled by
the X chromosomes. This is particularly amazing when you consider
that there are about 30 pairs of chromosomes of approximately the
same size in leps, and so the X chromosomes should have only about
3-4% of genes if they are sprinkled randomly around the chromosomes.
There is plenty of room to speculate on why that might be the case
(also found to be true for Drosophila, though there are only 4 pairs
of chromosomes there). However, a more basic question is whether that
is really true, since there might be a bias in the likelihood that
someone will publish a case of X linkage.
 
I imagine that most people have hit the delete button by now, but in
case any of you have not, this leads me to my point in bringing up
this esoteric-sounding point on Leps-l. It turns out that it is
relatively straightforward to establish whether there is significant
X-linked inheritance of a trait. But it requires patient and
solicitous rearing of butterflies, something that I just don't have
the time to do these days, but which a number of Leps-l subscribers
might be in a position to do. If we can get enough data from a
variety of different closely-related species pairs, then it might
give us some really interesting insights into what is going on with
species and speciation.
 
By "straightforward to establish", I mean that all you need to do is
show that the two types of hybrid females are different from each
other in a particular trait and resemble their respective fathers.
First generation hybrid female butterflies  (females are XY) get
their X chromosome from their father, and their Y chromosome from
their mother. Male hybrids (which are XX) get X's from both their
father and their mother. If you have species "a" crossed with species
"b" then you have two kinds of crosses, depending on if it involves
the female of "a" or the female of "b" (and of course the opposite
male). The two types of hybrid females are thus XaYb and XbYa. Male
hybrids would XaXb (=XbXa) for both crosses, and so should be
identical to each other (perhaps intermediate or perhaps like one
parent if that gene is dominant).
 
And all it takes is doing the two types of crosses and rearing up the
caterpillars of each cross, and you could have some cool information
on where the genes for the differences between those species are
located. If there is significant hybrid inviability, which is true as
a logical corollary of one commonly used definition of what species
are, then you won't have great hybrid survival, though you don't need
many offspring to make it though to say whether F1 hybrid females
resemble their father. Also many species are kept apart by barriers
before mating, and when you put them together artificially they mate
even though they don't in nature.
 
Anybody game to try?
 
 
        Felix
 
ps: On weekends I have some chance of a faster response to discussions.
 
 
 
>Felix and others,
>
>I am agnostic about the status of Papilio joanae. I have two vague points
>about mtDNA, hoping they will be made more precise by someone expert in
>their use in this case.
>
>mtDNA seems to introgress pretty readily, moving from one species to
>another if any hybridization occurs. Introgressive hybridization (Edgar
>Anderson first studied it in plants) seems faster for mtDNA than nuclear
>DNA for most eukaryotes. Is this true for papilionids?
>
>Ignoring introgression, mtDNA evolves at very different rates in
>different organisms. By evolution here I mean mutation and substitution
>of DNA bases. As far as I know, and this is very limited, the best study
>in Drosophila suggests that mtDNA evolves more quickly in mtDNA than
>nuclear DNA, but this is not always true. What is true in swallowtails?
>
>The answers to these questions may help to clear up the amount of
>separate evolution P. joanae has done from its relatives.
>
>Patrick Foley
>patfoley at csus.edu
 
 
 ------------------------------------------------------------
 
   For subscription and related information about LEPS-L visit:
 
   http://www.peabody.yale.edu/other/lepsl