Cold Duck May 19, 2000

Well, the Internet let me down. I was going to call this week's column 'Odds and Ends.' And I thought it would be kind of neat to find out exactly where that phrase originated. I figured it would come from clothing, that it referred to an odd sized bolt of cloth and the leftover ends. But I could not find anything on the 'net. I did find out that no one knows what 'the whole nine yards' really means. Or the controversy that surrounds 'mind your P's and Q's.' But no odds and ends.

So, we will just have to work without a title for a while. But I thought I would discuss some research that fits in with what I have talked about recently. A few weeks ago I talked about the pufferfish, Fugu, and its extremely small genome. Most of us know that genome size is not correlated to structural complexity or the estimated number of genes. In fact, organisms in the same genera can vary almost 1000-fold in the size of their genomes. This was termed the C value paradox back in the '50s.

A recent paper examines this paradox by using new technology. It has been observed that Drosophila, which has a small genome, spontaneously loses nonessential DNA at a higher rate than mammals. So, perhaps the size of the genome represents some sort of equilibrium between the increase in genome size by insertion, say by retrotransposons (see last week's column), and the decrease in genome size by deletion.

To answer this, a very specific type of sequence was looked at. Mobile elements, such as many non-LTR retrotransposons, are pretty inefficient in recreating themselves in the genome, often truncating the 5' end. These forms are "dead-on-arrival" (DOA; the extremely scientific term that was actually used in the paper). They are nonessential and are no longer able to retrotranspose. So they should be able to serve as a surrogate for neutral mutations, such as nucleotide substitutions. One can also examine the rate of nucleotide loss or insertions in these regions. In addition, by examining several different species, one can determine the various lineages between all the retrotransposons. Substitutions in an active branch of the lineage will be found in multiple DOA elements across different species (i.e. the active retrotransposon will make multiple copies of itself as time goes on, each having the substitutions diagnostic for that retrotransposon). However, changes in a DOA element itself will be unique, since it cannot copy itself elsewhere.

So, a new non-LTR retrotransposon, Lau1, was identified from a form of cricket, Laupala. Some of the individual species examined are L. kanaele, L. molakaiensis, L. melewiki. Now, I'm not certain but I'll bet some poor slob had to go to Hawaii to collect those crickets. Imagine having to walk all over the tropical islands, looking for crickets. "Nope, already got that one.... Time for a Mai Tai... Sorry, Boss. Didn't find any crickets on Kauai today. I'll try harder tomorrow. Maybe I'll try Molakai."

Anyway, they examined the nucleotide sequences of a large number of Lau1 sequences and identified many DOA lineages. Then they looked at the number of nucleotide substitutions, insertions and deletions. The number of substitutions (which serves as a surrogate for time passed) could be graphed against the number of deletions to get a measure of the rate of nucleotide loss. Laupala deleted DNA at about half the rate of Drosophila.

In addition, Laupala had an insertion rate per nucleotide substitution 40% higher than Drosophila. The average size of an insert in Laupala was two times larger than in Drosophila and the size of a deletion was four times smaller. But this was not all of it. It also appears that the rate of nucleotide substitution in Laupala is almost 4 times lower than Drosophila. So the overall rate of DNA loss in Laupala is closer to 42 times slower than in Drosophila. Not surprisingly, its genome is 11 times larger than Drosophila.

So, besides remaking the genome, non-LTR retrotransposons, and other mobile elements, may have a large effect on genome size, providing even more raw material for gene creation. Now I want to mention a really weird aspect of non-LTR retrotransposons that I ran into while researching last week's column. Nature has an article this week detailing horizontal gene transfer in prokaryotes. This is the non-sexual transfer of genetic material from one species to another.

Retroviruses, such as HIV, may represent one form of horizontal gene transfer in eukaryotes, since it appears that humans may have picked it up from other primates. But, outside of its devastating effect on the person harboring the retrovirus, I have not read anything about its ability to move and remold the genome it rests in. As shown above, mobile elements, such as LINEs, do appear to have the ability to remold the genome. Well, it turns out that some LINEs may have entered the mammalian genome through horizontal gene transfer – from snakes.

Non-LTR sequences tend to be fairly narrow in their distribution. For example, LINE-1 is usually only found in mammals. Several mobile elements have been found in different species of Drosophila that appear to have moved between them through some sort of horizontal transfer. This was shown by examining the high degree of similarity of the mobile elements, even though the Drosophila species housing them last shared a common ancestor 50 million years ago. But some sort of sexual transmission could not be ruled out.

Well, listen to this. An article in Gene published last year (Gene [1999] 238:171) examined Bov-B LINE sequences, which were originally described as an order-specific retrotransposable element. This non-LTR retrotransposon has only been found in true ruminants, such as cattle, goats, sheep, etc. They are not found in the ruminant's closest cousins, the camels, indicating that they originated in the Ruminata about 40-50 million years ago, when the split between camels and ruminants occurred. They are distributed throughout the chromosomes. And NO other mammal appears to harbor these sequences.

In fact, the Bov-B LINE sequences have not been found in any other warm blooded animal. But they have been detected in several cold-blooded ones, particularly snakes and lizards. The degree of similarity between the elements found in snakes and in ruminants is about 75%, much too high for descent by a common ancestor (The last common between ruminates and reptiles occurred several hundred million years ago). In fact, if a phylogenetic tree is constructed using the Bov-B LINE sequences, the Ruminants would form a sibling species with vipers. They would be closer genetically to vipers than pythons are!!

Now, obviously, cows are not snakes, they can not breed with snakes, so how did they pick up a sequence that comes from reptiles? How is there any kind of gene flow between the cold-blooded species and the warm blooded ones? Well, we understand horizontal gene transfer between bacteria. I mean, they pick up DNA and plasmids from each other all the time. But what we appear to have here is the transfer of a segment of DNA from one vertebrate to another. But it is a very special sort of DNA. It can replicate itself. One can think of it as some sort of primitive retrovirus that does not have a means to package itself and survive outside a cell. It just replicates itself and inserts itself back into the host DNA. And it contains everything needed to create multiple copies of itself.

So if there was some mechanism for transferring a non-LTR mobile element from a snake to a primitive ruminant, the element would be able to carry on with its own replication, just in a new host. There are several blood-sucking parasites that live on both snakes and ruminates. And we know that such vectors are able to transmit all sorts of things between species (One of the reasons I am glad we do not have a tremendous mosquito problem here). It is possible that some insect picked up the mobile element from a snake and transferred it to a ruminant ancestor tens of millions of years ago.

Now, I would think that this sort of horizontal transfer of genes would not occur very frequently. Too much such shuffling would produce a tremendous amount of non-Mendelian inheritance of characteristics, which we do not see. But, the introduction of some self-replicating forms of mobile elements could progress horizontally, since one would not expect them to present a phenotype to be selected for or against. Or do they? I have already discussed the apparent ability of these elements to alter the size and complexity of the genome. Next week I'll look some more at these fascinating elements, particularly their effect on active, expressed proteins. It seems that the introduction of these mobile elements could have a substantial effect on both non-coding, non-essential DNA and on genes that are actively being expressed.

Okay, the title. My father loves sweet Rhine wines. Very sweet. It was not until I was in college that I ever tasted a dry white wine. I thought all white wines had a cloying taste that took lots of milk to erase. Anyway, one red wine he would drink, every so often between the Blue Nun, was some form of Cold Duck. It is a sort of sparkling red wine. My father always loved to tell me the origin of the name Cold Duck. It seems that there would often be some residual wine left over in casks, just sloshing around. Not enough to bottle individually. The German vintners, being good businessmen, would not just toss this away. No, they mixed them all together so they could bottle the dregs. (He never told me but I imagine producing sparkling versions helped overcome any problems with mixing different varietals.) Anyway, these dregs from the butts of casks were called Kalte Ende (you can figure that out). Of course, this phrase would not be fit for polite company, too many suggestive overtones. Not good marketing. The similar German word, Ente, was substituted, so that delicate patrons could ask for a bottle of Kalte Ente. A direct translation is Cold Duck. A fitting description for the end of my column!