Help with literature review on experimental evidence for mechanisms of eukaryotic HGT

To the collected brainpower at PS:

One question much discussed here concerns whether the distribution of genes in Earth life is better explained by a nested hierarchy with a single root, or design. (These are not exclusive, of course, but let’s just take Tree of Life versus some ID pattern for the time being.) Design as a hypothesis in biological systematics suffers from being poorly articulated, or not articulated at all. Nonetheless, attempts are being made – e.g., Winston Ewert’s dependency graph proposal, some preliminary thought experiments by Eric Holloway [go to the BioLogos forum, Phylogenetic Signal threads] – to put something up for consideration.

Meanwhile, in research for a book I’m writing on common descent, I’ve been looking at the growing literature on eukaryotic horizontal gene transfer (HGT). Here are two recent papers:

One evidential dimension missing from these papers, but critical to forming Bayesian priors on eukaryotic HGT, is the actual mechanism of transfer.

HELP needed. Please, if you know of any experimental literature showing, in real time, the transfer of gene sequences from donor (either prokaryote or eukaryote) to host (eukaryote), let me know. I’ve asked a post-doc who should know, and he doesn’t know of any examples.

Thanks. Kindly post the citations in this thread.


At first blush, nuclear mitochondrial DNA (numt’s) could serve as a model for understanding HGT in general. For example:

I would think that there are several possible mechanisms, ranging from homologous recombination to transposon/retroviral mechanisms. As with most problems in biology, there probably isn’t a simple and concise answer to the question.

1 Like

Why would you only accept real time, experimental evidence? This seems to require an a priori rejection of all historical evidence.

I don’t think this is rejecting the historical evidence, but asking for information on how it happens, and maybe how often.

At risk of stepping far beyond my expertise, I should think it would be exceedingly difficult to capture HGT in real time.

1 Like

If the question is about mechanisms for “transfer of gene sequences from donor (either prokaryote or eukaryote) to host (eukaryote),” then it’s not exceedingly difficult to observe and measure this in the lab.


Exactement! If you were here, Dan, I’d give you a big hug, only COVID-19 and all that.

If anomalous eukaryotic gene distributions cannot be explained by HGT, that raises the prior on the sort of picture Winston Ewert and Eric Holloway are developing.

In general, retroviruses should qualify as well known examples of transfer of gene sequences into eukaryotes, and as you know, the “actual mechanism of transfer” in those systems is not only well described but has been harnessed for use in the lab and the (pre)clinic.

You may find further examples in a very new review article in Trends in Parasitology, below. I have made a PDF available here.

1 Like

Cites, please. This is exactly what I need.

Do these data show incorporation into the germline, followed by expression?

Review article on homologous recombination in eukaryotes:

May be of some help.

In the case of Ewert, the claimed anomalous gene distributions are much more likely due to insufficient and inconsistent gene annotation. What Ewert needs to focus on is sequence data, not annotation data. Better species coverage would also be a plus.


Which is why I generally stay out of the biology discussions! :flushed:

Oh Garsh! :kissing_closed_eyes: :hugs:

If anomalous eukaryotic gene distributions cannot be explained by HGT, that raises the prior on the sort of picture Winston Ewert and Eric Holloway are developing.

Ewert’s approach is the best idea to come out of ID, ever (a testable hypothesis!). I had a discussion with Eric, and his approach is still very preliminary. I really think he ought to stick with accepted phylogenetic methods and not try to reinvent the wheel.

1 Like

OK, another dumb statistician question, but can anomalous sequences be the result of HGT but not detectable as HGT? My intuition is “yes”.

Are you asking about retroviruses? If so, then we know that their MO is insertion into the genome and expression from LTRs. That’s Wikipedia-level knowledge.

And can this happen in germ cells? Yes. I would say, yes, of course, but I guess there is a universe in which sperm and oocytes would be resistant to retroviral infection.

A couple abstracts below, but know that it should not be necessary to provide references in support of the claim that retroviruses can invade the germline.


3 posts were split to a new topic: The Cheese Stands Alone?

Just a few more thoughts . . . [and echoing some of @sfmatheson’s post]

Given the context of Ewert’s and Holloway’s work, I am assuming we are not talking about retroviral or transposon sequences, but about classic eukaryotic genes.

The first thing we have to consider is how often naked DNA from a different species makes its way into the nucleus of a host cell.

The production of germline cells may play a role as well. How separate are the germline and somatic cells? If memory serves, simpler eukaryotic species have more fluidity between germline and somatic tissue involving recruitment of functioning somatic cells from functioning tissue. This may be why HGT in nematodes is much more common (is this true?) than in more complex eukaryotes. How common is it to find naked DNA from other species in human sperm or egg cells? This may be a much more common event in cells that line the human gut or other mucosa, or in immune cells that actively ingest foreign cells, but I don’t see how it would be as common in tissues that produce egg and sperm.

1 Like

Paul, I have pointed you (and PS) to a body of literature that shows, in real time, transfer of functional genes from the chloroplast to the nucleus. The key author for your literature (and PS) search is Bock.

In addition, plant scientists have been using exactly this sort of system for more than 30 years to genetically modify plants. Of course, I refer to the uses of Agrobacterium tumefaciens (as well as Agrobacterium rhizogenes). The literature on these systems is voluminous, and I am not going to even try to narrow it down here (although if you wait a month, I may arrive a a new crop of current work for my course). Needless to say, there is excellent evidence that these bacteria have “been used” over evolutionary time scales to incorporate new genetic information into host plants. Also needless to say, the mechanisms for this system are very, very well-understood. Your postdoc might want to spend a week on the internet and sort through the many, many excellent reviews and research papers on this system.

(Just incidentally, on the off chance that you are missing this because you may work with Sanford, Agrobacterium is far, far superior to the gene gun for plant transformation.)


Yes. Retroviral and transposon insertion is relevant only if genes / proteins such as those in the Cummings et al. MS (link above) are carried along and integrated in the germline.

We should probably keep in mind that the context of the question is eukaryotes. In plants, one mechanism of HGT is extremely well known and is used in labs and in biotech all the time: transformation and genome integration by the bacteria of the Agrobacterium genus.

The basic notion of gene transfer is not deeply mysterious and mechanisms by which it happens are so well known that they have been commandeered by humans. There are surely mechanisms we don’t know about, and there are elaborate defenses erected against the invaders, which we’re still learning about, but the basic phenomenon is not mysterious nor is it in doubt.

The review article I linked above should be a really good start but my comment here is about the vast knowledge we have of viral and bacterial vectors that move gene sequences into eukaryotes. Not only can these processes be observed and measured, they have been retooled and put into catalogs.

Both you and Dan were asking/commenting about “transfer of gene sequences from donor (either prokaryote or eukaryote) to host (eukaryote)." Between retroviruses and Agrobacterium, you have solid proof of principle along with so much mechanistic detail that it’s now codified into product manuals. I hope you don’t need cites for that.

1 Like