Talmudic Question #48
Can you think of an example of an intact protein molecule being taken up by a free-living bacterium?
We jumped the gun and departed from the usual format by sending this question in advance to a few people. This netted us some intriguing thoughts from Jon Beckwith, Winfried Boos, Ian, Booth, and Andrew Wright, plus notes from a few others who told us they couldn't come up with much. Now it's your turn. Please don't be deterred from adding your ideas and comments.
Here's what we have so far:
- Bacteriocins. Many are small peptides, but some are as big as 20 kDa, so at least these qualify as respectable proteins. And they certainly penetrate, or else they wouldn't work on such targets as DNA, rRNA, and tRNA. The uptake mechanism is known in some cases to involve porins.
- Phage proteins. DNA phages inject so-called "pilot proteins" needed for the translocation of their DNA. In the classic Hershey-Chase experiment, a small proportion of the 35S‑labeled proteins of the original phage was associated with the bacteria after infection. This didn't bother Hershey and Chase but later turned out to be of interest. However, does "injection" by a phage count?
- Periplasmic proteins such as Maltose Binding Protein can penetrate into the periplasm when added in solution to E.coli cells permeabilized with calcium. But they are not known to enter the cytoplasm, so this doesn't count, does it?
- Other. Andrew Wright wrote:"Another "artificial system" is REMI (restriction enzyme mediated insertion) where restriction enzymes are introduced into cells (yeast and Dictystelium) by electroporation. I don't know if this works with bacteria but my guess is that it should. Of course electroporation cuvettes are not floating around in the environment but other conditions in nature either now or in the distant past might have mimicked this. I imagine many types of proteins could be introduced in the same way – if attached to DNA or some other macromolecule."
Why would we want to know? This qualifies as another Talmudic Question itself. But not to prolong matters, here are some of our answers:
- Organelles, such as mitochondria and chloroplasts, take up proteins that are encoded by the nuclear genome and synthesized on ribosomes in the cytoplasm. When and how did they acquire this skill? Should we address the question to their ancestral relatives, the rickettsiae and the cyanobacteria? And how about bacterial endosymbionts?
- Could proteins involved in DNA repair/mutagenesis when taken up from the environment play a role in bacterial evolution?
- Could soluble proteins be taken up to act as signaling agents between species?
Myxococcus xanthus appears to assimilate some proteins from its neighbors/environment as evidenced by the rescue of some mutants by wildtype. Perhaps the best documented example of this is the outstanding paper by Nudleman, Wall, and Kaiser in Science. 2005 Jul 1;309(5731):125–7."Cell-to-cell transfer of bacterial outer membrane lipoproteins." The abstract for this paper is below:
Myxococcus xanthus cells can glide forward by retracting type IV pili. Tgl, an outer membrane lipoprotein, is necessary to assemble pili. Tgl mutants can be transiently "stimulated" if brought into end-to-end contact with tgl+ donor cells. By separating the stimulated recipient cells from donor cells, we found that Tgl protein was transferred from the donors to the rescued recipient cells. Mutants lacking CglB lipoprotein, which is part of a second gliding engine, could also be stimulated, and CglB protein was transferred from donor to recipient cells. The high transfer efficiency of Tgl and CglB proteins suggests that donor and recipient cells briefly fuse their outer membranes.
COMMENT ON COMMENT:
David, this is amazing. We, and surely our readers, are grateful to you for pointing this out. The remaining question is: "how unique is this?"
Elio
Isn't transport into organelles analogous to transport out of bacteria? The protein is moved from the cytoplasm into the intramembrane space and then into the organelle. If so this would count either since it isn't the acquisition of whole protein import but the equivalent of protein export it just happens to be going into an organelle.
COMMENT ON COMMENT:
It depends whose cytoplasm you consider. Getting a protein into an organelle is to introduce it into its "cytoplasm." So, it's still going in the inward direction.
But this reminds me that we should also wonder why bacteria, being good at secreting proteins, have forgotten how to do it the other way, if they ever knew.
What does Bob B. have to say about this?
Elio
Elio, Milton Saier once wrote a paper discussing the possibility of Bdellovibrio "forcing" the operation of secretion machinery backwards during predation. I don't know if he has followed up on those ideas:
Saier MH Jr. (1994). "Protein uptake into E. coli during Bdellovibrio infection. A process of reverse secretion?" FEBS Lett. 337: 14–7
Abstract: Bdellovibrio bacteriovorus is a small bacterial parasite that infects other Gram-negative bacteria, resides in the periplasm of the host cell, and utilizes host macromolecules as a source of nutrients. Evidence is summarized suggesting that B. bacteriovorus secretes proteases and nucleases synthesized in its own cytoplasm that are targeted to the cytoplasm of the host cell. Possible mechanisms for this trans-trimembrane protein transport process are discussed.
Dear Elio,
In some experiments performed by Bryant and King (1984), it was shown that P22 can inject a functional protein into the cytoplasm of the host. In the experiments, phage particles lacking the injection protein could be rescued by coinfection with replication defective phage that otherwise contained a normal complement of injection protein. Other examples in phage abound. Many phage contain protein covalently attached to their DNA termini and are essential for replication; phi29 being one such example. The phage N4 injects a DNA dependent RNA polymerase during its infection. This is all the more amazing because the size of the protein in the virion particle is bigger than the tail tube down which it must travel to get into the cytoplasm.
Of course these are all examples of proteins being "put there" by viruses, so maybe they don't count as protein molecules being "taken up" by free living cells: the cells are passive recipients and perhaps not active consumers, as it were. Surely it is only a matter of time before examples of the latter are characterized...
‑hermes
Talmudic question 48 is quite interesting and has me on quite the reading kick regarding endosymbionts, mitochondria, plastids, and other organelles. Perhaps 'willing uptake' (to separate from either laboratory introduction or toxic events such as bacteriocins,
phage proteins, and bdellovibrio predation) of free proteins by free bacteria is very rare and hasn't evolved. One would expect development of such uptake systems primarily in situations where bacteria are interacting with other organism in either a symbiotic interactions (endosymbiosis, consortia, and biofilms) or in predation/pathogenesis. Otherwise, diffusion and the relatively low number of secreted proteins would make it unlikely the bacteria would encounter alien proteins to uptake.
For the evolution of an uptake system the bacteria would need a constant and sufficient source of specific proteins. Also the proteins taken in would likely need to give some
functional advantage (missing metabolism) beyond mere carbon, nitrogen, and sulfur sources. Many bacteria leave behind huge amounts of biomass when they leave a biofilm.
But such makes me wonder what happens in long culture conditions where biofilm production is a necessity. In these cases the bacteria go through cycles of growth, senescence, and lysis. Perhaps the latter growth phases requires uptake of proteins in addition to peptides.
COMMENT ON COMMENTS:
I found your notions particularly astute. Thinking of the ecological imperatives, especially biofilms, seems like the right way to focus on such questions.
Thanks,
Elio
Oh I forgot...
What about pilus retraction?
Just found this interesting article. Apparently yeast mitochondrial targeting sequences can be imported into Paracoccus denitrificans and required a membrane potential.
Import of a mitochondrial presequence into P denitrificans: Insight into the evolution of protein transport
David Roise and Merritt Maduke in FEBS Letters 337 (1994) 9–13
Recently, pointed out to me by a labmate:
Endocytosis of Gemmata obscuriglobus, a bacteria in the phylum Planctomycetes
http://www.pnas.org/content/107/29/12883
Good point. A story that will unfold further, one would think.
Elio