An Amazing Protein

We’ve seen that proteins represent the most amazing design material, so let’s pause for a moment and consider just how amazing the particle protein (Ffh) is by listing some of its nifty properties.

It’s multifunctional. Even though the particle protein is about 450 amino acids in length, it carries out no less than seven distinct functions: it binds the ribosome, binds the signal particle proteinsequence, binds GTP in a coordinated fashion, hydrolyzes GTP, binds the RNA, binds the receptor, and hands the signal sequence off to the translocon.  All of these functions are integrated into the overall adaptor function, connecting the ribosome with the membrane.  What’s more, the seven functions are not divided up to be carried out by seven domains.  On the contrary, each of the three domains has multiple functions.

It’s modular. The particle protein from E. coli can be plugged into the mammalian system and replace its counterpart (SRP54) except when it came to the final hand-off:

The 54-kDa subunit of the mammalian signal recognition particle (SRP54) binds to the signal sequences of nascent secretory and transmembrane proteins and facilitates their cotranslational targeting to the membrane translocation apparatus in the endoplasmic reticulum (ER). A 48-kDa Escherichia coli protein that shares extensive sequence similarity with SRP54 was identified in homology searches. Recent genetic experiments by Phillips and Silhavy [Phillips, G. J. & Silhavy, T. J. (1992) Nature (London) 359, 744-746] have shown that depletion of this protein, designated Ffh (fifty-four homolog), leads to a significant secretory defect in vivo. We demonstrate here that Ffh is structurally and functionally related to SRP54 by virtue of its ability to mimic closely its mammalian counterpart in several established biochemical assays, thereby suggesting that it plays a direct role in protein export. Ffh assembled efficiently with mammalian SRP components into a chimeric ribonucleoprotein [“SRP(Ffh)”] and bound at the site normally occupied by SRP54. Like SRP54, the Ffh moiety of the chimeric particle specifically recognized the signal sequence of preprolactin in a photocrosslinking assay. Moreover, Ffh could also act in concert with other SRP components to arrest elongation of preprolactin upon recognition of the signal sequence. In all of these assays, Ffh had approximately the same specific activity as SRP54. In contrast, SRP(Ffh) did not promote the translocation of preprolactin across the membrane of microsomal vesicles, suggesting that Ffh cannot mediate an interaction with a membrane component that is required for the translocation of nascent chains.  – Here

It’s highly conserved.  I gathered amino acid sequences from several of the most distantly related bacteria and archaea (Bacillus subtilis, Escherichia coli K12, Thermotoga maritima, Nostoc punctiforme, Aquifex aeolicus, Treponema pallidum, Chlorobium tepidum, Deinococcus radiodurans, Halobacterium sp., Sulfolobus solfataricus, Pyrobaculum aerophilum, Methanosarcina acetivorans, and Archaeoglobus fulgidus.).  I then used a program called CLUSTALW to align the various amino acid sequences.  It turns out that 51 positions have the same amino acid, 41 positions have one of two amino acids, 32 positions have one of three amino acids, 18 positions have one of four amino acids, and 6 positions have one of five amino acids.  This sequence conservation is not really clustered in any spot, but is instead spread through the protein.

ffh

The above is a color-coded representation of the amino acid sequence conservation in Ffh.  Every block represents 10 amino acids, starting with the first 10 amino acids (the yellow box in the upper left) and then proceeding left- to-right,  one block (10 amino acids) at a time until amino acids 450-460 (the box in the lower right corner).  For example, amino acids 31-40 had 5 conserved positions, so the block is scored light blue (according to the legend).

It’s not easily co-opted. As far as I been able to tell, neither the Ffh (the particle protein) nor FtsY (the receptor) play any other role apart from connecting the ribosome to the translocon.  In fact, it’s possible that this function is quite isolated from all other cellular functions.  In 2003, one study looked at the Ffh, FtsY, and SecY genes in dozens of different organisms whose genomes had been fully sequenced [38] and the vast majority contained no homologs for these genes.  For example, the bacterium E. coli contains one Ffh gene and computer searches of the genome failed to turn up any other gene that could be construed as a homolog.  The same goes for FtsY and SecY.  So why is this significant?  One would think that the Ffh would have been duplicated many times over during this organism’s evolutionary history and some of these duplicates might have acquired a mutation that would have imparted an advantage to E. coli.  In other words, we would expect these genomes to contain several homologs of such ancient genes, leading the researchers themselves to describe the missing homologs as a “mystery.”  Let’s emphasize this mystery by playing the numbers.

Another bacteria closely related to E. coli are Salmonella, famous for their ability to cause food poisoning.  It is believed that these two bacteria diverged from each other about 100 million years ago [39].  Thus, by looking at E. coli and Salmonella, we are looking at 200 million years worth of evolution.  How many times might the Ffh gene have been duplicated during this time? E. coli are able to divide about every 20 minutes, so if we imagine 3 rounds of division per hour for 200 million years, that’s roughly 5.3 x 10^12 cell divisions.  According to one study of Salmonella, the frequency of cells carrying a duplication of a particular site ranges from approximately 10-4 to 3 X 10-2 [40].  If we opt for the lowest frequency, that means there would have been over 500 million times both E. coli and Salmonella had duplicated their Ffh genes.  Yet apparently, every time this happened, the duplicate was lost rather than acquiring some new function.  And this is just considering two lineages that split about 100 million years ago.  Imagine the shear number of duplications that must have occurred among the millions of bacterial species over billions of years.

It’s universal. All cells, whether eukaryotic, bacterial, or archaeal, possess the particle protein.  This makes sense if the SRP is a necessary device for connecting the ribosomes to membranes.  This also means that the SRP system was most likely in place, in LUCA (adding to the complexity of these creatures).

ffhbunny

Advertisements

One response to “An Amazing Protein

  1. Hey Mike,

    Your calculation 5.3 x 10^12 takes into account only one individual, not entire population. We don’t know average size of population throughout the history, but even current size would give us some impression about orders of magnitude your calculation is too low.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s