+1: An ancient family of SelB elongation factor-like proteins with a broad but disjunct distribution across archaea

Yay, one more paper is accepted: An ancient family of SelB elongation factor-like proteins with a broad but disjunct distribution across archaea. Gemma C Atkinson, Vasili Hauryliuk and Tanel Tenson, BMC Evolutionary Biology 2011, 11:22.

The nitty-gritty: we found a SelB relative which is not likely to bind and deliver selenocysteinyl-tRNA. It does something else, and we do not know what. tRNA binding domain and ribosome-binding interfaces are intact, but the G domain is messed up, so it might be a translational GTPase with no GTPase activity! For more details check out Gem's blog or... just read the paper, it is open access!

All ribosomes are equal, but some ribosomes are more equal than others

There are many ways of regulating translation - different mRNA structures, modifications of the canonical set of translation factors, specialized factors and so on. Well, you can also have different ribosomes and they may have different functions.

Here is a review about different ribosomal flavors. Main points:

1. rRNA can be modified differently under different conditions, thus resulting in ribosomes with different properties, such as thermal stability, affinity between 30S and 50S, etc. Here it an example of ITC used for studying these rRNA-modified ribosomes.

2. Ribosomes can have different rRNA and proteins when produced under different conditions. The most striking example is Haloarcula marismortui with 3 rRNA operones, out of which one codes extremely divergent copy which is expressed at high temperatures. You mess with it and bug becomes temperature sensitive.

3. Profs of functional differences of these ribosomes - here we do not have much. Stability - yes, see above. But function...

Well, we do have a bunch of proteomics data showing that in Saccharomyces cerevisiae different paralogues of r-proteins localize differently and are specifically involved in translation of some mRNAs.

A quick reminder - Saccharomyces cerevisiae had a whole genome duplication (WGD), thus they generally have loads of paralogues and thus are used to study evolution of proteins after duplication. Apart from yeast, WGD has happend in many other lineages (bony fish, plants), and it would be interesting to see what happens to r-proteins during WGD...

Back to the functionality of different ribosomes. One interesting possible functional regulation is discussed. Knocking out non-essential ribosomal protein Rps25 makes ribosomes incapable of translating some IRESes, though no effect on normal cap-dependent translation. Is expression of Rps25 regulated during viral invasion? No evidence of that as yet.

4. There are two possible ways of using different ribosomes:

First it can be that when the cell changes its ribosomal set, changes one flavor for another, no mixing - a global rewiring of the translational machinery. This seems to be the case for rRNA modification in bacteria - appropriate enzyme is induced  under certain conditions and all the ribosomes are modified, viola. Same for different rRNA genes in archaea.

Alternative approach is to have many different ribosomes for different mRNAs. This is seemingly what we have in yeast (see above). Specific localization of different ribosomes and use of different mRNA-specific factors would then ensure proper coupling of appropriate ribosome with the right mRNA. Different localization of different paralogues of r-proteins in Saccharomyces cerevisiae is shown experimentally, and these proteins have different requirements for assembly into the 80S.

PS: what all the ribosomes have in common is their color. They all are yellow.

References:

1. Gilbert VW. Functional specialization of ribosomes? (2011) Trends. Biochem. Sci. 2011 PIMD 21242088

2. Lopez-Lopez at al. Intragenomic 16S rDNA divergence in Haloarcula marismortui is an adaptation to different temperatures. (2007) J. Mol. Evol. 65, 687–696

3. Esguerra J. et al. Functional importance of individual rRNA 20-O-ribose methylations revealed by high-resolution phenotyping. (2008) RNA 14, 649–656

4. Komilli at al. Functional specificity among ribosomal proteins regulates gene expression. (2007) Cell PIMD 17981122

4. Kellis at al. Proof and evolutionary analysis of ancient genome duplication in the yeast Saccharomyces cerevisiae (2007) Nature v. 131 pp 557-571

Darwin meets Gibbs: making a temperature-resistant protein

In order to perform its function, a protein should be properly folded. Therefore stability of this protein's native state is crucial for its function. Denatured protein can be toxic for the cell and requires specialised machinery to degrade it, thus compromising the cell's fitness. Having a denatured protein is not equal to just not having a functional one, it is equal to not having a functional one and having some costly junk.

Since stability is so crucial for protein function, it must leave its trace in the patterns of amino acid conservation. Bioinformatic studies show that there is a strong correlation between the Surface Accessible Area (SAA) of the residue and its conservation, or, simply speaking, conserved residues are mostly buried inside the protein. That sounds logical – the core should be more important for protein stability than its outer shell. The outer residues, on the other hand, can be rearranged in order to change the protein's binding selectivity and evolve new function. But lets get back to the core.

Basic thermodynamics relationships link protein stability to parameters like Gibbs free energy (ΔG), enthalpy (ΔH), entropy (ΔS), heat capacity (C, or, to be specific heat capacity at constant pressure, C_p) and absolute temperature (T). And adaptation to extreme temperatures gives us a striking example of thermodynamics shaping protein evolution. But first let us start with some basic theory - I know that sounds painful, but please stay with me for a moment!

Gibbs free energy is divided into enthaplic and entropic components (ΔG = ΔH - TΔS). By the definition of Gibbs energy, in order for the protein to be stable, ΔG of folding should be negative, and when it is positive, the protein unfolds.

Both of the components of ΔG change with temperature. Enthalpy changes linearly, with the proportionality coefficient being heat capacity (ΔC_p, ΔH(T) = ΔH(T₀) + ΔC_p(T-T₀)). Heat capacity is the amount of heat needed to change the temperature of protein by one degree.

Entropy also changes with temperature, though in a bit more complex way (ΔS = ΔS(T₀) + ΔC_pln(T/T₀)). When we combine the two components of ΔG, we get this:

ΔG(T) = ΔG(T₀) + ΔC_p(T-T₀) - ΔC_pTln(T/T₀)

This is a very interesting relationship. It gives ΔG(T) its characteristic shape with a maximum corresponding to the T of maximums stability, and two denaturation temperatures (on the graph below I plot –ΔG, rather than ΔG just so that the plot looks nicer).

We are all familiar with protein denaturation at high temperatures (we all boiled eggs!), but at lower temperatures? Well, this one happens as well, but very, very slowly, as all the reactions tend to at low temperatures, so we do not notice it that much. However, indeed, some proteins are better off when stored at -20C^o, than at -80C^o.

Heat capacity is intimately linked to the above-mentioned solvent accessible area (SAA). The reason for that is that it is the water surrounding the protein that gives it its heat capacity. Water molecules next to the protein are restricted in their freedom; they are essentially frozen, and ice, as we know, has tremendous heat capacity. When the protein denatures, its SAA increases, and so does the heat capacity. Heat capacity change upon denaturation in turn is determining the shape of folding ΔG dependence on temperature (see equation above).

And now we are primed to discuss how the extrermophylic proteins cope with high temperatures. One can imagine two obvious solutions. First, they could increase their stability (ΔG) (curve A). However, this would result in a bit too stable proteins that will be very hard to degrade, and this is not good for metabolism. Also, they will be too rigid, and flexibility is necessary for protein function. Second, they could move their temperature of maximum stability (curve B).

In reality they do something completely different! They decrease the ΔC_p instead, flattening the ΔG curve.

So how do they decrease the ΔC_p? Well this is all about the nature of the denatured state. ΔC_p is proportional to ΔSAA of protein unfolding, but proportionality is different for hydrophobic residues (these freeze water well, thus proportionality coefficient is high) and hydrophilic ones (these are similar to water in their nature, and thus do not restrict its movement too much, and the proportionality coefficient is lower).

Thermophilic proteins enrich the normally hydrophobic protein core with polar residues, forming salt bridges and dipole-dipole pairs. This results in a more rigid structure, thus you still pay in flexibility somewhat, therefore if there is no need for extreme temperatures, hydrophobic core is better.

Modifying the protein core is not an easy task since you need to compensate for one substitution with another (say, you have a positively charged residue, and now in order to compensate it you need a negatively charged one). Moving in the other direction (thermophilic to mesophilic) would be equally tricky. Therefore keeping your temperature stable – just like we do! – allows avoiding all these complicated thermodynamic matters altogether.

References:

Fu H, Grimsley G, Scholtz JM, & Pace CN (2010). Increasing protein stability: importance of ΔC_p and the denatured state. Protein science : a publication of the Protein Society, 19 (5), 1044-52 PMID: 20340133

Franzosa EA, & Xia Y (2009). Structural determinants of protein evolution are context-sensitive at the residue level. Molecular biology and evolution, 26 (10), 2387-95 PMID: 19597162

Drummond DA, & Wilke CO (2008). Mistranslation-induced protein misfolding as a dominant constraint on coding-sequence evolution. Cell, 134 (2), 341-52 PMID: 18662548

Geiler-Samerotte KA, Dion MF, Budnik BA, Wang SM, Hartl DL, & Drummond DA (2011). Misfolded proteins impose a dosage-dependent fitness cost and trigger a cytosolic unfolded protein response in yeast. Proceedings of the National Academy of Sciences of the United States of America, 108 (2), 680-5 PMID: 21187411

Loladze VV, Ermolenko DN, & Makhatadze GI (2001). Heat capacity changes upon burial of polar and nonpolar groups in proteins. Protein science : a publication of the Protein Society, 10 (7), 1343-52 PMID: 11420436

DePristo MA, Weinreich DM, & Hartl DL (2005). Missense meanderings in sequence space: a biophysical view of protein evolution. Nature reviews. Genetics, 6 (9), 678-87 PMID: 16074985

How to swap a gearbox for a new model right on the highway

Protein biosyntheses is central a hub for cellular physiology: proteins are essencial for all the cellular processes. Therefore changing something really important in translational machinery is really hard: you still need to continue producing proteins! Swapping an important translational factor for another one? That sounds impossible, but this is exactly what happend with eEF1A - eukaryotic factor that brings aminoacylated tRNA to the ribosome. Moreover, it happened several times!

It was indeed swapped for a similar, yet different protein EFL (EF-Like) several times during eukaryotic evolution. The main difference between EFL and eEF1A is in theis GTPase cycle. eEF1A, just like its bacterial counterpart EF-Tu, needs a specialized factor in order to regenerate it from the GDP to GTP-bound state (Guanine nucleotide Exchange Factor, GEF). EFL does not need a GEF, so it is in a sense simpler.

Loosing a GEF seems to be a common theme in the evolution of translational GTPases. Mitochondrial EF-Tu lost its GEF (EF-Ts) in Saccharomyces cerevisiae, though retained that in human and S. pombe! Moreover, it is possible to select mutants in yeast eEF1A which would confere GEF-independence, turning into something like EFL.

Sometimes regulating GTPases is just too much to ask for and Nature cuts corners.

References:

Keeling PJ, & Inagaki Y (2004). A class of eukaryotic GTPase with a punctate distribution suggesting multiple functional replacements of translation elongation factor 1alpha. Proceedings of the National Academy of Sciences of the United States of America, 101 (43), 15380-5 PMID: 15492217

Rosenthal LP, & Bodley JW (1987). Purification and characterization of Saccharomyces cerevisiae mitochondrial elongation factor Tu. The Journal of biological chemistry, 262 (23), 10955-9 PMID: 3301847

Chiron S, Suleau A, & Bonnefoy N (2005). Mitochondrial translation: elongation factor tu is essential in fission yeast and depends on an exchange factor conserved in humans but not in budding yeast. Genetics, 169 (4), 1891-901 PMID: 15695360

Ozturk SB, & Kinzy TG (2008). Guanine nucleotide exchange factor independence of the G-protein eEF1A through novel mutant forms and biochemical properties. The Journal of biological chemistry, 283 (34), 23244-53 PMID: 18562321

mitochondrial translation - complete mess

Mitochondria have their own genome, their own translational machinery and their own mRNAs which code a handful of proteins. Most of the proteins come from the cytoplasm, but some get translated inside the mitochondria. And this is done in amazingly weid way...

Bacteria - mitochondrial ancestors - have 3 initiation factors, IF1, 2 and 3, and all of these are absolutely necessary for the bacterial viability.

Mitochondria do not have IF1 (all mitochondria), and unlike mammals, yeast ones do not have IF3! Also they seem to use mitochondria-specific initiation factor AEP3. Do mammals have AEP3 homologue? Worth checking... To make things more complicated, mitochondria have their own special mRNA-specific factors... and again, there is a catch. Yeast mRNA have long 5' UTRs (untranslated regions), and mammals have short, they basically have leaderless mRNAs (I'd love to see a good reference for that! THIS is a little bit out of date...) - therefore I would expect that these mRNA-specific IFs work differently in yeast in mammals. So yeast and mammalian mitochondrial translation seems to have very, very different translational apparatus. Isn't is weird?

Another amazing thing is that translation and protein localization are tightly linked in mitochondria. mRNA-specific initiation factors (i.e. Pet111p) and above-mentioned AEP3 interact with the mitochondrial membrane (most of the mitochondrially-translated proteins are membrane proteins), so that translation is localized where the proteins should go.

Biochemistry and Evolution: heterologous systems

We have loads of different critters co-inhabiting with us, and this can be very useful if you need to publish some papers and progress your career. There are at least 2 ways of doing so.

First, you can do phylogenetic analysis of these and ask questions, such as what is conserved in particular protein? Can we relate these conserved residues to some conserved function? What is not conserved? How is this protein related to other proteins? Can we retrace functional evolution via sequence evolution and vice versa, can we predict some functions for the not yet well studied protein knowing something about its sequence? Gem here is talking about stuff like that.

However, there is a problem with this sort of approach. You end up presenting your data in a format that can be criticized by reviewers or readers, or both. They might wonder about your bootstrap values, accuse you of LBA and in general scoff your alignments and so on. Also, you endup testing your hypotheses, or having them tested by other people, and there is a possibility of being wrong when you test things you predict.

Therefore there is an alternative way of benefiting from the diversity of live in order to produce papers. That would be using heterologous systems. This is how it goes.

You take an interacting protein pair from 2 bugs, A and B. It does not matter what bugs are these. Some bugs. As an example you can take proteins EF-G and RRF which split the ribosome after it has finished making the protein. Then you mix them and see if they can do the job in different combinations (AA, AB, BA and BB). If they all work, then you take bug C, and repeat it until you find a combination that does not work. And if you are not bored, you can take bugs D and F and do it again. So far you produced two papers.

The only question remaining is - did you learn anything? From the evolutionary point of view your result is boring - you have found some sequence co-variation, which results in factors from the same organism working together, but not cooperating with a stranger. This is an obvious result which so far generated no hypotheses (having an alignment on the side could generate one though - looking for the co-variation in the 2 proteins would suggest where the differences are).

And when we talk about alignments, we are talking about multiple ones. Aligning 2 genes from the bugs A and B you study has very little information about the conservation of the differences you are looking at (but this is exactly what is done in the EF-G / RRF paper cited above and in the later paper from the same group working now on IF2 from E. coli and bovine mitochondria - heterologous system strikes again!).

References:

Having a tree to go with alignment will make it all even more informative, cause then you can see the directions of sequence changes, but hey - now we are getting in trouble! We are about to do it properly.

Rao AR, & Varshney U (2001). Specific interaction between the ribosome recycling factor and the elongation factor G from Mycobacterium tuberculosis mediates peptidyl-tRNA release and ribosome recycling in Escherichia coli. The EMBO journal, 20 (11), 2977-86 PMID: 11387230

Seshadri A, Singh NS, & Varshney U (2010). Recycling of the posttermination complexes of Mycobacterium smegmatis and Escherichia coli ribosomes using heterologous factors. Journal of molecular biology, 401 (5), 854-65 PMID: 20561528