Quinone binding in Type I reaction centers from Heliobacteria

I have been recently reading about the interesting photosynthetic reaction center found in Heliobacteria. I think one of the most striking things about it is the debate and discussions about whether the Heliobacterial reaction center (RC) binds a quinone or not. The most recent and better purification strategy showed that there was 1.6 menaquinones per RC. They suggest that this could be interpreted as 64% of the centers having both quinones bound, 32% with one quinone bound and one empty site, and 4% with no quinones at all. Whatever the case the most likely interpretation about the whole debate―that has been going on for several decades as a matter of fact, is that the menaquinones in the Type I RC from Heliobacteria are loosely bound...  the question is why and what is the relevance of this feature.

I wanted to see how the phylloquinones in Photosystem I are bound to the protein. You can see the figure below that most of the quinone is pretty much in a hollow space.

The head of the quinone interacts with the protein by a single hydrogen bond and by hydrophobic interactions to a tryptophan and phenylalanine side chain, see below.

Not only that, but the tail is surrounded by five chlorophylls and a carotenoid molecule, approaching within 4 Å of the quinone, see below.

To top it all, these are held together by some of the smaller peripheral subunits:

So, no wonder why the quinones of PSI are so well bounnd.

In the case of the Heliobacterial reaction center, the tryptophan and phenylalanine are not conserved; there is only one carotenoid molecule (4,4'-diaponeurosporene) in comparison to 22 β-carotenes in Photosystem I; there are only about 20 chlorophylls in compared to about 96 in Photosystem I from Thermosynechoccocus; and there are no peripheral subunits. So, if the folding of the Heliobacterial reaction center protein is similar to Photosystem I, then the quinones will be very exposed.

I think a really interesting possibility is whether the Heliobacterial Type I could have a quinone reduction activity like in Type II reaction centers under certain conditions: this is the more likely when you consider that the PshB protein that should hold the terminal electron acceptors F(A) and F(B) is also loosely bound.

Ethanol production using heterocyst-forming cyanobacteria

Cyanobacteria are of great biotechnological interest for their potential to produce biofuels driven by oxygenic photosynthesis. In other words, you can make biofuels from sunlight, water, and CO₂. One approach is to produce ethanol: in order to do this a couple of enzymes need to be introduced using genetic engineering to metabolize pyruvate to ethanol. Arguably, the most successful of this is that by company Algenol were they introduced pyruvate decarboxylase and alcohol dehydrogenase from the alpha-proteobacterium Zymomonas mobilis into the marine unicellular cyanobacterium Synechococcus sp. PCC 7642. This approach was patented and in the patent it is said that rates of 1.7 µmoles of ethanol / mg chlorophyll-a / hour were obtained, which in my opinion is a very modest rate. Certainly, it is below 1% of the rates of Photosystem II activity under saturating light. Algenol is now producing 10000 gallons / acre / year, what appears to be a promising yield... and if the life cycle analysis they published is correct, the energy balance is positive: in other words, there is more energy in the ethanol produced than it was invested to drive the company and purify the ethanol.

Here I want to propose an alternative approach that I think will generate better yields. This is based on quantitative proteomic results in multicellular cyanobacteria capable of differentiating heterocysts. Under nitrogen starvation multicellular filamentous cyanobacteria differentiate 5-10% of their cells into a cell type specialized in atmospheric nitrogen fixation, nitrogen-fixing cells are called heterocysts. Heterocysts contain nitrogenase and other oxygen intolerant enzymes and for that reason photosynthetic oxygen evolution is inactivated or slowed down in the heterocysts. However, the surrounding cells are still capable of oxygenic photosynthesis and they transfer reductant to the heterocysts in exchange for fixed nitrogen in the form of glutamine.

A, Filamentous cyanobacteria, the arrow points to a heterocysts. B, Isolated heterocyts.

The proteomic work by Ow et al. (2009) indicated that heterocysts from Nostoc punctifurme contained very large amounts of the enzyme pyruvate kinase which uses phosphoenolpyruvate to generate ATP and pyruvate. In this study it was shown that pyruvate kinase was at least 3.8 times more abundant in the heterocysts compared to the vegetative cells. This predicts that heterocysts might have naturally higher concentrations of pyruvate. The reason why pyruvate is in higher concentrations in heterocysts is because it is the precursor of 2-oxoglutarate, which is the precursor of glutamate. In heterocysts glutamate reacts with ammonia (the product of atmospheric nitrogen reduction by nitrogenase) to produce glutamine.

My idea is to express pyruvate decarboxylase and alcohol dehydrogenase from Zymomonas or Saccharomyces in the heterocysts of a multicellular cyanobacterium, such as the fresh water Nostoc sp. PCC 7120, Nostoc punctiforme or Anabaena variabilis or a marine version like Anabaena sp. 90. The technology to do this is already in existence. Since the metabolism of heterocysts is super-ramped up to provide nitrogen for 90-95% of the cells, I am pretty sure that the yields of ethanol could be pretty high.

It does not come without challenges because probably the diversion of pyruvate to generate ethanol might impair to certain extent nitrogen-fixation, however it has been shown that heterocysts can compensate a loss of reducing equivalents by boosting up cyclic photosynthesis and probably their own metabolism. It might be that a lack of pyruvate could enhance carbon-fixation, a bonus. In any case, we will not know if this is a sound idea until we try it out.

Trichoplax adhaerens, a weird animal with a piece of Photosystem I

I was doing a BLAST of the PsaA subunit of Photosystem I (PSI) restricted to Metazoans and found that the organism Trichoplax adhaerens had a fragment of 141 amino acids of this subunit. The predicted amino acid sequence from T. adhaerens has 97% similarity to that of the diatom Synedra acus.

T. adherens is a  placozoan, a basal eumetazoan, in other words a very ancient animal.

It encompasses the region of PsaA from amino acid 615 to 755 using the numeration of the crystal structure from Thermosynechococcys elongatus 1jb0. This region contains the last two transmembrane helices of the PsaA subunit where some of the redox active chlorophylls are coordinated (see figure 1).

Figure 1. Cartoon model of the PsaA subunit from T. elongatus. In sand color the section of protein that is encoded in the PsaA fragment from T. adherens.

Because of the high similarity of the sequence to that of the diatom, and due to the fact that animals do not have photosystems, we can safely assume that it is an event of horizontal gene transfer... possibly too, contamination.

I will check if there are more fragments of photosynthetic proteins in this animal and report here if I find something else.