630 - Diastereomer-dependent substrate reduction properties of a dinitrogenase containing 1-fluorohomocitrate in the iron-molybdenum cofactor

The normal nitrogenase central cofactor contains homocitrate near the central metal atom; it is required for formation of the cofactor. This study looks at compounds similar to homocitrate (diastereomers of fluorohomocitrate) incorporated into the cofactor, and how they affect the enzyme's activity.

What They Saw
They purified enzyme from Klebsiella pneumoniae and Azotobacter vinelandii I think; it's hard to tell because this paper (and others they cite) are really badly written with regard to methods. (They seriously put in a footnote saying "The experimental details of the synthesis of 1-fluorohomocitrate may be obtained from [author initials]." How is that acceptable, especially in a PNAS paper?). Somehow they got FeMo-cofactor with different analogs of homocitrate, and tested their activity with different substrates.

With threo-fluorohomocitrate, there was hardly any nitrogen fixation, similar to with citrate, but with erythro-fluorohomocitrate, there was about 3.5x less than with plain homocitrate, so it was somewhat active. Both were about half as good at acetylene reduction (measured by ethylene), and just as good at cyanide reduction (measured by methane).

In terms of hydrogen production, if there was a difference between plain homocitrate and the fluoros, it wasn't very big; they produced almost as much as the normal one. The same was true of other analogs (homoisocitrate, isocitrate, etc), though some were impaired (such as citrate, producing only about half as much).

The addition of inhibitors, carbon monoxide (CO) or carbonyl sulfide (COS), affected these different enzymes differently. The amount of hydrogen from homocitrate increased (maybe significantly) but decreased some or none for analogs, depending on the analog, up to 63%. The inhibitors reduced acetylene reduction from 35-100% in all cases, and cyanide reduction a little or a lot too.

Cyanide itself affected hydrogen production. (Dang, this paper is so badly written, it's giving me a headache trying to figure it out.) With all tested analogs (and homocitrate), cyanide inhibited hydrogen production 85-95%. Adding CO with cyanide prevented this inhibition with homocitrate and partially with fluorohomocitrate, but not much the other analogs. CO also inhibits cyanide reduction, at least with some analogs (especially the more active ones).

What This Means
This study can tell a lot about the biochemistry of different substrates binding to the enzyme and the enzyme acting on them. I wonder if the results are skewed somewhat because they only measured some of the products, not all possible ones (i.e. might some analog-based cofactors produce ethane from acetylene in addition to ethylene?). In terms of application, even if some of these analogs have desirable properties, it seems like it would be difficult to get them incorporated in vivo.

Reference:

Madden, M. S., Kindon, N. D., Ludden, P. W. & Shah, V. K. Diastereomer-dependent substrate reduction properties of a dinitrogenase containing 1-fluorohomocitrate in the iron-molybdenum cofactor. PNAS 87, 6517–6521 (1990).

542 - Purification and Characterization of the vnf-encoded Apodinitrogenase from Azotobacter vinelandii

This study looks at the vanadium nitrogenase apoprotein and how it works with its own cofactors or those of the other versions.

What They Saw
Azotobacter vinelandii strains possessing or lacking genes for one or more of the nitrogenases were grown and their protein was extracted. They also extracted central cofactors for each type separately.

Without nifB, a strain can't produce any of the cofactors, so it's easier to get apoproteins. They took vanadium aponitrogenase (Vnf version) and activated it with FeVco. Not surprisingly, they only saw activity in strains containing vnf when vanadium was present and Mo or ammonium was not. Extracts of active strains lost activity when treated with heat or exposed to air for 45 minutes, though the holoenzyme seemed more stable to heat.

When they purified the enzyme as much as possible, the delta subunit (vnfG-encoded) seemed only loosely attached to the others, unlike in A. chroococcum where it purifies together with the others; though it's not clear that conditions were the same. It was necessary for active enzyme though, and seems to be involved in inserting the cofactor into the enzyme, but also something else.

They tried replacing FeVco with FeMoco in the V nitrogenase (or vice versa in the Mo version). With the V version and FeVco, carbon monoxide inhibited about 70% of the acetylene reduction activity but no hydrogen production activity. The Mo nitrogenase with FeVco had a fraction of the V version's acetylene activity but no nitrogen fixation; sadly they didn't test hydrogen production.

With the V nitrogenase and FeMoco, it had a bit (1/6th) of the acetylene reduction activity, which seemed insensitive to CO. Ethane production was proportionally higher (1:4 instead of 1:12). Hydrogen production was reduced about 40%. Nitrogen fixation was pretty much abolished.

With the correct cofactors, the V nitrogenase had about 31% the acetylene reduction activity, 34% of the hydrogen production activity, and 22% of the nitrogen fixation activity of the Mo nitrogenase, but it's not clear if these in vitro assays allow for accurate comparisons.

It didn't seem like FeFeco allowed any activity in the V aponitrogenase.

What This Means
It seems like nitrogen fixation requires a very specific environment, and messing with it in various ways (mutations, different cofactors) messes it up while allowing the enzymes to still do less strict activities.

Reference:

Chatterjee, R., Allen, R. M., Ludden, P. W. & Shah, V. K. Purification and Characterization of the vnf-encoded Apodinitrogenase from Azotobacter vinelandii. J. Biol. Chem. 271, 6819–6826 (1996).

248 - NAD-, NMN-, and NADP-dependent modification of dinitrogenase reductases from Rhodospirillum rubrum and Azotobacter vinelandii

What They Knew
Nitrogen fixation is a demanding process, using a lot of energy, so bacteria regulate it tightly, shutting it off whenever fixed nitrogen is available already. One diazotroph, Rhodospirillum rubrum, regulates its nitrogenase by ADP-ribosylating the dinitrogenase reductase component, using a protein called Dinitrogenase Reductase ADP-ribosyl Transferase, or DRAT. This takes ADP-ribose from NAD when ammonium is present. Another protein, Dinitrogenase Reductase-Activating Glycohydrolase (DRAG), reverses the process.

What They Wanted to Know
Ponnuraj and colleagues studied the nitrogenases of R. rubrum and Azotobacter vinelandii to see how specifically each interacted with the DRAT/DRAG system and various ADP-ribosyl donors.

What They Did
They took different ADP-ribosyl containing molecules (NAD, NADP, NADH, NMN), combined each with A. vinelandii's dinitrogenase reductase (DNR) and R. rubrum's DRAT, then ran them with SDS-PAGE along with samples lacking the molecules, to see which molecules could be used to donate ADP-ribosyl. They also exposed some of each sample to DRAG and ran that alongside to see if it could remove the modification.
Results from this were confirmed with another test, seeing if modified or de-modified DNR could function with dinitrogenase to reduce acetylene.

To see how small a modification works to inactivate the system, they removed a phosphate from phosphoribosylated DNR and tested it again.

In addition to testing A. vinelandii's DNR, they tested R. rubrum's too.

To see more specifically what was going on with A. vinelandii's DNR, they used MALDI-TOF mass spectroscopy.

What They Observed
Based on gels and activity assays, NAD, NMN, and NADP all seemed pretty good at donating to DRAT to inactivate A. vinelandii's DNR. NAAD not so much, or anything else they tried. DRAG seemed able to re-activate DNR with all of them too.

Surprisingly, with R. rubrum's DNR, only NAD seemed to be a good donor. Mass spec results confirmed their expectations about what was going on biochemically.

What This Means
R. rubrum uses DRAT and DRAG to regulate its nitrogenase activity based on whether fixed nitrogen is available already and whether its environment is illuminated or not. This helps save energy, so it doesn't have to break down the whole system and reconstruct it with every little environmental change.

It's not clear how relevant it is for A. vinelandii, though, because that organism doesn't appear to have the genes to produce DRAT/DRAG proteins. It's somewhat interesting that R. rubrum's proteins are able to modify A. vinelandii's nitrogenase, arguably even better than they can with R. rubrum's, but this actually makes some sense, that R. rubrum would have tighter control over its nitrogenase regulation system. Though apparently some in vivo studies suggest it might not be as tight as it seemed here.

This doesn't necessarily mean that A. vinelandii doesn't have a system for post-translational regulation, just that it isn't exactly this one. I haven't found what it is yet, if there is one. And other studies seem to imply that A. vinelandii might not have such tight control (010). I wonder why.

Reference:

Ponnuraj, R. K., Rubio, L. M., Grunwald, S. K. & Ludden, P. W. NAD-, NMN-, and NADP-dependent modification of dinitrogenase reductases from Rhodospirillum rubrum and Azotobacter vinelandii. FEBS Letters 579, 5751–5758 (2005).