What They Saw
They grew cells with or without ammonium, then added argon or hydrogen to their headspace, and measured whole-cell or purified hydrogenase activity. Oxygen or methylene blue were electron acceptors.
With ammonium, there was a little activity, but adding hydrogen gas increased it about 2.5 to 5 times. As a control, injecting the same amount of argon didn't change anything. In nitrogen-fixing cells, adding hydrogen didn't affect activity.
As with others, activity increased over time in the culture, even corrected by biomass; the hypothesis was that excess carbon inhibits it somehow.
If they added an mRNA or protein synthesis inhibitor (rifampin or chloramphenicol) before adding the hydrogen, activity didn't increase with either case, so it seemed like the regulation was transcriptional.
Also, since the effect was the same with methylene blue (which doesn't require electron transport chain components to act as electron acceptor), it seemed that the regulation was at the hydrogenase directly rather than a related component.
Comparing a couple of Mo nitrogenase-deficient strains (CA11 and CA30) to their parent, they saw that hydrogen didn't affect hydrogenase activity in CA much (in nitrogen-fixing conditions), but it did increase the activity a lot in the mutants. The hydrogenase protein abundance increased too. But in conditions with ammonium, CA and CA11 behaved pretty similar.