At the core of just about every plant, algae, and blob of green pond scum on Earth sits a molecular engine for harvesting sunlight. Its only emissions are oxygen – a gas we can all be incredibly thankful for today.

If not for the evolution of this vastly common form of photosynthesis (also known as oxygenic), complex life as we know it almost certainly would never have emerged, at least not in the shape it did.

But knowing exactly whom to thank for such a precious gift is far from straightforward. Most efforts to pin down the origins of an oxygen-splitting photosystem suggest a period around 2.4 billion years ago, a time that coincided with a flood of oxygen spilling into our oceans and atmosphere.

More primitive forms of photosynthesis are likely to have existed, though the ability to pluck oxygen from water would have truly given phototropic organisms an edge, implying this oxygen-producing version was a late adaptation.

Imperial College London molecular biologist Tanai Cardona argues we might have it all wrong, suggesting oxygenic photosynthesis could have been around when life was just getting started around 3.5 billion years ago.

"We had previously shown that the biological system for performing oxygen-production, known as photosystem II, was extremely old, but until now we hadn't been able to place it on the timeline of life's history," says Cardona.

Several years ago, Cardona and his colleagues compared genes in two distantly related bacteria; one that was capable of photosynthesizing without producing oxygen, called Heliobacterium modesticaldum, and a phototropic microbe called a cyanobacterium.

They were surprised to find that in spite of last sharing a common ancestor billions of years ago, and the fact each bacterium harvested sunlight in different ways, an enzyme critical to their respective processes was uncannily similar.

H. modesticaldum's ability to split water strongly suggested microbes might have been capable of generating oxygen from photosynthesis far sooner than contemporary models suggested.

This latest study takes their research a step further, estimating the rate at which proteins essential to photosystem II have evolved over the ages, allowing the team to calculate back to a moment in history when a functional version of the system might have arisen.

"We used a technique called Ancestral Sequence Reconstruction to predict the protein sequences of ancestral photosynthetic proteins," says the study's first author, Thomas Oliver.

"These sequences give us information about how the ancestral photosystem II would have worked and we were able to show that many of the key components required for oxygen evolution in photosystem II can be traced to the earliest stages in the evolution of the enzyme."

As a point of comparison, the team applied the same technique to enzymes known to be crucial to life from the onset, such as ATP synthase and RNA polymerase.

They found strong evidence that photosystem II has been around for as long as these 'foundation' enzymes, placing them among the first-ever microbial life forms around 3.5 billion years ago.

"Now, we know that photosystem II shows patterns of evolution that are usually only attributed to the oldest known enzymes, which were crucial for life itself to evolve," says Cardona.

Just how well these enzymes would have functioned is a task for future research. Without signs of oxygen levels rising so far back in time, it's unlikely to have been an efficient process or one that necessarily conveyed a huge advantage.

Knowing the building blocks were in place, however, could affect the way we determine priorities in searching for life on other planets, suggesting oxygen on a planet barely a billion years old may constitute signs of life.

The discovery also provides researchers with a starting place for designing synthetic forms of photosynthesis.

"Now we have a good sense of how photosynthesis proteins evolve, adapting to a changing world, we can use 'directed evolution' to learn how to change them to produce new kinds of chemistry," says Cardona.

"We could develop photosystems that could carry out complex new green and sustainable chemical reactions entirely powered by light."

This research was published in BBA-Bioenergetics.