NEW YORK (GenomeWeb) – Using quantitative proteomics, a team led by researchers from the University of California, San Diego, has shed light on a question that has challenged biologists for nearly a century.
In a study published this week in Nature, the researchers presented a model in Escherichia coli that provides an explanation for why, in instances of high growth, cells will use fermentation for energy production instead of the more efficient respiration pathway, even when sufficient oxygen is available.
This phenomenon was first identified in the 1920s by Nobel laureate Otto Warburg, who observed it in cancer cells, and has since been observed in a variety of cells during periods of high growth, including fast-growing bacteria, yeast, and human stem cells.
Over the years, a variety of explanations for the phenomenon have been put forth, UCSD researcher Terence Hwa, senior author on the Nature paper, told GenomeWeb.
"I would say that anything you can think of has been proposed," he said. "For instance, for bacterial, [it was proposed] that they want to take in excess glucose for competitive reasons so that sugars are not around for other species to eat. Ecological reasons, genetic reasons — everything has been looked at."
One of these theories, he noted, has been the solution that he and his colleagues hit upon — that cells turn to fermentation in high-growth situations because, while it is less efficient than respiration in terms of the carbon required, it is more efficient in terms of the amount of protein machinery needed.
This conclusion was hinted at years ago by researchers who noted that proteins involved in fermentation have faster kinetics than those involved in respiration. Then, in 2009, researchers from The Netherlands' Vrije Universiteit Amsterdam and Delft University of Technology put forth the theory that the shift to the less-efficient fermentation stemmed from the need to increase protein content under high growth conditions.
For cells to grow quickly, they need to have large amounts of ribosomes, and this, Hwa noted, means that existing ribosomes must direct energy toward production of these new ribosomes, which, in turn makes for fewer ribosomes making proteins. Under such conditions, he said, "if you can, you would rather use a pathway where you don't need to make as many proteins in order to generate whatever [energy] you need."
Respiration, he noted, is more efficient in terms of the carbon required to produce a given amount of ATP, but it is less efficient in terms of the total amount of protein machinery required to produce that amount of ATP.
"The protein machinery [in respiration] is big," Hwa said. "[The cell] needs lots of ribosomes to make all of that machinery. When you are growing fast you don't care so much about saving carbon, so saving ribosomes becomes the predominant issue."
Mass spec-based proteomics allowed the researchers to actually measure the amounts of proteins involved in the respiration and fermentation pathways, providing a quantitative underpinning to the theory. Using a Sciex TripleTOF 5600 they collected quantitative measurements of all the proteins involved in the two pathways.
"We used proteomics to established the [protein] numbers, and our result was that fermentation is twice as efficient from a proteome standpoint," Hwa said.
He noted that, while proteomics can struggle with obtaining accurate measurements of low-abundance proteins, this limitation was not important for their work as they were interested in the total amount of proteins involved in each pathway, so the abundance of individual proteins was trivial.
"The ones we [couldn't reliably measure] we did some estimates for," Hwa said, "but it just doesn’t change the ballpark figure that much."
The researchers' next step was to see if they could develop a model that would predict how the balance between respiration and fermentation would change based on changes in growth rates.
"If the protein cost is important then there are certain consequences [that should follow]," he said."For example, if we force ribosomes to make useless proteins, that should make the protein cost even more expensive, and the cell should create even more fermentation products."
"Our model tells exactly how it should do it and the data changed according to how we predicted it," he added.
The researchers performed additional tests of their model, inhibiting ribosome function by antibiotics, for instance, which likewise increase the cost of protein production. In this case, as well, the cell shifted more heavily toward fermentation as their model predicted.
The model predicted E. coli behavior "marvelously," Hwa said, adding that, in the case of the relationship between respiration and fermentation in this organism, he believed the Nature paper was "the final word."
"We did a pretty exhaustive study," he said.
Whether the results can be generalized to other organisms, on the other hand, remains to be seen.
"Is this also the case for [yeast], stem cells, [or] cancer? Of course, we cannot make any extrapolations from bacteria to cancer, but I think our study does provide a perspective of why this happens," he said. "So for people who are working on this problem in cancer, it is a direction they could look into."
Hwa said that his lab does not plan, itself, to apply its findings to cancer research, though he noted that "if the opportunity comes up, I could collaborate with cancer biologists."
"But for me it is more an illustration of, if you want to study the proteome cost issue, this is a blueprint of how this could be done for other organisms," he said.