This story originally ran on July 21.
A study published in this month's issue of Nature suggests that metalloproteomes – the full collection of metal-containing proteins in an organism – may be significantly larger and more diverse than previously thought.
Combining liquid chromatography with inductively coupled plasma mass spectrometry and tandem mass spectrometry, researchers at the University of Georgia, the Scripps Research Institute, and the Lawrence Berkeley National Laboratory demonstrated that the metalloproteome of Pyrococcus furiosus is likely at least twice as large as was thought, Michael Adams, professor of biochemistry and molecular biology at the University of Georgia and first author on the paper, told ProteoMonitor.
The results suggest that metal-binding proteins – considered a fairly mature field of study – are more significant and less well-characterized than previously assumed, Adams said.
"That's one of the main messages to come out of this – metals play a much more dominant role than we thought," he said. "Traditionally metals have tended to be almost an afterthought. That's to say, people are focusing on a particular enzyme and they purify the enzyme and then find it contains metal. There are some really good stories in the literature of people working with enzymes for decades without even knowing there was a metal present. So the metalloprotein literature is based mainly on what people have found after the event."
In the study, the scientists approached the question from the opposite direction, first analyzing metals found in the cytoplasm of P. furiosus using liquid chromatography and inductively coupled plasma mass spectrometry and then attempting, via tandem mass spec, to determine what proteins these metals had bound to.
Proteins containing five metals – cobalt, iron, nickel, tungsten, and zinc – had been previously purified from P. furiosus, but Adams and colleagues found a total of 21 different metals in the microorganism's cytoplasm, including lead, titanium, and uranium. Of these 21, 18 were found as part of macromolecular complexes as opposed to as free ions.
To determine if uptake of the unexpected metals was the result of a biological function as opposed to inadvertent assimilation, the researchers put the samples through an anion exchange separation. After the separation, they identified ten metals taken up by stable cytoplasmic metalloproteins, five of which – molybdenum, manganese, vanadium, lead, and uranium – were not previously known in P. furiosus.
They then put these samples through second-level chromatography, ending up with 790 fractions that ICP-MS analysis revealed to contain a total of 343 distinct metal peaks. Of these 343 metal peaks, 158 didn't match any predicted metalloprotein, with the five metals previously known to be present in P. furiosus accounting for 83 of the unassigned peaks, and the five previously not known to be present accounting for the remaining 75 unassigned peaks.
"About half the metal peaks we identified we couldn't, from a bioinformatics perspective, assign a particular protein to that metal," Adams said, "So by definition that metal had to be associated with a new type of metal-containing protein."
The researchers then selected eight of the 158 unexpected metal peaks and purified the peaks through multistep chromatography until they were left with a homogeneous protein containing the metal, which they then identified using MALDI-MS. This process yielded four proteins containing misincorporated lead and uranium and four novel nickel- and molybdenum-containing proteins – further demonstrating gaps in the characterization of the P. furiosus metalloproteome.
"Hopefully this raises the awareness of metals," Adams said of the research. "People will be much more aware that metals are perhaps much more prevalent in biological systems and should perhaps be one of the first things researchers look at rather than the last thing, if at all."
More prosaically, it may also allow researchers to culture a variety of microorganisms that have been heretofore difficult or impossible to grow due to incomplete knowledge of what metals are needed for the culture media, he noted.
Adams is now examining a variety of other microorganisms, particularly deep subsurface organisms and those related to bioenergy production, as part of his investigation of the metalloproteome's diversity.
His team is also working to streamline and miniaturize the methodology for use with HPLC, as well as to develop bioinformatics tools that will allow the identification of metalloproteins in metal peaks without having to go through multistep chromatographic purification.
"The issue is even with large-scale purification, when we purify these proteins we still end up with only micrograms of some of them. So we can do the analytical purification, but for the very low-abundance proteins you're going to have to have some sort of statistical analysis," Adams said.
"What we're trying to do now is make this more of an analytical procedure and come up with a computational framework to try and predict which protein is associated with which metal just by co-occurrence," he added. "We're about to submit a paper to BMC Bioinformatics on exactly this, having a sort of computational framework where you look at all the fractionation space and you look at all the proteins that are found in that space and all the metals found in that space and you define or quantitate their co-occurrence."
The ultimate goal, he said, is "to establish the methodology to be able to determine metal utilization and metal distribution very quickly in different tissues or biological samples, [for instance] a normal cell versus a cancer cell. People with their tissue or their microbe can, in a very short amount of time determine the types of metals their organisms contain and how much of that organism's proteome it devotes to a certain metal."