Amid the circus of proteomics product introductions at the Drug Discovery Technology Conference, held this week in Boston, Agilent Technologies unveiled a liquid chromatography polyclonal affinity antibody column that it claims is capable of removing the six most prevalent proteins in human blood serum, with minimal retention of less abundant proteins. (For other products, see table, p. 8.)
The six proteins are: albumin, immunoglobulin G, immunoglobulin A, alpha-1-antitrypsin, transferrin, and haptoglobin. They make up some 85 percent of the protein mass in human blood serum, and often are blamed for researchers’ inability to see less abundant but potentially more significant proteins when screening serum for biomarkers.
While not all researchers would agree that removing the proteins is the best solution to this problem, it’s a “darn good idea,” according to Gilbert Omenn, a University of Michigan professor and one of Agilent’s beta testers. “That’s a pretty good set to get rid of, because they aren’t likely disease markers,” said Omenn, who heads HUPO’s Human Plasma Proteome Project. “So for methods where you have a limit in how much you can load — or just to get rid of the big smear on gels and the overlap of many fractions on liquid separation — it’s a terrific advance.”
The big issue in studying serum, as well as cellular material or tumor lysates, said Omenn, is “that there’s a very large dynamic range of concentration of proteins of interest:” Ideally, plasma proteomics researchers would like to detect proteins across eight to 10 orders of magnitude. This is impossible even if Agilent’s column was to work as well as the company claims it does, but “you move down the concentration curve at least one order of magnitude” using the column, Omenn said. He added that another order of magnitude would be gained by the ability to load 10 times as much sample onto an LC column or 2D gel once the offending proteins had been removed.
Not all scientists who work with the plasma proteome agree that removing the six proteins that Agilent targets will really help things. “You can keep removing proteins all day and you don’t dig that much more deeply in the proteome, because there’s always another protein that will replace it as a very abundant protein,” said Joshua Adkins, a postdoctoral researcher who works on the human serum proteome and has “dabbled” in high abundance protein removal at the Battelle Institute of Pacific Northwest National Laboratory in Seattle. Adkins published a paper last year (see PM 12-16-02) with Joel Pounds, in which he described the removal of IgG by fusing it to another product, protein A/G — which he said “works great.” He has also tried removing albumin and tranferrin, using both monoclonal and polyclonal antibodies. Ultimately, however, Adkins is unconvinced that this exercise is worth the trouble. “So they’ve removed six, maybe we should start to remove 10 or 20 — where do you stop?” he said.
With competing products such as protein A/G beads, Cibacron blue binding dye, and chicken IgY antibodies already on the market to remove select proteins from serum, as well as a new albumin and IgG remover from Amersham (see p. 2), Agilent certainly isn’t the first to make a product that removes unwanted proteins — although it is the first to attempt to remove so many in one fell swoop. It is also not yet clear that the column’s specificity is as impressive as Agilent says it is. “In my experience, polyclonal antibodies [which Agilent uses] did a fantastic job of removing albumin, but they also had the most likelihood of pulling out other proteins in a specific/non-specific fashion,” Adkins said.
Jerome Bailey, technology program manager of the bioreagents group at Agilent, admitted that pulling out extra proteins that may be bound to the targeted proteins was a problem, although he downplayed its effect. “You never totally avoid [removing other proteins] ever, but we’re using affinity purified polyclonals, and that’s as good as you can do,” he said. Bailey noted that tests had found some apolipoprotein in the elution columns, but that “anything else that is retained is not quantitative.” Omenn also said that the word is still out on how well Agilent’s column avoids excessive retention, since his own lab’s tests have not been completed.