By Dean Haycock

Special To BioWorld Today

"What can you infer about the Creator by studying his works?" J.B.S. Haldane was reportedly asked. "He has an inordinate fondness for beetles," answered the biologist, referring to the prevalence of the insects on the planet.

Today, someone thinking in molecular terms might cite "an inordinate fondness for carbohydrates." Once thought to be "hydrates of carbon" — a description later found to be inaccurate — carbohydrates are the most abundant organic compounds found in nature. They are made and eaten by nearly all plants and animals. Sugars, starches, cellulose, chitin, and many other cellular products are carbohydrates. They serve as energy sources, structural components and as part of DNA and RNA molecules. Streptomycin and some other antibiotics are carbohydrate derivatives.

But where are all the other carbohydrate-based drugs one might expect, given the widespread and important functions carbohydrates play in the body? Where are the carbohydrates for treating cancer, inflammatory and infectious diseases, and other disorders?

At least 17 companies are working on them. Most are in late stage development. A few are in the clinic.

The structural complexity of many carbohydrates presents a significant challenge to chemists. Several groups have developed new procedures designed to simplify the synthesis of complex carbohydrates. These approaches combine traditional methods of chemical synthesis and the use of enzymes.

"I think that the status of the synthesis technology is already adequate for high-value products like therapeutic and medical products. But there are important markets that can be addressed in a consumer product arena as well," said Jim Paulson, chief scientific officer and general manager of Glytec, a unit of San Diego-based Cytel Corp. that specializes in manufacturing carbohydrates for corporate partners.

In May, Cytel received a patent for a new way to enzymatically synthesize complex carbohydrates, also called oligosaccharides. Dubbed Sugar Nucleotide Cycling (SNC), this process uses recombinant glycosyltransferase enzymes to link up carbohydrate components in a way closer to that found in nature than in a traditional chemistry lab. The process allows Cytel to achieve yields nearing 100 percent for certain carbohydrates.

Now, Cytel scientists have teamed up with researchers from the Institute for Biological Sciences of the National Research Council, in Ottawa, Ontario, to provide the low-cost, high-yield technology needed to manufacture complex carbohydrate drugs and other commercial products in a commercially feasible way.

Warren Wakarchuk, a research officer at the Institute for Biological Sciences, and his co-authors describe their advance in "The synthesis of sialylated oligosaccharides using a CMP-Neu5Ac synthetase/sialyltransferase fusion," which appears in the August issue of Nature Biotechnology.

The researchers succeeded in combining two enzymes used in the SNC process into one molecule. The enzymes, sialyltransferase and a sugar-nucleotide synthetase, are essential for the large-scale enzymatic synthesis of certain carbohydrates called sialyated oligosaccharides. "Linking them together seemed to be a logical step, given that both of them express very well and that we were trying to avoid having to do two purification steps," Wakarchuk said.

Carbohydrates Made In Kilogram Quantities

The strategy of fusing proteins has been used before but never applied to the synthesis of oligosaccharides.

The custom-assembled, bifunctional enzyme was constructed using genes cloned from Neisseria meningitidis bacteria. The genes were inserted into Escherichia coli bacteria, which produced large amounts of the fusion protein. Importantly, the overexpressed product could be purified using a relatively simple procedure. Bacterial enzymes of this type have been easier to produce on a large scale than mammalian enzymes.

"There is a long history of people having a lot of problems using enzymes derived from mammals in the kind of bacterial expression system that we used. It seemed to us that the bacterial enzymes we had been working with are expressed far better than their mammalian counterparts," Wakarchuk said.

Starting with milk sugar, sialic acid and a few other essential ingredients, the researchers used the fusion protein to produce a carbohydrate called alpha-2,3-sialyllactose, which is present in human breast milk. At first, the authors demonstrated that they could produce carbohydrate on the 100-gram scale. Cytel has since begun using the process to produce the carbohydrate on the 15-kilogram scale. In a field where yields of tens of milligrams of product were the norm, production of kilogram quantities at affordable cost represents a significant advance.

"I think this technology is an important step in the direction of achieving costs consistent with developing carbohydrates as consumer products," Paulson said.

Approach Avoids Insolubility Snag, Cuts Costs

The "two-enzymes-in-one" approach gets around the problem of insolubility presented by one of the components, the sialyltransferase. It also reduces costs since it is only necessary to grow one bacterial culture instead of two.

Paulson is thinking about the benefits of fusing the other two enzymes involved in the synthesis of the breast milk carbohydrate.

"A combination of the other two enzymes would make sense." Paulson said "When you start thinking about ton scale rather than kilogram scale, that is very important from a cost consideration."

Carbohydrates like that described in the paper require one particular type of nucleotide sugar. To make other types of carbohydrates, other building blocks will be needed.

"For a global synthesis technology, you need the ability to make six different nucleotide sugars. Just to expand this idea to other nucleotide sugars would be the first major step. The six sugars would give you more combinations of potential structures, just as amino acids do [in protein synthesis]," Paulson said.

At the moment, Wakarchuk's group has no further plans to work on this particular fusion protein but he indicated that his group is constantly looking for new enzymes that will be useful for carbohydrate synthesis. "Obviously now the idea of making other enzyme fusions will probably be explored not only by us but by other people," Wakarchuk told BioWorld Today. *