Yearning For New Yeasts

What’s holding up approvals for modified microorganisms?
By Susanne Retka Schill | March 13, 2015

The new yeasts being developed for cellulosic ethanol face confusing and somewhat puzzling hurdles, starting with the regulatory language. Agencies don’t approve genetically modified organisms (GMOs). The U.S. EPA reviews the Microbial Commercial Activity Notices that developers are required to submit, and if the agency has no questions, the submission is “dropped from review.” For yeasts that will end up in feed, the Food and Drug Administration also doesn’t call it an approval in the Generally Recognized as Safe program, known as GRAS, but rather says, “FDA has no questions,” or, in the opposite case: “The notice does not provide a basis for a GRAS determination.”

Digging further into the regulations raises even more complexities. Kristi Smedley, vice president of the Center for Regulatory Services Inc., helps companies navigate the FDA regulatory process. “FDA has put the ethanol industry on notice that anything that goes into ethanol production, in which the DDGS is used for feed, should be cleared or authorized as a feed ingredient,” she says. She adds that microorganisms and other processing aids for converting feedstocks like corn stover—with no coproducts going into animal feed—are not regulated by FDA.

Regulatory approval—ahem, clearance—is a big issue, however, for first-generation corn ethanol plants considering bolt-on cellulosic processes that coferment or return the solids left after fermentation to the distillers grain stream. Currently, cellulosic process developers are talking about using the standard yeast and getting a smaller yield boost from the liberated C6 sugars (glucose) in the corn fiber fraction, until such time that modified yeasts or new C5 fermenters are cleared for use in feed.   
The most formal of the three paths toward getting a new feed ingredient recognized is a food additive petition through the FDA, which includes a notice of filing published in the Federal Register and a formal review process. “You end up with a regulation in the Code of Federal Regulations,” Smedley says.

A more informal, and thus more popular, route starts with applying for a new feed definition with the Association of American Feed Control Officials, which leads many to initially think they can circumvent FDA entirely. “That’s not the case at all,” Smedley says. While a more informal process, the memorandums of understanding between AAFCO and FDA put the agency in the driver’s seat. FDA provides the scientific review function, decides whether an application is suitable for the AAFCO process and makes its recommendations to the AFFCO Ingredient Definitions Committee.

The AAFCO submission is extensive, Smedley says. “You have to submit a significant amount of information to FDA to prove that your product is safe for the intended use. Safety is defined as safe for the animal that is consuming the product and safe for the human that would consume the eggs, meat or milk from animals that have consumed the product.” The company has to show the manufacturing process is safe as well and that the new ingredient is consistently produced. For yeast, that means the organism is stable—the yeast produced two years ago was the same a year ago and this year. The AAFCO submission has to cover all of the same points that must be addressed for a feed additive petition, she adds, “but the feed additive petition is just a little more rigorous and there’s also an environmental component.”

Going For GRAS
The FDA’s voluntary notification program for GRAS determination is quite different, and wrongly considered a shortcut by some, Smedley says. “There is no prescriptive way of doing it,” she explains, turning to the language of the regulation that says “generally recognized as safe may be based on the views of experts qualified by scientific training and experience to evaluate the safety of substances directly or indirectly added to food.” Part B of that section goes on to say the scientific procedures used for GRAS “shall require the same quantity and quality of scientific evidence as required to obtain approval of food additive regulation.”

In practice, FDA has interpreted the expert view to include information published in peer-reviewed scientific literature, which is primarily used for ingredients that have been developed through academic research. Another way people have historically demonstrated expert agreement on a new ingredient is to put together a panel, Smedley continues. In the case of a modified yeast, that panel might include experts in the ethanol industry and its coproducts, a microbiologist and someone understanding the safety concerns. The panel could review a company’s documentation or even help draft a dossier demonstrating the product is safe. “The bottom line is there is no prescriptive way. FDA hasn’t said this is how you make a GRAS determination. Under the law, you can market products backed on that GRAS determination.”  That includes self-determined GRAS.

Within the ethanol industry, insiders have noted, self-determined GRAS has been the most popular route taken for a number of additives that end up in distillers grains. The acceptance of self-determined GRAS is dependent upon a high degree of comfort level within the industry that there’s little risk of a liability issue. But, with genetic traits already an issue in new varieties of corn and DDGS exports embargoed in foreign ports on the basis of GMO contamination, it’s no surprise that the industry wants some assurance that GM yeasts will be considered safe, if they end up in the feed.

FDA has a voluntary GRAS determination notification program, where companies submit their documentation for review, Smedley says, “to see if they raise questions—not if they would agree to it.” The human food side, the Center for Food Safety and Applied Nutrition, listed its first GRAS notification in 1998 and has entered 562 ingredients since then. About 40 of those are pending and 420 received FDA letters saying the agency had no questions.

The feed side under the Center for Veterinary Medicine started the voluntary GRAS determination program in 2010 and since then there have been 18 applications, of which three are pending. Five determinations say: “FDA has no questions,” five say: “Notice does not provide a basis for a GRAS determination,” and the other five were withdrawn.

Smedley explains the regulators on the food side have adopted the practice of phoning the applicant if questions were arising, rather than continuing the practice of publishing a negative letter explaining what was lacking, thus giving the applicant a chance to withdraw the notice. When the agency’s feed side began reviewing GRAS notifications, the regulators didn’t give any notice of pending questions and, instead, published detailed, negative letters outlining deficiencies in the GRAS determinations that Smedley says frankly, shocked the industry. “What business would want to get a negative letter like that?” In 2011, the notices that didn’t receive negative letters were withdrawn and it wasn’t until late 2012 before an enzyme notification cleared the FDA with no questions. Nearly a year later, another enzyme cleared and since then, three other notifications have been filed. One of those pending is a modified Saccharomyces cerevisiae to be used in the fermentation of corn to produce ethanol, filed by DSM Nutritional Products. Whether that yeast is intended for first- or second-generation processes is not indicated.

Mascoma has gone the route of getting its modified S. cerevisiae yeasts approved via the AAFCO/FDA process. In February 2012, EPM reported Mascoma’s bioengineered yeast had received a new feed ingredient definition. At the time, Mascoma CEO Bill Grady said he believed it was the first bioengineered yeast accepted for use in corn ethanol production of distillers coproducts for animal feed. Lallemand, Mascoma’s manufacturing and marketing partner, acquired Mascoma late last year and two months later, in early January, AAFCO added another GM yeast from Mascoma to the definitions list. Both yeasts produce enzymes intended for yield enhancements in first-generation corn ethanol plants. The AAFCO definitions name the modified organisms and the organisms used as the source of the genetic insertions to express named enzymes. The definition also states that distillers products for use in animal feed contain no live bioengineered yeast.

The small number of yeasts successfully reviewed or pending might be an indication the tide is about turn for the new yeasts, but for two huge unknowns. Smedley explains that GRAS determinations on the human food side of FDA have become highly controversial and one opposition group has won a lawsuit that may force changes to the system. That process may take a couple of years to play out, but whatever happens will eventually impact the animal food GRAS process.

Just as big, or potentially bigger, is FDA’s final rule for the Food Safety Modernization Act, due out in August. The rule is expected to include new good manufacturing practice regulations affecting everyone in the feed industry. No one in the industry is predicting what changes the FSMA rule might bring to new ingredient approvals. Some say it is quite early in the FSMA process and changes affecting the introduction of novel yeasts and new processing aids may be further down the road. Some are hoping, though, that any changes will make the process less confusing.

EPA’s Microbrial Review
Companies about to introduce GM microorganisms must file a Microbial Commercial Activity Notice with the U.S. EPA under the Biotechnology Program in the Toxic Substances Control Act. Since 1998, just over 50 MCANs have been dropped from review, clearing the organism for manufacture. There is a second review process for environmental release applications, which do require approval from EPA, and there are exemptions for R&D in contained structures. 

Most MCANs listed on the EPA website are for modified organisms producing various enzymes, many from well-known companies offering familiar enzymes to the ethanol industry. Details on the microbial strains and, in many cases, the company names themselves are withheld as confidential business information. In addition to ethanol process enzymes, other microbial products are intended for the manufacturing of detergents or unspecified chemicals. About half the MCANs listed show a date for the required notice of commencement of manufacture—nearly all for enzyme production and just two yeasts.

While it may seem that the agency is efficient in publishing determinations within 90 days of listing, an ethanol industry insider points out that if the agency has a question, the 90 days starts over, and a second question might reset the clock.

Author: Susanne Retka Schill
Senior Editor, Ethanol Producer Magazine
701-738-4922
sretkaschill@bbiinternational.com