Is the food additive trehalose really unsafe? - "Contrary to the facts" -SFSS conducted a fact-check on the Japanese statements concerning the controversial article in British journal "Nature".



First of all, we would like to refer the following Japanese articles that were released in response to the Nature article:

"Article Observation: Trehalose added to food was the cause of Clostridium epidemics (original content in Japanese)" Shinichi Nishikawa, NPO All-About Science Japan Representative Director, Yahoo Japan News, January 7, 2018, Sun,
"British scientific journal pointed out the risk. Is a food additive, trehalose, really safe? (original content in Japanese)" Nikkan Gendai, January 24, 2018,

As basic information, Dr. Nishikawa explained in his article that trehalose is a disaccharide naturally contained in foods such as mushrooms and fermented foods using yeast (bread and beer), and Clostridium difficile is a human intestinal bacteria that can cause virulent epidemic outbreaks usually due to changes in the intestinal microbial balance by antibiotics.

However, a significant amount of safety data is required for a food additive to be approved for use not only in Japan, but also in the US and Europe. It could be premature to suspect the safety of trehalose based on the Nature article. The SFSS conducted a fact-check (fact verification) on the following statements in 2 Japanese articles previously mentioned and the data from the Nature article.

The title of the first article:
"Article Observation: Trehalose added to food was the cause of Clostridium epidemics"

Excerpt from the second article:
"Trehalose that has been used as food additive causes epidemic Clostridium difficile (CD) outbreaks." "Trehalose is .... suspected to cause food poisoning by becoming a source of energy for Clostridium difficile (CD) in the intestines."

Please refer to the policy and criteria of the SFSS fact-check here. Since this was the first time to use a scaled judgment (rating), the judgment criteria table is shown below.

[Judgment criteria]
The SFSS shall assign one of the following judgment ratings as a conclusion of the fact check:

Level 0 (accurate): Statement is based on scientific evidence and is clear and accurate.
Level 1 (uncertain evidence): Scientific evidence is not found as a result of investigation.
The author of the statement needs to demonstrate scientific evidence.
Level 2 (inaccurate): Not proved to be contrary to the facts, but lacks scientific evidence in important factual description of the statement. The expression is inappropriate and misleading.
Level 3 (contrary to the facts): Statement lacks scientific evidence and is contrary to the facts.
Level 4: (Fake news): Statement lacks scientific evidence and is intended for disinformation.

Evidence verification No.1

The authors of the Nature article state that the widespread adoption and use of trehalose in the diet coincides with the emergence of virulent Clostridium difficile (RT027 and RT078 strains), thus trehalose in the human diet has played a significant role in the emergence of these epidemics as shown in Fig.6:


Although trehalose was not approved as GRAS in the US until 2000, an epidemic outbreak of RT 027 strain was reported in Minneapolis in 1988 and an epidemic outbreak of RT 078 strain in Arizona was reported in 1991. It is clear that these CD strains were not induced by trehalose added to food products (or it could be interpreted that they were induced by trehalose existing in nature?) In addition, an epidemic outbreak of RT027 strain in more than 30 hospitals in Canada was reported in 2003 before trehalose was approved as food in Canada in 2005. (The information is obtained from Hayashibara Co. Ltd, which obtained the regulatory approval in Canada).

It can be wondered if the authors of the Nature article are suggesting such outbreaks were caused by trehalose that would have been illegally brought to Canada. Moreover, if most of the outbreaks spread within the hospitals, a food ingredient contained in regular food products consumed outside of the facilities is less likely to be directly attributable to the outbreaks. It can be more reasonable to conclude that the virulent CD strains were due to the imbalance of the patients' intestinal microbial flora from other reasons specific to the hospital environment.

The weakest point of Fig. 6 is that Japan, where the largest amount of trehalose as a food additive has been distributed since 1995, is not listed. It appears that the authors have hypothesized that trehalose is a cause for the epidemic outbreaks of virulent CD strains. If this is the case why have such outbreaks not been reported in Japan, where trehalose has been consumed for over 20 years at much higher per capita amounts than where the CD outbreaks have occurred?
*Although virulence CD strains were reported in 2001, there were no outbreaks reported in Japan.

[Conclusion of fact verification derived from aforementioned evidence]
We rated the following article as Level 3 (contrary to the facts).
"Article Observation: Trehalose added to food was the cause of Clostridium epidemics"
"Trehalose that has been used as food additive causes epidemic Clostridium difficile (CD) outbreaks.

Evidence verification No.2


The authors of the Nature article infected two groups of mice with CD. Before being infected the mice were administered either trehalose or no treatment (control group) after the mice were treated with antibiotics. The mortality of the two groups were compared. They stated that the addition of trehalose caused increased mortality compared with the mice not given trehalose. (As shown in the graph at right, Figure 3 in the article) The graph appears to support the conclusion that trehalose intake enhances Clostridium difficile, which worsens bowel inflammation.

However, there are two weaknesses in their arguments. One is the fact that trehalose is most likely metabolized in the small intestine of mice and becomes glucose when orally administered. It has been demonstrated that glucose and the glucose from the digestion of trehalose are essentially taken up into the blood at the same rate when orally administered in the same amount. (Kinya T et al., Nutr Res, 23, 287 -298, 2003). In other words, this test was more likely demonstrating the effect of glucose rather than trehalose.

Another weakness of their argument is that the same amount of other carbon sources, other than trehalose, were not provided to the control group. Suppose the argument in the previous paragraph is valid, the same amount of glucose could have caused overgrowth of CD and increased the mortality of the control group. This invites skepticism as to whether any other carbon source was purposely excluded since an overgrowth of CD and a subsequent increase the mortality, would have occurred.


In fact, as shown in the table on the right, the authors presented the data to confirm whether the listed carbon sources are utilized by non-RT027CD strains (CD630) and the RT027 strain (CD2015) in the beginning of the article, respectively. The carbon sources labeled (+) are utilized by the CD bacteria, which indicates that RT027 CD strain utilized not only trehalose, but also glucose as previously pointed out. In addition, the chart indicates that carbon sources such as fructose, sorbitol and methionine were also utilized only by the RT027 CD strain and not by the non-RT027 CD strain.

It is known that carbon sources including carbohydrate sugars are generally utilized by intestinal bacteria. Although the results from the animal and in-vitro studies indicated that trehalose enhances virulent CD bacteria, it is very unlikely that a human would be in a situation where trehalose is the only source of energy. This makes the hypothesis unrealistic. Additionally, the intestinal microbial flora is completely different between a mouse and a human. The conclusion based on the evidence provided is deemed uncertain to accept at face value. Suppose carbohydrate sugars including trehalose are a direct cause of virulent CD bacteria growth, then food items such as mushrooms, bread and beer, which naturally contain trehalose (Please refer to the Carbohydrate Section in the Standard Tables of Food Composition in Japan by the Ministry of Education Ministry of Food and Nutrition) could also cause of epidemic outbreaks of CD bacteria. This is obviously not the case in Japan where no such epidemics have been reported, which strongly suggests the hypothesis is false and not trustworthy.

[Conclusion of fact verification derived from aforementioned evidence]
The article containing the quote that, "Trehalose is .... suspected to cause food poisoning by becoming a source of energy in the intestines", is rated as Level 3: (contrary to the facts), almost reaching Level 4: (Fake news) since it appears to stimulate consumer's anxiety on the safety of a food additive from flawed conclusions.

After in-depth assessment, SFSS came to the conclusion that the following statements in the articles are rated Level 3: (contrary to the facts) using our fact verification process based on scientific evidence:
"Trehalose added to food was the cause of Clostridium epidemics." and, "Trehalose is .... suspected to cause food poisoning by becoming a source of energy in the intestines."

Takeshi Yamasaki, D.V.M, Ph.D. (Chairman of SFSS) takes full responsibility for the wording and content of this article.

First Draft: February 6, 2018 14:30
Updated: February 7, 2018 14:40 (the first image was replaced.)

*Please refer to the policy and criteria of SFSS fact-check (content in Japanese) click here
*Please refer to overview of the SFSS organization (content in Japanese) click here