Bacteria to the Rescue!
By: Carol A. Rouzer, VICB Communications
Published: June 30, 2014
Therapeutically modified bacteria increase the production of a lipid mediator that reduces obesity in mice
A rapidly growing body of evidence indicates the importance of the symbiosis between humans and the trillions of microbes that colonize their bodies. Indeed, the gut microbiome (microbes inhabiting the gastrointestinal tract) has been shown to have an important impact on many aspects of human health and disease through its influence on neurological, metabolic, and immune functions. These findings inspired Vanderbilt Institute of Chemical Biology member Sean Davies, his Vanderbilt colleagues Owen McGuinness and Kevin Niswender, and Denis Coulon at the University of Bordeaux, to see if therapeutically modified bacteria can be used to combat the obesity epidemic [Z. Chen et al., (2014) J. Clin. Invest., published online June 24, DOI:10.1172/jci72517] (Figure 1).
Figure 1. Mouse models of obesity are frequently used to investigate the causes and potential treatments of obesity in humans. An obese (left) and normal lean mouse (right) are shown. Image reproduced through the courtesy of Wikimedia Commons. U.S. Federal Government, public domain.
The N-acylethanolamines (NAEs) comprise a class of compounds with known anorexigenic activity. Produced in the small intestine in response to feeding, these molecules act on the PPARα (peroxisome proliferator-activated receptor alpha), TRPV1 (transient receptor potential cation channel subfamily V member 1), and GPR119 (G protein-coupled receptor 119) receptors to reduce food intake, delay gastric emptying, inhibit fat absorption, and increase fatty acid oxidation. Ironically, consuming a high fat diet suppresses the production of NAEs, resulting in a loss of their beneficial effects when they are most needed. NAEs are synthesized by the action of a specialized phospholipase D on a precursor N-acyl-phosphatidylethanolamine (NAPE). In turn, NAPEs are generated through the action of an N-acyltransferase, using phosphatidylethanolamine and either a fatty acid or a fatty acyl-CoA as substrates (Figure 2). This relatively simple biochemical pathway led the Davies lab to hypothesize that bacteria transformed with an N-acyltransferase to promote high levels of NAPE production could be used to increase NAE formation in vivo, thereby exerting an anti-obesity effect. They tested this hypothesis using Escherichia coli bacteria to treat C57Bl/6J mice, which become obese when fed a high fat diet.
Figure 2. Biochemical pathway for the biosynthesis of N-acylethanolamines (NAEs). An N-acyltranferase catalyzes the condensation reaction of a fatty acid or a fatty acyl-CoA with a phosphatidylethanolamine precursor to produce an N-acylphosphatidylethanolamine (NAPE) intermediate. A phospholipase D (NAPE-PLD) then cleaves the bond between the phosphate and ethanolamine groups of the NAPE to produce the NAE and phosphatidic acid.
In pilot studies, the investigators transformed the C41-DE3 laboratory strain of E. coli with the N-acyltransferase from Aradidopsis thaliana. These bacteria were then administered to lean mice for seven days by oral gavage. The treated mice exhibited a two-fold increase in NAPE levels in the colon, and smaller increases in other parts of the gastrointestinal tract. Consistent with the hypothesis, the mice also consumed 15% less food. These effects were not observed in mice treated with bacteria that had been transformed with an empty vector. The results suggested that NAPEs produced by the N-acyltransferase-transformed bacteria were being converted to NAEs, which then exerted an anorexigenic action on the mice.
The investigators followed up on these initial promising results with more extensive studies using a probiotic strain of bacteria, E. coli Nissle 1917 (EcN). They first transformed the bacteria with the Photorhabdus luminescens luciferase gene. Then, the bacteria were transformed with an expression vector containing the gene for the A. thaliana N-acyltranferase, or an empty vector, producing the pNAPE-EcN and pEcN strains of E. coli, respectively. The investigators used mass spectrometry to confirm that pNAPE-EcN produced high levels of NAPEs (Figure 3).
Figure 3. Construction of NAPE-producing bacteria. The E. coli Nissle 1917 strain was transformed first with an expression vector for the P. luminescens luciferase and then with either a N-acyltranferase expression vector (pNAPE-AT) or an empty vector (pEmpty) to produce the pNAPE-EcN and pEcN strains of bacteria, respectively. Production of NAPEs by pNAPE-EcN was confirmed by mass spectrometry.
The researchers administered either pNAPE-EcN or pEcN to C57Bl/6J mice through their drinking water for a period of eight weeks. Luminescence from P. luminescens luciferase expressed by the bacteria provided a way to confirm that they were retained in the intestinal tracts of the mice (Figure 4). During that time, the mice also consumed a high fat diet, leading to substantial weight gain. However, mice treated with pNAPE-EcN ate less food, gained less weight, and accumulated less body fat than those treated with pEcN or simply water. The pNAPE-EcN-treated mice also exhibited better glucose tolerance and had lower plasma leptin and insulin levels than either control group. In contrast, pNAPE-EcN treatment had no effect on lean body mass, muscle strength, coordination, or speed, and the treated mice showed no signs of gastrointestinal distress. Furthermore pNAPE-EcN treatment caused no significant change in the types of bacteria occupying the gut, suggesting that the administered E. coli occupied only a very small niche in the microbiome. These results suggested that the reduced weight gain in pNAPE-EcN-treated mice was not due to a nonspecific effect on the gastrointestinal tract or the general microbial population. Rather, the effects most likely resulted from an elevation of NAE formation from NAPEs produced by the transformed bacteria.
Figure 4. Results from bioluminescence imaging of P. luminescens luciferase-expressing bacteria in mice. The two mice on the left were administered water only, while the two mice on the right were administered water containing pNAPE-EcN. Following four days of treatment, luminescence was evaluated using an IVIS imaging system. The IVIS system provides a heat map of the luminescence intensity, with increasing intensity represented by a color change from blue to green to yellow to orange to red. Image kindly provided by Zhongyi Chen and Sean Davies. Copyright 2014.
Following the eight week pNAPE-EcN treatment, the investigators stopped administration of the bacteria while maintaining the mice on their high fat diet. They used luminescence from the luciferase expressed by the transformed bacteria to detect them in the feces of the treated mice. Bacterial excretion continued for about four weeks. During this time, and for an additional two weeks, the pNAPE-EcN-treated mice continued to show reduced food intake, body weight, and fat mass, but then, the mice began to gain weight like the pEcN- and water-treated mice. The investigators concluded that continued colonization of the gastrointestinal tract by pNAPE-EcN sustained the anti-obesity effects of the bacteria after administration was discontinued. However, colonization was not permanent, and the effects were lost as soon as the bacteria were no longer present.
NAPEs produced in the gastrointestinal tract are converted to NAEs, which are then carried by the portal circulation directly to the liver. The researchers confirmed that pNAPE-EcN treatment resulted in a 44% increase in liver NAE levels. They also observed reduced accumulation of fat in the livers of pNAPE-EcN-treated mice, which was consistent with increased expression of the mRNA for enzymes of fatty acid oxidation but not fatty acid synthesis in those livers. A reduction in expression of inflammatory genes in the livers of pNAPE-EcN-treated mice was also consistent with the presence of fewer immunoreactive cells. When pEcN-treated mice were pair-fed the same amount of food consumed by the pNAPE-EcN-treated mice, they showed some reduction in weight gain as compared to pEcN-treated mice allowed to eat at will; however, the pair-fed mice did not exhibit the gene expression changes observed in pNAPE-EcN-treated mice. These findings suggest that the effects of pNAPE-EcN treatment on gene expression are at least partially independent from those on feeding and weight gain.
The effects of pNAPE-EcN treatment on obesity in C57Bl/6J mice fed a high fat diet were striking. However, the investigators acknowledged that these mice are not necessarily a good model for human obesity, which has a polygenic origin and does not necessarily require consumption of a high fat diet. Thus, they tested the effects of pNAPE-EcN treatment on female TallyHo/Jng mice, which become obese on a normal diet. Again, they found that pNAPE-EcN treatment reduced weight gain in these mice. The mice also exhibited the same changes in liver triglyceride accumulation, inflammation, and gene expression as were observed in the pNAPE-EcN-treated obese C57Bl/6J mice; however, the small numbers of TallyHo/Jng mice available precluded demonstration of statistical significance in these parameters.
Together the data support the hypothesis that therapeutically transformed bacteria can be used to alter the metabolic environment of an individual with potentially beneficial outcomes. Clearly, this hypothesis must be tested further in animal models, and eventually in the clinic. However, there is no doubt that it establishes an intriguing paradigm for exploitation of the relationship between humans and their microbial symbionts.