This article has been updated to clarify Monsanto's relationship with Miragen and miR-168a's effect on low-density lipoprotein.
Adding to the evidence that shows mammals do not take up dietary plant microRNAs, researchers from Monsanto and Miragen Therapeutics last week published new data showing that small RNAs from rice were not detectable in mice following tightly controlled feeding experiments.
The findings directly contradict a highly publicized Cell Research paper from last year, in which a team from Nanjing University led by Chen-Yu Zhang reported that exogenous plant miRNAs were found in the blood of mice and Chinese individuals. Among these was miR-168a, which is highly abundant in rice and is capable of inhibiting low-density lipoprotein receptor adapter protein 1, or LDLRAP1, in the liver and decreasing low-density lipoprotein clearance.
Given that animals in their study were found to have elevated LDL levels, the investigators concluded that exogenous rice miRNAs were passing through the gut and regulating their targets, which raised red flags among those concerned about the safety of genetically modified foods given the interest by some companies in developing crops with altered miRNA profiles.
But to Sergey Ivashuta, RNA technology lead at Monsanto and one of the authors of the new study, the Zhang et al. paper was intriguing because it hinted at the possibility of a new route of administration for therapeutic miRNAs and other nucleic acid drugs, he told Gene Silencing News.
The results in Cell Research were "extraordinary," he said. Given an existing relationship between scientists at Monsanto and miRNA drug developer Miragen, the two companies teamed up again to see if they could replicate the findings from the Zhang-led group.
As described in a paper in Nature Biotechnology, Ivashuta and his colleagues undertook a mouse-feeding study using rice-containing chow diets or a purified synthetic chow lacking plant grain or forage.
After a two-week acclimation period on the synthetic chow, the mice fasted and then were placed on diets of either synthetic chow devoid of plant matter, a nutritionally balanced chow that contained about 40 percent rice, or a chow that consisted primarily of rice.
The investigators sequenced rice small RNAs from both the rice-containing chow and rice grains used in the food preparation and "observed the expected distribution and abundance of rice miRNAs," with miR-168a coming up as the most abundant in both samples.
After all three groups of mice had completed their feeding regimen, small RNAs were sequenced from the animals' livers and plasma using the Illumina HiSeq system.
"We observed the expected endogenous miRNA profile and miRNA length distribution in mouse plasma and liver and rice samples, indicating consistent quality of the small RNA sequencing procedure," the scientists wrote in their paper.
However, an analysis of small RNAs from the livers and plasma of mice fed diets of primarily chow or nutritionally balanced rice-containing chow revealed no measurable uptake of any rice grain miRNAs, including miR-168a.
Of the more than 10 million total sequence reads per library, fewer than 10 reads identical to known rice miRNAs per library were observed in five out of eight samples from mice fed on rice-containing chow.
Notably, a similarly low number of rice miRNA-mappable reads were found in some of the samples from mice fed diets that contained no grains or forage from plants, leading the researchers to suspect sequence errors or cross-contamination.
Further investigation revealed that a low number of mouse small RNA reads are identical to several rice miRNA sequences, they noted. "Even so, most of the rice-like sequences were identical to the miR-414 sequence, which is not detectable in rice grain. In addition, these plant-like sequences were present in similar quantities in all mouse libraries, regardless of diet regimen, including animals fed on synthetic chow, strongly suggesting that these reads are sequencing artifacts or that they resulted from contamination."
To validate their findings, the Monsanto and Miragen team conducted qPCR analysis on the mouse plasma and liver samples for miR-168a. Despite being easily detectable in rice grain, they saw no apparent uptake of the miRNA in the animals fed either of the three diets.
Next, they spiked a mouse liver sample with RNA isolated from rice and used the mix for small RNA library construction.
"Analysis of sequencing results from the mouse-rice RNA library and the library prepared from rice grain RNA indicates that rice miRNAs can be detected at expected or higher frequencies in the mouse-rice RNA library," according to the Nature Biotechnology report. "Sequencing of the mouse-rice RNA library at a depth of just 5 million reads readily obtained the 30 most abundant miRNAs found in deep sequencing of rice grain, accounting for 94.7 percent of the total rice grain miRNAs. Numerous, less abundant miRNAs were also found in the mouse-rice library, indicating a broad capability to detect plant miRNAs."
Based on their findings, Ivashuta suspects that the results obtained by Zhang's team were the result of contamination.
"They found some traces of the plant microRNA in animal samples and they got excited," he said. Because miR-168a shares sequence identity with a gene involved in low-density lipoprotein biology and they observed a change in LDL levels in their samples, that group "connected the dots … and made a sensational claim."
As for the elevated LDL levels observed in Zhang's study, Ivashuta pointed to the fact that those mice were fed rice alone, which is known to affect cholesterol levels. In animals fed a balanced diet but with essentially the same amount of rice, "you don't see any changes," he said.
Publication of the Monsanto/Miragen work comes just a few months after two other research groups independently published reports contradicting the Zhang paper (GSN 7/11/2013).
The first came out of Johns Hopkins School of Medicine and showed no differences between miRNA levels in pigtailed macaques before and after they were fed a plant miRNA-rich fruit and protein shake.
The second study was published by Brigham and Women’s Hospital investigators, who examined the uptake of plant miRNAs in humans, mice, and honeybees.