My own paper. A 15-minute, under-$1 fiber tip method pulled vesicles averaging about 189 nm out of 20 everyday fruits and vegetables. It shows how many common foods carry them, not what they do once you eat them.
the research library exosomes everywhere
the high-EV diet
Food is not only fuel. Some of it arrives carrying messages.
I kept seeing food vesicles used to sell things before anyone had counted how many are in the food, what cooking does to them, or whether they make it past a stomach. So I read the primary papers and put the measured numbers in one place. Some are solid, most are early, and a few of the most quoted claims did not hold up when other labs tried them.
this page, by tier
- 0 human evidence
- 3 early human
- 10 lab + animal
- 0 background
the short version
what the research found
the headline numbers
-
Raw cow's milk held about 1.01 × 10¹⁰ vesicle particles per mL, and homogenizing, heating and every pasteurization method tested cut that by more than 60%.
-
All five home cooking methods tested on broccoli lowered its vesicle count, and only about 10% of broccoli vesicles survived simulated digestion.
-
Two cow's milk microRNAs made it through a computer-controlled model of the human gut at roughly 10⁹ to 10¹⁰ copies per 300 mL, with the biggest loss in the stomach.
-
In mice, ginger vesicles were taken up by Lactobacillaceae gut bacteria, where one vesicle microRNA switched off a bacterial gene and raised the host's IL-22.
-
After five adults drank up to a liter of milk, one lab reported absorbed microRNA. A second lab re-tested the same blood samples three ways and found none.
the nano angle
This is my home field. I have pulled these vesicles out of 20 everyday fruits and vegetables myself. They are fat-wrapped particles, mostly 30 to 150 nanometers across, and nearly every food vesicle claim you have seen traces back to a handful of papers that isolated and counted them.
my honest bottom line
Food vesicles are real, they can be counted, and in animals they reach the gut and change gut bacteria. The biggest limitation is the step everyone wants to be true: food vesicle cargo reaching human blood and changing human genes has not held up in three separate attempts to repeat it.
evidence at a glance
13 papers
- human evidencenone yet0
- early human3
- lab + animal10
- backgroundnone yet0
Includes 7 nano-angle papers and 4 counterpoints.
every title links to the original
the papers
Sorted from the strongest human evidence down to the lab work and background reading. Counterpoints are marked, and they stay in on purpose.
The same human blood samples were re-tested three ways: a 223-test array, single qPCR tests and small RNA sequencing. None showed a microRNA change after milk, and no uniquely cow microRNA turned up. The authors concluded the data do not support dietary microRNA transfer.
Five adults drank 0.25, 0.5 and 1 liter of cow's milk on different days. Blood curves suggested two milk microRNAs were absorbed while a control microRNA did not move. These are the numbers the next paper tried to reproduce.
Athletes ate fruit rich in three plant microRNAs, and none of them showed up in their blood afterward. Mice and honeybees fed plant microRNAs also showed next to no delivery.
Frying, boiling, microwaving, scalding and steaming all lowered broccoli vesicle counts, protein and cargo. About 10% of the vesicles survived simulated digestion, and only a small fraction crossed a lab model of the gut lining.
Raw cow's milk held 1.01 × 10¹⁰ vesicle particles per mL. Every industrial step tested (homogenization, low-temperature heat, HTST and UHT pasteurization) cut that by more than 60%, and homogenizing and heating each did damage on their own.
show all 13 papersshow fewer
Vesicles the size and shape of exosomes were isolated from honey, royal jelly and bee pollen. In the lab they slowed and killed Staphylococcus aureus, broke up its biofilm, and were taken up by human stromal cells, where honey and royal jelly vesicles sped up cell migration.
Ginger vesicles were picked up mostly by Lactobacillaceae bacteria in the mouse gut. A microRNA inside them switched off a bacterial gene, the bacteria made more indole-3-carboxaldehyde, and the mice made more IL-22, a signal that supports the gut lining. The colitis result depended on IL-22.
In a computer-controlled model of the human stomach and gut, roughly 10⁹ to 10¹⁰ copies of two milk microRNAs per 300 mL of store-bought milk made it through. The stomach caused the biggest drop, and about 10⁸ to 10⁹ copies were still measurable further down.
Vesicles from grape, grapefruit, ginger and carrot all carried protein, fats and microRNA. Ginger vesicles raised heme oxygenase-1 and the calming signal IL-10 in gut immune cells, and mice that ate them had more active stem cell crypts.
Grape vesicles were taken up by the stem cells that renew the gut lining, helped those cells grow in organoids, and reduced gut injury in mice given a chemical that causes colitis. The vesicle fats alone were enough to reach the same cells.
Monkeys had blood drawn before and 1, 4 and 12 hours after a plant-rich meal. A few faint signals appeared with one test, but a more exact test suggested they were noise, and there was no consistent rise after eating.
The paper that started the whole idea: a rice microRNA was reported in human blood and shown to lower a liver protein that clears LDL in mice. Later labs could not reproduce the absorption step, which is why the next papers exist.
keep reading
more from the library
This page summarizes published research for education. It is not medical advice, and nothing here describes what any nanonourish™ product does. Talk with your care provider before changing your diet or starting a supplement, especially if you are pregnant, nursing or taking medication. Summaries by Dr. Kaylan Jackson, PhD, a chemist, not a physician. Citations checked against PubMed and Crossref, September 2026.