Anandamide is usually described as a brain molecule. The highest levels measured in mammalian tissue were not found in the brain but in the uterus, in the days around an embryo attaching, and there the level has to come down rather than up. Four years later another group showed that a newborn mouse does not start suckling at all if the same receptor is blocked during the first day of life. Three studies, all in mice, about a system that is in place before anything else has begun.

The level that has to come down

The finding comes from The Hormel Institute at the University of Minnesota and was published in Proceedings of the National Academy of Sciences on 15 April 1997. Schmid, Paria, Krebsbach, Schmid and Dey measured anandamide in the mouse uterus through the early phases of pregnancy.

The levels ranged from 142 to 1345 pmol per micromole of lipid phosphorus, corresponding to 1 to 7 micrograms per gram wet weight. The authors' phrasing is that these are the highest levels yet discovered in a mammalian tissue.

What is genuinely unexpected is the direction. The level is not high when the uterus is at its most receptive. It is high when it is at its least:

  • Down-regulation of anandamide is associated with receptivity to implantation.
  • Up-regulation is associated with the uterus being refractory, that is, not accepting.
  • Levels are highest in the non-receptive phase and at the sites between implantation sites, and lowest at the implantation site itself.

The authors suggest that the low level right there may be a way for the implanting embryo to protect itself from the detrimental effects of the ligand. They also tested it directly: anandamide reduced blastocyst hatching from the zona pellucida in vitro, and the synthetic cannabinoid agonist CP 55,940 inhibited implantation when given systemically. Both effects were reversed by the CB1 antagonist SR141716A, so they run through the receptor.

Their own conclusion is phrased as a hypothesis, not a finding: aberrant anandamide synthesis or aberrant receptor expression in the uterus or the embryo could underlie early pregnancy failure or female infertility. That step remains unanswered.

The first meal

If the system is in place before implantation, the next question is what it does afterwards. The answer came in 2001 from the Hebrew University of Jerusalem, with Raphael Mechoulam among the authors, in the European Journal of Pharmacology.

Fride and colleagues gave the CB1 antagonist SR141716A to newborn mouse pups, either as a single dose on the first day of life or daily for a week from day two. They describe the effect on milk ingestion and growth as devastating themselves. The first 24 hours turned out to be the critical ones. Death followed within four to eight days. Co-administration of THC almost entirely reversed the antagonist's effect in the week-long regimen.

Two follow-ups have narrowed the finding:

2003 the same group showed in the same journal that the window is precisely that first day. Injection into pups two or five days old had a much smaller effect or none at all. Mice without the CB1 receptor did not take milk on their first day of life, exactly like treated normal pups, but began showing milkbands from day two. That the knockout animals got going while the treated ones did not led the authors to propose a further receptor, which they called CB3. That proposal is their own and has not been established since.

2007 they gave a neutral CB1 antagonist instead, VCHSR1, to one-day-old pups, in Pediatric Research. It produced the same dose-related effect on weight gain, milkbands, body temperature and survival. The difference is technical but decisive: a neutral antagonist only blocks the receptor, whereas an inverse agonist also lowers its resting activity. That both produce the same effect argues that what is being knocked out is an endogenous tone, not a quirk of the first compound.

And the reward in being near someone

The third study moves the question from the body to the company it keeps. Wei and colleagues at the University of California, Irvine published a study in PNAS in 2015 of the mouse nucleus accumbens, the part of the brain that handles motivated behaviour.

Social contact increased anandamide mobilisation there. Isolation decreased it. Using pharmacological and genetic experiments they then showed that anandamide mobilisation and the consequent CB1 activation are both necessary and sufficient for social contact to be rewarding, measured by socially conditioned place preference.

Then oxytocin was tied in. Oxytocin drives anandamide mobilisation in the nucleus accumbens. Block the oxytocin receptors and the response fails to appear. Activate the oxytocin neurons in the paraventricular nucleus of the hypothalamus chemogenetically and it is stimulated. And interrupt the degradation of anandamide, genetically or pharmacologically, and it offsets the effect of oxytocin receptor blockade.

Oxytocin is often called the love hormone. This study describes the link after it: the signal comes from the oxytocin, but what makes it rewarding sits further along the chain.

One of the authors is an inventor on patents protecting one of the compounds used in the study. It appears in the study's own conflict of interest statement and belongs in any account of it.

Where it runs out

All of the above is mice. That goes for the implantation, for the suckling and for the reward. None of the studies measured humans, and there is no automatic transfer.

The hypotheses the authors formulate themselves also remain hypotheses. The link to early pregnancy failure in humans is a proposal from 1997. CB3 is a proposal from 2003 that was never established. That deficits in the oxytocin and anandamide chain contribute to social impairment in autism is a proposal from 2015. None of them is settled.

What can be said is that the system usually described as an afterthought from cannabis research sits in the middle of three of the most basic events in a mammal's life: attaching, eating for the first time and seeking company. That is an unusual position for a system first described as late as the 1990s.

Read on

What chronic stress does to the endocannabinoid system, on what happens to the same system when the load becomes prolonged.

This article describes published research on the body's own endocannabinoid system. It is not medical advice and contains no claims about health effects of cannabidiol or any other product.