Heritability of Substance Abuse Meets Epigenetics?

March 5, 2014

Virginia Hughes has a nice piece out on generational transmission of……experiences. In this case she focuses on a paper by Dias and Ressler (2014) showing that if you do fear conditioning to a novel odor in mice, the next two generations of offspring of these mice retain sensitivity to that odor.

This led me to mention that there is a story in substance abuse that has been presented at meetings in the past couple of years that is fascinating. Poking around I found out that the group of Yasmin Hurd (this Yasmin Hurd, yes) has a new paper out. I’ve been eagerly awaiting this story, to say the least.

Szutorisz H, Dinieri JA, Sweet E, Egervari G, Michaelides M, Carter JM, Ren Y, Miller ML, Blitzer RD, Hurd YL. Parental THC Exposure Leads to Compulsive Heroin-Seeking and Altered Striatal Synaptic Plasticity in the Subsequent Generation.Neuropsychopharmacology. 2014 Jan 2. doi: 10.1038/npp.2013.352. [Epub ahead of print] [PubMed, Neuropsychopharmacology]

This study was conducted with Long-Evans rats. The first step was to expose both male and female rats, during adolescence, to Δ9tetrahydrocannabinol (THC) at a dose of 1.5 mg/kg, i.p. every third day from Post Natal Day 28-49. No detectable THC was still present in the animals 16 (and 28) days later. The animals were bred at PND 64-68. Parallel Vehicle exposed rats were the comparison.

The resulting pups were fostered out to surrogate mothers in new “litters” consisting of approximately equal male/female pubs and an equal number from the THC-exposed and Vehicle-exposed parents. So this rules out any effects the adolescent THC might have on parenting behavior (that would affect the pups) and mutes any effect of littermates who are offspring of the experimental or control parents.

TransGenerationalTHCheroinThe paper shows a number of phenotypes expressed by the offspring of parents exposed to THC in adolescence. I’ve picked the one that is of greatest interest to me to show. Figure 1d from the paper depicts behavioral data for a heroin intravenous self-administration study conducted when the offspring had reached adulthood. As you can see, under Fixed-Ratio 5 (5 presses per drug infusion) the animals with parents who were exposed to THC pressed more for heroin than did the control group. They were equal in presses directed at the inactive lever and exhibited equal locomotor activity during the self-administration session. This latter shows that the drug-lever pressing was not likely due to a generalized activation or other nonspecific effect.

The paper contains some additional work- electrophysiology showing altered Long Term Depression in the dorsal striatum, differential behavior during heroin withdrawal and alterations in glutamate and dopamine-related gene expression. I’ll let you read the details for yourself.

But the implications here are stunning and much more work needs to be completed post-haste.

We’ve known for some time (centuries?) that substance abuse runs in families. The best studied case is perhaps alcoholism. The heritability of alcoholism has been established using human twin studies, family studies in which degree of relatedness is used and adoption studies. Establishing that alcoholism has a heritable component led to attempts to identify genetic variations that might confer increased risk.

The findings of Szutorisz and colleagues throws a new wrinkle into the usual human study designs. It may be possible to identify another factor- parental drug exposure- which explains additional variability in family outcomes. This would probably help to narrow the focus on the genetic variants that are important and also help to identify epigenetic mechanism that change in response to actual drug use.

On the pre-clinical research side…..wow. Is it via the male or female…or is it both? Does the specific developmental window of exposure (this was adolescent) matter? Does the specific drug matter? Is the downstream effect limited to some substances but not others? Is there a general liability for affective disorder being wrought? Does the effect continue off into subsequent generations? Can it be amped up in magnitude for the F2 generation (and onward) if the F0 and F1 generations are both exposed?

I think if this finding holds up it will help to substantially advance understanding of how An Old Family Tradition can become established. As I posted before:

In his classic song the great philosopher and student of addictive disorders, Hank Williams, Jr., blames a traditional source for increasing the probability of developing substance abuse:

….Hank why do you drink?
(Hank) why do you roll smoke?
Why must you live out the songs you wrote?
Stop and think it over
Try and put yourself in my unique position
If I get stoned and sing all night long
It’s a family tradition!

13 Responses to “Heritability of Substance Abuse Meets Epigenetics?”

  1. mtomasson Says:

    Very intriguing study, no doubt. My skeptic nerves twitch when i don’t see a mechanism (besides, *jazz hands* epigenetics). They look at gene expression in the striatum, cool. But, since they see changes in the expression of relevant genes, it would be a straight shot to nail this completely if they showed epigenitic changes at these loci. Why not show us some methylation changes, or some such? Coming soon? Will reserve judgement until I can get full text..but for now a tentative…<_<


  2. Grumble Says:

    Do the children of people who used pot in their teens and 20s have a greater risk of becoming opiate addicts than the children of people who didn’t use pot?

    Liked by 1 person

  3. Dave Says:

    You can apply these findings to many metabolic diseases as well (obesity, diabetes), and probably conditions like myopia etc. it has been known for a long-time that fetal-programming has a significant role to play in determining ones risk for many of these conditions, and the latest “trend” is that this is due to “epigenetics”. Most simplify this term to mean DNA methylation when, in reality, it should refer to any environment-gene interaction (geneticists like to simplify things that can be measured on chips!). So, we should be focusing on transcription factor activity, chromatin accessibility, methylation and more, and should be looking at these with one eye still on genomic polymorphisms. How do common (or rare) variants influence the epigenetic landscape? This holistic view probably explains why the search for gene mutations underlying things like alcoholism and schizophrenia have mostly failed. It was always going to be much more complicated than that!!!!!


  4. becca Says:

    On the one hand, duh. Everyone who studies epigenetics, even tenuously, could tell you that “(runs in families) + (identical twins > fraternal twins) + (common problem) – (already identified genetic locus clearly responsible) = epigenetic effects in play”. It’s true-by-definition.
    On the other hand, I eagerly await all those follow up studies you mention in the preclinical side. Cool things to unravel for sure! Not sure how helpful it is on the clinical side yet. Teens don’t refrain from doing things they know might set up their future offspring for difficulties in far more problematic ways than this. I mean, how hard is it to bribe an 18 year old military recruits into doing things like cleaning up nuclear waste?


  5. drugmonkey Says:

    “everyone who studies epigenetics” thinks there are druggable targets to manipulate, becca.


  6. becca Says:

    No. Everyone who writes grants on epigenetics pretends there will very soon be drugable targets that actually make sense. You, more than most, should know better not mix up what people *think* with what they write *in a grant*, and we all know the difference between “this is coming right along, technologically speaking” with “we have this in our hot little FDA approved hand”.

    We can apply drugs that impact epigenetic programming, like HDAC inhibitors. Some of them we can demonstrate to have net beneficial impacts, even though we have no idea which of the pleiotrophic impacts are crucial. Heck, a quick pubmed shows they’ve even been used in substance abuse in animal models. But that doesn’t mean we know how they work. An HDAC might be a druggable target, but not as the phrase has traditionally be implied, and not in any sense that relates to “we understand what we’re doing”


  7. drugmonkey Says:

    if you think anything other than a minority of “druggable targets” equal “we understand what we’re doing” you do not understand science.


  8. Dave Says:

    Interesting article in Nature on this topic this week. You neuro guys will like this too:



  9. anonymous postdoc Says:

    All anonymous postdoc is saying is, if something seems too good to be true, consider the possibility that it IS too good to be true. These studies take major time and effort investment, which makes independent replication rarer…and increases the probability of human-driven Type 1 error, to avoid having wasted all that time and effort. I believe in some of these effects and may even be working in this field….but that does make me suspect of some of the higher profile findings. This shit is messy, and pretending it is clean does not serve science.

    We have a lot of neuroscientists wanting to pretend that they are reproductive biologists…but they aren’t, and show no excitement about collaborating with actual reproductive biologists. I would say that the degree to which a person has shown enthusiasm about, or participated in, such a collaboration could be a good indicator of how much of their transgenerational epigenetics story to believe.


  10. becca Says:

    I don’t think “druggable target” needs to mean “we have a perfect understanding of what targeting this will do”, and I realize we rarely understand what we’re doing entirely.

    However, if you think of rational drug design as a process by which we increase the molecular *and* functional specificity of drugs, it’s at least currently impossible to rationally design drugs that target epigenetic processes. You can target the *enzymes* as specifically as you like, but the biological effects will be enormously pleiotropic. Of course, that’s true with plenty of immune system modulators and a zillion other drugs, but it’s not true to the same degree for e.g. proton pump inhibitors as it is for HDAC inhibitors. They are just very different types of processes and we have very different levels of understanding of their functions.

    I think that our inability to convincingly predict how epigenetic drugs will impact biological systems will impact clinical applications. I suspect that the simple fact we understand less of how they are working will (and probably should) tilt things to a higher burden of proof of cost-benefit analysis. That is, I fully expect HDACs to applied to glioblastoma multiforme far sooner than for THC use/abuse. And that has nothing to do with whether the drugs would be more likely to exert beneficial effects on targets relevant to those two conditions.


  11. Curtis Says:

    How does the dose of 1.5 mg/kg compare to the amount of THC normally consumed by humans? For example, does anyone know how much THC is in your “average” joint?


  12. drugmonkey Says:

    I recall that Huestis has reported human’s smoking MJ to hit 60-200 ng/ml peak plasma levels. 1.5 mg/kg IP in a rat is going to come in at the lower end of this range, iirc.


  13. cabrogal Says:

    What’s more, if you examine the songs of Hank Williams, Hank Williams Jr and Hank Williams III you can see evidence right there of amped up substance abuse through to the F2 generation.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: