The psychologist Oliver James has made some waves recently, coinciding with the publication of his new book, with the suggestion that nurture might be 'outdoing' nature when it comes to various concepts from intelligence to mental health. At times the recent 'debates' in this area have not been pretty as arguments about 'what the science actually says' with regards to [structural] genetics vs. environment have tended to get a little heated, and the word 'blame' being banded around quite a lot. This set in the context of views and opinions of certain relevant disciplines.
With all that in mind the paper by Ziegler and colleagues  (open-access) perhaps offers an olive branch between the various camps on whether genes or environment play the more important role in relation to something like mental health. Reporting results on "MAOA [monoamine oxidase A] methylation changes during the course of exposure-based cognitive behavioral therapy (CBT) in PD [panic disorder]" the authors describe how: "In a psychotherapy-epigenetic approach, responders and non-responders to a 6-week standardized CBT as defined by the number of panic attacks showed differential dynamics of MAOA methylation during the course of treatment: response was associated with a significant increase in MAOA methylation up to the level of healthy controls, while non-response rather went along with a further decrease in MAOA methylation."
In a sort of two-stage study, Ziegler et al first compared the methylation status of the MAOA gene in a small-ish sample of female participants diagnosed with PD (n=28) compared with age- and sex-matched 'not PD' controls. Actually, it wasn't just a case of looking at the gene and saying 'yes it's methylated' or not but rather looking at those little islands on a gene where a methyl group can be added and reporting findings (for 13 of them). Authors reported that as a group, the PD participants showed MAOA methylation differences compared to controls - overall hypomethylation of the gene and "at CpGs 3, 6–9, 12 and 13." Hypomethylation normally means something akin to increased gene expression; as the authors note: "As in a functional in vitro assay decreased methylation has been shown to activate MAOA expression... MAOA hypomethylation might result in a decreased availability of monoamines in the synaptic cleft and thereby confer an increased risk for PD." Authors also reported something of an interesting association between baseline PD severity and MAOA methylation levels.
In the second part of the study: "MAOA methylation was furthermore analyzed at baseline (T0) and after a 6-week CBT (T1) in the discovery sample parallelized by a waiting time in healthy controls, as well as in an independent sample of female PD patients (N=20)." The particular type of CBT administered has been detailed in other results  as we are also told that: "Pharmacological treatment remained unmodified during the course of CBT" and "patients were instructed to keep smoking behavior constant during the time course of therapy." Researchers reported that: "overall [in the] patient group irrespective of responder/non-responder status, as well as in the control group, MAOA methylation did not change significantly from T0 to T1 for average methylation or at any individual CpG site." When however the CBT group were sub-categorised as 'responders' or 'non-responders' on the basis of 'the number of panic attacks at T1 compared with T0' there was something a little more interesting to see. So: "responders displayed an increase in average methylation after therapy (mean change±s.e., 3.37±2.17%), while non-responders decreased in average methylation (mean change±s.e., −2.00±1.28%; P=0.001)."
Without getting too carried away with the Ziegler results and accepting that furrowed brows still accompany talk about epigenetics, the potential implications of this study could be pretty huge. I'm not completely enthralled by the 'talking therapies' it has to be said, and the 'bigging up' of the idea that they are some sort of panacea when it comes to mental health and wellbeing as a whole despite their usefulness in certain contexts. Plurality people, plurality. I am however interested when an association is made between their use and gene expression as a function of responder status even if only one gene and one condition has so far been examined with little other genetic or biochemical factors taken into account. My stance on the whole 'genes vs environment' bit too is that it is rather too simplistic to say just 'one or the other' when it comes to complicated things like human behaviour and the vast heterogeneity that underlies it. For some people it could be more of a genetic thing  underlying certain characteristics of a particular condition, for others it might be more non-genetic factors. The N=1 seems to be an important point.
If indeed it does turn out that the talking therapies (among various other 'environmental' factors) can impact on gene expression, there could be a few implications. The rise and rise of something like mindfulness (minus the hype) might also find a similar effect and could perhaps be [scientifically] pitted against CBT and other similar intervention hopefuls both in terms of behavioural outcomes and methylation status. Same goes for other potential 'methylation-modifiers' such as exercise for example in light of some changing attitudes in this area (see here). There is also the prospect that with some further science to do, the types of modification to gene expression could eventually be translated into a more biological intervention. Y'know, as per other discussions about the methionine cycle as a source of those methyl groups (see here) or the various agents that can affect methylation practices. Perhaps even looking at the various medications that psychiatry already has in its arsenal as having an epigenetic effect too?
More investigation is required.
 Ziegler C. et al. MAOA gene hypomethylation in panic disorder-reversibility of an epigenetic risk pattern by psychotherapy. Transl Psychiatry. 2016 Apr 5;6:e773.
 Okbay A. et al. Genetic variants associated with subjective well-being, depressive symptoms, and neuroticism identified through genome-wide analyses. Nature Genetics. 2016. April 18.
Ziegler C, Richter J, Mahr M, Gajewska A, Schiele MA, Gehrmann A, Schmidt B, Lesch KP, Lang T, Helbig-Lang S, Pauli P, Kircher T, Reif A, Rief W, Vossbeck-Elsebusch AN, Arolt V, Wittchen HU, Hamm AO, Deckert J, & Domschke K (2016). MAOA gene hypomethylation in panic disorder-reversibility of an epigenetic risk pattern by psychotherapy. Translational psychiatry, 6 PMID: 27045843