A one-sided review of stress

The subject of this 2016 Italian/New York review was the stress response:

“The stress response, involving the activation of the hypothalamic-pituitary-adrenocortical axis and the consequent release of corticosteroid hormones, is indeed aimed at promoting metabolic, functional, and behavioral adaptations. However, behavioral stress is also associated with fast and long-lasting neurochemical, structural, and behavioral changes, leading to long-term remodeling of glutamate transmission, and increased susceptibility to neuropsychiatric disorders. Of note, early-life events, both in utero and during the early postnatal life, trigger reprogramming of the stress response, which is often associated with loss of stress resilience and ensuing neurobehavioral (mal)adaptations.”


A key point in the review was the intentional dismissal of the role of GABA in favor of the role of glutamate:

“The changes in neuronal excitability and synaptic plasticity induced by stress are the result of an imbalance of excitatory (glutamatergic) and inhibitory (GABAergic) transmission, leading to long-lasting (mal)adaptive functional modifications. Although both glutamate and GABA transmission are critically associated with stress-induced alteration of neuronal excitability, the present review will focus on the modulation of glutamate release and transmission induced by stress and glucocorticoids.”

No particular reason was given for this bias. I inferred from the review’s final sentence that the review’s sponsors and funding prompted this decision:

“In-depth studies of changes in glutamate transmission and dendrite remodeling induced by stress in early and late life will help to elucidate the biological underpinnings of the (mal)adaptive strategies the brain adopts to cope with environmental challenges in one’s life.”

The bias led to ignoring evidence for areas the reviewers posed as needing further research. An example of relevant research the reviewers failed to consider was the 2015 Northwestern University study I curated in A study that provided evidence for basic principles of Primal Therapy that found:

“In response to traumatic stress, some individuals, instead of activating the glutamate system to store memories, activate the extra-synaptic GABA system and form inaccessible traumatic memories.”

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4812483/ “Stress Response and Perinatal Reprogramming: Unraveling (Mal)adaptive Strategies”

A skin study that could have benefited from preregistration

This 2016 German human skin study found:

“An age-related erosion of DNA methylation patterns that is characterized by a reduced dynamic range and increased heterogeneity of global methylation patterns. These changes in methylation variability were accompanied by a reduced connectivity of transcriptional networks.”


The study could have benefited from preregistration using an approach such as Registered Reports. As it was, the study gave the impression of a fishing expedition.

For example, the initial subjects were 24 women ages 18-27 and 24 women ages 61-78. The barbell shape of the subjects’ age distribution wouldn’t make sense if the researchers knew they were going to later use the epigenetic clock method. The researchers did so, although the method’s study noted “The standard deviation of age has a strong relationship with age correlation” and provided further details in “The age correlation in a data set is determined by the standard deviation of age” section.

The researchers recruited a second group of subjects, 60 women aged 20-79, “that also included intermediate ages.” No discrete numbers were provided, but from eyeballing Figure S1 in the supplementary material, the ages of the second group appeared to be evenly distributed.

The subject groups were lumped together to make findings such as:

“We observed a significant age-related hypermethylation of CpG island-associated probes. This effect was strongly enriched during two specific age windows, at 40–45 and 50–55 years. Considering that our samples were exclusively derived from female volunteers, it seems reasonable to link the latter window to menopause, which is also known to distinctly accelerate skin aging.”

The study didn’t state that the second group of subjects were screened for either menopause or for use of hormone therapies, such as skin creams that sell in the US for $.42 a day. If the ages of the second group of subjects were evenly distributed, 6 of the 108 subjects would be ages 50-55. It wasn’t “reasonable to link” a small number of subjects to conditions for which they hadn’t been screened.

http://onlinelibrary.wiley.com/enhanced/doi/10.1111/acel.12470/ “Reduced DNA methylation patterning and transcriptional connectivity define human skin aging”

Oxytocin research null findings come out of the file drawer

In 2016 Belgian researchers released their previously unpublished studies:

“Is there a file drawer problem in intranasal oxytocin research?

We submitted several studies yielding null-findings to different journals but they were rejected time and time again.

The aggregated effect size was not reliably different from zero [including all of the researchers’ previously unpublished intranasal oxytocin studies].”

Neuroskeptic comments:

“By publishing these results, Lane et al. have ensured that future meta-analysts will be able to include the full dataset in their calculations.”

http://blogs.discovermagazine.com/neuroskeptic/2016/03/17/open-the-file-drawer/ “Psychologists Throw Open the File Drawer”

See Testing the null hypothesis of oxytocin’s effects in humans for more on the topic.

 

Problematic research into epigenetic effects of paternal stress on male offspring

This 2016 Chinese rodent study and its accompanying commentary Don’t stress dad — it’s bad for your kids’ health were caught up in an agenda.

The first problem I noticed was that the hyperglycemic effects found only in the male offspring weren’t consistently labelled as sex-specific. Try to find that fact in the paywalled commentary with its intentionally misleading headline, or in the news coverage with headlines such as “Stressed mouse dads give their offspring high blood sugar.”

That the effects were male-only was briefly noted in the study, yet “male” was absent from the “stress-F1 mice” label used after the initial mention. The male and female symbols in the diagrams were likewise applied to the parents but not the offspring in the study, and its misleading graphic was subsequently used by the news coverage.

The researchers provided no mechanisms that plausibly linked the effects to offspring sex. There was plenty of time between the May 3, 2015 submission and the February 18, 2016 publication to clarify this and other items. I wonder what the reviewer noted.

The second problem was that the highest number of male “stress-F1 mice” tested was 6. I didn’t see any disclosures of what led to the scarcity of subjects, or of the likely impact of using so few.

A related limitation was that the male “stress-F1 mice” were killed as young adults. Whether or not the hyperglycemic effects carried through to old age or to another generation wasn’t determined.


I’m leery of studies like this one that didn’t have a Limitations section, and especially so when the news coverage overlooked obvious limitations. It was difficult to place the findings in a context other than promoting that a male’s stress may also adversely affect their offspring.

One of the problems that research caught up in an agenda create is that non-headline findings are overlooked. Other than sex-specific effects, the study found that the putative preconception cause of hyperglycemia didn’t cause other symptoms:

  • “No significant growth defects were observed in male offspring from stress-F0 fathers (stress-F1 mice) during their early lives.
  • Insulin sensitivity was not changed in stress-F1 mice.
  • Serum glucagon, leptin, and pro-inflammatory cytokines (tumor necrosis factor α [TNFα], interleukin-6 [IL-6]) were unaffected.
  • Body weight, food intake, locomotor activity, CO2 production, O2 consumption, and respiratory exchange ratios also remained unchanged.
  • Liver weight, liver weight/body weight ratios, hepatic triglyceride content, and the histological phenotypes were also comparable.
  • The methylation pattern and expression of microRNAs were not affected in the fetal brains of stress-F1 mice.”

The handling of the study reminded me of Transgenerational epigenetic programming with stress and microRNA where most of the news coverage similarly focused on it being a male’s stress, not a female’s, that affected the developing embryo. The important part lost from news coverage of that study was it demonstrated how a damaging influence can begin immediately after conception, but the symptoms didn’t present until adulthood.

http://www.sciencedirect.com/science/article/pii/S1550413116300067 “Paternal Psychological Stress Reprograms Hepatic Gluconeogenesis in Offspring”

The current paradigm of child abuse limits pre-childhood causal research

As an adult, what would be your primary concern if you suspected that your early life had something to do with current problems? Would you be interested in effective treatments of causes of your symptoms?

Such information wasn’t available in this 2016 Miami review of the effects of child abuse. The review laid out the current paradigm mentioned in Grokking an Adverse Childhood Experiences (ACE) score, one that limits research into pre-childhood causes for later-life symptoms.


The review’s goal was to describe:

“How numerous clinical and basic studies have contributed to establish the now widely accepted idea that adverse early life experiences can elicit profound effects on the development and function of the nervous system.”

The hidden assumption of almost all of the cited references was that these distant causes can no longer be addressed. Aren’t such assumptions testable here in 2016?

As an example, the Discussion section posed the top nine “most pressing unanswered questions related to the neurobiological effects of early life trauma.” In line with the current paradigm, the reviewer assigned “Are the biological consequences of ELS [early life stress] reversible?” into the sixth position.

If the current paradigm encouraged research into treatment of causes, there would probably already be plenty of evidence to demonstrate that directly reducing the source of the damage would also reverse the damaging effects. There would have been enough studies done so that the generalized question of reversibility wouldn’t be asked.

Aren’t people interested in human treatments of originating causes so that their various symptoms don’t keep bubbling up? Why wouldn’t research paradigms be aligned accordingly?


The review also demonstrated how the current paradigm of child abuse misrepresents items like telomere length and oxytocin. Researchers on the bandwagon tend to forget about the principle Einstein expressed as:

“No amount of experimentation can ever prove me right; a single experiment can prove me wrong.”

That single experiment for telomere length arrived in 2016 with Using an epigenetic clock to distinguish cellular aging from senescence. The seven references the review cited for telomere length that had “is associated with” or “is linked to” child abuse findings should now be viewed in a different light.

The same light shone on oxytocin with Testing the null hypothesis of oxytocin’s effects in humans and Oxytocin research null findings come out of the file drawer. See their references, and decide for yourself whether or not:

“Claimed research findings may often be simply accurate measures of the prevailing bias.”

http://www.cell.com/neuron/fulltext/S0896-6273%2816%2900020-9 “Paradise Lost: The Neurobiological and Clinical Consequences of Child Abuse and Neglect”

A review that inadvertently showed how memory paradigms prevented relevant research

This 2016 Swiss review of enduring memories demonstrated what happens when scientists’ reputations and paychecks interfered with them recognizing new research and evidence in their area but outside their paradigm: “A framework containing the basic assumptions, ways of thinking, and methodology that are commonly accepted by members of a scientific community.”

1. Most of the cited references were from decades ago that established these paradigms of enduring memories. Fine, but the research these paradigms precluded was also significant.

2. All of the newer references were continuations of established paradigms. For example, a 2014 study led by one of the reviewers found:

“Successful reconsolidation-updating paradigms for recent memories fail to attenuate remote (i.e., month-old) ones.

Recalling remote memories fails to induce histone acetylation-mediated plasticity.”

The researchers elected to pursue a workaround of the memory reconsolidation paradigm when the need for a new paradigm of enduring memories confronted them directly.

3. None of the reviewers’ calls for further investigations challenged existing paradigms. For example, when the reviewers suggested research into epigenetic regulation of enduring memories, they somehow found it best to return to 1984, a time when dedicated epigenetics research had barely begun:

“Whether memories might indeed be ‘coded in particular stretches of chromosomal DNA’ as originally proposed by Crick [in 1984] and if so what the enzymatic machinery behind such changes might be remain unclear. In this regard, cell population-specific studies are highly warranted.”


As an example of relevant research the review failed to consider, the 2015 Northwestern University study I curated in A study that provided evidence for basic principles of Primal Therapy went outside existing paradigms to research state-dependent memories:

“If a traumatic event occurs when these extra-synaptic GABA receptors are activated, the memory of this event cannot be accessed unless these receptors are activated once again.

It’s an entirely different system even at the genetic and molecular level than the one that encodes normal memories.”

What impressed me about the study was the obvious nature of its straightforward experimental methods. Why hadn’t other researchers used the same methods decades ago? Doing so could have resulted in dozens of informative follow-on study variations by now, which is my point in item 1 above.

The 2015 French What can cause memories that are accessible only when returning to the original brain state? was another relevant but ignored study that supported state-dependent memories:

“Posttraining/postreactivation treatments induce an internal state, which becomes encoded with the memory, and should be present at the time of testing to ensure a successful retrieval.”


The review also showed the extent to which historical memory paradigms depended on the subjects’ emotional memories. When it comes to human studies, though, designs almost always avoid studying emotional memories.

It’s clearly past time to Advance science by including emotion in research.

http://www.hindawi.com/journals/np/2016/3425908/ “Structural, Synaptic, and Epigenetic Dynamics of Enduring Memories”

The link between scientific value and content is broken at PNAS.org

Should we expect content posted on the Proceedings of the National Academy of Sciences of the United States of America to have scientific value?

This 2016 Singapore study was a “PNAS Direct Submission” that claimed:

“This paper makes a singular contribution to understanding the association between biological aging indexed by leukocyte telomeres length (LTL) and delay discounting measured in an incentivized behavioral economic task.

LTL is an emerging marker of aging at the cellular level, but little is known regarding its link with poor decision making that often entails being overly impatient.”


Whether measured at the level of a human or of a blood cell, in 2016 there wasn’t incontrovertible evidence to support:

  • “Biological aging indexed by leukocyte telomeres length
  • LTL is an emerging marker of aging at the cellular level”

Using an epigenetic clock to distinguish cellular aging from senescence found:

Cellular ageing is distinct from cellular senescence and independent of DNA damage response and telomere length.”

If that study was too recent, the researchers and reviewer knew or should have known of studies such as this 2009 study that found the correlation between a person’s chronological age and blood cell telomere length was r = −0.51 in women and r = −0.55 in men.

A study of biological aging in young adults with limited findings was cited for evidence that “the seeds of biological aging are widely thought to be planted early in life.” That study didn’t elucidate the point, however, as it didn’t fully link its measurements when the subjects were 38 years old with measurements taken during the subjects’ early lives.

Problematic research with telomere length was cited for evidence that “other factors, such as the early family environment, lifestyle, and stress, also have considerable impact on cellular aging.” The researchers had to be willing to overlook that study’s multiple questionable practices in order to cite it as evidence for anything.

Nevertheless, the study used a one-to-one correspondence of telomere length and cellular aging.


The researchers speciously modeled a relationship between telomere length and the behavioral trait “poor decision making that often entails being overly impatient.” That overreach was further stretched to the breaking point:

“We then asked if genes possibly modulate the effect of impatient behavior on LTL.

The oxytocin receptor gene (OXTR) polymorphism rs53576, which has figured prominently in investigations of social cognition and psychological resources, and the estrogen receptor β gene (ESR2) polymorphism rs2978381, one of two gonadal sex hormone genes, significantly mitigate the negative effect of impatience on cellular aging in females.”


The “significantly mitigate” finding was “fun with numbers” that produced false effects rather than solid evidence. Consider that:

  1. The study’s model disregarded the probability that “Cellular ageing is independent of telomere length.”
  2. The researchers provided no mechanisms that plausibly linked performance “in an incentivized behavioral economic task” with telomere length.
  3. The researchers didn’t characterize any causal mechanisms whereby two gene variants affected the task performance’s purported effect on telomere length.

What’s the real reason the reviewer forwarded this paper to PNAS.org?

http://www.pnas.org/content/113/10/2780.full “Delay discounting, genetic sensitivity, and leukocyte telomere length”