The History of the Decline of Rainfall in the Maya Empire

Posted on by
Reply

Ancient dogmatic texts and Hollywood blockbusters have a lot to answer for. While many of us remain in denial regarding the ongoing ecological catastrophe, prophecies of doomsday remain in vogue. Perhaps this reflects a sub-conscious desire to go out with a bang at the hands of some cosmic threat, rather than depart with a whimper, with no-one but ourselves to blame. As we approach the latest in a long line of unspecified Armageddons (but no doubt involving human sacrifice, and dogs and cats living together), we might take a break from misinterpreting the Maya calendar, and consider what befell the Maya civilization at the end of its Classic period.

Maya civilization dates back to at least 1800 B.C., the beginning of its so-called Preclassic Era. Preclassic archaeological sites are found throughout Guatemala and the Yucatan Peninsula of modern-day Mexico, with others being discovered as far away as Honduras, El Salvador and central Mexico. The Classic era was defined by the development of sophisticated and populous city-states, dating from about 250 A.D. These included the famous city of Tikal, Guatemala, which reached its zenith in the Classic era, serving as an important centre of trade and governance. However, Tikal and many other Maya cities were slowly abandoned between 800 and 1000 A.D., a period known as the Terminal Classic Period, or TCP.

What befell the Maya is one of the great mysteries of archaeology. Proposed causes include the spread of infectious disease, the collapse of trade routes, revolution, or military defeat, but none of these factors alone can satisfactorily explain the Maya collapse. An alternative possibility is that recurrent episodes of drought ultimately led to depopulation and the decline of the cities. There is evidence for multiple droughts in the Yucatan Peninsula during the TCP, but similar droughts were not unheard of throughout the golden age of Maya civilization. Strong evidence supporting the drought hypothesis remained elusive…until recently.

Two researchers at the University of Southampton, Martín Medina-Elizalde and Eelco Rohling, took a closer look at the physical data on the TCP droughts, and used it to quantify annual precipitation (principally rain and drizzle). The data was derived from four sources:

1)      A stalagmite (a cave formation that rises from the floor, due to the dripping of mineralised solutions from the roof) named Chaac, after the Maya rain god.

2)      The shells of aquatic snails in Lake Chichancanab.

3)      The shells of backwater ostracods (a type of crustacean) in Lake Punta Laguna.

4)      The sediment of Lake Chichancanab.

The first three materials all contain calcium carbonate (CaCO3), a molecule which contains three oxygen atoms. Oxygen can exist as one of three stable isotopes, which differ in their number of neutrons. These isotopes are 16O, 17O, and 18O, with eight, nine and ten neutrons respectively, and eight protons in each case. The superscript numbers for each isotope represent the mass of the atom, derived from the number of particles (protons+neutrons) in the atomic nucleus. Rainwater is enriched for 16O, as 18O more readily evaporates from seawater. As a result, the more it rains, the more 16O is present in the sea. Hence, by measuring the ratio of 18O to 16O (or δ18O) in a layer of calcium carbonate, it is possible to determine the amount of rainwater that fell during the layer’s formation. The researchers of the current study examined the δ18O within the stalagmite Chaac , as well as in the snail and crustacean shells. From this, they were able to deduce that the Yucatan Peninsula had been subjected to several long periods of drought.

The fourth record is based on a separate natural phenomenon. Gypsum, a soft mineral, is found in Lake Chichancanab. Evaporation of water in the lake causes the concentration of gypsum in the remaining water to increase. By examining the amount of gypsum present in sediment at the bottom of the lake, it is possible to infer changes in the amount of rainfall.

The authors used the available data to construct a simulation of δ18O fluctuations in Lake Chichancanab. This system was then used to examine changes in δ18O, following perturbations to the simulation. By adjusting the amount of summer rainfall to that predicted by the stalagmite, the model predicted lake δ18O values phenomenally close to those of the shell samples. In addition, the model predicts a 30% reduction in the Lake Chichancanab water levels during droughts in 830 and 928 A.D., in agreement with the high gypsum concentration observed today. Therefore this study has managed to integrate all the available sources of physical data into a single coherent model.

The reduction in rainfall predicted by the model is fairly mild; this suggests that groundwater availability in the Yucatan Peninsula may be particularly sensitive to small changes in rainfall. The periods of drought observed are associated with a decrease in 16O; such decreases today are counter-balanced by heavy rain caused by storms and hurricanes. Hence, a decrease in such storms may have brought about the observed periods of drought.

Classic Maya Civilization did not go out with a bang, and neither will the modern world. What we face may lack the spectacle of planets colliding, but shares the potentially deadly effects. The discussed study not only informs us of the region’s past, but has very real implications for the future – small decreases in the annual rainfall of the Yucatan Peninsula are predicted in the near future, which may again have a dramatic effect on groundwater availability. This could have very serious consequences; the Maya population of the lowlands dropped from four million people in 800 A.D. to a few hundred thousand people a mere 150 years later. To anticipate our future, we need to look to the decline of the Maya, not to their calendar.

Dealing with Anxious Owners

Posted on by
Reply

There is considerable evidence supporting the ability of animals to provide anxious individuals with emotional support, as well as serving as a communicative bridge to people with autism. Pet owners are reported to have lower blood pressure than those without animal companions, and are also less likely to have depression. Less understood, however, is how the personality and well-being of pets may be affected if their owners suffer from anxiety.

In this perspective article, I will address the findings of two recent papers from the lab of Kurt Kotrschal, head of the Konrad Lorenz Forschungsstelle (Research Centre; KLF), University of Vienna. Research at the KLF focuses on understanding the social mechanisms of several species of bird, with particular emphasis on cognition, and the relationship between steroid hormones and behaviour. In recent years, the KLF has also turned its attention to studying social relationships between humans and animals. Ultimately, the team hopes that a greater appreciation of the social dynamics between man and beast may provide insight into the complexities of human relationships.

Both studies assessed how behaviour is influenced by the respective personalities of owner and pet. The earlier study concerned human-dog companionships, while the later study applied similar methods to the study of cat behaviour. In the former study, all participating dogs were unneutered males from single dog households, while the later study included both male and female cats. The behaviour of owner and pet were observed within their home on multiple occasions. This was accompanied by certain challenges for owner and pet, including presenting the dog with an intimidating stranger, or the cat with a soft owl toy.

Owners also filled in questionnaires regarding their attitude towards their pet, and completed a NEO-FFI personality test. This test measures personality in five dimensions – the ‘five factor inventory’ – namely neuroticism, extroversion, openness, agreeableness and conscientiousness. Those high in neuroticism tend to be overwhelmed by frequent negative emotions, and are more emotionally labile. The extroversion scale measures willingness to spend time in the company of others. Individuals who score highly on openness have broad interests, being creative and intellectual, while those with lower scores are more conservative and conventional in thought and deed. Agreeableness represents compassion and empathy towards others. Finally, the conscientiousness scale encompasses traits such as diligence, the ability to plan, and will-power.

Dog and cat personality was also assessed by the researchers, based on each animal’s behaviour. Dog personality was described in terms of sociality, anxiety, aggression and attentiveness. Cat personality, meanwhile, was expressed in terms of three traits: activity level, anxiety, feeding behaviour and play style.

The most apparent finding was that the more neurotic the owner, the more likely they were to see their dogs as givers of social support. In turn, their dogs were friendlier and exhibited less signs of stress, as well as having lower baseline levels of the cortisol, a stress hormone. However, such pairings performed poorly at a joint co-ordination test, reflective of the observation that neurotic owners were less likely to engage in shared activities with the dog, spending more time petting it instead. Cats interacted less with more neurotic owners; the authors argue that neurotic owners seeking social support may approach their cats more, and consequently the cats themselves feel less of a need to engage with their owners.

On the other hand, extroverted owners were inclined to view their dogs as companions with which to share activities. These dogs had relatively higher levels of cortisol, suggesting a more stressful lifestyle. This in itself suggests a need for owners to ensure they make adequate time to pet their dog, as well as play together.

Sex-specific differences were also identified in both studies. Women are more likely than men to speak to their cats, and this may be related to an increased tendency of cats to approach female owners. A similar trend was not reported for dogs; rather, male dogs of male owners were found to be more sociable than those of female owners. This could be interpreted in two different ways – possibly the dogs (all unneutered males) assume a more dominant role with women. Alternatively, the behaviour may arise from differences in the way in which men and women interact with their pets. Furthermore, dogs of male owners showed a greater elevation in cortisol following an encounter with a potential threat. Interestingly, the more time male owners invested in shared activities with their dog, the greater the rise in cortisol observed in the owner following the potential threat. It is not clear whether owners more likely to spend time with their dogs are also more susceptible to sharing their dog’s stress, or whether the act of spending time with the dogs results in greater attachment, and thus more concern for the dog.

The authors note that the owners included in the study were less neurotic than the general population, while being more extroverted and conscientious. When we consider that the participants were all volunteers, recruited either in person or via newspaper ads, this is perhaps not surprising. Individuals with strong neurotic leanings are arguably much less likely to welcome such intrusion into their homes and lives, skewing the personality profiles of the study. Alternatively, it may be that it is dog ownership itself which reduces the neuroticism of owners compared to the general population. It may be possible to address this issue, by conducting a similar study with the co-operation of veterinary practices. Upon bringing their pets to these practices for routine check-ups, dog owners may be asked if they wish to complete a short questionnaire, and permit the taking of saliva samples from the dogs. In this way, survey uptake may be increased, reducing bias inherent in the methods used by the study in question.

Are social relationships between pets and humans akin to those between humans? The results of earlier studies suggest this is so, as evidenced by similarities in the major bonding mechanisms between man and wolf, and the development of mutual attachment. Mammals share a number of ‘social tools’, and cope with stress in similar ways. Strong attachments between dogs and cats are frequently observed, as anecdotally represented by the touching tale of Bobbi and Bob Cat. What can these studies begin to tell us of human interactions? Might having neurotic friends help boost self-esteem and reduce stress, as potentially observed for dogs and their emotionally-needy owners? While directly extrapolating these findings to intra-human relationships is dubious at best, these could form the basis for hypotheses to be tested in future psychological experiments.

Web Design 101

Posted on by
Reply

Spiders produce silk for a variety of purposes, including capturing and immobilising prey, cocooning eggs, building nests, parachuting and as an early-warning tripwire. The silk fibres are phenomenal materials, tougher than Kevlar*, with a tensile strength comparable to steel†. Moreover, spiders spin these silks into complex webs, and the underlying geometry of these structures is important in providing rigidity. However, it is not known whether the properties of silk itself influence web performance. This is not just a matter of passing interest to dedicated arachnophiles; dramatic improvements in human architecture could arise from a greater understanding of the relative contributions of material and structure to web properties.

The laboratory of Professor Markus Buehler, at Massachusetts Institute of Technology, set out to address this question through a combination of computer simulations and experimentation (Nature. 2012 Feb 1; 482(7383):72-6). Virtual webs were constructed, using experimentally-derived properties for the constituent fibres. One critical property is the stress-strain relationship of the fibres: stretching the fibres creates stress (the amount of force per unit area), which causes the silk to elongate (strain). The stress-strain response of silk is non-linear; that is, the strain is not proportional to stress. Although very resistant to small amounts of stress, fibres become softer in response to larger forces. However, as the force increases, the fibres become stiffer again, and show reduced extension. Ultimately, the fibre is unable to bear additional tension, and breaks.

In order to assess whether web properties arise from architecture or material, virtual webs consisting of three different materials were created: 1) spider silk; 2) a perfectly elastic material, in which strain increases proportionally with stress; and 3) a material which is elastic initially, but under greater stress becomes plastic (undergoes non-reversible elongation). The authors also created virtual webs with threads in two different arrangements: 1) spiral (resembling the pattern on a snail’s shell) or 2) radial (with threads arising from a central point, like a ship’s wheel). The durability of these webs was assessed by stretching individual threads. For both spiral and radial webs, damage was localised to the area surrounding the thread, leaving most of the web intact. For radial webs made of spider silk, the initial softening of the material reduced the stress on the entire structure, and ensured that only the targeted thread stiffened as it was stretched further. Eventually this thread alone breaks, sacrificing itself to preserve the integrity of the web. When a perfectly elastic material is substituted for spider silk, threads adjacent to the tugged strand absorb a greater amount of the tension, resulting in greater damage to the web. Webs made of the elastic-plastic material were damaged to an even greater extent, as softening of the threads distributed the tension throughout the structure.

Then what if the web encounters a more global challenge, in the form of a strong breeze? As for the effects on individual threads, radial and spiral webs reacted similarly to gusts of wind. However, while virtual webs made of spider silk caught strong winds like a parachute, perfectly elastic and elastic-plastic webs were not so easily-deflected, and instead were prone to snapping.

This suggests that the characteristic softening/stiffening behaviour of spider silk is critical for preserving integrity of the web. By ensuring that damage is localised to small regions, the spider is spared the metabolic demands of having to rebuild the entire web every time the wind blows. Along with its aforementioned strength and toughness, these properties conspire to make spider silk a truly outstanding example of evolution’s potential. The authors go so far as to suggest that human engineers may also consider the possibility of emulating webs in their designs, by incorporating load-bearing elements that may be sacrificed to save the rest of the structure.

* Silk is able to absorb more energy before fracturing.

† The ability to of high-grade steel and silk fibres to withstand being stretched.

Fomenting Aspiration

Posted on by
Reply

Two fascinating studies were published back-to-back in the latest issue of Science magazine (3 February 2012 Vol. 335, Issue 6068), concerning how women may be encouraged to aspire to high-ranking positions in science and government.

Despite efforts to promote women in the workplace, the upper echelons of many professions continue to be dominated by men. The reasons are manifold, including continued discrimination against women, and the potential difficulties involved in combining career and family. Moreover, it has been reported that, in general, women are less inclined than men to enter into competitive situations. Whereas certain men excel in a competitive environment, many equally-qualified women will avoid risking failure, by opting out of the race. To increase the number of qualified women applying for upper-level positions, a number of different strategies have been adopted by various public bodies.

One approach is to use a quota system, in which a certain number of positions are reserved for women. Ultimately, it is hoped that such interventions will negate the need for quotas in the future, as the presence of women in leadership roles may inspire future generations of women to aspire to the same roles. Believing that one has the ability to perform can become a self-fulfilling prophecy; self-belief is correlated with future aspirations and effort. Alternatively, female leaders may be reasonably expected to implement policies that aid other women in reaching the upper ranks of the relevant profession.

The first of the two studies retrospectively examined the impact of reserving leadership positions for women in selected villages of West Bengal. The 73rd and 74th amendments to the Indian Constitution provide that one-third of randomly-selected village councils be reserved for a female ‘pradhan’, or leader. As such, the research team was able to compare female aspirations between villages that had never had a female pradhan, and villages in which a female pradhan had served one or two terms. By interviewing adolescents and married couples, the researchers were able to gauge both the educational aspirations of teenagers, and the expectations of parents. These data were combined with village-specific data on the mathematical and reading ability of schoolchildren, the occupational destinations of young people, and the amount of time adolescents spent on domestic chores, culminating in a comprehensive overview of how female pradhans influenced village life.

The study found that parents of all villages had higher aspirations for their sons than daughters, but this gender gap was narrowed slightly in villages that have had a female pradhan for two terms. While fathers primarily expressed an increased desire for their daughters to become pradhan themselves, mothers expressed a broader range of educational and occupational aspirations for their female offspring. This finding is consistent with reports that women have more subjective identification with female role models than men, and that men’s perceptions of women’s ability to lead is increased by the presence of a female pradhan. Very similar trends were observed for the aspirations of adolescents themselves; while in general boys were found to be more ambitious than girls, the aspirations of girls rose in reserved villages. More dramatically, the gap in educational attainment between boys and girls was fully closed (and sometimes reversed) in villages that underwent two terms of female leadership. Moreover, adolescent girls spent less time on household work.

As mentioned, female leaders may help women both by acting as a role model, and by introducing policies that facilitate female advancement. While it is impossible to completely tease apart these effects, the authors suggest that the decreased time spent on chores may relate to investments in infrastructure made by female pradhans. The authors also present several lines of evidence, which argue that viewing the pradhan as a role model is likely to underlie changes in the aspirations of parents and adolescents. Firstly, the career prospects of women aged 16-30 are no different between unreserved villages and villages with female pradhans. Secondly, there is no evidence that female leaders introduced educational reforms that would have given girls an advantage. Thirdly, while changes in infrastructure tended to be implemented by female pradhans in the first term, increased aspirations for daughters were not observed until the second. Thus, this change may be related to increased confidence in the ability of female leaders, which is not observed until the second cycle.

The second study addressed how effective quotas and other interventions are at encouraging women to enter competitions, in a controlled experimental setting. Women and men were asked to add as many sets of five two-digit numbers together as possible in three minutes, receiving a financial reward for each correct calculation. As expected, on average men and women were equally as capable at this task. However, if the experiment is altered so that participants now compete against one another (with a large reward for the winners, and nothing for the losers), the performance of women suffers. Furthermore, given the choice, women are half as likely as men to compete.

The authors then investigated how affirmative action policies influenced both the likelihood that women would compete, and the outcome of the contest. For the quota intervention, at least one of the two winners had to be female. For repetition of the competition, the contest was repeated until at least one woman won. Finally, weak or strong variants of preferential treatment were employed; all women received a head start of one or two extra points, respectively. Ultimately, only preferential treatment resulted in a significantly greater number of women opting to compete. Crucially, the decision of men to compete was not affected by any of these women-favouring policies.

Such policy interventions have been criticised as inefficient at selecting the best candidate for the job. However, in the described experimental set-up, the average ability of the winners was not significantly altered by the introduction of affirmative action. This is largely due to very few better-qualifying men being passed over in favour of women, which is encouraging. That said, five men lost out under the quota condition, and one man was a victim of the strong preferential treatment condition.

Finally, as colleagues frequently continue to work together after one has received a promotion while the other has lost out, it is important that policy interventions do not influence post-competition cooperation. This was measured by a simple coordination game; participants are paired, and informed of the other’s gender, and whether they won or lost under policy interventions. Each member of the pair then picks a number from 1 to 7. If both members pick a high number, they both stand to win a large amount of money. If they both go for lower numbers, they each receive a smaller sum. However, if one member picks a high number while the other picks a low number, the latter guarantees her or himself a small sum, while his/her partner wins nothing. As such, there is the potential to cooperate for mutual gains, or to inflict punishment on the other player. Good will won the day – efficiency (defined as a percentage of the highest possible payoff) was consistently close to 75%, with no significant differences between winners and losers, the policy intervention used, or men and women.

Overall, these studies suggest that affirmative action can encourage women to compete, and that women successful under these policies can act as a positive role model to future generations. In addition, this does not affect the willingness of men to compete, or to cooperate with women favoured by affirmative action. There are of course, a few caveats; the findings from the second study arise from a very artificial set-up, which cannot begin to emulate the complexities of the real world. Moreover, there are many work environments which thrive on competition and not cooperation – how would competition-adverse women (or men) handle such situations, and can anything be done to assist?

Could more be done to encourage more well-qualified women to compete with their male peers? It is noted that “in general, men increase their performance in competitive environments more than women”; clearly some equally well-qualified men also miss out on promotion by avoiding competition. Individuals with anxiety disorders also tend to avoid competition for fear of failure, and the greater prevalence of women with anxiety disorders may go some way to explaining the dearth of women in leadership positions. While affirmative action goes some way to helping women acquire these jobs, they may be ineffective in treating the root cause, which is not a trait specific to women.

The primary role of affirmative action is to redress the balance of gender in leadership roles, to more accurately reflect the proportions of women and men in society. Egalitarian hopes are pinned on such measures being temporary in nature, as it is important that the pendulum not be allowed to swing too far in the opposite direction. Further attention needs to be given to treating people as individuals, and helping everyone achieve their potential. Only in this way will we see discrimination, whether negative or positive, become a thing of the past.

Proceedings of the 44th Carnival of Evolution

Posted on by
Reply

The 44th Carnival of Evolution is now up at The Atavism, and I encourage everyone reading this to attend! This month’s edition presents a phenomenal volume of exciting articles, and it’s fantastic to see such a level of support for the Carnival. The Self-Writing Mechanism article, “On the Origin of Anxiety” has been included in the first session, entitled “Symposium on the evolution of novelty”. Self-Writing Mechanism extends its sincerest thanks to the organisers, David Winter of The Atavism and Bjørn Østman of the blogs Pleiotropy and the Carnival of Evolution. If anyone reading this has an article idea for the next carnival, I strongly urge you to submit, and keep the Carnival going!

To Breed for Speed

Posted on by
Reply

In the present genomic era, our ability to glean meaningful information from ancient DNA samples is steadily increasing. Combining recovered genetic data with contemporaneous written records can yield considerable insight into the past, and we are thus beginning to see cross-talk between the academic disciplines of history and the biological sciences. At the end of last year, researchers at McMaster University, Canada, and the University of Tübingen, Germany, reported on the sequencing of Yersinia pestis genomes from Black Death victims. Historical documents indicate that these victims were interred between 1348 and 1350, revealing the age of the ancestral microbes, and facilitating studies into the evolution of its pathogenicity. Now, a paper published in Nature Communications (Bower et al. (2012). Jan. 24; 3:643) describes how Thoroughbred horse genetics have been influenced by trends in British racing.

The Thoroughbred breed originated in England in the late 17th century, in response to growing interest in racing. All living Thoroughbred horses can trace their ancestry to one of three stallions imported from the Middle East; the Byerley Turk, the Darley Arabian and the Godolphin Arabian. These stallions were bred to native British or imported mares, and from 1791, all matings were recorded in the General Stud Book. With breeders selecting for the qualities of a good racehorse, successive generations of Thoroughbreds became larger and faster. However, the early 19th century saw a shift in racing, from bouts of endurance to sprints over shorter distances. What influence did this transition have on Thoroughbred genetics?

The current report, spearheaded by Dr. Emmeline Hill at University College Dublin, investigated the prevalence of two forms (alleles) of a single gene, called myostatin (or MSTN). These alleles differ at a single position, with one containing a cytosine residue (the C-allele), and the second substituting this for thymine (the T-allele). As for humans, horses inherit one set of genes from each parent, and hence can possess two C-alleles (C/C), two T-alleles (T/T) or one of each (C/T). C/C Thoroughbred horses are best suited for sprinting, while their T/T brethren excel in feats of endurance. As may be expected, the C/T heterozygotes are better adapted to middle-distance races. Hill and colleagues were only able to detect the T-variant in donkeys and zebras, suggesting that this allele represents the original MSTN gene. The C-allele is, however, found in horse breeds other than Thoroughbreds, at varying frequencies. This indicates that it did not arise solely within the Thoroughbred lineage. As out-breeding of Thoroughbreds with other horses has not occurred since 1791, the C-mutation must have been introduced early on. From where did it arise?

Comparing the MSTN-flanking DNA sequences of Thoroughbreds and other breeds revealed a high degree of similarity between C-allele-carrying chromosomes of Shetland horses, and those of Thoroughbreds. This suggests that the C-alleles in the Thoroughbred population may have been introduced by British mares.

Following this examination of latter-day horse DNA, the researchers turned to samples of historically-important stallions obtained from museums and private collections. These Thoroughbred stallions, born between 1764 and 1930, all possessed two copies of the T-allele (T/T). Again, this suggests that the C-allele was initially of maternal origin. Finally, by tracing the ancestry of today’s Thoroughbreds, the authors found that a significantly greater proportion of C/C than T/T horses are descended from Nearctic, sire of one of the modern era’s most important studs, Northern Dancer. The sire of Nearctic, Nearco, is considered one of the greatest race horses of the 20th century. Therefore, it seems that while T/T stallions dominated in the past, in recent times C/C Thoroughbreds have superseded their cousins as successful racehorses.

This paper encapsulates how ancient DNA samples can both further our understanding of the past, and identify facets that may be subject to change in the future. Understanding how other genes interact with alleles of MSTN in contributing to speed will be of considerable interest to breeders of Thoroughbreds. The technology has already been commercialized; the company Equinome Ltd., established by Dr. Emmeline Hill, will screen horses for multiple DNA markers associated with performance, to forecast each equine’s racing potential. Arguably as importantly, the ability to rapidly genotype horses should also aid selection for genetically divergent crosses and healthy progeny.

Keystone Symposia: Epigenomics & Chromatin Dynamics Joint Meeting 2012

Posted on by
Reply

In accordance with Keystone Symposia’s Media and Communications Policy, all references to presentations within this post were made with the prior approval of the respective speakers. Self-writing mechanism extends its sincerest thanks to the scientific organizers of the 2012 Keystone Symposia ‘Epigenomics’ and ‘Chromatin Dynamics’ meetings, and those presenters who granted permission for their work to be included here.

If all the DNA in human cells were stretched out as one linear molecule, it would be approximately two metres long. As such, a cell with a diameter 200,000 times as small needs to compact its DNA into a manageable structure. This structure is called chromatin, and it consists of tightly-wrapped units of DNA called nucleosomes. These nucleosomes consist of about 147 base pairs of DNA and several proteins called histones, interlinked by unwrapped DNA. This forms a structure akin to beads (the nucleosomes) on a string (the free DNA). However, the nucleosome performs more than just a structural role, as chemical modifications of the histone proteins (and the DNA itself) can affect the activity, or expression level, of the underlying DNA. In the same way that mutations in the DNA sequence can be inherited, changes to chromatin structure may also be passed on. As changes to chromatin do not affect the DNA sequence, they are considered ‘epigenetic’ – ‘epi’ means above or over, and so epigenetic changes refer to alterations ‘above’ the level of the gene sequence.

The Keystone ski resort in Colorado, USA, recently played host to the joint Epigenomics and Chromatin Dynamics meeting. The dual conference, held between January 17th-22nd 2011, marks one of the first meetings of Keystone Symposia’s 40th anniversary year. Running in parallel, the Epigenomics meeting addressed the application of whole-genome (that is, every gene in the body) techniques to the study of epigenetic inheritance, while the Chromatin Dynamics meeting focused on efforts to understand how chromatin structure affects repair, replication and expression of the DNA template, with a particular emphasis on how disruption of these usual interactions may result in cancer. Professor Richard Young, of the Whitehead Institute for Biomedical Research, USA, began proceedings with his keynote address. Following a brief history of epigenetics, Young discussed some of his recent findings. By creating a map of chemical modifications of histones present throughout the genome of embryonic stem cells, he and his lab were able to assign each gene to a particular ‘state’, based on their specific combination of histone marks. Certain of these modifications activate gene expression (‘transcription’), while others have a silencing effect. Examination of the chromatin state map revealed long stretches of genes with the same type of transcriptional state, which tend to be retained in all cells. Active gene clusters tend to consist of genes required for fundamental cell processes (so called ‘housekeeping’ genes), while stretches of silenced genes are primarily gene families with functions specific to certain specialised cells.

These findings were followed by an account of the role of the LSD1 protein in the change of embryonic stem cells into other cell forms (a process called differentiation). LSD1 removes methyl groups from a particular histone, called H3. If LSD1 is inhibited, differentiated cells retain some of the characteristics of stem cells, as LSD1 is required to repress expression of stem cell-specific genes. In undifferentiated stem cells, the activity of LSD1 is prevented by a second chemical modification of histones, involving the addition of an acetyl group; upon differentiation, these acetyl groups are removed, allowing LSD1 activity.

Finally, Professor Young discussed recent progress in the development of anti-cancer drugs. It has become apparent that in many cancers, a particular gene, called MYC, is over-expressed. The degree of over-expression is correlated with poor clinical outcome, diminished response to chemotherapy, the capacity for cancer proliferation, and the likelihood of relapse. Studies using mice in which MYC is over-expressed have suggested that inactivation of this gene may be effective in halting cancer progression, but unfortunately inhibitors of the c-Myc protein degrade too quickly to be of any use in the clinic. Young discussed a potential alternative, involving inhibition of a second protein, Brd4. Brd4 may help mediate the effect of c-Myc on gene expression, through recruiting components of the necessary transcription machinery. Both c-Myc and Brd4 are highly over-expressed in multiple myeloma, a cancer of white blood cells, and are found together at the regulatory regions of active genes. The molecule JQ1 (thieno-triazolo-1,4-diazepine) binds to Brd4, and inhibits its activity. It was recently reported that JQ1 has a selective effect on the expression of genes regulated by c-Myc in multiple myeloma, as well as reducing MYC mRNA (the message transcribed from the gene) and c-Myc protein. At the whole organism level, survival is prolonged in a mouse model of multiple myeloma. This is great for mice – more importantly for us, JQ1 also reduces the viability of patient-derived multiple myeloma samples, while not affecting the viability of healthy blood cells.

In addition to chemical modifications of histones, the underlying DNA may also be modified. DNA methylation involves the addition of a methyl group to a cytosine residue, one of the bases (A, T, C and G) that confer the coding potential of DNA. Such modification is associated with silencing of the expression of DNA. DNA methylation has a checkered evolutionary past, with modern eukaryotes exhibiting enormous variation in the extent of DNA methylation – certain species use DNA methylation for several purposes, while others appear to have lost this process completely. Professor Daniel Zilberman of the University of California, Berkeley, presented fascinating recent data from his lab, which adds to our understanding of the evolutionary history of DNA methylation. The Dnmt1 family of enzymes (methyltransferases) catalyses the maintenance of DNA methylation in many organisms, but it is notably absent from certain distantly-related algae species. Nevertheless, Zilberman reported the identification of distinct patterns of DNA methylation in these organisms, which may depend on a novel methyltransferase. The patterns are highly regular, with spacing reminiscent of that of nucleosomes. Such investigations into these lesser-studied organisms promise to grant us considerable insight into the history of life on Earth.

Finally, Professor Peter Jones of USC Norris Comprehensive Cancer Center, Los Angeles, discussed a novel technique that can simultaneously map nucleosome positioning and DNA methylation – Nucleosome Occupancy Methylome Sequencing: NOMe-Seq. As nucleosomal DNA is wrapped into a tight configuration, it is inaccessible to proteins. The free DNA between nucleosomes, however, lies naked and exposed. NOMe-Seq exploits such differential accessibility to proteins, by exposing DNA samples to a methyltransferase that acts at GC sequences. This sequence is not methylated in human DNA, and therefore the presence of a methylated cytosine within the GC context indicates non-nucleosomal DNA. On the other hand, methylated cytosines in a CG context are the product of natural DNA methylation. By extracting the DNA and subjecting it to a technique called bisulphite purification, the presence of all methylated cystosines can be inferred, revealing the absence of nucleosomes (methylation at GC) or the presence of DNA methylation (methylation at CG). Might this technique facilitate comprehensive studies into the possible relationship between DNA methylation and nucleosomes hinted at in algae?

In respect to the candid nature of the talks and posters, this has been but a brief introduction to the exciting talks at this meeting. This meeting brought together many excellent scientists in a well-organised and enjoyable setting, concordant with Keystone Symposia’s reputation for quality. If this meeting is any indication, we can look forward to many exciting developments within the fields of epigenetics and chromatin in the coming years.

 

Overcoming Father’s Silence: Angelman Syndrome

Posted on by
Reply

We all have two copies of most genes, inheriting one from our father, and the other from our mother (a notable exception being genes on the male-specific Y chromosome). Usually, both copies of a gene are active in each cell. As such, a deleterious mutation in one gene may not be a serious problem, as its healthy counterpart will compensate for its lack of function. Unfortunately, this is not always the case. For some genes, activation depends on whether they were inherited by the mother or the father. Such parental-specific effects are known as genomic imprinting. In such cases, genetic disease may arise if the normally-active gene is mutated or deleted. The best known examples of such diseases are perhaps Prader-Willi and Angelman syndrome. Prader-Willi syndrome results from the loss of a region on the paternal copy of chromosome 15, and is associated with over-eating, obesity, low muscle tone, infertility and in some cases, learning difficulties. Angelman syndrome, named for its original description by Dr Harry Angelman, was previously known as ‘happy puppet syndrome’. As suggested by its original (now considered pejorative) nominer, Angelman syndrome is associated with frequent laughing or smiling, and jerky movements. More critically, it also presents with seizures, sleep disturbances, and intellectual and developmental delay. Unlike Prader-Willi syndrome, all the symptoms of Angelman syndrome arise from the mutation of a single gene, called UBE3A (encoding ubiquitin protein ligase E3A). In neurons, the paternal copy of the UBE3A is inactive, and thus mutation of the maternal allele eliminates functional ubiquitin protein ligase E3A from the nervous system.

A consortium of three laboratories, all based at the University of North Carolina, set out to identify a potential therapy for Angelman syndrome, through testing the idea that activating the silenced paternal copy of UBE3A in neurons may restore normal neuronal function. Their succinct results, published in the latest issue of Nature (Huang et al. (2012) Jan. 12. 481 p185-189) describe the identification of drugs that can achieve the desired goal. The lab of Benjamin Philpot provided mutant mice, which express a fusion gene of UBE3A and YFP inherited from the father. YFP encodes ‘yellow fluorescent protein’, and therefore if the paternal UBE3A is activated, the resulting protein will be detected by its fluorescence. Neurons were removed from the cortices of these mice, to be cultured on plates. Pharmacologist Bryan Roth then exposed these cultured neurons to thousands of small molecules, and by screening the neurons for fluorescence, his lab was able to determine which drugs could prevent silencing of the paternal gene.

Their screen identified a single compound, called irinotecan. Irinotecan is known to inhibit the function of type I topoisomerases, enzymes that relax the DNA double helix when it becomes over-wound. In a secondary screen focusing on other topoisomerase inhibitors, the authors found that the FDA-approved topotecan was even more effective than irinotecan in activating paternal UBE3A. Importantly, it was also demonstrated that the resulting UBE3A protein was catalytically active in these cultured neurons, suggesting that the paternal gene can fully substitute for the loss of its maternal equivalent.

How might topotecan and other topoisomerase inhibitors activate UBE3A? The paternal gene is normally repressed in neurons through expression of its antisense transcript; that is, a gene that has the opposite orientation to UBE3A. When both of these genes are expressed, the resulting mRNA messages are exactly complementary to each other, and can bind. These double-stranded messages are then degraded by the cell – they essentially cancel each other out. Topotecan increases expression of the UBE3A gene, while simultaneously decreasing expression of its antisense twin.

Finally, the authors extend their findings to live mice. Topotecan was administered directly into the brains of UBE3A-YFP mice, where it activated the paternal gene as it had in cell culture. Topotecan itself was found to disappear shortly after cessation of its delivery, suggesting it is rapidly removed or metabolised by the brain. When the authors introduced topotecan into the spinal cord, they were amazed to find that the number of UBE3A-YFP fluorescent neurons was high, even at 12 weeks following the end of drug treatment. This suggests that possibly even a short treatment with the drug may be enough to permanently activate the usually silenced paternal gene.

While these results are very exciting, the authors themselves have been quick to point out that they are not evidence that topoisomerase inhibitors are suitable for treating Angelman syndrome in humans. That awaits more long-term animal studies, followed by careful and controlled clinical trials. It has also been noted that topotecan reduces expression of certain paternal genes associated with Prader-Willi syndrome. Therefore, there is a risk that long-term treatment with topoisomerase inhibitors may result in symptoms presented by this genetic disorder. That said, topotecan is routinely used to treat cancer patients, and there are no reported cases of complications arising from this therapy. Nevertheless, it would seem prudent to proceed with caution when manipulating the well-conserved patterns of genomic imprinting, that have presumably been retained by evolution for good reason.

The Emerging Brain

Posted on by
Reply

When many of us think of evolution, we tend to envisage selection for and against random mutations within genes. These mutations may alter the proteins encoded by the genes, potentially giving rise to new functions. This is not the whole story, as evolution may also occur through changes in the expression level of existing genes – that is, how much protein they ultimately generate. Such changes may occur through either mutations in the regulatory regions of the gene (so-called cis-factors, as cis is Latin for ‘the same side’), or through alterations in the activity of other molecules that control the gene’s expression levels (trans-factors, from the Latin for ‘the opposite side’, as in Trans-Atlantic). While a cis-mutation usually affects that gene alone, a change in the activity of a trans-factor may affect a host of genes.

Dramatic, human-specific changes in gene expression patterns are believed to partially underlie the evolution of the human brain, but until recently it was not known whether this resulted from cis- or trans-changes. Furthermore, our understanding of how gene expression patterns change throughout development was limited. A study published recently in PLoS Biology (Somel et al. (2011) Dec. 9, 12 pp1-16) attempts to address both of these areas, by comparing the genes expressed in the brains of rhesus macaques, chimpanzees and humans of different ages. First and foremost, all brains used were obtained from subjects that had died suddenly, for reasons unrelated to the study. And as importantly for human donors, informed consent was obtained in writing from either the donor or their next of kin.

Philipp Khaitovich, and his lab at the CAS-MPG Partner Institute for Computational Biology in Shanghai, identified genes that showed species-specific expression patterns, and divided them into three categories based on their developmental profiles. Type I genes were expressed at the same rate throughout life, while type II genes showed changes in expression over time. Type III genes also showed altered expression through development, but unlike type II genes, these changes did not follow the same pattern in each species. The expression patterns of type I and II genes were found to have diverged in a similar way to gene sequences between humans and chimpanzees, thereby reflecting the known divergence from our most recent common ancestor. On the other hand, expression divergence was found to be far greater for human type III genes, suggesting that more changes in developmental expression patterns had occurred in our immediate ancestors, following the split from the chimpanzee branch.

What underlies this dramatic change in developmental expression profiles in the human lineage? The authors next compared changes in gene regulatory regions (cis-factors) to the observed changes in expression patterns. Species-specific changes in type I and II genes correlated with mutations in regulatory regions, as did type III genes of macaque and chimp brains. Conversely, human type III genes showed no such correlation between sequence and expression pattern. Turning to trans-factors, the authors found that these had also diverged dramatically in the human branch, paralleling the changes in developmental expression patterns. Certain trans-factors, called microRNAs (miRNAs), regulate gene expression by binding to the mRNA message, causing its degradation. The target sites for 39 such miRNAs were enriched within human type III genes, and expression of 12 of these miRNAs inversely correlated with their potential targets. This correlation suggests these miRNAs may be down-regulating type III gene expression within the brain. When the authors experimentally introduced certain of these miRNAs into cells, they found more neuron-related genes were down-regulated than in cells treated with unrelated miRNAs, in agreement with the neuronal bias within the type III gene set.

Overall, these results suggest a scenario in which altered expression of certain human miRNAs influenced development of the proto-human brain, through changing the temporal profile of neuronal genes. This work isn’t without its limitations – through comparing genes conserved between humans, chimpanzees and macaques, the study will exclude any genes that have diverged so much as to be unsuitable for comparison. The omitted genes may include those encoding miRNAs, which are known to evolve rapidly; the authors acknowledge that certain species-specific miRNAs could conceivably exert greater effects on development than their more conserved brethren. Similarly, this study is restricted to gene expression following birth; ethical considerations ensure that our understanding of expression during this critical stage will largely await the development of non-invasive technology. Nevertheless, this study represents a substantial step forward in our understanding of the mechanisms that have differentiated man from ape.

Behind Closed Doors

Posted on by
Reply

Much has been said recently about laboratory safety, in the wake of multiple tragic incidents in which young people lost their lives while at work. It is apparent that in many academic settings, lab safety is considered the domain of the principle investigator, and in consequence, the culture of safety frequently ends up paying lip service to regulations set by the university and national agencies. What has been less discussed is the relationship between lab accidents and the accepted culture of long-working hours at the lab bench.

Considerable criticism has also been made of how many big-name scientists expect their underlings to put in long hours undertaking experimental work, running to weekends and evenings. For the majority, such demands are unendurable, as work-life balance is important for physical and mental wellbeing. The argument has been made that to be successful at any career, whether that be lawyer, farmer or hotel proprietor, working at evenings and weekends is to be expected. For bench scientists, and indeed medical doctors, there is one important difference to other professions: a mistake can mean the difference between life and death.

My concern is not with those unfortunates unable to keep up with the sickening pace in many research labs. These individuals will hopefully find the courage to depart, and find fulfilling careers with greater rewards and stability elsewhere, in less of a pressure-cooker environment. Ultimately, the loss of creative and passionate people at the expense of drones will be to the detriment of science, but that is not the topic of this post. The concern is with those either arrogant or ambitious enough to willingly sacrifice their health and well-being for the sake of a shot at success. While they may be doing something they love, surely it is not healthy for a single activity to become an all-consuming passion, at the expense of other interests, relationships and sufficient sleep.

In an environment that potentially contains dangerous machinery, hazardous chemicals and pathogens, it is vital that workers are able to concentrate fully on their experiments. While we will never know whether tiredness played a role in the untimely death of Michele Dufault in a Yale chemistry lab accident, the fact that the incident occurred late at night suggests it may have been a factor. An insightful piece of journalism by Heidi Ledford illustrates the dangers of insufficient downtime – Nestoras Mathioudakis, a research fellow in the laboratory of neurosurgeon Alfredo Quiñones-Hinojosam, found himself unable to focus his right eye, after an existence spent mainly at work forced him to subsist on frozen meals. How can such an environment, with physically impaired and chronically-tired researchers be considered safe? With such long working hours and little sleep, mistakes are inevitable. How to quantify all the little chemical spills and splashes that may have long-term health consequences? And importantly, in a laboratory setting these potential hazards are not restricted to the individual, but can affect the safety of neighbouring scientists. Regrettably, there is a need for greater regulation.

If they wish to be successful, academics are not free to determine the direction of their research – this is ultimately determined by the prejudices and preferences of journal editorial boards and grant review committees. Neither should they be free to violate health and safety regulations. Universities should take a greater interest in not only who is present outside of standard office hours, but why they are present. If we generously suggest that researchers are entitled (expected?) to be present in the lab from 8am until 7pm, no questions asked, it should be acceptable for both the researcher and the relevant supervisor to provide justification for any time spent in the lab outside of that time. Time-course assays are unlikely to always fit within such civilised hours, and at particularly busy times, it may be advisable for the researcher in question to spend some extra hours in the lab. The problems start when these ‘busy times’ become ‘all the time’. Not only does the risk of accidents increase, there is also increased risk of detrimental effects on mental health. I refer the reader to the tragic suicides of three individuals in a single laboratory, that of Elias James Corey (Felix Chau, Fung Lam and Jason Altom), as examples of how pressure to perform can have tragic consequences.

Records need to be kept on all cases of ‘overtime’ and the underlying reasons therefore. Furthermore, the passive collection of such records is insufficient; they need to be actively reviewed periodically by an independent body, with the authority to instigate investigations into labs in which members routinely work long hours without justification, with the goal of shifting lab culture to one in which researchers make efficient use of the time available during the day. Should this not occur, we can expect further fatal accidents and suicides in the future.