Neurobiology: what marmosets can teach us

Marmosets are big vocalizers

Marmosets are big vocalizers

C. Heise, Alamy Stock Photo

Marmosets are drawing new attention as models of human disease and social behavior

The Society for Neuroscience annual meeting is underway in San Diego, California with nearly 30,000 attendees and around 13,500 posters. Some of the research being presented relates to this small, long-tailed New World monkey–Callithrix jacchus, native to the dense forests of Eastern Brazil and a promising model organism in neuroscience. Mice continue to be helpful models in neuroscience. They have a prefrontal cortex but, for example, they lack the dorsolateral prefrontal cortex that humans and primates share and that is involved in cognition, learning and memory.

One large-scale project in Japan, Brain Mapping by Integrated Neurotechnologies for Disease Studies (Brain/MINDS), is focused on marmosets. In this month’s Technology Feature, we hear from some Brain/MINDS scientists about their work and their plans. We also talk with scientists elsewhere to find out more about why they believe marmosets are promising models. A number of these researchers head to the wild to study marmosets in their natural habitat.

“Humans talk a lot,” says neuroscientist Cory Miller of the University of California at San Diego who studies social and vocal behavior in primates. “For apes, we talk a preposterous amount, other apes don’t talk as much,” he says. Miller, who has also studied vervet monkeys and macaques as well as simakobu monkeys, cotton-top tamarins, macaques, and capuchins, appreciates how vocal and social marmosets are. This monkey’s vocal and social traits as well as the fact that females usually give birth twice a year, usually to non-identical twins, all likely evolved as an adaptation to life in the dense forest and intense predation. “Everything eats marmosets,” says Miller.

Marmosets are cooperative breeders: both parents take care of the young along with an extended family of relatives and unrelated group members. These monkeys can learn to complete all sorts of cognitive tests. These aspects are but some of this animal’s attributes drawing attention in labs around the world.

Marmosets are not complete strangers to neuroscience labs, says Miller, who points to work by Angela Roberts at the University of Cambridge, Marcello Rosa at Monash University, Xiaoqin Wang at Johns Hopkins University, Miller’s post-doctoral advisor  and many others. This work has spanned a wide breadth: basic neuroscience, motor control, social behavior, neuropsychiatry.

Marmosets are highly social primates

Marmosets are highly social primates

L. Serebrennikov/Alamy Stock Photo

Being social

As a model for human behavior and the human brain, labs draw on primates, for example macaques. But, says a neuroscientist who wishes to remain unnamed,  many universities do not wish to have macaque labs on campus and there are too few positions for postdoctoral fellows leaving labs, ready to  continue primate research. ‘Marmosets will save the primate model in biology’ is a phrase the scientist has heard from a number of neuroscientists.

Overall, the idea, says UCSD’s Miller, is not to repeat with marmosets what has already been done with macaques, but rather to have a complementary primate model that will allow an expansion of the types of questions that can be asked.

A number of marmoset researchers connect work they do in the lab with work in the wild. Miller’s group along with scientists at the Federal University of Pernambuco in Recife, Brazil are equipping nearby wild marmosets with a collar to record vocalization data, ambient sound and information, such as how high the animals sit in the trees when they vocalize.

Atsushi Iriki, who directs the laboratory for symbolic cognitive development at Riken Brain Science Institute and who is also part of Brain/MINDS has also spent time in Brazil observing marmosets in the wild and he noted how different they are from wild Japanese macaques that are more familiar to him. What struck him was how cooperative marmosets are in their group as they spend all day together, as they move around or forage for food. They are always keeping track of one another in the dense bush by calling out to one another and they are always helping one another. They might, for example, share food with one another or help the younger ones jump from one branch to another.

Atsushi Iriki, a RIKEN neuroscientist in Brazil observing marmosets in the wild.

Atsushi Iriki, a RIKEN neuroscientist, in Brazil observing marmosets in the wild.

Iriki has encountered wild macaques in Japan while hiking, biking or skiing. Macaques vocalize but much less than marmosets do and usually only when they are in tense situation such as when they are frightened, trying to scare others or competing with others in some way. “Another difference is that marmosets are female dominant, while macaques are male dominant like most of other primate species,” says Iriki.

Given that it is much harder transport primates than people, Charles Jennings who directs the neurotechnology development program at the McGovern Institute for Brain Research at MIT has proposed the idea of setting up ‘hotel space’ for researchers at major primate research centers so they can work with marmosets on site. There may even be ways to telecommute to work with marmosets remotely.  “It would seem almost necessary for people who want to do work with the actual animals but are at an institute that doesn’t have the resources,” says his colleague Rogier Landman who is in Guoping Feng’s lab at the McGovern Institute.

The idea of research ‘hotel space “might be ideal, especially academically, if it could be realized,” says Iriki. But, he says, this approach also means that agreements about cross-border collaboration, data-sharing, data-security and intellectual property must be put in place.

Transgenic marmosets will help scientists parse molecular and genetic mechanisms in comparative primatology, says Iriki. He studies fundamental neuroscience and the emergence of human traits such as intelligence in primate evolution. He began studying marmosets around ten years ago and continues to also work with macaques and people.

Landman and others who have started working with marmosets are finding that these monkeys are hard to train. New types of cognitive tests are needed, says Landman, ones that are adapted to the needs of these animals. Iriki believes that some of these issues crop up because existing tasks and tests have been refined and adapted to the needs of macaque monkeys. For some tasks, such as vocal communication, social cooperation, altruism and others, Iriki sees advantages with marmosets in that their behavior is more akin to human behavior.

In Iriki’s view, the human brain has evolved in an interaction with an ecosystem and has been shaped by tool-making and the use of language. He believes that higher-order cognition both in humans and primates may derive from information-processing mechanisms related to navigating a spatial environment. Some of these aspects can be studied in the lab. Japanese macaques in the wild don’t use tools but can be trained in the lab to of a rake to extend their reach to grab food. Analysis of voxel-based morphometry of MRI imaging has shown that such tool-use in experimental settings increases cortical gray matter in monkeys.

Iriki cautions against an anthropocentric view of marmoset intelligence. These primates might appear less intelligent than other monkeys in some ways. Then again, he jests, humans might appear to monkeys as animals with intermediate intelligence.

 

 

Glycoscience: a tea party no longer

Later this year or early next Richard Cummings plans to launch The Human Glycome Project. It will happen during a workshop that he is currently organizing and which is open to scientists from near and far. The workshop is slated to be held at the Radcliffe Institute for Advanced Study at Harvard University. Also in the works is a Harvard-based center for glycoscience that reaches out to potential collaborators at all Boston-area universities and academic medical centers.

Cummings, who hails from Alabama and who moved from Emory University School of Medicine to Harvard Medical School last fall, loves glycans, which are the ubiquitous carbohydrates made by all cells, and which can be linked to lipids or proteins. Both in humans and in a variety of animal species, the universe of glycolipids and glycoproteins is extraordinary, he says.

In Cummings’ box of plans is the development a human reference glycome so the growing research community committed to these macromolecules can explore the diversity of the human glycome and develop methods and standards with which to do so. He also envisions comparative glycomics, the comparison of human, porcine and bovine glycomics to tease out differences and similarities. “It wasn’t possible before, really,” he says. But dreaming big in glycoscience is now becoming possible.

Glycobiology has been hampered by complicated methods, which his and other labs have been addressing over the years. In his recent work, published the June issue of Nature Methods, the Cummings lab uses household bleach to release glycans from tissue and cells. He started this research at Emory School of Medicine and continued at his new lab at Harvard Medical School. He also directs the Center for Functional Glycomics, a virtual center that he already led at Emory and that is funded by the National Institutes of Health to explore protein-glycan interactions and to develop new tools and technologies to explore glycoconjugate functions.

When people now stop by the Cummings lab they can, for example, leave with four grams of carbohydrates in a 50ml tube full of white powder. “Those are all the carbohydrate structures in the pig lung,” he says. With this material on hand scientists can use nuclear magnetic resonance techniques for glycan analysis.

Cummings and his team want to enable more labs around the world to study glycoscience by shipping material to colleagues upon request.

Hear Rick Cummings talk about the offer here (14 seconds)



Glycans are difficult to synthesize but now it is possible to harvest them from natural sources such as eggs, meat or plants. “We can make them at such large scale now, we‘re going to just give them away,” he says. Once purified, glycans can be archived, printed on microarrays to explore glycan recognition by lectins, antibodies, bacteria or viruses, or sequenced with mass spectrometry, nuclear magnetic resonance techniques or other methods.

As researchers become aware of the role of carbohydrates in health and disease, the field of glycoscience is broadening, says Cummings. Glycans are being recognized as one of the four major classes of macromolecules, alongside nucleic acids, proteins and lipids.

In the 1970s and 1980s, this field was just getting its start and it was considered merely another part of biochemistry. When carbohydrate researchers got together at meetings, it was more like “tea parties” with 50 to 100 attendees, says Cummings. Glycoscience was far from the spotlight. The community began using the term glycobiology, which Raymond Dwek coined in 1985 and which resonated with researchers. And then, he says,  “all of us kind of chose the term glycomics at some point to distinguish ourselves scientifically from proteomics and the other ‘omics.”

Hear Rick Cummings talk about the history of the field here (40 seconds)


Studying glycan function preceded the study of carbohydrate structure, says Cummings, a situation not unlike molecular biology. For example, work by the chemist Linus Pauling on sickle cell disease occurred before the responsible mutation had been identified and before it was possible to sequence DNA. “We really didn’t know the gene until years later,” says Cummings. The molecular biology arena exploded when it became possible to clone and to synthesize oligonucleotides. “We’re at that point now in glyco-science,” he says.

These days it’s increasingly difficult for scientists to overlook glycans, says Cummings. Access and collaboration are what is needed next to grow the field now that researchers are more than willing to, as he says, “dip their little toes in the glycoscience waters.” That being said, he does still hear disparaging comments about glycoscience, but he takes the remarks as a matter of pride. “So you can think of glycans as being like that little awkward kid on the playground who grew up to be a sizable individual whom no one bullies anymore.”

Sharing data to advance structural biology

In our May editorial, we highlight two new archives: for raw X-ray crystallography (the Structural Biology Data Grid, or SBDG) and for cryo-EM (EMPIAR). These archives join the long-established Biological Magnetic Resonance Data Bank, or BMRB (which hosts biomolecular NMR spectral data) as important resources which will facilitate greater transparency and accelerate progress in structural biology.

Note that neither archive is intended as a “data dump”: datasets in the SBDG must be tied to a journal publication and must be sponsored by the principle investigator of the work, and datasets in EMPIAR must be tied to an Electron Microscopy Data Bank (EMDB) EM density map entry.

Though we do not mandate raw X-ray or cryo-EM data deposition at this time, we applaud these efforts and welcome feedback from the structural biology community about how these archives are bolstering community needs.

An archive for raw EM data

Earlier this week we published a Correspondence describing EMPIAR, a public archive for raw 2D electron microscopy (EM) image data.

While the established Electron Microscopy Data Bank (EMDB) hosts the 3D EM map data required by most journals for publication, the EM community has long been calling for an archive to host the raw 2D image data underlying the 3D maps, as highlighted in our Method of the Year 2015 feature. EMPIAR, a pilot project from the Protein Data Bank in Europe (PDBe), now fills this need.

At Nature Methods we support this archive as a welcome development in the rapidly growing 3D EM field that will enhance transparency, reproducibility, and facilitate the development and refinement of data analysis tools. Though we do not require that our authors deposit their 2D EM image data in EMPIAR, we do encourage it. We urge researchers to make use of the archive and provide feedback to the developers in order to ensure that it is meeting the needs of the field.

Any interested readers without a subscription or site license may read the full text of the Correspondence here.