Mechanically gated channels help pollen during plant sex

October 22, 2015 § Leave a comment

Pollen, which consists of grains with the plant male gametes tucked inside, dries as it matures to increase its chances of survival. When carried over to neighboring plants, for example by insects or wind, pollen comes into contact with fluid in the female organ of a plant and then swells rapidly.

This swelling process a sensitive procedure; a pollen grain can die if it is not carefully rehydrated. Until now, the mechanism regulating this fluid uptake was unknown. Researchers writing in the journal Science today report that they have discovered an ion channel that helps pollen grains sense and respond to changes in internal water pressure.

Elizabeth Haswell at Washington University in St. Louis has been studying mechanical signals in plants for more than a decade. In the paper just published, she and her colleagues describe the discovery of ion channels on pollen membranes that monitor and respond to osmotic changes.

If the fluid content inside a membrane becomes too great, pores open to allow ions to leave. Water follows, relieving the pressure. The mechanosensitive ion channel, known as MSL8, senses pressure and makes adjustments as necessary. An incorrect amount of this protein deceases the pollen’s ability to fertilize.

By using RNA analysis, Haswell’s team determined that MSL8 transcripts are found in floral tissue but not in leaf or root tissue. They then fluorescently marked the proteins to show that the proteins were present on the plasma membranes of mature pollen grains.

After rehydrating, pollen grows a tube to carry its sperm cell to the eggs. Haswell and colleagues found that pollen without MSL8 germinated more effectively but generated so much pressure that the tube burst, impairing fertilization. Conversely, pollen that overexpressed MSL8 did not generate enough pressure for the pollen tube to break through the cell wall, rendering the pollen infertile.

This delicate osmotic balance demonstrates mechanical signals aiding in the developmental process. Researchers previously established that bacteria use stretch-activated channels to relieve internal pressure in response to environmental stress signals. The findings by Haswell and colleagues now indicate a previously unknown use for mechanically gated ion channels: reproduction. To cope with “the uncertain and potentially severe conditions of [the] pollen journey, pollen has developed some equally severe compensatory mechanisms, including this fascinating desiccation and rehydration process,” Haswell says. Other strategies include multiple nuclei and a tough cell wall.

While the function of MSL8 seems clear, the mechanism by which it operates — directly, by releasing osmolytes, or indirectly, through regulatory pathways — will be a target for further study. Haswell’s team also is interested in several related ion channels and in studying how membranes survive the dehydration/rehydration process.


 This blog post was written by Alexandra Taylor who is a science writing intern at the American Society for Biochemistry and Molecular Biology.

A rare horse disease has similarities with human neurodegenerative disorders

September 29, 2015 § 1 Comment

’Horse with chronic grass sickness, showing marked weight loss and narrow stance. Photo provided by Bruce McGorum.

’Horse with chronic grass sickness, showing marked weight loss and narrow stance. Photo provided by Bruce McGorum.

Each year in the U.K., about 2 percent of horses die from grass sickness. No one knows what causes the disease, but it does occur almost exclusively in grass-fed animals, including ponies and donkeys. A similar disease is thought to afflict dogs, cats, rabbits, hares, llamas, and possibly sheep.

In an attempt to understand what happens at the molecular level of equine grass sickness, researchers recently reported in the journal Molecular & Cellular Proteomics their analysis of tissue samples taken from horses stricken with the disease. They found misfolded and dysregulated proteins in the tissues that resembled those found in human neurodegenerative conditions, such as Alzheimer disease, Parkinson disease and Huntington disease.

Animals with grass sickness usually suffer gut paralysis. The animals roll, sweat, drool and have trouble swallowing. Animals acutely afflicted with the disease usually have to be euthanized.

The disease is known to attack the neurons, but the causative agent is not known. To get a look at what goes on at the molecular level, Thomas Wishart at The Roslin Institute in Scotland, teamed up with Bruce McGorum at the University of Edinburgh’s veterinary school. The investigators applied proteomic techniques to samples taken from horses that came down with grass sickness.

“We do know which tissues are most consistently affected” by the disease, says Wishart. “We considered that a proteomic analysis would provide a snapshot of the molecular processes in play within those samples at that point in time.”

He points out that the work described in the MCP paper “is the first application of modern proteomic tools and in-silico analytical techniques to equine neuronal tissues and to an inherent neurodegenerative disease of large animals that is not a model of human disease.”

The investigators found that the expression levels of 506 proteins were changed in the ganglia taken from horses felled by grass sickness. Moreover, some of the proteins were misfolded, aggregated or in the wrong places. The proteins included amyloid precursor protein, the microtubule associated protein tau and several components of the ubiquitin proteasome system. These proteins have been implicated in human neurodegenerative disorders.

Finding this similarity between human and horse neurodegenerative diseases, says Wishart, suggests the aggregated or misregulated proteins are “more likely to be end-stage regulators or late consequences rather than initiators of the degenerative cascades.”

As equine grass sickness can be hard to diagnose in some horses, a next step for the investigators is to see if they can come up with a noninvasive diagnostic test.


An artificial ribosome that doesn’t need to fall apart to work

July 29, 2015 § 1 Comment

By joining the RNA of the large and small ribosomal subunits into one molecule, researchers engineered a functional, tethered ribosome (Ribo-T). Ribo-T proves that reversible dissociation and association of ribosomal subunits is not required for efficient protein synthesis.  Credit: Erik Carlson

By joining the RNA of the large and small ribosomal subunits into one molecule, researchers engineered a functional, tethered ribosome (Ribo-T). Ribo-T proves that reversible dissociation and association of ribosomal subunits is not required for efficient protein synthesis.
Credit: Erik Carlson

Frustration has its perks. In a paper just out in Nature, researchers describe making an artificial ribosome because they couldn’t get normal ribosomes to do what they wanted. In creating this artificial ribosome, called Ribo-T, the investigators unwittingly turned conventional molecular biology wisdom on its head: Unlike regular ribosomes, Ribo-T doesn’t need to fall apart and come together again to support protein synthesis.


Alexander Mankin from the University of Illinois at Chicago says his group and that of Michael Jewett at the Northwestern University were trying to teach normal ribosomes new tricks, like getting it to translate “difficult-to-make” proteins or to take in unnatural amino acids to make special polymers. “We were frustrated with our inability to test or alter the functions of the ribosome,” says Mankin.


Trying to tweak the existing ribosomal RNA, which does much of the work of protein synthesis in the ribosome, didn’t go anywhere. Changes to it killed the cell.


So Mankin, Jewett and their teams considered making a portion of the ribosome that would be able to guide the ribosome into making the special polymers. But the problem is that the ribosome, made up of two subunits, falls apart and comes together in every cycle of protein synthesis. How would they stop the re-engineered portion of the ribosome from being swapped out by the normal subunit?


That’s when the idea of a tether came in. But “dissociation of ribosomal subunits was believed to be a prerequisite for efficient translation, and it was unclear whether ribosome with the tethered subunits would be functional,” says Mankin. Still, the investigators decided to give it a shot.


After many tries, one design worked: the Ribo-T. Mankin, Jewett and colleagues engineered a ribosomal RNA that combined sequences from the two subunits of the ribosome into a single unit. Short RNA linkers separated the two subunit RNAs in the contiguous stretch of nucleic acid.


And Ribo-T worked even better than anticipated. Not only did Ribo-T make proteins in a test tube, it also made proteins in bacterial cells that lacked naturally occurring ribosomes and keep the cells alive. Mankin still sounds surprised: “We have created probably the first-ever-on-Earth organism which lived with the ribosome where two subunits are combined into a single entity.”


He adds that Ribo-T could pave the way to exploring properties of the ribosome and to make a independent protein-synthesis system in cells that does not interfere with the ribosomes that take care of expression the rest of the cellular proteins. But, for now, the investigators are focusing on what sparked off the whole project in the first place: Getting Ribo-T to carry out the tasks that are difficult for normal ribosomes to do.



Super-fast spins hurt lipoproteins

June 19, 2015 § 1 Comment


Spinning lipoproteins extremely fast in ultracentrifuges may do more harm than good. Image from

Spinning lipoproteins extremely fast in ultracentrifuges may do more harm than good. Image from

Sometimes the end doesn’t justify the means. In a recent paper in the Journal of Lipid Research, investigators describe how spinning high-density lipoproteins fast, a typical way to isolate them quickly, damages them. The finding suggests that the current understanding of the hydrodynamic properties and composition of HDL “is incorrect,” states William Munroe at the University of California, Los Angeles.

HDL, known as the “good cholesterol,” is an important lipoprotein in diagnosing cardiovascular disease. Its abundance in the bloodstream is considered to be a sign of good cardiovascular health because HDL carries away cholesterol.

Ever since the discovery in 1949 that lipoproteins can be separated and isolated in an ultracentrifuge, spinning lipoproteins like HDL at speeds 40,000 rpm or greater has been the norm. Samples often get spun at speeds of 65,000 to 120,000 rpm within 48 hours to hasten the isolation process.

But there have been whispers in the lipid community that the high speeds damage the molecules. So a trio of researchers at UCLA, led by Verne Schumaker, decided to see how speed affects HDL. “The phenomenon of HDL potentially exhibiting sensitivity to the ultracentrifuge speed is sometimes mentioned between lipoprotein researchers,” says Munroe, who is the first author on the paper. “However, there was little in the literature describing this phenomenon.”

In their JLR paper, Munroe, Schumaker and Martin Phillips showed that damage to HDL began as soon as the ultracentrifuge speed hit 30,000 rpm. Using mouse plasma samples, the investigators demonstrated that the damage got worse as the rotor went faster. Proteins, which are integral to the lipoproteins, got ripped out of the protein-lipid complexes, leaving few intact particles. “With enough gravitational force or time, this protein-deficient HDL undergoes further damage to lose lipid,” notes Munroe.

To try to circumvent the damage, the investigators tested out an alternative method for isolating HDL. They poured a potassium bromide density gradient over their sample. Next, they spun the gradient with the sample at a low speed of 15,000 rpm. Admittedly, the isolation took longer at 96 hours, but at least the amount of HDL that rose to the top of gradient was significantly higher than when using the conventional method.

Based on their findings, the investigators now want “to identify HDL-associated proteins that previous identification studies may have missed because certain proteins may have been completely lost from the recovered HDL particle during its isolation by ultracentrifugation,” says Munroe. “This may give insight into additional roles the HDL may participate in besides reverse cholesterol transport.”

An unexpected finding in prehistoric bison bones

May 19, 2015 § Leave a comment

Cranial bone sample from Bison latifrons (13.009a) with inside surface shown. Presence of material that is consistent with meninges connective tissue is shown. Structure that may be the site of a former blood vessel is found in the upper right of the bottom image (boxed area).

Cranial bone sample from Bison latifrons.

On October 14, 2010, construction workers excavating a site for a reservoir dam near Snowmass Village in Colorado stumbled across bones. The bones belonged to a woolly mammoth. More careful digging revealed close to 5,000 bones from different Ice Age animals. Camels, mastodons, and bison were among them. In a recent paper from the journal Molecular & Cellular Proteomics, researchers reported the analysis of proteins found in the bones of an extinct species of giant bison from the site. From their analysis, they described an unexpected feature of ancient collagen.

The bones at the Snowmass Village fossil site (which is also known as the Ziegler reservoir site) were remarkably well-preserved. The high altitude of the site, which was a lake in the Ice Age, kept it at relatively cool temperatures over the past 130,000 to 150,000 years. The cooler temperatures probably contributed to the preservation the buried materials; even some of the ancient plant material buried at the site was still green at the time of the discovery.

Kirk Hansen at the University of Colorado, Denver, heard of the Snowmass Village discovery in 2010 “while listening to public radio on my way into work.”  Hansen is a protein biochemist whose expertise is in the extracellular matrix. He called the Denver Museum of Nature & Science, which was directing the excavation of the bones, to see if he could help with analyzing samples.

Hansen’s laboratory carries out mass spectrometry analyses and he was aware of existing mass spectrometry work on fossilized proteins. Some studies have suggested that red blood cells can be preserved in ancient bones, but the validity of these interpretations have been questioned.  Skeptics also have wondered about inadvertent contamination of ancient samples with modern proteins.

But, Hansen says, “I thought that the methods we were developing to improve characterization of proteins from the extracellular matrix could be used on these well-preserved samples.” Hansen knew he would get good quality samples from the Snowmass Village site when, he says, “one of the scientists described the smell of the bone fossils as ‘very organic. ’”

Mindful of the issue of contamination, Hansen and colleagues were careful with the samples given to them by the museum. The samples were skull bones from an extinct species of giant-horned bison from the Pleistocene era called Bison latifrons. “We took extra precautions by using new chromatography columns and ensuring the samples were placed in only new vials,” he says.

The investigators carried out mass spectrometry on the proteins left in the bison bones. The biggest challenge was in the data analysis. Some of the proteins had degraded, as expected of old proteins, giving a “laddering” effect in the peptides, and numerous peptides were changed by post-translational modifications.

But the investigators sorted through the data and identified extracellular matrix proteins and plasma proteins. Thirty-three of the ancient bison proteins mapped over to modern bovine proteins, showing the evolutionary kinship.

In particular, Hansen and colleagues sequenced in detail the collagen from the bison samples. The extracellular matrix protein, which forms a fibrous ropelike structure, bore modifications seen in other studies of ancient collagen, such as proline hydroxylation.

But one modification was new and unexpected—hydroxylysine glucosylgalactosylation. “This was the first discovery of a preserved glycan, to the best of my knowledge,” says Hansen. “Finding it in a sample that is over 100,000 years old was surprising.”

Bioarcheologist Matthew Collins at the University of York in the U.K., who specializes in studying ancient collagen, is most impressed with the finding of the hydroxylysine glucosylgalactosylated residue. Glycosylation is a key structural feature of collagen, crosslinking chains together to stablilize its ropelike structure. But it was assumed the seemingly labile glycosylated residues would not withstand the test of time.

“You’d imagine, over this period of time, you would have lost the sugars. That’s one of the reasons why we never bothered to look for them: We didn’t expect to find them. This work elegantly shows that I was wrong!” says Collins. “We’re now going back and looking at our samples for glycosylated residues.”

As Hansen and colleagues were working on the bison samples, data came from a young Siberian woolly mammoth called Lyuba. Her proteins bore similar modifications to those of the bison proteins.  “Finding these modifications in modern tissue samples usually requires some form of enrichment,” says Hansen. But, with these two fossils, “the modifications were relatively easy to find.” He says the discoveries suggest that collagen with hydroxylysine glucosylgalactosylation might be enriched over time because it creates a stable complex.

Hansen and his team’s next aim is to study the relationships between collagen modifications and collagen fiber architecture. The ramifications of the work will go beyond the study of ancient proteins. As Hansen explains, “Once we make progress in this area, we will have a better understanding of the microenvironment’s role in tumor progression and the ability to rationally design biomaterials for tissue engineering applications.”

Blood for all: Making universal blood through enzymes

April 23, 2015 § 3 Comments

Making universal blood. Image provided by Stephen Withers and David Kwan.

Making universal blood. Image provided by Stephen Withers and David Kwan.

Universal blood is an appealing notion because it could be transfused into anyone regardless of blood type. Researchers have been kicking around the idea of using enzymes to create universal blood since the early 1980s, “but a major limitation has always been the efficiency of the enzymes,” says Stephen Withers at the University of British Columbia. “Impractically large amounts of enzyme were needed.”

Now in a paper just out in the Journal of the American Chemical Society, Withers, David Kwan, Jay Kizhakkedathu at the Centre for Blood Research and others describe the development of an improved enzyme that takes us a step closer to having universal and others describe the development of an improved enzyme that takes us a step closer to having universal blood.

Blood comes in four major types: A, B, AB and O. The difference between them lies in the sugar structures that festoon the surface of red blood cells. Both blood type A and B have the same core sugar structure as blood type O, but differ in an additional sugar at the tip of the sugar structure. Type A has an N-acetylgalactosamine residue. Type B has a galactose residue. Type AB has a mix of both residues. The moiety carrying the additional residue can be tacked onto the core structure in various ways, giving rise to subtypes of A and B blood.

The additional residue presents trouble during blood transfusions: It can trigger life-threatening immune responses. Type A people can’t take type B blood; type B people can’t take in type A blood. Type AB can take A or B. Only type O blood can be freely shared without the fear of immune responses.

The idea from the 1980s has been to use enzymes to remove the moieties with the terminal N-acetylgalactosamine or the galactose residues to leave the core sugar structure on red blood cells, just like in type O blood. But to date, sugar hydrolases have not been sufficiently efficient to make the idea practical.

So Withers’ group, which had some success in engineering different classes of sugar enzymes, tackled the creation of more efficient sugar hydrolases by directed evolution. In directed evolution, researchers carry out iterative rounds of mutations on a gene to ultimately produce a protein that performs better than the original gene product.

Kwan, Withers and the rest of the team carried out directed evolution on the family 98 glycoside hydrolase from a strain of Streptococcus pneumoniae. Kwan explains that the structure of the enzyme is known, which helped the investigators design their variants.

Kwan adds that the enzyme also is good at cleaving most A and B subtypes with the exception of a few A subtypes. The investigators decided to engineer the enzyme so it had better activity against those A subtypes. By directed evolution, the investigators got an engineered enzyme with a 170-fold greater efficiency than the original enzyme.

However, the engineered enzyme still doesn’t remove every single moiety with the N-acetylgalactosamine. Withers says that the immune system is sensitive enough to small amounts of the moiety to start an immune response. He says, “Before our enzyme can be used clinically, further improvements by directed evolution will be necessary to effect complete removal” of all moieties with the terminal N-acetylgalactosamine residues.

The investigators are now looking to tackle the remaining subtypes of type A blood that the S. pneumoniae hydrolase struggles to cut. Withers says, “Given our success so far, we are optimistic that this will work.”

Why high-fat, low-carb diets help some epileptics

March 19, 2015 § Leave a comment

A series of pie charts depicting the calorific contributions from carbohydrate, protein and fat in four diets: the typical American diet; the Atkins diet during the induction phase; the classic ketogenic diet in a 4:1 ratio of fat to combined protein and carbohydrate (by weight); and the MCT oil ketogenic diet. Image from

A series of pie charts depicting the calorific contributions from carbohydrate, protein and fat in four diets: the typical American diet; the Atkins diet during the induction phase; the classic ketogenic diet in a 4:1 ratio of fat to combined protein and carbohydrate (by weight); and the MCT oil ketogenic diet. Image from

Antiepileptic drugs don’t work for one-third of patients. Instead, those patients usually eat ketogenic diets, high in fat and low in carbohydrates, which somehow stops seizures. In a paper just out in Science, researchers report how ketogenic diets work at the molecular level and say they designed a new drug that mimics the molecular effects of ketogenic diets.

“Current antiepileptic drugs are designed to target molecules that regulate electrical currents in neurons,” says Tsuyoshi Inoue at Okayama University in Japan, who led the work.  He adds that they “focused on antiepileptic actions of ketogenic diets” so they could find other ways to target epilepsy, particularly through metabolism.

Ketogenic diets force the body to rely on ketone bodies instead of glucose as an energy source. Inoue and colleagues explored what happens in single neurons in mouse brain slices when their energy source is switched from glucose to ketone bodies.

“We found that a metabolic pathway, known as astrocyte-neuron lactate shuttle, regulates electrical activities in neurons,” says Inoue.

An enzyme in the pathway is lactate dehydrogenase. When the energy source went from glucose to ketone bodies, the investigators realized that the switch in energy source inhibited the pathway via lactate dehydrogenase and caused the neurons to become hyperpolarized.

When the investigators inhibited lactate dehydrogenase in a mouse model of epilepsy, the animals suffered fewer seizures.

Next, the investigators used an enzymatic assay to see which existing antiepileptic drugs act on lactate dehydrogenase. They found a drug called stiripentol, used to treat a form of epilepsy called Dravet syndrome, inhibits lactate dehydrogenase. The investigators modified the drug’s chemical structure and found an analog that better suppresses seizures than the original.

Inoue says that inhibitors of lactate dehydrogenase can be “a new group of antiepileptic drugs to mimic ketogenic diets.”

Boar semen biomarkers predict litter size

February 27, 2015 § Leave a comment

"Suckling pigs" by GregManninLB - Transferred from [1]: 2008-03-24 20:37 . . GregManninLB . . 894×667 (90721 bytes). Licensed under Public Domain via Wikimedia Commons -

Researchers found biomarkers that can predict pig litter sizes. Photo by GregManninLB – Transferred from [1]: 2008-03-24 20:37. Licensed under Public Domain via Wikimedia Commons

Pig business is big business. In 2012 alone, China, the world’s largest producer of pork, produced 50 million tons of the meat. Getting pigs to reliably produce large litters is of great economic importance in places like China, the United States and the European Union. In a new study just published in Molecular & Cellular Proteomics, researchers describe biomarkers on boar sperm that can predict the animal’s fertility.

Myung-Geol Pang at Chung-Ang University in South Korea led the MCP study. He says one challenge the pork industry faces is the reliance on inefficient methods to evaluate semen quality that could directly affect litter size.” Pang explains that current semen analyses look at some quantitative aspects of the sperm, “but the sensitivity of such analyses remains a subject of debate.”

So Pang and his colleagues wanted to come up with a different way to determine boar semen quality. Pang notes that pigs are a good model system in biomedical research so work in pigs could have applications in human fertility.

In particular, because Pang and colleagues wanted to find protein biomarkers that would predict litter size, the investigators undertook a proteomic analysis of boar semen. They collected semen from 18 stud boars at a Korean pig farm. Using methods like gel electrophoresis and mass spectrometry, Pang and colleagues analyzed all the proteins in the samples. They then matched their data to the number of pups sired by each stud boar.

Based on that information, the investigators grouped biomarkers that predicted whether a boar would sire a large number of pups (about 12) or a small number (10 or fewer). L-amino-acid oxidase, mitochondrial malate dehydrogenase 2 and calmodulin were among the proteins notably expressed in semen that gave large litters. Proteins such as Ras-related protein Rab-2A, spermadhesin AQN-3 and NADH dehydrogenase were abundant in semen that produced small litter sizes. In total, the investigators found 11 protein biomarkers that predicted litter size. “These biomarkers may be particularly important in the animal industry for the prediction and selection of better litter sizes,” says Pang. “Also, we cannot ignore their possible application to humans.”

Pang says, to further improve their analyses, they now need to do large-scale mining of the mRNA markers to make sure that the mRNA expression levels correlate with the expression levels of the protein biomarkers they found. He says, “We believe that this next study will provide valuable biomarkers of male fertility and contraception” that accurately predict fertility.

Disease detection on a dongle

February 4, 2015 § Leave a comment

A smartphone dongle with a disposable microfluidic cassette. [Credit: Tassaneewan Laksanasopin]

A smartphone dongle with a disposable microfluidic cassette.
[Credit: Tassaneewan Laksanasopin]

A handheld device powered by a smartphone can diagnose HIV and syphilis in Rwanda. In a paper just out in Science Translational Medicine, researchers describe the design and deployment of the dongle that carries three immunoassays to test for HIV and syphilis from a fingerprick of blood. It is the first diagnostic device for sexually transmitted diseases that brings together different assays into one field test.

Samuel Sia at Columbia University, who spearheaded the work, says that the World Health Organization recommends that researchers develop new tests for HIV and syphilis because the two diseases carry the highest risks for mother-to-child transmission in pregnant women. “We have verified that with our field surveys of healthcare workers in developing countries,” notes Sia.

Field tests used today to diagnose HIV and syphilis are not always accurate. The tests usually require an ELISA instrument that costs thousands of dollars, and it can take more than 2 hours to generate results (in developing countries, because of infrastructure issues, patients often wait at the healthcare center for the test results to come out).

The dongle developed by Sia and colleagues is estimated to cost $34 to make, and it spits out a result in 15 minutes. The microfluidic device plugs into the audio jack of a smartphone, such as an iPhone or an Android. Sia and colleagues showed that a fourth-generation iPod touch can power the dongle 41 times on a single charge.

The microfluidic device accommodates five different assays embedded in disposable plastic cassettes divided into zones.  “Each zone has a different affinity-capture molecule,” says Sia. The molecules pick up antibodies against HIV and syphilis. The investigators also built in negative and positive controls.

When tested at healthcare centers in Kigali, the assays worked almost as well as the conventional ones. Most of the 96 patients on whom the device was tested said they preferred the dongle over the conventional ELISA tests because it was quicker. Some did say they liked the fact they could be tested for more than one disease in one shot.

Notably, “this trial was the first to have healthcare workers run the tests instead of our research team members,” says Sia. “We were all pleasantly surprised at how well the test performed the first time, but that is not to say there is no room for improvement.”

Besides improving the dongle, Sia says he and colleagues are exploring how make it into a commercial product.

Making the complex less complicated: Scientists create a simple, functional ion transporter

December 18, 2014 § Leave a comment

Researchers report the computational design, structure, and function of a transmembrane antiport protein. (Artwork conceived and produced by Nathan Joh)

Researchers report the computational design, structure, and function of a transmembrane protein. (Artwork conceived and produced by Nathan Joh)

Scientists have designed a stripped-down version of a protein that carries certain ions across a membrane. In doing so, the scientists have shown that the large, complex transporters found in all living things actually operate on simple principles. They’ve also demonstrated that we now know enough about these molecular machines to design them from scratch. The work is described in a paper just out in Science.

Natural transporters “are large and complex machines. And yet, transmembrane transport is an absolutely essential feature of cellular life so it must have evolved very early on,” says one of the scientists involved in the work, Gevorg Grigoryan at Dartmouth College. “How did such complex machines come into being?”

Grigoryan, William Degrado at the University of California, San Francisco, and others designed a bare-bones transporter, one that inserts into the lipid membrane to carry zinc or cobalt ions in one direction and protons in the opposite direction. They named their protein Rocker, after a feature engineered into the molecule that made it rock back and forth between different conformations to enable transport—similar to how natural transporters are thought to work. In Rocker’s case, it oscillated as it bound zinc ions at one end of the molecule and released them at the other

The researchers developed Rocker in two steps. First, they designed the minimalistic protein with the help of computer programs. “Because of the complicated design goals, having to balance membrane insertion, formation of the desired topology, ion binding in the membrane, and specific dynamic features of the molecule, we had to develop a novel computational design approach,” says Grigoryan, who led the computational work.

Next up was making the actual 25-amino acid protein in the lab and getting it to work as designed. The scientists showed that Rocker formed four-helix bundles in membranes, in agreement with the computational model. The amino acids within Rocker had to be precisely positioned so that the protein transported only zinc and cobalt, but not calcium, ions through it.

Every time Rocker transported a zinc or cobalt ion, it pushed three or four hydrogen ions in the opposite direction. The investigators used methods like X-ray crystallography, analytical ultracentrifugation, and nuclear magnetic resonance to study Rocker in action as it transported ions in and out of microscopic sacks made of lipids.

Although tiny in comparison to native transporters, “Rocker is essentially a reductive deconstruction of the transport process,” says Grigoryan.  By designing a minimalistic transporter protein from scratch, Grigoryan says, he and his colleagues showed that “selective transport itself does not necessarily require complex structure and demonstrated a plausible evolutionary mechanism by which transport could have originated.”

Down the road, Rocker can be used as a model system to understand the structural and thermodynamic factors for ion transport. It also can be used as a mold to design other types of transporters.

Andrei Lupas at the Max Planck Institute for Developmental Biology, who wrote the Perspective article accompanying the paper by DeGrado’s and Grigoryan’s teams, used a quote of Richard Feynman to drive home the importance of the work: “What I cannot create, I do not understand.”

Image from

Quote in the upper righthand corner as written by Richard Feynman at Caltech around 1988. Image from

Where Am I?

You are currently browsing the Research findings category at Wild Types.