Advanced paper could be foundation for inexpensive biomedical and diagnostic devices

Paper is known for its ability to absorb liquids, making it ideal for products such as paper towels. But by modifying the underlying network of cellulose fibres, etching off surface ‘fluff’ and applying a thin chemical coating, researchers have created a new type of paper that repels a wide variety of liquids – including water and oil.
The paper takes advantage of the so-called ‘lotus effect’ – used by leaves of the lotus plant – to repel liquids through the creation of surface patterns at two different size scales and the application of a chemical coating. The material, developed at the Georgia Institute of Technology, uses nanometer- and micron-scale structures, plus a surface fluorocarbon, to turn old-fashioned paper into an advanced material.
The modified paper could be used as the foundation for a new generation of inexpensive biomedical diagnostics in which liquid samples would flow along patterns printed on the paper using special hydrophobic ink and an ordinary desktop printer. This paper could also provide an improved packaging material that would be less expensive than other oil- and water-repelling materials, while being both recyclable and sustainable.
‘Paper is a very heterogeneous material composed of fibres with different sizes, different lengths and a non-circular cross-section,’ said Dennis Hess, a professor in the Georgia Tech School of Chemical and Biomolecular Engineering. ‘We believe this is the first time that a superamphiphobic surface – one that repels all fluids – has been created on a flexible, traditional and heterogeneous material like paper.’
The new paper, which is both superhydrophobic (water-repelling) and super oleophobic (oil-repelling), can be made from standard softwood and hardwood fibres using a modified paper process. In addition to Hess, the research team included Lester Li, a graduate research assistant, and Victor Breedveld, an associate professor in the School of Chemical and Biomolecular Engineering
Producing the new paper begins with breaking up cellulose fibres into smaller structures using a mechanical grinding process. As in traditional paper processing, the fibres are then pressed in the presence of water – but then the water is removed and additional processing is done with the chemical butanol. Use of butanol inhibits the hydrogen bonding that normally takes place between cellulose fibres, allowing better control of their spacing.
‘The desirable properties we are seeking are mainly controlled by the geometry of the fibres,’ Hess explained.
The second step involves using an oxygen plasma etching process – a technique commonly used in the microelectronics industry – to remove the layer of amorphous ‘fluffy’ cellulose surface material, exposing the crystalline cellulose nanofibrils. The process thereby uncovers smaller cellulose structures and provides a second level of ‘roughness’ with the proper geometry needed to repel liquids.
Finally, a thin coating of a fluoropolymer is applied over the network of cellulose fibers. In testing, the paper was able to repel water, motor oil, ethylene glycol and n-hexadecane solvent.
The researchers have printed patterns onto their paper using a hydrophobic ink and a desktop printer. Droplets applied to the pattern remain on the ink pattern, repelled by the adjacent superamphiphobic surface.
That capability could facilitate development of inexpensive biomedical diagnostic tests in which a droplet containing antigens could be rolled along a printed surface where it would encounter diagnostic chemicals. If appropriate reagents are used, the specific colour or colour intensity of the patterns could indicate the presence of a disease. Because the droplets adhere tightly to the printed lines or dots, the samples can be sent to a laboratory for additional testing.
‘We have shown that we can do the operations necessary for a microfluidic device,’ Hess said. ‘We can move the droplet along a pattern, split the droplet and transfer the droplet from one piece of paper to another. We can do all of these operations on a two-dimensional surface.’
For Hess, Li and Breedveld, creating a superhydrophobic suface was relatively straightforward because water has a high surface tension. For oils, which have a low surface tension, the key to creating the repellent surface is to create re-entrant – or undercut – angles between the droplets and the surface.
Previous examples of superamphiphobic surfaces have been made on rigid surfaces through lithographic techniques. Such processes tend to produce fragile surfaces that are prone to damage, Hess said.
The principal challenge has been to create high-performance in a material that is anything but geometrically regular and consistent.
‘Working with heterogeneous materials is fascinating, but it’s very difficult not just to control them, because there is no inherent consistent structure, but also to change the processing conditions so you can get something that, on average, is what you need,’ he said. ‘It’s been a real learning experience for us.’
The new paper has so far been made in samples about four inches on a side, but Hess sees no reason why the process couldn’t be scaled up. Though long-term testing of the new paper hasn’t been done, Hess is encouraged by what he’s seen so far. Georgia Institute of Technology

NTRK1: a new oncogene and target in lung cancer

To the list of oncogenic drivers of lung cancer that includes ALK, EGFR, ROS1 and RET, results of a University of Colorado Cancer Center study presented at ASCO 2013 show that mutations in the gene NTRK1 cause a subset of lung cancers.

‘We’re reconceptualising lung cancer as many, related diseases. And we need to learn to identify and treat each individually. We can treat the forms of the disease that depend on ALK and EGFR mutations. We’re getting very close to treating lung cancers that depend on ROS1 and RET. And now we show another oncogenic driver of the disease that begs its own targeted treatment,’ says Robert C. Doebele, MD, PhD, investigator at the CU Cancer Center and assistant professor of medical oncology at the CU School of Medicine.

The group, in collaboration with Pasi A. Jänne, MD, PhD from the Dana-Farber Cancer Institute, started with lung cancer tumour samples from 36 ‘pan-negative’ patients, meaning that no other driver oncogene had been identified. So if not EGFR, ALK and the like, what was driving the cancer? Doebele and colleagues took the question to Foundation Medicine (Cambridge, MA), which used targeted, next-gen sequencing to analyse the samples for possible mutations in a couple hundred potential oncogenes identified as drivers of other cancers. NTRK1 had been identified as a driver of thyroid cancer and so was included in the panel (though drug development had stalled due in part to the relative rarity of the thyroid disease). Sure enough, next-gen sequencing identified NTRK1 gene fusions as the potential driver in two of these samples.

Doebele and colleagues took the lead back to the CU labs, where Marileila Varella Garcia, PhD, developed a specific test for NTRK1 fusions based on fluorescence in situ hybridisation (FISH), similar to what is used for ALK, ROS1 and RET fusions. This allowed the group to validate the finding of NRTK1 as a novel oncogene in these patient samples.

But the study went a step beyond identifying the oncogene. Doebele describes the relative ease with which genes that are improperly activated can be silenced – ‘whether a drug is already is in clinical trials, or already approved for another cancer, or just sitting on the pharma shelf somewhere, many drugs exist that turn off these candidate genes,’ Doebele says.

In this case, Doebele describes ‘walking up the street to Array BioPharma (Boulder, CO), who happened to have several compounds specific for this gene.’

The group showed that mutated NRTK1 genes in cells treated with drug candidates ARRY-772, -523, and -470 and others was effectively turned off.

‘This is still preclinical work,’ Doebele says, ‘but it’s the first – and maybe even second and third! – important steps toward picking off another subset of lung cancer with a treatment targeted to the disease’s specific genetic weaknesses.’ University of Colorado Cancer Center

Researchers identify genetic signature of deadly brain cancer

A multi-institutional team of researchers have pinpointed the genetic traits of the cells that give rise to gliomas – the most common form of malignant brain cancer. The findings provide scientists with rich new potential set of targets to treat the disease.
‘This study identifies a core set of genes and pathways that are dysregulated during both the early and late stages of tumour progression,’ said University of Rochester Medical Center (URMC) neurologist Steven Goldman, M.D., Ph.D., the senior author of the study and co-director of the Center for Translational Neuromedicine. ‘By virtue of their marked difference from normal cells, these genes appear to comprise a promising set of targets for therapeutic intervention.’
As its name implies, gliomas arise from a cell type found in the central nervous system called the glial cell. Gliomas progress in severity over time and ultimately become highly invasive tumours known as glioblastomas, which are difficult to treat and almost invariably fatal. Current treatments, which include surgery, radiation therapy, and chemotherapy, can delay disease progression, but ultimately prove ineffective.
Cancer research has been transformed over the past several years by new concepts arising from stem cell biology. Scientists now appreciate that many cancers are the result of rogue stem cells or their offspring, known as progenitor cells. Traditional cancer therapies often do not prevent a recurrence of the disease since they may not effectively target and destroy the cancer-causing stem cells that lie at the heart of the tumours.
Gliomas are one such example. The source of the cancer is a cell found in the brain called the glial progenitor cell. The cells, which arise from and maintain characteristics of stem cells, comprise about three percent of the cell population of the human brain. When these cells become cancerous they are transformed into glioma stem cells, essentially glial progenitor cells whose molecular machinery has gone awry, resulting in uncontrolled cell division.
Goldman and his team have long studied normal glial progenitor cells. These cells produce glia, a category that includes both astrocytes – cells that support the function of neurons – and oligodendrocytes – cells that produces myelin, the fatty insulation that allows the long-distance conduction of neural impulses.
While Goldman’s group’s work has primarily focused on ways to use glial progenitor cells to treat neurological disorders such as multiple sclerosis, their understanding of the biology of these cells and mastery of the techniques required to sort, identify, and isolate these cells has also enabled them to explore the molecular and genetic changes that transform these cells into cancers.
Using human tissue samples representing the three principal stages of the cancer, the researchers were able to identify and isolate the cancer-inducing stem cells. Working with Goldman, lead authors Romane Auvergne, Ph.D. and Fraser Sim, Ph.D. then compared the gene expression profiles of these cancer stem cells to those of normal glial progenitor cells. The objective was to both pinpoint the earliest genetic changes associated with cancer formation and identify those genes that were unique to the cancer stem cells and were expressed at every stage of disease progression.
Out of a pool over 44,000 tested genes and sequences, the scientists identified a small set of genes in the cancerous glioma progenitor cells that were over-expressed at all stages of malignancy. These genes formed a unique ‘signature’ that identified the tumour progenitor cells and enabled the scientists to define a corresponding set of potential therapeutic targets present throughout all stages of the cancer.
‘One of the key things you are looking for in drug development in cancer is a protein or gene that is over-expressed, so that you can attempt to achieve therapeutic benefit by inhibiting it,’ said Goldman.
The researchers chose to test this hypothesis by targeting one such gene, called SIX1, which was highly over-expressed in the glioma progenitor cells. While this particular gene is active in the early development of the nervous system, it had not been observed in the adult brain before. However, SIX1 signalling has been associated with breast and ovarian cancer, raising the possibility of its contribution to brain cancer as well. This turned out to indeed be the case. When the researchers blocked – or knocked down – the expression of this gene, the tumour cells ceased growing, and implanted tumours shrank.
‘This study gives us a blueprint to develop new therapies,’ said Goldman. ‘We can now devise a strategy to systematically and rationally analyse – and eliminate – glioma stem and progenitor cells using compounds that may selectively target these cells, relative to the normal glial progenitors from which they derive. By targeting genes like SIX1 that are expressed at all stages of glioma progression, we hope to be able to effectively treat gliomas regardless of their stage of malignancy. And by targeting the glioma-initiating cells in particular, we hope to lessen the likelihood of recurrence of these tumours, regardless of the stage at which we initiate treatment.’ University of Rochester Medical Center

New Disease-to-Drug genetic matching puts snowboarder back on slopes

A recent article describes genetic testing of a rare blood cancer called atypical chronic neutrophilic leukaemia (CNL) that revealed a new mutation present in most patients with the disease. The mutation also serves as an Achilles heel, allowing doctors at the University of Colorado Cancer Center to prescribe a never-before-used, targeted treatment. The first patient treated describes his best snowboarding season ever.

‘I’m a crazy sports fan,’ says the patient. ‘I go 30 days a season. I may be the oldest guy snowboarding on the mountain, but I’m not the slowest!’

When he lost a few pounds from what eventually proved to be undiagnosed cancer, the patient was initially pleased. ‘I was lighter and could snowboard better – ride better, jump better,’ he says. Then he took a blood test and his white blood cell count was far in excess of the normal range. His doctor couldn’t find a cause and so they watched and waited. A couple months later, another blood test showed his white count was even higher.

‘That’s when I decided to go to the University of Colorado Hospital,’ he says. There he met Daniel A. Pollyea, MD, MS, CU Cancer Center investigator, assistant professor and clinical director of Leukemia Services at the University of Colorado School of Medicine, and co-author of what would become the recent study.

‘Pollyea said my illness didn’t fit into any major categories,’ the patient says. ‘I could see in his face that he’d run into something abnormal, something new. He was aggressive but didn’t force his own opinion. I saw him reach out to every source he could find – every other specialist he could get in contact with.’

‘He’d been sent from doctor to doctor being told incorrect information,’ Pollyea says. ‘By the time we saw him, his blood counts were going in a bad direction due to the progression of his leukaemia.’

Pollyea had worked on blood cancers since his fellowship training at Stanford University, and through his work there developed a relationship with researchers at the University of Oregon, which had an ongoing project in blood cancers that defied common classifications. Pollyea and his team took a sample from his patient and sent it to Oregon for testing, with the hopes that if they could identify a gene mutation causing this cancer, there might be a chance they could target the mutation with an existing drug.

Sure enough, sequencing showed a mutation in a gene that makes a protein called colony-stimulating factor 3 (CSF3R). Cells with this mutation have uncontrolled growth in the bone marrow, resulting in a leukaemia.

Further studies revealed a drug, ruxolitinib, could effectively target cells with this mutation. Approved to treat another condition, myelofibrosis, just months before, the drug hadn’t previously been considered as a treatment for this type of leukaemia. But with dwindling options, Pollyea and colleagues decided ruxolitinib was worth a try.

‘There were no good alternatives other than to use the ruxolitinib,’ Pollyea says. ‘Our patient became the first person with this condition who received this treatment. His white blood count came down, his other blood counts normalised, and his symptoms virtually disappeared.’

‘I had my best snowboarding season ever,’ says the patient. ‘Good, late season snow here in Colorado. Actually, I’d lived elsewhere and when I first got the disease I wondered if maybe something about moving to Colorado made it happen – you know, the altitude, the lack of oxygen. But now after working with Dr. Pollyea, I realise that I didn’t get sick because I live here, I got cured because I live here. Would I have had this kind of treatment anywhere else? I’m not so sure.’ University of Colorado Cancer Center

Over-produced autism gene alters synapses, affects learning and behaviour in mice

A gene linked to autism spectrum disorders that was manipulated in two lines of transgenic mice produced mature adults with irreversible deficits affecting either learning or social interaction.
The findings have implications for potential gene therapies but they also suggest that there may be narrow windows of opportunity to be effective, says principal investigator Philip Washbourne, a professor of biology and member of the University of Oregon’s Institute of Neuroscience.
The research, reported by an 11-member team from three universities, targeted the impacts of alterations in the gene neuroligin 1 — one of many genes implicated in human autism spectrum disorders — to neuronal synapses in the altered mice during postnatal development and as they entered adulthood. One group over-expressed the normal gene, the other a mutated version.
Mice with higher-than-normal levels of the normal gene after a month had skewed synapses at maturity. Many were larger, appearing more mature, than normal. In these mice, Washbourne said, there were clear cognitive problems. ‘Behaviour was just not normal. They didn’t learn very well, and they were slower to learn, but their social behaviour was not impacted.’
Mice over-producing a mutated version of the gene reached adulthood with structurally immature synapses. ‘They were held back in development and behaviour — the way they behave in terms of learning and memory, in terms of social interaction,’ he said. ‘These were adult mice, three months old, but they behaved like normal mice at four weeks old. We saw arrested development. Learning is a little bit better, they are more flexible just like young mice, they learn faster, but their social interaction is off. To us, this looked more like Asperger’s syndrome.
‘So with the same gene, doing two different manipulations — over-expressing the normal form or over-expressing a mutated form — we’ve gone to two different ends of the autism spectrum,’ said Washbourne, whose lab focuses on basic synapse formation and what goes wrong in relationship to autism. Work has been done in both mice and zebra fish.
‘We made these mice so that we can turn the genes on and off as we want,’ Washbourne said. ‘Using an antibiotic, doxycycline, it turns off these altered genes that we inserted into their chromosomes. While on doxycycline, the mice are absolutely normal.’
However, if the inserted gene was turned off after the completion of development, mice still showed altered synapses and behaviour. This result suggests that any kind of gene therapy may have to be applied to individuals with autism early on.
Effects seen in the social behaviour of mice with the mutated gene, he said, are not unlike observations reported by parents of many autistic children. While normal mice prefer to engage with new mice entering their world rather than familiar others, or even a new inanimate object, these mice split their time equally. ‘It’s not a deficit in memory regarding which mouse is which, it’s more a weighting of their interaction. Does that mean they are autistic? I don’t know, but if you talk to parents of autistic children, one of the frustrating things they report is that their children treat complete strangers in exactly the same way that they treat them.’
While the findings provide new insights, Washbourne said, any translation into treatment could be decades away. ‘A problem with autism is there are many different genes potentially involved. It could be that some day, if you are diagnosed with autism, a mouth swab might allow for the identification of the exact gene that is mutated and allow for targeted therapy,’ he said. ‘Genome sequencing already has turned up subtle mutations in lots of genes. Autism might be like cancer, with hundreds of potential combinations of faulty genes.’ University of Oregon

Nearly one-third of children with autism also have ADHD

In a study of the co-occurrence of attention deficit hyperactivity disorder (ADHD) and autism spectrum disorder (ASD) in early school-age children (four to eight years old), researchers at the Kennedy Krieger Institute found that nearly one-third of children with ASD also have clinically significant ADHD symptoms. The study also found that children with both ASD and ADHD are significantly more impaired on measures of cognitive, social and adaptive functioning compared to children with ASD only.

Distinct from existing research, the current study offers novel insights because most of the children entered the study as infants or toddlers, well before ADHD is typically diagnosed. Previous studies on the co-occurrence of ASD and ADHD are based on patients seeking care from clinics, making them biased towards having more multi-faceted or severe impairments. By recruiting patients as infants or toddlers, the likelihood of bias in the current study is significantly reduced.

‘We are increasingly seeing that these two disorders co-occur and a greater understanding of how they relate to each other could ultimately improve outcomes and quality of life for this subset of children,’ says Dr. Rebecca Landa, senior study author and director of the Center for Autism and Related Disorders at Kennedy Krieger. ‘The recent change to the Diagnostic and Statistical Manual of Mental Disorders (DSM-5) to remove the prohibition of a dual diagnosis of autism and ADHD is an important step forward.’

Participants in this prospective, longitudinal child development study included 162 children. Researchers divided the children into ASD and Non-ASD groups. The groups were further categorized by ADHD classification according to parent-reported symptoms of ADHD on the Hyperactivity and Attention Problems subscales of the Behavioral Assessment System for Children-Second Edition, a standard assessment specifically designed to identify the core symptoms of ADHD.

Results revealed that, out of 63 children with ASD in the study, 18 (29%) were rated by their parents as having clinically significant symptoms of ADHD. Importantly, the age range for children in the study (four to eight) represented a younger and narrower sample than has been previously reported in published literature. ‘We focused on young school-aged children because the earlier we can identify this subset of children, the earlier we can design specialized interventions,’ says Dr. Landa. ‘Tailored interventions may improve their outcomes, which tend to be significantly worse than those of peers with autism only.’

Researchers also found that early school-age children with co-occurrence of ASD and ADHD were significantly more impaired than children with only ASD on measures of cognitive and social functioning, as well as in the ability to function in everyday situations. They were also more likely to have significant cognitive delays (61 versus 25 percent) and display more severe autism mannerisms, like stereotypic and repetitive behaviours. The study findings suggest that children with the combined presence of ADHD and ASD may need different treatment methods or intensities than those with ASD only in order to achieve better outcomes.

Dr. Landa and her team recognise that this research supports the need for future prospective, longitudinal studies of attention, social, communication and cognitive functioning from the time that the first red flags of ASD are identified. Such research will lead to important insights about the relative timing of onset and stability of disruption to attention mechanisms and barriers to successful functioning in children with co-occurring ASD and ADHD. Kennedy Krieger Institute

Hormone levels may provide key to understanding psychological disorders in women

Women at a particular stage in their monthly menstrual cycle may be more vulnerable to some of the psychological side-effects associated with stressful experiences, according to a study from UCL.
The results suggest a monthly window of opportunity that could potentially be targeted in efforts to prevent common mental health problems developing in women. The research is the first to show a potential link between psychological vulnerability and the timing of a biological cycle, in this case ovulation.

A common symptom of mood and anxiety problems is the tendency to experience repetitive and unwanted thoughts. These ‘intrusive thoughts’ often occur in the days and weeks after a stressful experience.
In this study, the researchers examined whether the effects of a stressful event are linked to different stages of the menstrual cycle. The participants were 41 women aged between 18 and 35 who had regular menstrual cycles and were not using the pill as a form of contraception. Each woman watched a 14-minute stressful film containing death or injury and provided a saliva sample so that hormone levels could be assessed. They were then asked to record instances of unwanted thoughts about the video over the following days.
There is actually a fairly narrow window within the menstrual cycle when women may be particularly vulnerable to experiencing distressing symptoms after a stressful event.
‘We found that women in the ‘early luteal’ phase, which falls roughly 16 to 20 days after the start of their period, had more than three times as many intrusive thoughts as those who watched the video in other phases of their menstrual cycle,’ explains author Dr Sunjeev Kamboj, Lecturer in UCL’s Department of Clinical, Educational and Health Psychology.
‘This indicates that there is actually a fairly narrow window within the menstrual cycle when women may be particularly vulnerable to experiencing distressing symptoms after a stressful event.’
The findings could have important implications for mental health problems and their treatment in women who have suffered trauma.
‘Asking women who have experienced a traumatic event about the time since their last period might help identify those at greatest risk of developing recurring symptoms similar to those seen in psychological disorders such as depression and post-traumatic stress disorder (PTSD),’ said Dr Kamboj.

‘This work might have identified a useful line of enquiry for doctors, helping them to identify potentially vulnerable women who could be offered preventative therapies,’ continued Dr Kamboj. University College London

New, more accurate test for Down’s syndrome developed

Researchers at King’s College London and King’s College Hospital, part of King’s Health Partners Academic Health Sciences Centre, have developed a new, non-invasive blood test that can reliably detect whether or not an unborn baby has Down’s syndrome. The test can be given earlier in pregnancy and is more accurate than current checks.
Down’s syndrome, also referred to as trisomy 21, is a genetic disorder caused by the presence of all or part of an extra copy of chromosome 21 in a person’s DNA. Current screening for Down’s syndrome and other trisomy conditions includes a combined test done between the 11th and 13th weeks of pregnancy, which involves an ultrasound screen and a hormonal analysis of the pregnant woman’s blood. Methods such as chorionic villus sampling (CVS), which involves taking cell samples from the placenta, and amniocentesis (using a sample of amniotic fluid), are also used to detect abnormalities but they are both invasive and carry a risk of miscarriage.
Several studies have shown that non-invasive prenatal diagnosis for trisomy syndromes using foetal cell free (cf) DNA from a pregnant woman’s blood is highly sensitive and specific, making it a potentially reliable alternative that can be done earlier in pregnancy.
Kypros Nicolaides, Professor of Fetal Medicine at King’s College London and Head of the Harris Birthright Research Centre for Fetal Medicine at King’s College Hospital, and colleagues have now demonstrated the feasibility of routine screening for trisomies 21, 18, and 13 by cfDNA testing. Testing done in 1005 pregnancies at 10 weeks had a lower false positive rate and higher sensitivity for foetal trisomy than the combined test done at 12 weeks. Both cfDNA and combined testing detected all trisomies, but the estimated false-positive rates were 0.1 percent and 3.4 percent, respectively.
‘This study has shown that the main advantage of cfDNA testing, compared with the combined test, is the substantial reduction in false positive rate. Another major advantage of cfDNA testing is the reporting of results as very high or very low risk, which makes it easier for parents to decide in favour of or against invasive testing,’ said Professor Nicolaides.
A second Ultrasound in Obstetrics & Gynecology study by the group, which included pregnancies undergoing screening at three UK hospitals between March 2006 and May 2012, found that effective first-trimester screening for Down’s syndrome could be achieved by cfDNA testing contingent on the results of the combined test done at 11 to 13 weeks. The strategy detected 98 percent of cases, and invasive testing was needed for confirmation in less than 0.5 percent of cases.
The authors conclude that screening for trisomy 21 by cfDNA testing contingent on the results of an expanded combined test would retain the advantages of the current method of screening, but with a simultaneous major increase in detection rate and decrease in the rate of invasive testing. Kings College London

Genes help shape-shifting skin cancer cells to spread

Researchers have identified a set of genes that allow melanoma cells, a type of cancer cell, to change rapidly between two shapes to escape from the skin and spread around the body. This new research – funded by the Wellcome Trust, Cancer Research UK and the US National Institutes of Health – could pave the way for scientists to develop desperately needed drugs for malignant melanoma, the deadliest form of skin cancer, which kills more than 2200 people every year.
The most dangerous aspect of melanoma is its ability to spread, or become malignant, to other parts of the body in the later stages of disease. This most often includes the liver, lungs and brain.
Dr Chris Bakal, a Wellcome Trust research fellow at the Institute of Cancer Research, London, explains: ‘We already knew that metastatic melanoma cells, or cells that are able to spread through the body, have to be able to adopt different shapes so that they can squeeze their way between healthy cells and move around the body.
‘The cells have to become rounded to travel through the bloodstream or invade soft tissues such as the brain, but they take on an elongated shape to travel through harder tissues like bone. But until now, we knew hardly anything about how the cells assume either of these shapes and how they switch between the two.’
To investigate this, researchers at the Institute of Cancer Research, London, and Weill Cornell Medical College in Houston started out by looking at fruit fly cells. They found that under normal conditions, the fruit fly cells grew in five different shapes. By switching off specific genes, they were able to change the mix of shapes among the fruit fly cells and identify several different genes that control a cell’s shape-shifting ability.
When they looked in human melanoma cells, they found that the human versions of these genes had a similar effect. In particular, they noted that switching off a gene called PTEN increased the proportion of cells that were elongated rather than rounded.
PTEN is a gene that is also involved in stopping healthy cells from becoming cancer cells, a so-called ‘tumour suppressor’ gene. This particular gene is switched off in around 1 in 8 melanoma patients and in almost half of melanoma patients who carry a mutation in another cancer gene called BRAF.
‘We think that metastatic melanoma cells lose their PTEN function so that they can increase their shape-shifting ability, which in turn enables them to move to many different tissues within the body. It’s early days, but taken together our findings offer new opportunities to develop drugs to try and stop the spread of melanoma,’ Dr Bakal added.
Dr Julie Sharp, Senior Science Communications Manager at Cancer Research UK, said: ‘This is still early research, but it gives us a better grasp of the way cancer cells behave in the body. By mimicking these conditions, our researchers are learning more about melanoma and bringing us closer to beating it. Wellcome Trust

Epigenetic factor likely plays a key role in fuelling most common childhood cancer

Changes in an epigenetic mechanism that turns expression of genes on and off may be as important as genetic alterations in causing pediatric acute lymphoblastic leukemia (ALL), according to a study led by scientists at St. Jude Children’s Research Hospital.
The results suggest the mechanism called cytosine methylation plays a previously under-appreciated role in the development of leukaemia. Cytosine methylation involves adding or removing methyl groups to cytosine, which is a building block of DNA.
The study is the most comprehensive effort yet to identify and understand genetic and epigenetic factors that work together to cause ALL, the most common childhood cancer. ALL is a cancer of white blood cells known as lymphocytes. Scientists at St. Jude and Weill Cornell Medical College collaborated on the project.
Researchers used a variety of techniques to examine hundreds of thousands of methylation sites across the genome in normal and leukemic lymphocytes, including samples from more than 160 children with ALL. Investigators found that known ALL subgroups, which are defined by chromosomal alterations, have unique methylation profiles. Those profiles correlated with different patterns of gene expression.
‘It is well known that different leukaemia subgroups have distinct patterns of gene expression that are important in the development of leukaemia,’ said Charles Mullighan, MBBS (Hons), MSc, M.D., an associate member of the St. Jude Department of Pathology. Mullighan and Ari Melnick, M.D., Gebroe Professor Hematology/Oncology at Weill Cornell Medical College, are the study’s co-corresponding authors.
‘We have assumed that the underlying genetic changes are important determinants of those gene expression profiles. We now know that changes in methylation state also have key roles in influencing gene expression,’ Mullighan said.
The study used tissue samples from 137 St. Jude patients with B-cell leukaemia and 30 children with T-cell leukaemia. The patients represented all major ALL subgroups.
‘The data show that aberrant epigenetic gene programming can now be considered a hallmark of acute lymphoblastic leukaemia, occurring in all patients regardless of the presence of genetic mutations,’ Melnick said. ‘This offers the opportunity for development of epigenetic targeted therapies for patients with ALL that could be broadly applicable to many patients.’
For comparison, researchers also checked B and T cells from 27 healthy children. Investigators found that leukaemia cells shared a core group of abnormally methylated genes. The genes included ones involved in regulating the cell division and proliferation. ‘This remains to be tested, but the findings suggest that alterations in methylation are an important early step in the development of leukaemia,’ Mullighan said.
The research provides further evidence that genetic and epigenetic events are both important in establishing different subgroups of ALL. For this study, researchers conducted genome-wide sampling of methylation, gene expression and DNA structural abnormalities, including the gain or loss of DNA. Shann-Ching Chen, Ph.D., St. Jude Pathology, developed many of the methods used to integrate and analyse the results. Chen and Maria Figueroa, now of the University of Michigan and formerly of Cornell, are the study’s co-first authors.
The study also found that more than one-third of 71 genes targeted by genetic alterations are also abnormally methylated in ALL. The methylation changes involved known tumor suppressor or oncogenes genes including CDKN2A, CDKN2B, PTEN and KRAS. ‘The findings suggest these genes are inactivated or deregulated more frequently than suggested by simply analysing structural changes in the genome,’ Mullighan said. St. Jude Children’s Research Hospital