Photo: Steve FischSKIN CELLS INTO NEURONS
The inability to investigate living neurons directly poses a challenge for understanding disorders such as autism and schizophrenia on a cellular level. In a major step toward creating patient- and disease-specific neurons for study, researchers at the School of Medicine have identified a combination of four transcription factors that can convert human skin cells into functional neurons in four to five weeks' time. "We are now much closer to being able to mimic brain or neurological diseases in the laboratory," says senior scientist Marius Wernig (above), an assistant professor of pathology and member of the Institute for Stem Cell Biology and Regenerative Medicine.
Significantly, the method bypasses the intermediate "induced pluripotent stem cell" or iPS stage. There have been some indications in laboratory mice that the proteins used to coax cells into that flexible state might trigger immune reactions.
The research builds on earlier work by Wernig's group using three of the same four factors to convert mouse skin cells into neurons. (More recently, they used the same protocol to turn liver cells from mice into neurons.) However, whereas they previously found that 20 percent of mouse skin cells were transformed into working neurons, in the current study only 2 to 4 percent of human skin cells formed cells that looked and behaved like neurons. Wernig says that he and his colleagues are working to optimize the technique and conditions to improve efficiency.
Photo: Norbert Von Der Groebe/Stanford Hospital and ClinicsMEASURING PAIN OBJECTIVELY
In both clinical and research settings, the standard for pain assessment is the patient's self-report. But that has several drawbacks: In addition to being subjective, the approach may compromise treatment for patients who can't communicate their pain levels—in particular, the very young and the very old. What's more, studies have shown a cultural bias against chronic pain sufferers, who are often perceived as feigning their symptoms. "While self-reported pain provides useful clinical information and proves to be an effective approach in most situations, it can fail certain vulnerable populations," wrote Sean Mackey (above), chief of the division of pain management and associate professor of anesthesia, co-author of a paper published in PLoS ONE. "There is, therefore, a need to develop a pain assessment tool that is based on physiology, and requires no communication on the part of patients."
To that end, Mackey and his collaborators trained a computer algorithm to identify what pain looks like in the brains of healthy subjects from fMRI scans. The computer correctly classified scans of subjects who received a moderately painful heat stimulus (versus those receiving a non-painful stimulus) 81 percent of the time. The researchers stressed that future studies are needed to determine whether the method will work on other kinds of pain and whether it's sensitive enough to distinguish physical pain from emotional states such as anxiety or depression.
