Editors' Choice

Science  28 Jul 2017:
Vol. 357, Issue 6349, pp. 367
  1. Neurodegeneration

    The yin and yang of TDP-43 reduction

    1. Stella M. Hurtley

    TDP-43 in microglia reduces brain amyloid, but with a cost.


    The gene that encodes TDP-43 is implicated in fronto-temporal lobar degeneration and amyotrophic lateral sclerosis. Microglia are the brain's innate immune cells, constantly surveying for anything that might interfere with correct brain functions. Paolicelli et al. suggest that TDP-43 is a regulator of microglia function. In an inducible conditional knockout mouse line crossed to an Alzheimer's mouse model, the depletion of TDP-43 from brain microglia resulted in a decrease in brain amyloid. Such a reduction in amyloid might be construed as beneficial. However, the mice also exhibited a reduction in synaptic markers and displayed signs of neurodegeneration. Thus, TDP-43 plays a role in modulating microglia-mediated clearance. In neurodegeneration, such abnormal phagocytic activity could be detrimental if it promotes excessive microglial-mediated synaptic removal, despite enhanced amyloid clearance.

    Neuron 10.1016/j.neuron.2017.05.037 (2017).

  2. Workforce

    Support is needed to leave the lab

    1. Melissa McCartney

    As doctoral recipients in STEM (science, technology, engineering, and mathematics) fields increasingly move to nonacademic positions, graduate training programs struggle to design strategies and resources that prepare graduates for a broad job market. St. Clair et al. investigated factors that affect the career search confidence and job search strategies of doctoral trainees. Using data from an NIH-funded survey of biomedical doctoral students and postdoctoral fellows, the team found that trainee perceptions of institutional support show that how trainees perceive their support systems (or lack thereof) has a strong direct effect on their career search efficacy, as well as the types of career development strategies that they adopt. Although this research shows the importance of programmatic support, what constitutes such a positive culture remains unknown.

    PLOS ONE 10.1371/journal.pone.0177035 (2017).

  3. Cancer

    Diagnostic tumor methylation profiles

    1. Catherine Griffin

    Diagnosis of specific types of cancer is vital for decisions regarding treatment and monitoring plans. However, acquiring sufficient tissue from a tumor biopsy to perform a histological examination and diagnosis can be difficult because of limited tumor accessibility or ambiguous tissue samples. Methods that allow minimally invasive biopsies are required to improve patient care. Hao et al. identified DNA methylation profiles associated with four common cancers that can be used to identify tumors from smaller tissue samples with accuracy comparable to that of current diagnostic methods. These methylation markers can also be used to identify the tissue of origin of metastasized cancers, which can be critical for treatment, and as predictors of prognosis and survival to improve standards of diagnosis.

    Proc. Natl. Acad. Sci. U.S.A. 10.1073/pnas.1703577114 (2017).

  4. Virology

    Tumors in green turtles

    1. Gemma Alderton

    A new approach allows for the culture of a herpesvirus that may cause cancer in green sea turtles.


    Green turtles are an endangered species, and, like many marine species, they can have cancer. Green turtles can develop a fatal condition, fibropapillomatosis, which involves the development of multiple tumors in the skin and various other organs. Fibropapillomatosis is thought to be caused by a herpesvirus, ChHV5, but this remains unproven because the virus is not easily cultured. Work et al. report a method to culture ChHV5 in sections of green turtle skin and in three-dimensional matrices that resemble the structure of skin. This method should allow the study of ChHV5 (and other viruses that cannot be grown in culture), thereby answering the question of whether it is the cause of fibropapillomatosis.

    J. Virol. 10.1128/JVI.00404-17 (2017).

  5. Nanoscience

    Getting random with thermal noise

    1. Marc S. Lavine

    Software-generated random numbers depend on an initial seed value, and thus a sequence can be replicated if one knows this initial seed. In contrast, one can generate true random numbers by tapping into random natural phenomena, such as noise caused by thermal fluctuations. Gaviria Rojas et al. built static random access memory (SRAM) cells from single-walled carbon nanotubes fixed onto polymer substrates. When the SRAM is first supplied with power, it exists in a metastable binary state. Whether it is on or off is then determined by thermal fluctuations. The authors leveraged this digitization of random noise to create a simple generator of true random number that is compatible with printable and flexible electronics.

    Nano Lett. 10.1021/acs. nanolett.7b02118 (2017).

  6. Immunology

    DNA methylation makes for tired T cells

    1. Seth Scanlon

    T cell exhaustion is a phenomenon in chronic infections and cancer in which T cells express inhibitory receptors and stop working properly. One clinically applied approach to reversing this process has been to block the inhibitory receptor PD-1. However, T cells may resist such treatment as a result of transcriptional and epigenetic reprogramming. Ghoneim et al. found that progressive de novo DNA methylation is an essential process inaugurating CD8+ T cell exhaustion. This results in a distinct gene repression signature that is preserved during PD-1 blockade. The inhibition of DNA methylation before PD-1 blockade enhanced T cell rejuvenation and improved disease outcomes in mice. This approach offers hope of more effective treatments for viral infections and cancer.

    Cell 170, 142 (2017).

  7. Nanomaterials

    Smoothing graphene by encapsulation

    1. Phil Szuromi

    Graphene is often depicted as a flat sheet, but in reality some surface roughness is present, especially for graphene placed on substrates, and this roughness scatters electrons and decreases carrier mobility. Thomsen et al. used diffraction-tilt measurements in a transmission electron microscope to determine the roughness of suspended graphene samples. The root-mean-square roughness of free-standing graphene was 114 pm, whereas it was 21 pm for graphene supported on hexagonal boron nitride (hBN) and only 12 pm for graphene encapsulated by hBN. First-principles calculations indicate that flexural acoustic phonon modes that roughen the graphene become localized mainly in hBN layers.

    Phys. Rev. B 10.1103/PhysRevB.96.014101 (2017).