Tilting the T cell

Illness or stress can seriously deplete the T cell population in the immune system. But T cells can then bounce back with a vengeance, often becoming 'effector-like' cells that can react against the body's own tissues. In the April 16 Cell, Cecile King et al. show that a similar mechanism could precipitate autoimmune disease.

The researchers found that the NOD mouse, which develops autoimmune diabetes, has low levels of T cells. They also found that T cells in these animals were nonetheless proliferating constantly, as if to replenish the stores. The researchers then when on a hunt for the basis of this T cell imbalance.

Scanning for genes that were in the NOD locus, they emerged with a candidate: the cytokine IL-21, which enhances T cell proliferation. The authors found that NOD mice expressed high levels of IL-21, and produced numerous IL-21 receptor–activated T cells. Many cells had a tendency to express effector cytokines, suggesting they could be pathogenic. If NOD mice contain high levels of these proliferating cells, why do they have so few T cells? The authors found that, although IL-21 promoted T cell proliferation, the cytokine does not support survival. Although many of these IL-21-tweaked T cells die, presumably some live on to destroy pancreatic β-cells and promote diabetes. —CS

Another 'umab'

The monoclonal antibody and anti-cancer drug trastuzumab (Herceptin) interferes with the activation of ErbB2 in tumors that overexpress this receptor. But ErbB2 can be activated in tumors not only by overexpression, but also through ligand-mediated stimulation of other ErbB receptors. Dimerization then triggers the signals that lead to cancer cell survival and proliferation. In the April Cancer Cell, Matthew Franklin et al. show that another antibody, pertuzumab, can interfere with ErbB2 signaling, regardless of how the receptor is activated. The investigators undertook a structural study of ErbB2 bound to pertuzumab. They found that, consistent with biochemical studies, pertuzumab binds to a region required for dimerization and is thereby well-situated to destroy tumors with low levels of ErB2. The new antibody is now in phase 2 trials. —CS

Spewing PolyQ

The eukaryotic proteasome fails to digest long polyglutamine (polyQ)-containing proteins, report Prasanna Venkatraman et al. in the April 9 Molecular Cell. The finding could help explain how polyQ-containing proteins accumulate and aggregate into inclusions, which mar the brains of individuals with Huntington disease and other polyQ-repeat disorders.

Proteasomes chew most proteins up into 2–24 amino acid bits that are then destroyed by cellular peptidases. Although PolyQ-containing proteins are degraded primarily by the proteasome, the fate of the polyQ sequences themselves was unknown. Using fusion peptides with 10–30 glutamine residues, the researchers found that isolated proteasomes could not digest the repeats. To verify that this was due to an inability of the proteosome's active sites to bind polyQ, they showed that ancestral archaeal proteasomes, which have less specific active sites, could digest these peptides.

What happens to the rejected bits of polyQ? Since cells regularly clear polyQ sequences, unidentified cellular peptidases must chew them up. But the released fragments are more likely to aggregate than full-length polyQ-containing proteins. Moreover, whereas small peptides exit the proteasome efficiently, longer polyQ stretches might not. Clogging the central channel of the proteasomes could lead to the erosion of proteasome function that seems to occur in polyQ-repeat disorders. The results dovetail with observations in cells, such as the accumulation of ubiquitinated polyQ proteins, tagged for destruction, in inclusions as a patient ages. Proteins associated with some diseases can have more than 300 glutamines in a stretch. —CS

Let me out!

Credit: Simon Fraser/Science Photo Library

A newborn will emit its first scream shortly after birth, but the fetus is communicating with its mother even more urgently before that. According to Jennifer Condon et al. a fetal lung protein called surfactant protein A (SP-A) may signal that it is time to give birth.

In humans, surfactant protein production begins two-thirds of the way through a pregnancy, and then rapidly rises. In the April 6 PNAS Condon et al. found that SP-A also rose before birth in mice. The authors saw a corresponding rise in NFκB in the maternal uterus, and in the secretion of IL-1β from fetal macrophages. They propose that SPA-1 triggers migration of fetal macrophages to the uterus. Secretion of IL-1β from these cells then activates NFκB, which in turn promotes labor.

In support of this scheme, the authors found that SP-A stimulated IL-1β and NFκB expression in macrophages cultured from the amniotic fluid. Amniotic injection of SP-A caused preterm delivery; and injection of an SP-A antibody or NFκB inhibitor delayed labor by more than 24 hours. Given that delivery does eventually occur, other signals—perhaps also from the fetus—come into play. —CS

Twist on ALS

The commonly held view of how nerve cells die in amyotrophic lateral sclerosis (ALS) is through extracellular accumulation of toxic levels of the neurotransmitter glutamate. Research in the April 1st online Journal of Neurochemistry suggests that this mechanism might be overrated.

Juan Carlos Corona and Ricardo Tapia implanted microdialysis probes in the spinal cords of rats and tested the effect of drugs that modify glutamate-mediated neurotransmission. They found that increasing the synaptic concentration of glutamate by blocking its uptake or stimulating its synaptic release had no effect on motor neuron survival. By contrast, direct application of AMPA, a glutamate-receptor agonist, led to hindlimb paralysis and neuron death. These results suggest that the synaptic concentration of glutamate that can be reached by manipulating the endogenous pool of the transmitter is insufficient to kill the motor neurons. The neurotoxic effect might therefore require high levels of receptor activation, such as those afforded by the administration of AMPA in this study.

In addition to challenging the idea that motor neuron death results from increased extracellular glutamate, this study sheds light on the pathogenesis of sporadic ALS, which accounts for more than 90% of clinical cases and whose study has been largely neglected in favor of genetic forms of ALS. —JCL

Written by Juan Carlos López and Charlotte Schubert