Improved control of engineered T cells
Scanning electron micrograph of human T cell. Image courtesy of the National Institute of Allergy and Infectious Diseases.
Chimeric antigen receptor T (CAR-T) cells engineered to recognize cancer cells have produced remissions of B-cell leukemia in clinical trials. However, serious side effects such as uncontrolled cytokine release and depletion of normal B cells can result from the inability to control CAR–T-cell activity in vivo. David Rodgers et al. (pp. E459–E468) report a method for controlling CAR–T-cell activity using an antibody-based intermediate. The authors engineered T cells to recognize a yeast peptide, which was introduced into an antibody against a tumor marker to create a switch molecule that mediates the interaction between T cells and cancer cells. Coadministration of the CAR-T cells and a switch against the B-cell tumor marker CD19 cleared tumors in a mouse model of B-cell leukemia. Tumor regression, cytokine release, and T-cell phenotype could be controlled by varying the switch dosage. In a related article, Jennifer Ma et al. (pp. E450–E458) developed a similar method for controlling the activity and specificity of CAR-T cells using a different antibody-based switch. The authors engineered CAR-T cells to recognize the compound fluorescein, which was conjugated to an antibody against CD19. Coadministration of CAR-T cells and this semisynthetic switch cleared tumors in a mouse leukemia model to a degree comparable to CAR-T cells currently in clinical trials. The authors found that adjusting the antibody dosing regimen enabled tumor clearance while minimizing CAR–T-cell toxicity, as measured by body weight loss and cytokine release. The same CAR-T cells could also target cancer cells expressing a different marker, CD22, when a fluorescein-tagged antibody was used against CD22 instead of CD19. Thus, a single universal CAR-T cell combined with distinct switch molecules might be used to treat a variety of cancers, including heterogeneous and recurrent tumors, according to the authors. — B.D.
Grammar acquisition in children
Volunteer reads to children. Image courtesy of iStockphoto/monkeybusinessimages.
Human language acquisition presents the question of whether a child acquires aspects of a language by discovering structure in the language of the community in which the child lives or by imposing structure on that language. To investigate the relationship between grammatical knowledge of parents and their children, Chung-hye Han et al. (pp. 942–947) explored the variability of a grammatical feature among Korean speakers, particularly parents and their children. In an experiment with 31 adult Korean speakers, the authors found that participants’ placement of the verb within a negative sentence varied, and that the verb placement was consistent for individual participants, suggesting that the participants were not maintaining multiple grammatical constructs. In another experiment, the authors assessed variability in the verb placement of 22 Korean children between the ages of 4 and 5.5 years, as well as the verb placement of 21 of the children’s mothers. The authors found that the children’s verb placement was independent of that of their mothers. The findings suggest that in a language with variable grammatical structures, people may consistently adopt one structure according to internal processes and independent of parental linguistic influences, according to the authors. — P.G.
Genetic effects of early human expansion
Sunset over a grass hut in the Kalahari Desert, southern Africa.
Human genomes carry hundreds of potentially deleterious mutations, but it is unclear how the number of deleterious mutations varies among populations. Brenna Henn et al. (pp. E440–E449) characterized the distribution of deleterious mutationsacross human populations using genomic data from Namibian, Congolese, Algerian, Pakistani, Cambodian, Siberian, and Mexican populations. Within individuals, the authors observed an increase in the number of alleles predicted to be deleterious with increasing distance from sub-Saharan Africa. The correlation was strongest for alleles predicted to have large, but not extreme, deleterious effects, whereas the number did not vary significantly among populations for the most severe mutations. The reduction in heterozygosity for deleterious mutations relative to neutral heterozygosity was significantly larger in Africans than in non-Africans. Furthermore, for moderately deleterious genes, this reduction did not correlate with distance from Africa. The results are predicted by a spatially explicit model of human expansion out of Africa. Further, the results suggest that genetic drift has been has been stronger in non-African populations than African populations, and that mildly and moderately deleterious mutations evolved as if they were neutral during the expansion out of Africa, according to the authors. — B.D.
Evolutionary forces and leukemia
Distinct evolutionary forces dictate leukemia incidence at different life stages.
Cancer is thought to arise from a sequence of mutations that confer a fitness advantage to tumor cells. Although cancer-driving mutations accumulate with age, the rates of certain cancers such as leukemia are higher in young children than in young adults. To explain this puzzling observation, Andrii Rozhok et al. (pp. 1050–1055) examined through computational modeling the evolution of populations of hematopoietic stem cells (HSCs), which accumulate mutations to give rise to malignant blood cells in leukemia patients. The simulations relied on published data showing that HSCs divide less frequently, but grow in population size, with increasing age. In young children, evolution in the small pool of HSCs was driven primarily by drift: random fluctuations in the numbers of gene variants. The drift-driven expansion of genetically identical cells, combined with the rapid division of HSCs, promoted the accumulation of cancer-driving mutations at young ages. During early adulthood, the accumulation of mutations was suppressed by the decrease in cell division rates combined with stabilizing selection—a type of natural selection that decreases genetic diversity. Later in life, positive selection acted on HSCs to promote the accumulation of mutations that conferred a fitness advantage in the declining microenvironment of aged tissues. According to the authors, the findings suggest that leukemias of childhood and adulthood may be different diseases forged by distinct evolutionary forces. — J.W.