Search for: All records

ABSTRACT The development of vaccines for fungal diseases, including cryptococcosis, is an emergent line of research and development. In previous studies, we showed that aCryptococcusmutant lacking theSGL1gene (∆sgl1) accumulates certain glycolipids called steryl glucosides (SGs) on the fungal capsule, promoting an effective immunostimulation that totally protects the host from a secondary cryptococcal infection. However, this protection is lost when the cryptococcal capsule is absent in the∆sgl1background. The cryptococcal capsule is mainly composed of glucuronoxylomannan (GXM), a polysaccharide microfiber consisting of glucuronic acid, xylose, and mannose linked by glycosidic bonds forming specific triads. In this study, we engineered cells to lack each of the GXM components and tested the effect of these deletions on protection under the condition of SG accumulation. We found that glucuronic acid and xylose are required for protection, and their absence abrogates the production of IFNγ and IL-17A by γδ T cells, which are necessary stimulants for the protective phenotype of the∆sgl1. We analyzed the structure of the GXM microfibers and found that although the deletion ofSGL1only slightly affects the size and distribution of these microfibers, it significantly changes the ratio of mannose to other components. In conclusion, this study identifies the structural modifications that the deletion ofSGL1and the consequent accumulation of SGs impart to the GXM structure ofC. neoformans. This provides significant insights into the protective mechanisms mediated by SG accumulation on the capsule, with important implications for the future development of an efficacious cryptococcal vaccine.IMPORTANCECryptococcus neoformansis an encapsulated fungus that causes invasive fungal infections with high morbidity and mortality in susceptible patients. With increasing drug resistance and high toxicity of current antifungal drugs, there is a need for alternative therapeutic strategies, such as a cryptococcal vaccine. In this study, we identify the necessary capsular components and their structural organization required for a cryptococcal vaccine to protect the host against challenge with a virulent strain. These capsular components are glucuronic acid, xylose, and mannose, and they work together with certain glycolipids called steryl glucosides (SGs) to stimulate host immunity. Interestingly, SGs on the capsule may favor the formation of small capsular microfibers organized in specific mannose triads. Thus, the results of this paper are important because they identify a mechanism by which SGs affect the structure of the cryptococcal capsule, with important implications for the future development of a cryptococcal vaccine using capsular components and SGs. 

We show how any PAC learning algorithm that works under the uniform distribution can be transformed, in a blackbox fashion, into one that works under an arbitrary and unknown distribution ‍D. The efficiency of our transformation scales with the inherent complexity of ‍D, running in (n, (md)d) time for distributions over n whose pmfs are computed by depth-d decision trees, where m is the sample complexity of the original algorithm. For monotone distributions our transformation uses only samples from ‍D, and for general ones it uses subcube conditioning samples. A key technical ingredient is an algorithm which, given the aforementioned access to D, produces an optimal decision tree decomposition of D: an approximation of D as a mixture of uniform distributions over disjoint subcubes. With this decomposition in hand, we run the uniform-distribution learner on each subcube and combine the hypotheses using the decision tree. This algorithmic decomposition lemma also yields new algorithms for learning decision tree distributions with runtimes that exponentially improve on the prior state of the art—results of independent interest in distribution learning. 

We give a pseudorandom generator that fools m -facet polytopes over {0, 1} n with seed length polylog( m ) · log n . The previous best seed length had superlinear dependence on m . 

We give an nO(log log n)-time membership query algorithm for properly and agnostically learning decision trees under the uniform distribution over { ± 1}n. Even in the realizable setting, the previous fastest runtime was nO(log n), a consequence of a classic algorithm of Ehrenfeucht and Haussler. Our algorithm shares similarities with practical heuristics for learning decision trees, which we augment with additional ideas to circumvent known lower bounds against these heuristics. To analyze our algorithm, we prove a new structural result for decision trees that strengthens a theorem of O’Donnell, Saks, Schramm, and Servedio. While the OSSS theorem says that every decision tree has an influential variable, we show how every decision tree can be “pruned” so that every variable in the resulting tree is influential.