While the extent and impact of horizontal transfers in prokaryotes are widely acknowledged, their importance to the eukaryotic kingdom is unclear and thought by many to be anecdotal. Here we report multiple recent transfers of a huge genomic island between Penicillium spp. found in the food environment. Sequencing of the two leading filamentous fungi used in cheese making, P. roqueforti and P. camemberti, and comparison with the penicillin producer P. rubens reveals a 575 kb long genomic island in P. roqueforti—called Wallaby—present as identical fragments at non-homologous loci in P. camemberti and P. rubens. Wallaby is detected in Penicillium collections exclusively in strains from food environments. Wallaby encompasses about 250 predicted genes, some of which are probably involved in competition with microorganisms. The occurrence of multiple recent eukaryotic transfers in the food environment provides strong evidence for the importance of this understudied and probably underestimated phenomenon in eukaryotes.

Wild-type p53 is a tumor suppressor gene which can activate or repress transcription, as well as induce apoptosis. The human p53 proline-rich domain localized between amino acids 64 and 92 has been reported to be necessary for efficient growth suppression. This study shows that this property mainly results from impaired apoptotic activity. Although deletion of the proline-rich domain does not affect transactivation of several promoters, such as WAF1, MDM2 and BAX, it does alter transcriptional repression, reactive oxygen species production and sequence-specific transactivation of the PIG3 gene, and these are activities which affect apoptosis. Whereas gel retardation assays revealed that this domain did not alter in vitro the specific binding to the p53-responsive element of PIG3, this domain plays a critical role in transactivation from a synthetic promoter containing this element. To explain this discrepancy, evidence is given for a prolinerich domain-mediated cellular activation of p53 DNA binding.

DNA microarray technology enables investigators to measure the expression of several thousand mRNA species simultaneously in a biological specimen. However, the reliability of the microarray technology to detect transcriptional differences representative of the original samples is affected by the quality of the extracted RNA. Thus, it is of critical importance to standardize sample-handling protocols and to perform a quality assessment of RNA preparations. In this report, 59 human tissue samples were used to evaluate the relationships between RNA quality and gene expression. From Affymetrix® GeneChip® array data analysis of these samples, we compared the performance of the 28S/18S ratio, two computer methods (RIN and Degradometer) and our in-house RNA Quality Scale (RQS) in assessing RNA quality. The optimal RNA reliability threshold was determined for each method using statistical discrimination measures. We showed that RQS, RIN and Degradometer have a similar capacity to detect reliable RNA samples whereas the 28S/18S ratio leads to a misleading categorization. Furthermore, we developed a new approach, based on clustering analyses of full chip expression, to control RNA quality after hybridization experiments. The combination of these methods, allowing monitoring of RNA quality prior to and after the hybrizidation experiments, ensured reliable and reproducible microarray data.3

After validation in an independent cohort of patients, our gene signature could be used as a decision tool to assist oncologists in selecting colorectal cancer patients who could benefit from FOLFIRI chemotherapy, both in the adjuvant and the first-line metastatic setting.