Polarized epithelial non-human (canine) cell lines stably transfected with human or murine complementary DNA (cDNA) encoding for various efflux transporters (P-gp/MDR1, MRP1, MRP2, and Bcrp1) were used to study transepithelial transport of Lopinavir (LVR) and compare results with the MDCKII-Wild type cells. These transmembrane proteins cause multidrug resistance by decreasing the total intracellular accumulation of drugs. Lopinavir efflux was directional and was completely inhibited by MK-571, a selective MRP family inhibitor in the MDCKII-MRP2 cell line. Similarly, LVR efflux was also inhibited by P-gp inhibitors P-gp 4008 and GF120918 in the MDCKII-MDR1 cell line. The efflux ratios (Efflux rate/ Influx rate) of LVR in the absence of any efflux inhibitors in the MDCK-Wild type, MDCKII-MDR1, MDCKII-MRP1 and MDCKII-MRP2 cell monolayers were 1.32, 4.91, 1.26 and 2.89 respectively. The MDCKII-MDR1 and MDCKII-MRP2 cells have significantly increased LVR efflux ratio relative to the parental cells due to the apically directed transport by MDR1 and MRP2 respectively. The efflux ratios in MRP2 and MDR1 transfected cell lines were close to unity in the presence of MK-571 and P-gp 4008 respectively; indicating that LVR efflux by MRP2 and P-gp was completely inhibited by their selective inhibitors. MDCKII-MRP1 cells did not exhibit a significant reduction in the LVR efflux relative to the parental cells, indicating that LVR is not a good substrate for MRP1. Transport studies across MDCKII-Bcrp1 cells indicated that LVR is not transported by Bcrp1 and is not a substrate for this efflux protein. In conclusion, this study presents direct evidence that LVR is effluxed by both P-gp and MRP2 which may contribute to its poor oral bioavailability and limited penetration into the CNS. KeywordsMDCKII-MDR1; MDCKII-MRP2; MDCKII-MRP1; MDCKII-Bcrp1; MDCKII-WT; Pglycoprotein (P-gp); multidrug resistance protein (MRP); breast cancer resistance protein (BCRP); Lopinavir (LVR); uptake; transport; permeability; efflux ratio (ER)

No abstract

Lopinavir (LVR) is extensively metabolized by CYP3A4 and is prevented from entering the cells by membrane efflux pumps such as P-gp and MRP2. In an approach to evade the first-pass metabolism and efflux of LVR, peptide prodrugs of LVR [valine-valine-lopinavir (VVL) and glycine-valinelopinavir (GVL)] were synthesized. Prodrugs were identified with 1 H and 13 C NMR spectra and LC/ MS/MS was employed to evaluate their mass and purity. Solubility studies indicated that the prodrugs have much greater solubility as compared with LVR in water. In vitro evaluations were performed to determine affinities for efflux proteins (P-gp and MRP2) and CYP3A4 and permeabilities across intestinal barrier. Accumulation and transport data of VVL and GVL across MDCKII-MDR1 and MDCKII-MRP2 cells indicated evasion of prodrugs' efflux by P-gp and MRP2 significantly. Permeability studies across Caco-2 cells indicated that the prodrugs are transported by peptide transporters and have increased permeability as compared with LVR. VVL and GVL exhibited significantly better degradation rate constants as compared with LVR in rat liver microsomes. Enzymatic stability studies in Caco-2 cell homogenate indicated that the peptide prodrugs are first converted to the ester intermediate and then finally to the parent drug. Overall, the advantages of utilizing peptide prodrugs include chemical modification of the compound to achieve targeted delivery via peptide transporters present across the intestinal epithelium, significant evasion of efflux and CYP3A4 mediated metabolism and significantly better solubility profiles. Therefore, in vitro studies demonstrated that peptide prodrug derivatization of LVR may be an effective strategy for bypassing its efflux and enhancing its systemic bioavailability.

Abstract. The performance of parallel applications running on large clusters is known to degrade due to the interference of kernel and daemon activities on individual nodes, often referred to as noise. In this paper, we focus on an important class of parallel applications, which repeatedly perform computation, followed by a collective operation such as a barrier. We model this theoretically and demonstrate, in a rigorous way, the effect of noise on the scalability of such applications. We study three natural and important classes of noise distributions: The exponential distribution, the heavy-tailed distribution, and the Bernoulli distribution. We show that the systems scale well in the presence of exponential noise, but the performance goes down drastically in the presence of heavy-tailed of Bernoulli noise.

No abstract

No abstract

Due to its decentralized nature, Federated Learning (FL) lends itself to adversarial attacks in the form of backdoors during training. The goal of a backdoor is to corrupt the performance of the trained model on specific sub-tasks (e.g., by classifying green cars as frogs). A range of FL backdoor attacks have been introduced in the literature, but also methods to defend against them, and it is currently an open question whether FL systems can be tailored to be robust against backdoors. In this work, we provide evidence to the contrary. We first establish that, in the general case, robustness to backdoors implies model robustness to adversarial examples, a major open problem in itself. Furthermore, detecting the presence of a backdoor in a FL model is unlikely assuming first order oracles or polynomial time. We couple our theoretical results with a new family of backdoor attacks, which we refer to as edge-case backdoors. An edge-case backdoor forces a model to misclassify on seemingly easy inputs that are however unlikely to be part of the training, or test data, i.e., they live on the tail of the input distribution. We explain how these edge-case backdoors can lead to unsavory failures and may have serious repercussions on fairness, and exhibit that with careful tuning at the side of the adversary, one can insert them across a range of machine learning tasks (e.g., image classification, OCR, text prediction, sentiment analysis).

No abstract