Biology subjects

Durand, A.

Publications and source records attributed to Durand, A..

28 records · Page 2Linked to original sources

bioRxiv2022

Luminal progenitors undergo partial epithelial-to-mesenchymal transition at the onset of basal-like breast tumorigenesis

Defects in double-strand repair mechanisms - both through germline or somatic inactivation of repair genes - is a hallmark of basal-like breast cancers. In this genetically-unstable context, a recurrent shift in cell identity occurs within the mammary epithelium. Basal-like tumors have indeed been proposed to originate from luminal progenitor (LP) cells yet tumor-initiating events remain poorly understood. Here, we map state transitions at the onset of basal-like tumorigenesis, using a Brca-1 deficient mouse model launching tumorigenesis in multiple LP cells. Combining single-cell transcriptomics to spatial multiplex imaging, we identify a population of cycling p16-expressing cells, emerging from the luminal progenitor compartment, undergoing partial epithelial-to-mesenchymal transition and losing luminal identity. Pseudo-temporal analyses position these cells as a transitory pre-tumoral state between aberrant Brca1-deficient luminal progenitors and growing tumor cells. In patients, the p16 pre-tumoral signature is found in early stage basal-like tumors, that rarely recur. Concomitant to p16 activation, we show that LP cells undergo an epigenomic crisis attested by the general re-organization of their heterochromatin. They accumulate multiple H3K27me3 micro-foci - reminiscent of the formation of senescence-associated heterochromatin foci (SAHFs) - and lose their inactive X (Xi). Both p16 activation and heterochromatin reorganization are hallmarks of human basal-like breast tumors; we propose that these events occur during initial LP transformation and are scars of an initial transitory senescent-like state.

cancer biology↗

bioRxiv2021

The structure of HiSiaQM defines the architecture of tripartite ATP-independent periplasmic (TRAP) transporters

Tripartite ATP-independent periplasmic (TRAP) transporters are widespread in bacteria and archaea and provide important uptake routes for many metabolites 1-3. They consist of three structural domains, a soluble substrate-binding protein (P-domain), and two transmembrane domains (Q- and M-domains) that form a functional unit 4. While the structures of the P-domains are well-known, an experimental structure of any QM-domain has been elusive. HiSiaPQM is a TRAP transporter for the monocarboxylate sialic acid, which plays a key role in the virulence of pathogenic bacteria 5. Here, we present the first cryo-electron microscopy structure of the membrane domains of HiSiaPQM reconstituted in lipid nanodiscs. The reconstruction reveals that TRAP transporters consist of 15 transmembrane helices and are structurally related to elevator-type transporters, such as GltPh and VcINDY 6, 7. Whereas the latter proteins function as multimers, the idiosyncratic Q-domain of TRAP transporters enables the formation of a monomeric elevator architecture. Structural and mutational analyses together with an AlphaFold 8 model of the tripartite (PQM) complex reveal the structural and conformational coupling of the substrate-binding protein to the transporter domains. Furthermore, we characterize high-affinity VHHs that bind to the periplasmic side of HiSiaQM and inhibit sialic acid uptake in vivo. Thereby, they also confirm the orientation of the protein in the membrane. Our study provides the first structure of any binding-protein dependent secondary transporter and provides starting points for the development of specific inhibitors.

biochemistry↗

bioRxiv2021

MICRA-Net: MICRoscopy Analysis Neural Network to solve detection, classification, and segmentation from a single simple auxiliary task

High throughput quantitative analysis of microscopy images presents a challenge due to the complexity of the image content and the difficulty to retrieve precisely annotated datasets. In this paper we introduce a weakly-supervised MICRoscopy Analysis neural network (MICRA-Net) that can be trained on a simple main classification task using image-level annotations to solve multiple the more complex auxiliary semantic segmentation task and other associated tasks such as detection or enumeration. MICRA-Net relies on the latent information embedded within a trained model to achieve performances similar to state-of-the-art architectures when no precisely annotated dataset is available. This learnt information is extracted from the network using gradient class activation maps, which are combined to generate detailed feature maps of the biological structures of interest. We demonstrate how MICRA-Net significantly alleviates the Expert annotation process on various microscopy datasets and can be used for high-throughput quantitative analysis of microscopy images.

bioinformatics↗

bioRxiv2021

H3K27me3 is a determinant of chemotolerance in triple-negative breast cancer

Triple-negative breast cancer is associated with the worst prognosis and the highest risk of recurrence among all breast cancer subtypes1. Residual disease, formed by cancer cells persistent to chemotherapy, remains one of the major clinical challenges towards full cure2,3. There is now consensus that non-genetic processes contribute to chemoresistance in various tumor types, notably through the initial emergence of a reversible chemotolerant state4-6. Understanding non-genetic tumor evolution stands now as a prerequisite for the design of relevant combinatorial approaches to delay recurrence. Here we show that the repressive histone mark H3K27me3 is a determinant of cell fate under chemotherapy exposure, monitoring epigenomes, transcriptomes and lineage with single-cell resolution. We identify a reservoir of persister basal cells with EMT markers and activated TGF-{beta} pathway leading to multiple chemoresistance phenotypes. We demonstrate that, in unchallenged cells, H3K27 methylation is a lock to the expression program of persister cells. Promoters are primed with both H3K4me3 and H3K27me3, and removing H3K27me3 is sufficient for their transcriptional activation. Leveraging lineage barcoding, we show that depleting H3K27me3 alters tumor cell fate under chemotherapy insult - a wider variety of tumor cells tolerate chemotherapy. Our results highlight how chromatin landscapes shape the potential of unchallenged cancer cells to respond to therapeutic stress.

cancer biology↗

bioRxiv2020

β2-Arrestin germline knockout does not attenuate opioid respiratory depression

Opioids are perhaps the most effective analgesics in medicine. However, from 1999 to 2018, they also killed more than 400,000 people in the United States by suppressing breathing, a common side-effect known as opioid induced respiratory depression. This doubled-edged sword has inspired the dream of developing novel therapeutics that provide opioid-like analgesia without respiratory depression. One such approach has been to develop so-called biased agonists that activate some, but not all pathways downstream of the {micro}-opioid receptor (MOR), the target of morphine and other opioid analgesics. This hypothesis stems from a study suggesting that MOR-mediated activation of {beta}2-Arrestin is the downstream signaling pathway responsible for respiratory depression, whereas inhibition of adenylyl cyclase produces analgesia. To further verify this model, which represents the motivation for the biased agonist approach, we examined respiratory behavior in mice lacking the gene for {beta}2-Arrestin. Contrary to previous findings, we find no correlation between {beta}2-Arrestin function and opioid-induced respiratory depression, suggesting that any effect of biased agonists must be mediated through an as-yet to be identified signaling mechanism.

neuroscience↗

bioRxiv2020

Loss of resistance to punishment of cocaine use after prior experience

One behavioral feature of drug addiction is continued drug use despite awareness that this causes negative consequences. Attempts to model this feature in animals typically involve punishing drug self-administration with a brief electrical footshock and look for resistance to punishment. Though all individual animals eventually stop self-administering the drug with increasing intensity of punishment, some individuals do so at higher intensities than other individuals. The greater relative resistance to punishment of the former individuals is generally interpreted as evidence for a compulsion-like behavior. Here we show that resistance to footshock punishment is in fact not a stable individual behavioral feature. Specifically, when rats are retested for their resistance to increasing intensity of footshock punishment, they become much less resistant. As a result, they suppress their cocaine intake even when punished with an initially low and ineffective intensity. A series of original behavioral experiments reveals that this low resistance to footshock punishment is rapidly acquired after rats experience a punishment intensity that leads them to near-completely suppress their cocaine intake. Passive exposure to the same intensity does not induce this effect. Once acquired, low resistance to punishment persists during at least one month, but can nevertheless be extinguished by retesting rats on a daily basis. Interestingly, this acquired low resistance to footshock punishment does not generalize to a non-painful form of punishment (i.e., histamine) that is also seldom used in animal drug self-administration studies. We discuss some possible theoretical and methodological implications of these findings for future research on animal models of compulsion-like behavior.

neuroscience↗

bioRxiv2020

Pharmacokinetics trumps pharmacodynamics during cocaine choice: a reconciliation with the dopamine hypothesis of addiction

Cocaine is known to increase brain dopamine at supranormal levels in comparison to alternative nondrug rewards. According to the dopamine hypothesis of addiction, this difference would explain, at least in part, why the latter are eventually given up in favor of continued cocaine use during the transition to addiction. Though resting on solid neuroscientific foundations, this hypothesis has nevertheless proven difficult to reconcile with research on cocaine choice in experimental animals. When facing a choice between an intravenous bolus of cocaine and a nondrug alternative (e.g., sweet water), both delivered immediately after choice, rats do not choose the drug, as would be predicted, but instead develop a strong preference for the nondrug alternative, sometimes to the exclusion of continued drug use. Here we report converging evidence that reconciles this finding with the dopamine hypothesis of addiction. Briefly, our data suggest that cocaine is indeed supranormal in reward magnitude, as postulated by the dopamine hypothesis of addiction, but is less preferred during choice because its pharmacokinetics makes it an inherently more delayed reward than the alternative. Reframing previous drug choice studies in rats as intertemporal choice studies reveals that the discounting effects of delays spare no rewards, including supranormal ones, and that during choice, pharmacokinetics trumps pharmacodynamics. Finally, this study also reveals important gaps in our understanding of drug reward delays that need to be filled by future experimental and theoretical work.

neuroscience↗

bioRxiv2019

CD44 regulates epigenetic plasticity by mediating iron endocytosis

CD44 is a transmembrane glycoprotein that is linked to various biological processes reliant on the epigenetic plasticity of cells, including development, inflammation, immune responses, wound healing and cancer progression. While thoroughly studied, functional regulatory roles of this so-called cell surface marker remain elusive. Here, we report the discovery that CD44 mediates endocytosis of iron interacting with hyaluronates in tumorigenic cell lines and primary cancer cells. We found that this glycan-mediated iron endocytosis mechanism is enhanced during epithelial-mesenchymal transition, unlike the canonical transferrin-dependent pathway. This transition is further characterized by molecular changes required for iron-catalyzed oxidative demethylation of the repressive histone mark H3K9me2 that governs the expression of mesenchymal genes. CD44 itself is transcriptionally regulated by nuclear iron, demonstrating a positive feedback loop, which is in contrast to the negative regulation of transferrin receptor by excess iron. Finally, we show that epigenetic plasticity can be altered by interfering with iron homeostasis using small molecules. This comprehensive study reveals an alternative iron uptake mechanism that prevails in the mesenchymal state of mammalian cells, illuminating a central role of iron as a rate-limiting regulator of epigenetic plasticity.

cell biology↗

bioRxiv2018

A simple microbiome in esophagus and gills of the European common cuttlefish, Sepia officinalis

The European common cuttlefish, Sepia officinalis, is used extensively in biological and biomedical research yet its microbiome remains poorly characterized. We analyzed the microbiota of the digestive tract, gills, and skin in mariculture-raised S. officinalis using a combination of 16s rRNA amplicon sequencing and fluorescence spectral imaging. Sequencing revealed a highly simplified microbiota consisting largely of two single bacterial amplicon sequence variants (ASVs) of Vibrionaceae and Piscirickettsiaceae. The esophagus was dominated by a single ASV of the genus Vibrio. Imaging revealed a striking organization of bacteria distributed in a discrete layer that lines the esophagus. Imaging with specific probes confirmed the identity of these bacteria as Vibrionaceae. This Vibrio was also abundant in the microbiota of the stomach, cecum, and intestine, but occurred at lower density and in the lumen rather than in a discrete layer; it was present in only trace proportions in tank water and in the microbiome of shrimp that were used as feed for the cuttlefish. These Vibrio were resilient to treatment of animals with the commonly-used antibiotic, enrofloxacin. The gills were colonized by a single ASV in the family Piscirickettsiaceae, which imaging visualized as small clusters of cells. We conclude that bacteria belonging to the Gammaproteobacteria, especially Vibrionaceae, are the major symbionts of the cuttlefish Sepia officinalis cultured from eggs in captivity, and that the esophagus and gills are major colonization sites.\n\nIMPORTANCEMicrobes can play critical roles in the physiology of their animal hosts, as evidenced in cephalopods by the role of Vibrio (Aliivibrio) fischeri in the light organ of the bobtail squid and the role of Alpha- and Gammaproteobacteria in the reproductive system and egg defense in a variety of cephalopods. We sampled the cuttlefish microbiome throughout the digestive tract, gills, and skin and found dense colonization of an unexpected site, the esophagus, by a microbe of the genus Vibrio, as well as colonization of gills by Piscirickettsiaceae. We found these associations to be resilient to the treatment of animals with a common antibiotic, enrofloxacin. This finding expands the range of organisms and body sites known to be associated with Vibrio and is of potential significance for understanding host-symbiont associations as well as for understanding and maintaining the health of cephalopods in mariculture.

microbiology↗

bioRxiv2018

Machine Learning to Predict Osteoporotic Fracture Risk from Genotypes

BackgroundGenomics-based prediction could be useful since genome-wide genotyping costs less than many clinical tests. We tested whether machine learning methods could provide a clinically-relevant genomic prediction of quantitative ultrasound speed of sound (SOS)--a risk factor for osteoporotic fracture.\n\nMethodsWe used 341,449 individuals from UK Biobank with SOS measures to develop genomically-predicted SOS (gSOS) using machine learning algorithms. We selected the optimal algorithm in 5,335 independent individuals and then validated it and its ability to predict incident fracture in an independent test dataset (N = 80,027). Finally, we explored whether genomic pre-screening could complement a UK-based osteoporosis screening strategy, based on the validated tool FRAX.\n\nResultsgSOS explained 4.8-fold more variance in SOS than FRAX clinical risk factors (CRF) alone (r2 = 23% vs. 4.8%). A standard deviation decrease in gSOS, adjusting for the CRF-FRAX score was associated with a higher increased odds of incident major osteoporotic fracture (1,491 cases / 78,536 controls, OR = 1.91 [1.70-2.14], P = 10-28) than that for measured SOS (OR = 1.60 [1.50-1.69], P = 10-52) and femoral neck bone mineral density (147 cases / 4,594 controls, OR = 1.53 [1.27-1.83], P = 10-6). Individuals in the bottom decile of the gSOS distribution had a 3.25-fold increased risk of major osteoporotic fracture (P = 10-18) compared to the top decile. A gSOS-based FRAX score, identified individuals at high risk for incident major osteoporotic fractures better than the CRF-FRAX score (P = 10-14). Introducing a genomic pre-screening step into osteoporosis screening in 4,741 individuals reduced the number of required clinical visits from 2,455 to 1,273 and the number of BMD tests from 1,013 to 473, while only reducing the sensitivity to identify individuals eligible for therapy from 99% to 95%.\n\nInterpretationThe use of genotypes in a machine learning algorithm resulted in a clinically-relevant prediction of SOS and fracture, with potential to impact healthcare resource utilization.\n\nResearch in ContextO_ST_ABSEvidence Before this StudyC_ST_ABSGenome-wide association studies have identified many loci associated with risk of clinically-relevant fracture risk factors, such as SOS. Yet, it is unclear if such information can be leveraged to identify those at risk for disease outcomes, such as osteoporotic fractures. Most previous attempts to predict disease risk from genotypes have used polygenic risk scores, which may not be optimal for genomic-prediction. Despite these obstacles, genomic-prediction could enable screening programs to be more efficient since most people screened in a population are not determined to have a level of risk that would prompt a change in clinical care. Genomic pre-screening could help identify individuals whose risk of disease is low enough that they are unlikely to benefit from screening.\n\nAdded Value of this StudyUsing a large dataset of 426,811 individuals we trained and tested a machine learning algorithm to genomically-predict SOS. This metric, gSOS, had performance characteristics for predicting fracture risk that were similar to measured SOS and femoral neck BMD. Implementing a gSOS-based pre-screening step into the UK-based osteoporosis treatment guidelines reduced the number of individuals who would require screening clinical visits and skeletal testing by approximately 50%, while having little impact on the sensitivity to identify individuals at high risk for osteoporotic fracture.\n\nImplications of all of the Available EvidenceClinically-relevant genomic prediction of heritable traits is feasible using the machine learning algorithm presented here in large sample sizes. Genome-wide genotyping is now less expensive than many clinical tests, needs to be performed once over a lifetime and could risk stratify for multiple heritable traits and diseases years prior to disease onset, providing an opportunity for prevention. The implementation of such algorithms could improve screening efficiency, yet their cost-effectiveness will need to be ascertained in subsequent analyses.

genomics↗