Biology subjects

Azimi, S.

Publications and source records attributed to Azimi, S..

13 recordsLinked to original sources

bioRxiv2026

Robust and Quality-of-Life-Aware Treatment Protocols in NSCLC using Deep Reinforcement Learning

Under current systemic treatment of metastatic cancer, a drug is frequently prescribed at maximum tolerable dose (MTD) until either unacceptable toxicity or progression. Unfortunately, in many patients this treatment strategy leads to the development of treatment resistance. Evolutionary therapy approaches aim to forestall or delay treatment resistance in cancer by exploiting eco-evolutionary interactions. A well-known implementation is the adaptive therapy protocol of Zhang et al., in which tumour burden thresholds are used to guide strategic treatment holidays. Deep reinforcement learning (DRL) has recently been used to optimise these approaches. However, research combining DRL with evolutionary therapy approaches has so far focused on time to progression (TTP) as a performance metric, and has not quantified safety in terms of robustness to delayed treatment restart or included patient preferences regarding quality of life (QoL) in treatment design. In our study, we use a DRL agent informed by a mathematical two-population tumour growth model to design treatment schedules for patients with non-small cell lung cancer (NSCLC). The agent is trained on a virtual patient cohort using parameters previously fitted to data from patients with NSCLC treated with erlotinib. Beyond TTP, we focus on improving robustness to delayed treatment restart and on how individual preferences and values impact QoL experienced during treatment. We compare TTP, robustness and QoL under the DRL policy, the adaptive therapy protocol of Zhang et al., and MTD. We introduce a robustness metric ``margin-to-failure'' (MTF), and compare quality-adjusted-survival (QAS) across different patient preference profiles. Finally, we explore reward shaping to assess how QoL preferences can be incorporated into DRL-based treatment design. To evaluate our results, we consider different decision intervals, defined as the time between dosing adjustments. The DRL policy achieved greater median TTP, MTF, and QAS across all treatment decision intervals compared to the other two protocols. As decision intervals increased, TTP under DRL declined gradually towards that achieved under MTD. In contrast, the Zhang et al. protocol performed inconsistently and could result in premature progression. Additionally, a population-level policy trained on a cohort of virtual patients produced an interpretable treatment rule that extended TTP for most previously unseen patients and indicated that treatment should resume at a lower tumour burden when monitoring is less frequent. These findings show that DRL can balance the benefit of preserving drug-sensitive cells to suppress resistance against the risk of unsafe tumour regrowth. Reward shaping further showed how treatment strategies could be adjusted to reflect different patient preferences. Together, these results provide a biologically informed approach for designing robust and patient-centred evolutionary therapies in fast-growing cancers such as NSCLC.

systems biology↗

bioRxiv2026

Pseudomonas aeruginosa Intra-Population Diversity Shapes Host Airway Epithelial Responses

In chronic pulmonary infection airways are colonized by genetically and phenotypically diverse Pseudomonas aeruginosa populations, yet almost everything known about P. aeruginosa pathogenesis has been learned from single clinical isolates or laboratory reference strains studied in isolation. Whether a diverse population behaves as the sum of its members remains unclear. We infected differentiated primary cystic fibrosis (CF) airway epithelial cells (CF-pAECs) at air-liquid interface with whole P. aeruginosa populations collected from the sputum of three adults with CF, and, in parallel, with genetic variants retrieved from the same populations. We found that host responses to whole populations were not predicted by responses to their derivative variants, and mixed populations did not exhibit the average expected response of the members. Variation in host response was dominated by the tissue-remodeling mediators vascular endothelial growth factor (VEGF) and matrix metalloproteinase-9 (MMP-9), which differed distinctly between individual variants, whereas core pro-inflammatory cytokines (IL-6, IL-8, TNF-, IFN-{gamma}) varied comparatively little. Bacterial transcriptomes recorded during infection showed the same asymmetry. Clinical populations shared a transcriptional state distinct from PAO1, and mixed populations maintained stable expression of core regulatory, secretion, and DNA-repair loci (including hfq, xcpT, and ssb), while the derivative variants grown alone exhibited differential transcriptional profiles. Interactions among co-existing lineages therefore shape both bacterial physiology and host response, suggesting that the diverse population, not the single clone, is the appropriate unit of study in chronic infection.

microbiology↗

bioRxiv2026

Biophysical Constraints Dictate the Stability of Social Traits in Pseudomonas aeruginosa Aggregates

The maintenance of cooperative behaviors within microbial populations remains an evolutionary puzzle, particularly in spatially structured environments like those found in chronic infection. In Pseudomonas aeruginosa, quorum sensing (QS) directs the production of costly public goods that are vulnerable to exploitation by non-producing cheaters. However, the biophysical mechanisms that stabilize this cooperation in complex environments remain poorly understood. Here, we elucidate how the biophysical properties of the bacterial cell surface govern the micron-scale architecture of bacterial aggregates to promote or constrain the stability of cooperation. Using a polymer-structured growth environment, we demonstrate that cell surface hydrophobicity acts as a primary determinant of spatial organization, where hydrophilic wild-type cells assemble into "stacked" aggregates, whereas hydrophobic O-specific antigen-deficient cells form dense "clumps". In mixed populations, distinct surface properties exhibit phase-separation-like immiscibility, segregating the cells at the micron scale. We show that this segregation directly suppresses cheater fitness by physically sequestering cooperative clusters. Specifically, hydrophobic QS-deficient cells were effectively excluded by hydrophilic QS cooperators, limiting their access to public goods. Furthermore, invasion experiments revealed that hydrophilic cells are inherently fitter, capable of unidirectionally invading populations with a hydrophobic cell surface and independent of social dynamics. Conversely, the need to gain access to nutrient resources can fine-tune these barriers, enabling hydrophilic QS-deficient cells to cluster proximally to hydrophobic cooperators. These findings establish that bacterial cell surface traits impact strongly on microbial social interactions and offer new insights into the maintenance and loss of cooperative traits in chronic infections such as those found in chronic lung infections. Significance StatementCooperative behaviors are essential for the survival of pathogenic bacteria like Pseudomonas aeruginosa, yet they are continually threatened by "cheater" cells that exploit shared public good resources. We demonstrate that the physical properties of the cell surface, specifically hydrophobicity, act as a key architect of microbial social structure. In polymer-rich environments, differences in surface hydrophobicity drive a distinct "oil-and-water" segregation between cell populations. This micron-scale phase separation physically sequesters cooperative cells, insulating them from exploitation by cheaters. Consequently, spatial structure serves as a biophysical stabilizer of cooperation, limiting the metabolic advantages typically held by cheaters. Our findings demonstrate that microbial social evolution is governed not only by genetic strategies, but by fundamental biophysical constraints that dictate the spatial limits of exploitation.

microbiology↗

bioRxiv2025

Immuno-functionomics reveals geographical variation and a role for TLR8 in mRNA vaccine responses

The innate immune system plays a pivotal role in pathogen defense via pattern recognition receptor sensing, initiating responses upon infection or vaccination. Understanding its functional capacity is crucial for deciphering correlates of vaccine efficacy and understanding responses to infection. In this study, we developed a holistic approach to study immune function, generating >3100 readouts across 16 cell types, 18 pattern recognition receptors and 11 produced cytokines using spectral flow cytometry. To explore geographical variation, we studied the immune system of Europeans, urban, and rural Indonesians. We found differences in immune responses, such as increased IL1{beta} production in rural Indonesians and impaired IFN{gamma} production by innate lymphocytes after TLR8 stimulation. In Europeans vaccinated with mRNA-1273, baseline IFN{gamma} production by innate lymphocytes correlated with SARS-CoV-2 Spike-specific immune responses. In vitro mRNA vaccine stimulation also induced IFN{gamma} production, which was TLR8-dependent and reduced in rural Indonesians. This study highlights functional immune diversity and TLR8s potential role in mRNA vaccine responses. SummaryWe developed an approach to study the function capacity of the immune system, exploring geographical variation and mRNA vaccine responses. This revealed TLR8s potential role in responding to mRNA vaccines, and an impairment in this pathway in rural Indonesians.

immunology↗

bioRxiv2024

Migrastatic Therapy as a Potential Game-Changer inAdaptive Cancer Treatment

Adaptive therapy, which anticipates and counters the evolution of resistance in cancer cells, has gained significant traction, especially following the success of the Zhang et al.s protocol in treating metastatic castrate-resistant prostate cancer. While several adaptive therapies have now advanced to clinical trials, none currently incorporates migrastatics, i.e. treatments designed to inhibit cancer cell metastasis. In this study, we propose integrating migrastatics into adaptive therapy protocols and evaluate the potential benefits of using a game-theoretic spatial model. Our results demonstrate that combining adaptive therapy with migrastatics effectively delays the onset of metastasis and reduces both the number and size of metastases across most cancer scenarios analyzed. This approach not only extends the time to the first metastasis but also enhances the overall efficacy of adaptive therapies. Our findings suggest a promising new direction for cancer treatment, where adaptive therapy, in combination with migrastatic agents, can target both the evolution of resistance and the metastatic spread of cancer cells.

cancer biology↗

bioRxiv2024

Loss of O-specific antigen shapes Pseudomonas aeruginosa microbiogeography in murine preclinical pulmonary infection model

Chronic Pseudomonas aeruginosa infections are the hallmark of late-stage lung disease in individuals with cystic fibrosis. During chronic infection P. aeruginosa becomes the dominant bacteria in the airways. Within-host adaptation of P. aeruginosa leads to vast phenotypic and genetic population heterogeneity. In vitro studies show mutations in lipopolysaccharide (LPS) O-specific antigen changes the aggregate formation in P. aeruginosa, however role of these changes in aggregate assembly in vivo is not understood. Using a synthetic CF sputum media and a preclinical murine infection model we assessed how the PAO1 wildtype and O-specific antigen mutants interact with each other, and if P. aeruginosa population heterogeneity affects the colonization of the murine lungs. Our findings suggest that the presence of variants lacking O-specific antigen does not impact the population fitness and size in both in vitro and in vivo, however it can influence the aggregate volume in vivo.

microbiology↗

bioRxiv2023

Unraveling Genomic Diversity in Pseudomonas aeruginosa Cystic Fibrosis Lung Infection and Its Impact on Antimicrobial Resistance

Pseudomonas aeruginosa is an opportunistic pathogen which causes chronic, drug-resistant lung infections in cystic fibrosis (CF) patients. In this study, we explore the role of genomic diversification and evolutionary trade-offs in antimicrobial resistance (AMR) diversity within P. aeruginosa populations sourced from CF lung infections. We analyzed 300 clinical isolates from four CF patients (75 per patient), and found that genomic diversity is not a consistent indicator of phenotypic AMR diversity. Remarkably, some genetically less diverse populations showed AMR diversity comparable to those with significantly more genetic variation. We also observed that hypermutator strains frequently exhibited increased sensitivity to antimicrobials, contradicting expectations from their treatment histories. Investigating potential evolutionary trade-offs, we found no substantial evidence of collateral sensitivity among aminoglycoside, beta-lactam, or fluoroquinolone antibiotics, nor did we observe trade-offs between AMR and growth in conditions mimicking CF sputum. Our findings suggest that (i) genomic diversity is not a prerequisite for phenotypic AMR diversity; (ii) hypermutator populations may develop increased antimicrobial sensitivity under selection pressure; (iii) collateral sensitivity is not a prominent feature in CF strains, and (iv) resistance to a single antibiotic does not necessarily lead to significant fitness costs. These insights challenge prevailing assumptions about AMR evolution in chronic infections, emphasizing the complexity of bacterial adaptation during infection. ImportanceUpon infection in the cystic fibrosis (CF) lung, Pseudomonas aeruginosa rapidly acquires genetic mutations, especially in genes involved in antimicrobial resistance (AMR), often resulting in diverse, treatment-resistant populations. However, the role of bacterial population diversity within the context of chronic infection is still poorly understood. In this study, we found that hypermutator strains of P. aeruginosa in the CF lung undergoing treatment with tobramycin evolved increased sensitivity to tobramycin relative to non-hypermutators within the same population. This finding suggests that antimicrobial treatment may only exert weak selection pressure on P. aeruginosa populations in the CF lung. We further found no evidence for collateral sensitivity in these clinical populations, suggesting that collateral sensitivity may not be a robust, naturally occurring phenomenon for this microbe. Preprint servers: This manuscript has been submitted as a preprint to bioRxiv

microbiology↗

bioRxiv2021

Cell surface hydrophobicity determines Pseudomonas aeruginosa aggregate assembly

Bacteria live in spatially organized aggregates during chronic infections, where they adapt to the host environment, evade immune responses and resist therapeutic interventions. Although it is known that environmental factors such as polymers influence bacterial aggregation, it is not clear how bacterial adaptation during chronic infection impacts the formation and spatial organization of aggregates in the presence of polymers. Here we show that in an in vitro model of cystic fibrosis (CF) containing the polymers eDNA and mucin, O-specific antigen is a major factor in determining the formation of two distinct aggregate assembly types of Pseudomonas aeruginosa due to alterations in cell surface hydrophobicity. Our findings suggest that during chronic infection, interplay between cell surface properties and polymers in the environment may influence the formation and structure of bacterial aggregates, which would shed new light on the fitness costs and benefits of O-antigen production in environments such as CF lungs. ImportanceDuring chronic infection, several factors contribute to the biogeography of microbial communities. Heterogeneous populations of Pseudomonas aeruginosa form aggregates in cystic fibrosis airways, however, the impact of this population heterogeneity on spatial organization and aggregate assembly is not well understood. In this study we found that changes in O-specific antigen determine the spatial organization of P. aeruginosa cells by altering the relative cell surface hydrophobicity. This finding suggests a role for O-antigen in regulating P. aeruginosa aggregate size and shape in cystic fibrosis airways.

microbiology↗

bioRxiv2020

The evolution of virulence in Pseudomonas aeruginosa during chronic wound infection

Opportunistic pathogens are associated with a number of chronic human infections, yet the evolution of virulence in these organisms during chronic infection remains poorly understood. Here, we tested the evolution of virulence in the human opportunistic pathogen Pseudomonas aeruginosa in a murine chronic wound model using a two-part serial passage and sepsis experiment, and found that virulence evolved in different directions in each line of evolution. We also assessed P. aeruginosa adaptation to a chronic wound after 42 days of evolution, and found that morphological diversity in our evolved populations was limited compared to that previously described in cystic fibrosis (CF) infections. Using whole genome sequencing, we found that genes previously implicated in P. aeruginosa pathogenesis (lasR, pilR, fleQ, rpoN, and pvcA), acquired mutations during the course of evolution in wounds, with some mutations evolving in parallel across all lines of evolution. Our findings highlight that (i) P. aeruginosa heterogeneity may be less extensive in chronic wounds than in CF lungs; (ii) genes involved in P. aeruginosa pathogenesis acquire mutations during chronic wound infection; (iii) similar genetic adaptations are employed by P. aeruginosa across multiple infection environments, and (iv) current models of virulence may not adequately explain the diverging evolutionary trajectories observed in an opportunistic pathogen during chronic wound infection.

microbiology↗

bioRxiv2019

Allelic polymorphism shapes collective phenotypes in evolving Pseudomonas aeruginosa populations

Pseudomonas aeruginosa is an opportunistic pathogen that chronically infects the lungs of individuals with cystic fibrosis (CF) by forming antibiotic resistant biofilms. Emergence of phenotypically diverse isolates within CF P. aeruginosa populations has previously been reported, however, the impact of heterogeneity on social behaviors and community function is poorly understood. Here we describe how this heterogeneity impacts on behavioral traits by evolving the strain PAO1 in biofilms grown in a synthetic sputum medium for 50 days. We measured social trait production and antibiotic tolerance and used a metagenomic approach to analyze and assess genomic changes over the duration of the evolution experiment. We found that (i) evolutionary trajectories were reproducible in independently evolving populations; (ii) over 60% of genomic diversity occurred within the first 10 days of selection. We then focused on quorum sensing (QS), a well-studied P. aeruginosa trait that is commonly mutated in strains isolated from CF lungs. We found that at the population level (i) evolution in sputum medium selected for decreased production of QS and QS-dependent traits; (ii) there was a significant correlation between lasR mutant frequency, the loss of protease and the 3O-C12-HSL signal, and an increase in resistance to clinically relevant {beta}-lactam antibiotics, despite no previous antibiotic exposure. Overall, our findings provide insights into the effect of allelic polymorphism on community functions in diverse P. aeruginosa populations. Further, we demonstrate that P. aeruginosa population and evolutionary dynamics can impact on traits important for virulence and can lead to increased tolerance to {beta}-lactam antibiotics.

microbiology↗

bioRxiv2018

Combinatorial quorum sensing in Pseudomonas aeruginosa allows for novel cheating strategies

In the opportunistic pathogen Pseudomonas aeruginosa, quorum sensing (QS) is a social trait that is exploitable by non-cooperating cheats. Previously it has been shown that by linking QS to the production of both public and private goods, cheats can be prevented from invading populations of cooperators and this has been termed a metabolic incentive to cooperate. We hypothesized P. aeruginosa could evolve novel cheating strategies to circumvent private goods metabolism by rewiring its combinatorial response to two QS signals (3O-C12-HSL and C4-HSL). We performed a selection experiment that cycled P. aeruginosa between public and private goods growth media and evolved an isolate which rewired its control of cooperative protease expression from a synergistic (AND-gate) response to dual signal input, to a 3O-C12-HSL only response. We show that this isolate circumvents metabolic incentives to cooperate and acts as a combinatorial signaling cheat, with a higher fitness in competition with its ancestor. Our results show three important principles; first, combinatorial QS allows for diverse social strategies to emerge; second, that restrictions levied by private goods are not sufficient to explain the maintenance of cooperation in natural populations and third that modifying combinatorial QS responses could result in important physiological outcomes in bacterial populations.

microbiology↗

bioRxiv2018

A role for Fibroblast Growth Factor Receptor 1 in the pathogenesis of Neisseria meningitidis

Neisseria meningitidis remains an important cause of human disease. It is highly adapted to the human host - its only known reservoir. Adaptations to the host environment include many specific interactions with human molecules including iron-binding proteins, components of the innate and adaptive immune systems, and cell surface receptors such as the Epidermal Growth Factor Receptor (EGFR). Interaction of the meningococcus with EGFR has been elucidated in some detail and leads to intracellular signalling and cytoskeletal changes contributing to the pathogenesis of the organism. Here, we show that the meningococcus also recruits Fibroblast Growth Factor Receptor 1 (FGFR1) onto the surface of human blood microvascular epithelial cells (HBMECs). Furthermore, meningococci internalised into these cells recruit the activated form of this receptor, and that expression and activation of FGFR1 is necessary for efficient internalisation of meningococci into HBMECs. We show that Neisseria meningitidis interacts specifically with the IIIc isoform of FGFR1.

microbiology↗

bioRxiv2017

A simple mung bean infection model for studying the virulence of Pseudomonas aeruginosa

Here we highlight the development of a simple and high throughput mung bean model to study virulence in the opportunistic pathogen Pseudomonas aeruginosa. The model is easy to setup and infection and virulence can be monitored for up to 10 days. In a first test of the model, we found that mung bean seedlings infected with PAO1 showed poor development of roots and high mortality rates compared to un-infected controls. We also found that a quorum sensing (QS) mutant was significantly less virulent when compared with the PAO1 wild type. Our work introduces a new tool for studying virulence in P. aeruginosa, that will allow for high throughput virulence studies of mutants, and for testing the in vivo efficacy of new therapies at a time when new antimicrobial drugs are desperately needed.

microbiology↗