A synthetic peptide mimic kills *Candida albicans* and synergistically prevents infection

Candida spp. are a leading cause of hospital-acquired infections, with systemic candidiasis mortality exceeding 40%. Candida albicans accounts for about half of Candida bloodstream infections, and emerging multidrug-resistant pathogens such as Candida auris underscore the urgent need for new antifungals. Current approved drug classes (azoles, polyenes, echinocandins, flucytosine) have significant limitations including drug interactions, toxicity, and rising resistance due to target modification, overexpression, or efflux. Discovering new fungal targets is challenging given eukaryotic similarities to humans, and most pipeline agents remain close to existing classes. Combination therapy can mitigate resistance and host toxicity by acting on multiple targets. Natural antifungal peptides (AFPs) are potent but face production, cost, and stability constraints. Synthetic polymers that mimic cationic amphiphilic properties of AFPs offer a promising alternative; advances in reversible-deactivation radical polymerisation enable controlled synthesis, albeit without fully sequence-defined structures. Building on prior work identifying short amphiphilic polyacrylamides with good in vitro therapeutic index against C. albicans, this study defines the mode of action and evaluates in vitro and in vivo efficacy and synergy of the most promising compositions, focusing on enhancing efficacy, reducing toxicity, and limiting resistance development.

The study situates the need for new antifungals within increasing invasive fungal burden and resistance. Existing classes (azoles, polyenes, echinocandins, flucytosine) face resistance via target alterations and efflux and have toxicity/drug interaction issues. First-in-class agents (fosmanogepix, ibrexafungerp) are in trials but resistance risk remains. Combination strategies can resensitize resistant strains and reduce toxicity. Natural AFPs act via membrane disruption and intracellular stress but are limited by production and proteolysis. Synthetic antifungal polymers, including β-peptides, poly(2-oxazoline)s, polycarbonates, polyacrylamides, and peptidopolysaccharides, have shown promise. Photo-induced RDRP/PET-RAFT enables control over molecular weight and composition, though not precise sequences. The authors’ prior polyacrylamide library identified short (degree of polymerisation ~20) amphiphilic copolymers with strong anti-Candida activity and low mammalian toxicity, motivating mechanistic and therapeutic exploration here.

• Polymer design and synthesis: Four ternary amphiphilic polyacrylamides (LP, LH, CB, CX; DPn=20) differing in hydrophobic side chains (linear pentyl/heptyl; cyclic benzyl/hexyl) were synthesised via PET-RAFT and deprotected. Composition and MW were confirmed by 1H NMR and SEC (Đ 1.09–1.12; >98% conversion).
• Antifungal susceptibility: MICs (CLSI-modified broth microdilution) were determined against C. albicans SC5314 and multiple clinical drug-resistant isolates (azoles, amphotericin B, echinocandin resistance).
• Transcriptomics: C. albicans was exposed for 1 h to sub-inhibitory concentrations of polymers and controls (non-toxic poly-HEA, peptide LL-37). Microarrays were analysed by hierarchical clustering/PCA, and GO/KEGG enrichment identified pathways affected.
• Chemical-genetic profiling: Selected C. albicans deletion mutants spanning cell wall organization, glycosylation (O-/N-mannosylation), calcineurin and MAPK pathways, membrane/inositol signalling, efflux, and polyamine transport were screened for growth at sub-inhibitory polymer levels. Parallel tests in C. glabrata mutants were included.
• Membrane lysis assays: A GFP-expressing C. albicans strain assessed cytoplasmic leakage after treatment (LH vs amphotericin B vs tunicamycin). Tethered bilayer membranes reconstituted from lipids of C. albicans yeast/hyphae and human erythrocytes were used to quantify conductance changes upon LH or poly-HEA (and LL-37) exposure.
• Cell wall effects: TEM examined ultrastructure after sub-inhibitory treatments (LH, tunicamycin, poly-HEA). Isolated cell walls underwent acid hydrolysis and HPLC quantification of glucose (glucan) and mannose (mannan) to assess composition changes; och1Δ and caspofungin served as controls.
• Host-pathogen interactions: Human monocyte-derived macrophages (hMDMs) were used to quantify phagocytosis of C. albicans pre-treated with LH. PBMC cytokine responses (IL-1β, IL-6, TNF) to LH-pretreated fungi were measured.
• Morphogenesis: Hypha formation and microcolony assays under inducing conditions assessed LH and other polymers’ effects on filamentation kinetics.
• In vitro epithelial infection model (HECM): A-431 vaginal epithelial monolayers were infected with C. albicans in the presence of LH and/or antifungals (caspofungin, fluconazole). Damage was quantified by LDH release and propidium iodide staining; synergy pre-screen via checkerboard FIC index.
• In vivo efficacy: Galleria mellonella larvae were infected systemically with C. albicans and treated with LH, caspofungin, or combinations. Toxicity and survival up to 14 days were recorded.
• In vitro evolution and genomics: Serial passage at 1× MIC for 14 days with LH, fluconazole, caspofungin, or synergistic combinations. Post-evolution MICs were determined; whole-genome sequencing of evolved strains assessed aneuploidy, LOH, and copy number changes.

• Potent activity against resistant isolates: All four polymers were active against clinical C. albicans isolates with diverse resistance mechanisms. The most active polymer, LH, maintained or improved MICs versus wild type (e.g., 8–32 µg/mL [3–11 µM] for multiple resistant strains) and was unaffected by azole efflux mechanisms (Cdr/Mdr).
• Distinct mode of action: Transcriptomics revealed upregulation of ER protein processing/ERAD, glycosylation (including N-glycan biosynthesis), and MAPK stress responses; downregulation of ribosome, amino acid biosynthesis, oxidative phosphorylation. These signatures were specific to antifungal polymers (not poly-HEA or LL-37), suggesting ER stress and impaired protein glycosylation alongside cell wall/membrane stress.
• Genetic dependencies: Deletion mutants in O-mannosylation (PMT family), calcineurin (CRZ1, MID1), and MAPK (HOG1, PBS2) showed heightened susceptibility or no growth under polymer exposure. Mutants in inositol/phosphoinositide regulation (INP51) were hypersensitive. N-mannosylation mutants generally did not alter susceptibility. Efflux-related mutations did not impact activity.
• Membrane permeabilisation: LH caused dose-dependent GFP leakage and loss of viability (36% viability at 0.5× MIC; 3% at 1× MIC). Tethered membrane assays showed concentration-dependent conductance increases for C. albicans-derived lipids (stronger in yeast than hyphae) and lower effects on erythrocyte lipids compared with LL-37, indicating selective fungal membrane permeabilisation.
• Cell wall mannan disruption: TEM at sub-inhibitory LH showed disrupted N-mannan fibrils without overt membrane rupture. HPLC of purified walls showed decreased mannan proportion after LH treatment; controls behaved as expected (tunicamycin reduced mannan; caspofungin decreased glucan and increased mannan).
• Enhanced immune clearance: Pre-treatment with sub-inhibitory LH increased phagocytosis by primary human macrophages (significant reductions in non-phagocytosed fungi at 4–16 µg/mL). PBMC cytokine responses to LH-pretreated fungi showed reduced IL-1β, IL-6, and TNF at 4 µg/mL, consistent with altered mannan-mediated recognition.
• Reduced filamentation: Sub-inhibitory polymers, particularly LH, reduced hyphal length and microcolony size, potentially diminishing epithelial invasion.
• Synergy in vitro and on epithelial cells: Checkerboard assays identified synergistic or strongly additive interactions of LH with several agents, notably caspofungin and fluconazole (FIC ≤0.5–0.6). In the human epithelial cell infection model, LH alone did not prevent damage due to poor bioavailability; however, combinations were highly effective: 0.03 µg/mL caspofungin + 4 µg/mL LH reduced epithelial damage to 2% (no cytotoxicity), an eight-fold reduction in caspofungin MIC; 0.03 µg/mL fluconazole + 16 µg/mL LH reduced damage to 13%.
• In vivo efficacy: In Galleria mellonella, 100 mg/kg caspofungin alone protected 94.7% of larvae; 5 mg/kg caspofungin protected 10.5%. LH alone (250 mg/kg) did not protect. The combination (5 mg/kg caspofungin + 250 mg/kg LH) significantly improved survival to 42.1% at day 14 (p=0.003 vs 5 mg/kg caspofungin) and yielded 100% survival through day 6.
• Resistance evolution: Over 14 days at 1× MIC, tolerance emerged for single agents (LH, fluconazole, caspofungin) but not for LH+caspofungin and only minor for LH+fluconazole. LH-evolved strains exhibited increased MICs to multiple polymers and showed aneuploidies (e.g., chromosome 2 trisomy, previously linked to ER stress adaptation). No stable tolerant isolates emerged after combined LH+caspofungin or LH+fluconazole treatment.

The study demonstrates that amphiphilic synthetic peptide-mimicking polymers, especially LH, kill C. albicans via a multifaceted mechanism distinct from existing antifungal classes. Transcriptomic and genetic evidence indicates induction of ER stress and disruption of protein glycosylation, particularly impacting mannan-rich cell wall proteins, together with selective membrane permeabilisation. These combined actions reduce fungal viability, impair hyphal development, remodel the cell wall to enhance innate immune clearance, and are less susceptible to classic resistance mechanisms such as efflux or target modification. The synergism with caspofungin and fluconazole dramatically lowers required drug doses and overcomes limitations of LH bioavailability in epithelial models, translating to significant survival benefits in an invertebrate candidiasis model. Crucially, prolonged exposure to LH combinations did not yield genetically stable tolerance, suggesting that multi-target disruption alongside a standard-of-care agent can constrain evolutionary adaptation. These findings support incorporating such polymers as adjuvants to established antifungals to enhance efficacy, reduce toxicity through dose-sparing, and mitigate resistance development.

This work identifies a synthetic amphiphilic polyacrylamide (LH) as a potent anti-Candida agent with a distinct, multi-target mode of action involving ER stress, impaired protein glycosylation and mannan disruption, and membrane permeabilisation. LH retains activity against diverse drug-resistant clinical isolates, enhances macrophage-mediated clearance, reduces filamentation, and synergizes strongly with caspofungin and fluconazole to prevent epithelial damage in vitro and improve survival in an in vivo Galleria model. Importantly, combinations with LH did not produce tolerant strains during in vitro evolution. These results position LH as a promising lead for adjunctive antifungal therapy. Future work should optimise polymer architecture (e.g., monomer sequence/block order), improve bioavailability and delivery, evaluate pharmacokinetics/toxicity, and validate efficacy in vertebrate infection models, with the longer-term aim of standalone or combination clinical applications.

• Bioavailability constraints: LH alone did not protect epithelial cells in the human epithelial cell model, likely due to sequestration/adsorption by host components; activity of some combinations (e.g., with fluconazole) diminished in serum.
• Cytotoxicity at higher doses: Elevated LH concentrations can damage mammalian cells (e.g., >128 µg/mL in vitro), constraining therapeutic windows without combination strategies.
• Model limitations: Efficacy was demonstrated in vitro and in Galleria mellonella; vertebrate models and comprehensive PK/PD and toxicity profiles are lacking.
• Mechanistic resolution: While ER stress, glycosylation disruption, and membrane effects are implicated, precise molecular targets and uptake dynamics remain to be fully elucidated.
• Evolution experiments: Tolerance to single agents arose with aneuploidy changes; broader population-level dynamics and clinical relevance require further study.