De Novo Design of Miniprotein Inhibitors of Bacterial Adhesins

De Novo Design of Miniprotein Inhibitors of Bacterial Adhesins | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article De Novo Design of Miniprotein Inhibitors of Bacterial Adhesins Adam Chazin-Gray, Tuscan Thompson, Edward Lopatto, Pearl Magala, and 26 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7951484/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted You are reading this latest preprint version Abstract The rise of multidrug-resistant bacterial infections necessitates the discovery of novel antimicrobial strategies. Here, we show that protein design provides a generalizable means of generating new antimicrobials by neutralizing the function of bacterial adhesins, which are virulence factors critical in host-pathogen interactions. We de novo designed high-affinity miniprotein binders to FimH and Abp chaperone usher pili adhesins from uropathogenic Escherichia coli and Acinetobacter baumannii, respectively, which are implicated in mediating both uncomplicated and catheter-associated urinary tract infections (UTI) responsible for significant morbidity worldwide. The designed antagonists have high specificity and stability, disrupt bacterial recognition of host receptors, block biofilm formation, and are effective in treating and preventing murine models of uncomplicated and catheter-associated UTIs in vivo. Biological sciences/Microbiology/Antimicrobials/Antimicrobial resistance Biological sciences/Biochemistry/Proteins/Lectins Biological sciences/Drug discovery/Biologics/Antibody fragment therapy Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Main Multidrug-resistant (MDR) bacterial pathogens constitute a significant and growing threat to public health ( 1 ). To counter the alarming rise of these increasingly untreatable and deadly bacterial infections, accelerating the development of novel antimicrobial therapies is essential. Recent advances in protein design have opened the door to designing small (< 120 amino acids) de novo protein binders (minibinders) that precisely target virtually any epitope ( 2 , 3 ), enabling therapeutics against viruses ( 4 ), toxins ( 5 ), and other disease targets ( 6 ). However, a major challenge in applying these tools to combat bacteria is the accessibility of minbinders to intracellular and membrane-embedded target antigens due to their shielding by surface molecules such as lipopolysaccharides. Previous attempts have identified high-affinity protein binders to their integral outer membrane protein target antigens in vitro , only to find that the binders cannot access their targets in a cellular context (7, 8 ). We reasoned that bacterial adhesins would constitute ideal targets for designed minibinder therapeutics targeting bacteria because they lie well beyond bacterial surface molecules and play a key role in the virulence of MDR bacteria by mediating their attachment to and invasion of host epithelial cells as well as biofilm formation (Fig. 1a). Among the diverse family of bacterial adhesins, chaperone-usher pathway (CUP) pili adhesins lie at the tips of extracellular fibers, and recognize receptors in a lectin-substrate interaction that mediates host tropisms. CUP pili are ubiquitous among gram-negative bacteria and are critical for the establishment and persistence of numerous bacterial infections, including both uncomplicated urinary tract infections (UTIs) and catheter-associated UTIs (CAUTIs; Fig. 1a). UTIs are one of the most common bacterial infections among women and their treatment accounts for approximately 15% of antibiotic usage in the United States ( 9 , 10 ). Over 50% of UTIs, especially those acquired in hospitals, are caused by MDR bacteria, limiting the effectiveness of a dwindling number of traditional antibiotic treatments ( 11-13 ). Previous anti-adhesin therapeutic strategies have relied on small molecule ligand mimetic inhibitors ( 14 ) or monoclonal antibodies (mAbs) ( 15 , 16 ). However, these strategies require identification of the adhesin substrate, which is often unknown, or labor-intensive screening for inhibitory mAbs. Designed minibinders could overcome this therapeutic bottleneck by enabling the rapid development of antibiotic-sparing therapies, relying solely on the adhesin structure and identification of its putative binding pocket. To investigate this, we sought to design minibinders against three CUP adhesins: FimH from Escherichia coli , which accounts for 70-90% of UTIs, and Abp1D and Abp2D from Acinetobacter baumannii , which is prevalent in hospital-acquired CAUTIs and is often multidrug resistant. Results De novo design of minibinders targeting FimH Uropathogenic E. coli (UPEC) expresses type 1 pili, a prototypical CUP pilus, which are tipped with the mannose binding adhesin FimH ( 17 ). UPEC FimH is essential for bacterial attachment to the bladder epithelium via mannosylated glycoproteins ( 18 ). After attachment, UPEC forms intracellular bacterial communities that are recalcitrant to antibiotics and, if left untreated, can lead to recurrent UTIs and kidney infection ( 18 ). FimH consists of an N-terminal mannose-binding lectin domain and a C-terminal pilin domain that attaches to the rest of the pilus ( 18,19 ). The lectin domain is an allosteric protein that samples a conformational equilibrium between two well-characterized conformational states: a low-affinity state (LAS, also known as the “Tense” state) and a high-affinity state (HAS, also known as the “Relaxed” state) ( 20 ). In the LAS, the body of the FimH lectin domain interacts with the pilin domain, fixing it in place, and allosterically disrupts the mannose binding pocket, resulting in a low affinity for mannose ( 18,19 ). In the HAS, the lectin domain has little to no interaction with the pilin domain, allowing it to rotate freely relative to the pilin domain, and the mannose binding pocket loops are stably positioned to bind to mannose ( 18 ). We used RFdiffusion ( 3 ) to generate novel minibinder backbones against both states of FimH. We targeted designs to the binding pocket of published crystal structures (FimH in LAS: 3JWN; FimH in HAS: 1UWF) by specifying “hotspot” residues in the binding pocket as inputs to the model (Fig. 1b). We subsequently assigned sequences to these backbones using ProteinMPNN ( 21 ) and scored designs with AlphaFold2 (AF2) Initial Guess ( 22 ). The best sequences ranked by AF2 folding and interaction confidence metrics (predicted local distance difference test (pLDDT) and predicted aligned error (PAE) scores) were then iteratively refined by partially diffusing ( 23 ) the AF2 predicted structures with RFdiffusion, reassigning sequences with ProteinMPNN, and filtering with AF2. We obtained an oligonucleotide array encoding the top 10,000 designs by AF2 and Rosetta interface metrics. Experimental characterization of FimH minibinders FimH minibinders were screened via cDNA display, rather than a cell-based screening approach like yeast surface display, because FimH binds to mannose on the cell surface. We panned against the wild-type FimH lectin domain, which adopts the HAS, and a mutant lectin domain stabilized in the LAS ( 24 ). The top 96 designs enriched more than four-fold over the naive library and consisted of 78 binders identified through the HAS sorts and 39 binders identified through the LAS sorts. Twenty-one designs were enriched against both states. Eighty-one of the enriched sequences were designed against the LAS. We expressed these 96 FimH minibinders in E. coli and found that 79 exhibited monodisperse size exclusion chromatography (SEC) traces and expressed sufficiently for downstream characterization (see Materials and Methods for details). In an initial surface plasmon resonance (SPR) binding screen, nine of these designs exhibited sub-micromolar binding against at least one of the FimH states (table S3). All nine designs were originally designed against the LAS. To directly assess the ability of FimH minibinders to inhibit and disrupt bacterial adhesion, we performed a series of red blood cell (RBC) aggregation assays using E. coli strain KB23 expressing type 1 fimbriae. E. coli aggregates RBCs by binding to mannosylated glycoproteins on their surface via FimH. Of the 96 minibinders we tested, eight significantly inhibited RBC aggregation (Fig. S1), seven of which also had submicromolar affinities in the SPR screen. Minibinder F7 had the lowest minimum inhibitory concentration (69 nM) in the RBC aggregation inhibition assay and was therefore selected for further characterization (Fig. S2). F7 was designed against the LAS, expressed well in E. coli , and purified as a monodisperse species (Fig. 2a, 2b). Though it was designed against the LAS, F7 was identified by panning against the HAS. Further characterization by SPR confirmed that F7 binds to both conformations, but with a several-fold higher affinity to the LAS (K d =119 nM) than HAS (K d =713 nM) (Fig. 2c). F7 exhibited the expected circular dichroism (CD) spectra and proved to be highly thermostable, with a melting point above 75°C (Fig. 2d, 2e). Co-crystallization of F7 with a mutant FimH lectin domain stabilized in the LAS (L34K) ( 25 ) confirmed the designed binding interface with a C a RMSD of 0.69 Å to the design model (PDB Code: 9Q1V; Fig. 2f). The experimentally determined structure of FimH bound to F7 was much closer to the canonical LAS conformation than the HAS conformation (C a RMSD of 0.45 Å vs 2.26 Å). As expected, the minibinder occluded the binding pocket of FimH. Unexpectedly, it also wedged between one of the mannose-binding loops and the clamp loop, further displacing it from the binding pocket compared to the design model. The clamp-loop displacement induced by F7 closely resembles that of the recently determined structure of FimH bound to anti-FimH antibody mAb926, which was raised against the LAS and known to function through a parasteric mechanism (fig. S3) ( 26 ). Given that F7 preferentially binds the LAS-stabilized mutant FimH and that the crystal structure of FimH in complex with the minibinder resembles the LAS, we suspected that F7 binding may induce a shift from the HAS to the LAS in wild-type FimH lectin domain. We compared F7’s interaction with the L34K variant of the lectin domain (LAS) and wild-type lectin domain (HAS) by ( 15 N, 1 H)-HSQC NMR spectroscopy. The substantial structural differences between the LAS and HAS conformations of the isolated lectin domain are reflected in their distinct NMR spectra, which display very little peak overlap (fig. S4A, S4B). Addition of F7 to the LAS variant caused chemical shift perturbations localized at the binding interface, consistent with a binding event and/or local change in conformation, but not a global conformational change (fig. S5A). In contrast, the spectrum of the WT HAS domain showed widespread chemical shift changes, becoming similar to that of the LAS variant (fig. S5A). Notably, spectra of the F7-bound HAS and LAS variants now show extensive peak overlap, indicating a shared conformation (fig. S4C). Together, these data indicate that F7 binding to the isolated WT lectin domain induces a conformational change from the HAS to the LAS, shifting the conformational equilibrium in the opposite direction as binding the native ligand and existing mannoside therapeutics ( 20 ). We quantitatively assessed the ability of minibinder F7 to block bacterial adhesion to surfaces coated with bovine RNAseB—a model glycoprotein rich in Man 5 high-mannose type oligosaccharides that avidly binds FimH even in the LAS. We preincubated E. coli KB23 with either F7, alpha-methyl-mannoside, or a minibinder that showed no inhibition in RBC assays (A4) and then measured its binding to RNaseB. F7 displayed an IC50 of 1.9 µM (95% CI: 1.4-2.9 µM), a more than two-thousand-fold improvement over mannose (Fig. 3a). We then performed a detachment ELISA by first adhering bacteria to RNAseB-coated plates and subsequently introducing either F7, noninhibitory minibinder A4, or mAb926, which is known to effectively dissolve preformed biofilms (27). F7 but not A4 successfully detached UPEC with an IC50 of 1.7 µM (95% CI: 1.3-2.4 µM), but fell short of mAb926 by two orders of magnitude (Fig. 3b). We also sought to characterize F7’s inhibitory ability in a series of more biologically relevant RBC assays. F7 successfully inhibited RBC aggregation by E. coli with a minimum inhibitory concentration of 69 nM (Fig. 3c), exceeding by more than a thousand-fold that of alpha-methyl-mannoside (fig. S6). We tested the ability of F7 to reverse bacterial aggregation by adding it to pre-aggregated mixtures of RBC and E. coli . F7 successfully redispersed the RBCs within 15 minutes of introduction at 1 µg/ml (Fig. 3d). We confirmed the physiological relevance of F7's inhibitory ability by comparing the adherence of bacteria to T24 urinary bladder tissue culture cells in the presence and absence of minibinder. F7-treated cells had considerably fewer bacteria present than either untreated cells (Fig. 3e) or cells treated with A4 (fig. S7). Given the sequence diversity of FimH across disease-causing strains, we tested whether F7 is inhibitory not only to the FimH variant it was designed against but also to different FimH variants expressed by the most common highly-virulent and/or multi-drug resistant clonal groups of UPEC. F7 successfully inhibited RBC aggregation by strains from the clonal groups ST73, ST95, ST69, and ST131-H30 (fig. S8, table S4). F7 also inhibited aggregation by E. coli strain cas665 expressing Klebsiella pneumoniae FimH, which has approximately 85% sequence identity to E. coli FimH, suggesting that F7 may be cross-reactive with FimH variants from other bacterial pathogens. De novo design of minibinders targeting Abp adhesins CAUTIs caused by MDR bacteria pose a significant economic toll on the healthcare system in the United States ( 28 ). CAUTIs represent approximately 40% of hospital-acquired infections and have an increased risk of morbidity and mortality compared to uncomplicated UTIs ( 28,29 ). CAUTIs result in a serious deterioration of quality of life through pain, discomfort, disruption of daily activities, and increased healthcare costs, exacerbated by the rapid spread of drug resistance in uropathogens ( 28–32 ). One particularly concerning CAUTI pathogen is A. baumannii , which is classified by the CDC as a pathogen of urgent concern due to its high rates of multidrug resistance and increases in relative prevalence in CAUTIs compared to uncomplicated UTIs ( 33 ). The A. baumannii strain “ACICU” possesses two highly similar CUP pili, Abp1 and Abp2, that play critical roles in mediating CAUTI. The Abp1 and Abp2 pili are tipped with the Abp1D and Abp2D adhesins that allow these bacteria to bind to host-deposited fibrinogen on implanted catheters, which is a critical step for A. baumannii to establish a CAUTI ( 34 ). While the receptor binding domains of these adhesins share only 70% sequence identity, they share a higher degree of structural homology as seen in previously solved crystal structures (Abp1D: 8DF0; Abp2D: 8DEZ).(34) Previous structure-function studies identified the putative fibrinogen-binding pockets of these adhesins; however, the exact structural binding interactions between fibrinogen and the adhesins remain unknown ( 34 ). Similar to the dynamic conformational ensembles displayed by FimH, the binding affinity of the A. baumannii Abp1D and Abp2D receptor binding domains (RBD) are controlled via dynamics of the flexible anterior binding loop, illustrated by comparing the loop’s conformation in the Abp2D crystal structure to an AlphaFold2 model for this protein (fig. S9). In addition, while vaccination with Abp2D protected mice from A. baumannii CAUTI, these efforts to immunize mice with Abp2D failed to identify sufficiently inhibitory mAbs to effectively disrupt binding of ACICU to fibrinogen (IC50s > 100 µM) in a bacterial ELISA assay (fig. S10) ( 35 ). Thus, we were motivated to apply protein design tools to develop more effective Abp inhibitors ( 36 ). We used RFdiffusion ( 3 ) with “hotspot” conditioning to generate novel minibinder backbones specifically targeting the putative fibrinogen-binding sites of both the Abp1D and Abp2D adhesins (Fig. 1b). Taking into account the anterior loop conformational differences between the high-affinity and low-affinity states, we designed backbones against the crystal structures of Abp1D and Abp2D receptor binding domains, which display the anterior binding loop in the open (low-affinity) conformation, exposing the putative binding pocket. We assigned sequences to these backbones using ProteinMPNN, predicted their structures with AF2 Initial Guess, and then iteratively optimized high-quality designs as described above. The top ~5000 scoring designs against Abp1D (~1300) and Abp2D (~3700), with a maximum of 15 sequences per backbone, were obtained as a DNA oligo chip. Experimental characterization of Abp minibinders Abp adhesins do not bind mannose and are therefore amenable to screening minibinders using yeast surface display. Designed Abp binder sequences were amplified from the DNA oligo chip and cloned into a yeast display expression vector using yeast cloning. The resulting yeast display library was enriched for binding via two rounds of fluorescence-activated cell sorting (FACS) with 1 μM biotinylated Abp1D or Abp2D lectin domains using PE-conjugated streptavidin as a detection reagent. A third round of FACS sorting involved staining the enriched populations with 10-fold titrations of these adhesins from 1000 nM to 1 nM, revealing strong binding signal for the Abp2D-enriched population down to 1 nM, and binding signal down to 10 nM for the Abp1D-enriched population (fig. S11). Next-generation sequencing of sorted populations identified 31 highly enriched binder sequences across both populations, including some sequences that were enriched in both Abp1D and Abp2D sorts, suggesting that we identified cross-reactive binders. Interestingly, all 31 of these enriched binders were designed using the Abp2D crystal structure, which might be explained by the presence of a citrate molecule in the Abp2D crystal structure that opens its putative binding pocket and exposes more hydrophobic residues than in the Abp1D structure. Minibinder sequences that were enriched during yeast surface display with Abp adhesins were expressed in and purified from E. coli . Of the 31 enriched binders, twenty-four exhibited primarily monodisperse SEC traces and had sufficient yields for downstream characterization. An initial SPR binding screen of these 24 binders against both Abp1D and Abp2D identified 16 binders that exhibited sub-micromolar affinity to Abp2D, four of which also exhibited sub-micromolar affinity to Abp1D (table S5). A follow-up SPR experiment on the four cross-reactive binders revealed that Abp minibinder A7 (design model in Fig. 4a) exhibited high affinity to both Abp1D (K d =50.4 nM) and Abp2D (K d =3.5 nM; Fig. 4c). A7 expressed well (~50 mg/L of culture), behaved as a monodisperse species as assessed by SEC (Fig. 4b), exhibited the expected CD spectra (Fig. 4d), and was thermostable with a melting temperature greater than 85°C (Fig. 4e). Next, we confirmed the specificity of the designed adhesin binders using SPR. Minibinder A7 exhibited undetectable levels of binding to FimH LAS or HAS. Similarly, FimH minibinder F7 exhibited insignificant binding to Abp1D and Abp2D (fig. S12). Determining the structure of a co-crystal of A7 with Abp2D confirmed the designed interface, and exhibited a C a RMSD of 0.88 Å to the design model (PDB Code: 9Q1H; Fig. 4f). This structure confirms the designed binding mode that utilizes a helix-turn-strand motif to simultaneously stabilize the core of the minibinder, insert a loop deep into the putative fibrinogen binding pocket of Abp2D, and strand pair with an edge strand of this adhesin. As designed, A7 stabilizes the anterior binding loop of Abp2D in an open conformation, thus neutralizing the adhesin in the low-affinity conformation and matching the loop conformation in the Abp2D crystal structure. To assess the ability of A7 to inhibit bacterial adhesion, we performed a bacterial ELISA assay with A. baumannii ACICU. When preincubated with ACICU prior to being added to the fibrinogen-coated ELISA plate, A7 but not a BSA control significantly inhibited ACICU binding with an IC50 of 2.6 nM (95% CI: 1.8-3.0 nM; Fig. 5a). In a “detachment” ELISA, A7 successfully displaced adherent ACICU and exhibited an IC50 of 40.0 nM (95% CI: 25.5-52.1 nM; Fig. 5b). Since bacterial adhesins are known to contribute to biofilm formation (34), we also tested whether designed inhibitors could both prevent ACICU from forming biofilms and disperse preformed biofilms. Crystal violet staining of A. baumannii biofilms grown on PVC plates demonstrated that 1 µM of A7 could prevent the formation of ACICU biofilms, unlike a BSA control (p < 0.0001, one-way ANOVA; Fig. 5c). In addition, 1 µM of A7 but not a BSA control could almost completely disperse preformed ACICU biofilms after 2 hours of treatment (p=0.0021, one-way ANOVA), mimicking the phenotype of an ACICU Δabp1 Δabp2 double knockout strain (Fig. 5c; p=0.0003, one-way ANOVA). To explore whether minibinder A7 could disrupt A. baumannii ACICU biofilms in a more in vivo -like context, inhibition and detachment studies were performed with fibrinogen-coated silicone catheters ( 34,37 ). After coating silicone tubing with fibrinogen, ACICU was allowed to bind in the presence of A7 or BSA. We also tested the ability of A7 and BSA to disperse ACICU that had already bound to the silicone catheter. Immunostaining the tubing revealed that despite fibrinogen coating each piece of tubing, 100 nM of A7 but not the BSA control significantly prevented ACICU attachment to and dispersed prebound bacteria from the tubing, mimicking the phenotype of the ACICU Δabp1 Δabp2 strain (Fig. 5d, fig. S13, p=0.0052, one-way ANOVA). Clinical models of adhesin inhibitors We next explored whether these adhesin inhibitors would function in vivo in clinical models of UTI. To test anti-FimH minibinder F7’s ability to combat uncomplicated UTI, mice were injected intraperitoneally with either 100 µg of the minibinder, mAb926, or a buffer-only control 24 hours prior to bacterial challenge. Mice were then challenged with model uropathogenic E. coli strain CFT073 from the clonal group ST73 (10 6 CFU), and a second dose of protein was administered by intraperitoneal (IP) injection 24 hours after infection. Forty-eight hours post-infection, bacterial burden in the bladder was quantified (Fig. 6a). F7-treated mice showed a significant (~2 log) reduction in bacterial bladder titers compared to the mock control group (p = 0.034, one-way ANOVA) and reached similar levels of reduction as the antibody-treated group (Fig. 6b). We next tested anti-Abp minibinder A7 in a mouse model of CAUTI. Mice were catheterized and immediately infected with A. baumannii ACICU (10 8 CFU) premixed with 0.25 mg/mL of minibinder or buffer. After 1 hour, mice were treated with IP injections of 100 µg of A7 or a buffer-only control. After 3 hours, bacterial burden in the bladder and catheter of each mouse was quantified (Fig. 6c). Mice that were treated with A7 exhibited significantly reduced bacterial loads in the bladder (p = 0.0035, one-way ANOVA) and catheter (p = 0.0014, one-way ANOVA) compared to mock-treated controls (Fig. 6d). This reduction in bacterial burden aligns with previous studies showing only modest (~1 log) attenuation of bacterial burden by ACICU Δabp1 Δabp2, suggesting alternative adhesion mechanisms may be involved in this mouse model ( 34 ). This proof of concept experiment lays the ground for future antibiotic-sparing strategies against pathogens of great concern. Optimization of adhesin inhibitors We explored two routes to generating higher affinity designs. First, we used ProteinMPNN to resample sequences based on the designed backbone of minibinder F7, and selected variants with improved AF2 Initial Guess scores; this led to the identification of a new minibinder (C8) exhibiting ~10-fold higher affinity for FimH (LAS K d =15.0nM; HAS K d =243nM) and more than ~30-fold greater inhibitory activity in an inhibition ELISA with an IC50 of 37 nM (95% CI: 31-47 nM; Fig. S14, S15). Second, to enhance affinity via avidity effects, minibinder A7 was fused to de novo designed homo-oligomeric domains (38), generating multivalent binders. Oligomeric constructs such as oligomer C11 displayed significantly increased binding affinity to both adhesins, with K d s in the picomolar range for Abp1D and Abp2D (fig. S16). In cellulo, oligomer C11 exhibits a significantly lower inhibition ELISA IC50 (0.36 nM) than A7 (95% CI: 0.15-0.61 nM; fig. S17). Such affinity and avidity-increasing strategies should be generally applicable for further improving the in vivo efficacy of designed anti-adhesin minibinders. Discussion In this study we show that de novo protein design tools can be employed to design potent and specific minibinder inhibitors of bacterial adhesins, virulence factors critical for colonization and biofilm formation in a broad range of bacterial infections, including UPEC UTI and A. baumannii CAUTI. The choice of target was key to in vivo success, as the extracellular nature of these adhesins made them readily accessible to the designed proteins. Being able to specifically target minibinders to the substrate binding epitope using RFdiffusion with hotspot conditioning led to neutralizing binders without the need for antibody screening or knowledge of the native substrates. The conversion of FimH HAS to LAS by minibinder F7 shows that computational design allows not only the selection of target epitope but also target conformation. Targeting the substrate binding pocket of these adhesins should make it unlikely that resistance to these binders will arise quickly, as any mutation to the binding pocket may also disrupt adhesion. The specificity of these designed proteins for their targets should enable selective depletion of these pathogens without affecting commensal microbiota and limit pressure among the broader bacterial community to evolve resistance to the minibinders ( 39 ). Future studies should explore whether resistance can be evolved to escape these minibinders and how they affect the broader microbiome. Future development of the adhesion-neutralizing therapeutics demonstrated here will require characterization and possibly additional optimization of the pharmacokinetics and bioavailability of these designed proteins. Blocking adhesin activity using minibinders has potential utility against non-UTI bacterial pathogens such as Staphylococcus aureus , Streptococcus pneumoniae, and Salmonella ( 40,41 ). The strategies developed here should be applicable to combat the adhesion and colonization of pathogens beyond bacteria, including fungi and amoebae ( 42–44 ). Declarations Acknowledgments: Crystallographic diffraction data were collected at the Northeastern Collaborative Access Team beamlines at the Advanced Photon Source, which are funded by the National Institute of General Medical Sciences from the National Institutes of Health (P30 GM124165) and the ALS 4.2.2 beamline (P30 GM124169). This research used resources of the Advanced Photon Source, a US Department of Energy (DOE) Office of Science User Facility operated by Argonne National Laboratory under contract no. DE-AC02-06CH11357. This research used resources (FMX) of the National Synchrotron Light Source II, a U.S. Department of Energy (DOE) Office of Science User Facility operated for the DOE Office of Science by Brookhaven National Laboratory under Contract No. DE-SC0012704. The Center for BioMolecular Structure (CBMS) is primarily supported by the National Institutes of Health, National Institute of General Medical Sciences (NIGMS) through a Center Core P30 Grant (P30GM133893), and by the DOE Office of Biological and Environmental Research (KP1607011). Figures were made using Biorender.com. We would like to acknowledge Jacob M. Gershon, Hojae Choi, Yensi Flores Bueno, David S. Lee, Jeremiah Sims, Brian Coventry, and Jay Nix for materials, helpful discussions, and support; and L. Goldschmidt and K. Van Wormer for general operations. Funding: Defense Advanced Research Projects Agency grant HR0011-21-2-0012 (ACG, TRT) National Institutes of Health grant F30DK135390 (MRT) National Institutes of Health grant R01AI029549 (SJH) National Institutes of Health grant R37AI048689 (SJH) National Institutes of Health grant U19AI157797 (SJH, AHE) National Institutes of Health grant K99GM141364 (PM) National Institutes of Health grant R01AI171570 (REK, EVS) The Open Philanthropy Project Improving Protein Design Fund (ACG, TRT) Howard Hughes Medical Institute (DB) The Schmidt Science Fellows, in partnership with the Rhodes Trust (ACH) Author contributions: Conceptualization: ACG, TRT, EDBL Methodology: ACG, TRT, EDBL, PM, REK. PWE, KWD Investigation: ACG, TRT, EDBL, PM, DAS, EDBL, MC, AV, SB, DW, AH, KOT, AK, EJ, AKB, AM, PA, VT, IB, MRT, JSP, GI Visualization: ACG, TRT, EDBL, PM, KLH Writing: ACG, TRT, EDBL, KWD, KLH, DB, SJH, EVS Supervision: DB, SJH, EVS, AHE Ethics declarations: ACG, TRT, and DB are co-inventors on a provisional patent describing the adhesin minibinders (63/857,995). SJH is on the advisory board of Sequoia Vaccines Inc. SJH is a cofounder of Fimbrion Therapeutics that is developing FimH targeted therapies and may financially benefit if the company is successful. AHE received funding from Emergent BioSolutions, AbbVie, and Moderna that are unrelated to the data presented in the current study. AHE has received consulting and speaking fees from InBios International, Fimbrion Therapeutics, RGAX, Mubadala Investment Company, Moderna, Pfizer, GSK, Danaher, Third Rock Ventures, Goldman Sachs, and Morgan Stanley and is the founder of ImmuneBio Consulting. AJS and AHE are recipients of a licensing agreement with Abbvie that is unrelated to the data presented in the current study. MRT, KOT, JSP, KWD, AHE, and SJH are inventors on the US patent application US23/82426 regarding Abp2D monoclonal antibodies. Data and materials availability: The crystal structures have been deposited to the Protein Data Bank (PDB) with accession codes 9Q1V (FimH-F7), 9Q1H (Abp2D-A7). All other data are available in the main text or the supplementary materials References Van Duin D, Paterson DL. Multidrug-Resistant Bacteria in the Community. Infect Dis Clin North Am. 2020 Dec;34(4):709–22. Cao L, Coventry B, Goreshnik I, Huang B, Sheffler W, Park JS, et al. Design of protein-binding proteins from the target structure alone. Nature. 2022 May 19;605(7910):551–60. Watson JL, Juergens D, Bennett NR, Trippe BL, Yim J, Eisenach HE, et al. De novo design of protein structure and function with RFdiffusion. Nature. 2023 Aug 31;620(7976):1089–100. Cao L, Goreshnik I, Coventry B, Case JB, Miller L, Kozodoy L, et al. De novo design of picomolar SARS-CoV-2 miniprotein inhibitors. Vázquez Torres S, Benard Valle M, Mackessy SP, Menzies SK, Casewell NR, Ahmadi S, et al. De novo designed proteins neutralize lethal snake venom toxins. Nature. 2025 Mar 6;639(8053):225–31. Roy A, Shi L, Chang A, Dong X, Fernandez A, Kraft JC, et al. De novo design of highly selective miniprotein inhibitors of integrins αvβ6 and αvβ8. Nat Commun. 2023 Sep 13;14(1):5660. Storek KM, Auerbach MR, Shi H, Garcia NK, Sun D, Nickerson NN, et al. Monoclonal antibody targeting the β-barrel assembly machine of Escherichia coli is bactericidal. Proc Natl Acad Sci. 2018 Apr 3;115(14):3692–7. Gokce-Alpkilic G, Huang B, Liu A, Kreuk LSM, Adebomi V, Bueso YF, et al. De novo design of high-affinity miniprotein binders targeting Francisella tularensis virulence factor. Chua KP, Fischer MA, Linder JA. Appropriateness of outpatient antibiotic prescribing among privately insured US patients: ICD-10-CM based cross sectional study. BMJ. 2019 Jan 16;k5092. Fleming-Dutra KE, Hersh AL, Shapiro DJ, Bartoces M, Enns EA, File TM, et al. Prevalence of Inappropriate Antibiotic Prescriptions Among US Ambulatory Care Visits, 2010-2011. JAMA. 2016 May 3;315(17):1864. Madrazo M, Esparcia A, López-Cruz I, Alberola J, Piles L, Viana A, et al. Clinical impact of multidrug-resistant bacteria in older hospitalized patients with community-acquired urinary tract infection. BMC Infect Dis. 2021 Dec;21(1):1232. Watkins RR, Thapaliya D, Lemonovich TL, Bonomo RA. Gepotidacin: a novel, oral, ‘first-in-class’ triazaacenaphthylene antibiotic for the treatment of uncomplicated urinary tract infections and urogenital gonorrhoea. J Antimicrob Chemother. 2023 May 3;78(5):1137–42. Worby CJ, Schreiber HL, Straub TJ, Van Dijk LR, Bronson RA, Olson BS, et al. Longitudinal multi-omics analyses link gut microbiome dysbiosis with recurrent urinary tract infections in women. Nat Microbiol. 2022 May 2;7(5):630–9. Cusumano CK, Pinkner JS, Han Z, Greene SE, Ford BA, Crowley JR, et al. Treatment and Prevention of Urinary Tract Infection with Orally Active FimH Inhibitors. Sci Transl Med [Internet]. 2011 Nov 16 [cited 2025 May 6];3(109). Available from: https://www.science.org/doi/10.1126/scitranslmed.3003021 Hvorecny KL, Interlandi G, Veth TS, Aprikian P, Manchenko A, Tchesnokova VL, et al. Antibodies disrupt bacterial adhesion by ligand mimicry and allosteric interference [Internet]. Immunology; 2024 [cited 2025 May 6]. Available from: http://biorxiv.org/lookup/doi/10.1101/2024.12.06.627246 Lopatto EDB, Santiago-Borges JM, Sanick DA, Malladi SK, Azimzadeh PN, Timm MW, et al. Monoclonal antibodies targeting the FimH adhesin protect against uropathogenic E. coli UTI [Internet]. Microbiology; 2024 [cited 2025 May 6]. Available from: http://biorxiv.org/lookup/doi/10.1101/2024.12.10.627638 Le Trong I, Aprikian P, Kidd BA, Forero-Shelton M, Tchesnokova V, Rajagopal P, et al. Structural Basis for Mechanical Force Regulation of the Adhesin FimH via Finger Trap-like β Sheet Twisting. Cell. 2010 May;141(4):645–55. Foroogh N, Rezvan M, Ahmad K, Mahmood S. Structural and functional characterization of the FimH adhesin of uropathogenic Escherichia coli and its novel applications. Microb Pathog. 2021 Dec;161:105288. Le Trong I, Aprikian P, Kidd BA, Forero-Shelton M, Tchesnokova V, Rajagopal P, et al. Structural Basis for Mechanical Force Regulation of the Adhesin FimH via Finger Trap-like β Sheet Twisting. Cell. 2010 May;141(4):645–55. Kalas V, Pinkner JS, Hannan TJ, Hibbing ME, Dodson KW, Holehouse AS, et al. Evolutionary fine-tuning of conformational ensembles in FimH during host-pathogen interactions. Sci Adv. Dauparas J, Anishchenko I, Bennett N, Bai H, Ragotte RJ, Milles LF, et al. Robust deep learning–based protein sequence design using ProteinMPNN. Bennett NR, Coventry B, Goreshnik I, Huang B, Allen A, Vafeados D, et al. Improving de novo protein binder design with deep learning. Nat Commun. 2023 May 6;14(1):2625. Glögl M, Krishnakumar A, Ragotte RJ, Goreshnik I, Coventry B, Bera AK, et al. Target-conditioned diffusion generates potent TNFR superfamily antagonists and agonists. Science. 2024 Dec 6;386(6726):1154–61. Rodriguez VB, Kidd BA, Interlandi G, Tchesnokova V, Sokurenko EV, Thomas WE. Allosteric Coupling in the Bacterial Adhesive Protein FimH. J Biol Chem. 2013 Aug;288(33):24128–39. Magala P, Tuttle LM, Interlandi G, Carlucci LA, Mollica MY, Thomas WE, et al. Ultra-slow conformational dynamics and catch bond formation of a Bacterial Adhesin revealed by a single-domain variant of FimH. Hvorecny KL, Interlandi G, Veth TS, Aprikian P, Manchenko A, Tchesnokova VL, et al. Antibodies disrupt bacterial adhesion by ligand mimicry and allosteric interference [Internet]. Immunology; 2024 [cited 2025 Aug 4]. Available from: http://biorxiv.org/lookup/doi/10.1101/2024.12.06.627246 Kisiela DI, Avagyan H, Friend D, Jalan A, Gupta S, Interlandi G, et al. Inhibition and Reversal of Microbial Attachment by an Antibody with Parasteric Activity against the FimH Adhesin of Uropathogenic E. coli. Tsolis RM, editor. PLOS Pathog. 2015 May 14;11(5):e1004857. Molina JJ, Flores-Mireles AL. CAUTIon — not all UTIs are the same. Nat Rev Urol [Internet]. 2025 Aug 1 [cited 2025 Aug 15]; Available from: https://www.nature.com/articles/s41585-025-01065-z Yang X, Chen H, Zheng Y, Qu S, Wang H, Yi F. Disease burden and long-term trends of urinary tract infections: A worldwide report. Front Public Health. 2022 Jul 27;10:888205. Andersen MJ, Flores-Mireles AL. Urinary Catheter Coating Modifications: The Race against Catheter-Associated Infections. Coatings. 2019 Dec 29;10(1):23. Flores-Mireles A, Hreha TN, Hunstad DA. Pathophysiology, Treatment, and Prevention of Catheter-Associated Urinary Tract Infection. Top Spinal Cord Inj Rehabil. 2019 Jun;25(3):228–40. Saint S, Trautner BW, Fowler KE, Colozzi J, Ratz D, Lescinskas E, et al. A Multicenter Study of Patient-Reported Infectious and Noninfectious Complications Associated With Indwelling Urethral Catheters. JAMA Intern Med. 2018 Aug 1;178(8):1078. Hazen JE, Di Venanzio G, Hultgren SJ, Feldman MF. Catheterization of mice triggers resurgent urinary tract infection seeded by a bladder reservoir of Acinetobacter baumannii. Sci Transl Med. 2023 Jan 11;15(678):eabn8134. Tamadonfar KO, Di Venanzio G, Pinkner JS, Dodson KW, Kalas V, Zimmerman MI, et al. Structure–function correlates of fibrinogen binding by Acinetobacter adhesins critical in catheter-associated urinary tract infections. Proc Natl Acad Sci. 2023 Jan 24;120(4):e2212694120. Hultgren S, Timm M, Tamadonfar K, Nye T, Pinkner J, Dodson K, et al. Vaccination with Acinetobacter baumannii adhesin Abp2D provides protection against catheter-associated urinary tract infection [Internet]. In Review; 2023 [cited 2025 Aug 15]. Available from: https://www.researchsquare.com/article/rs-3213777/v1 Hultgren S, Timm M, Tamadonfar K, Nye T, Pinkner J, Dodson K, et al. Vaccination with Acinetobacter baumannii adhesin Abp2D provides protection against catheter-associated urinary tract infection [Internet]. In Review; 2023 [cited 2025 May 31]. Available from: https://www.researchsquare.com/article/rs-3213777/v1 Di Venanzio G, Flores-Mireles AL, Calix JJ, Haurat MF, Scott NE, Palmer LD, et al. Urinary tract colonization is enhanced by a plasmid that regulates uropathogenic Acinetobacter baumannii chromosomal genes. Nat Commun. 2019 Jun 24;10(1):2763. Edman NI, Phal A, Redler RL, Schlichthaerle T, Srivatsan SR, Ehnes DD, et al. Modulation of FGF pathway signaling and vascular differentiation using designed oligomeric assemblies. Cell. 2024 Jul;187(14):3726-3740.e43. Perez-Carrasco V, Soriano-Lerma A, Soriano M, Gutiérrez-Fernández J, Garcia-Salcedo JA. Urinary Microbiome: Yin and Yang of the Urinary Tract. Front Cell Infect Microbiol. 2021 May 18;11:617002. Rosenow C, Ryan P, Weiser JN, Johnson S, Fontan P, Ortqvist A, et al. Contribution of novel choline‐binding proteins to adherence, colonization and immunogenicity of Streptococcus pneumoniae. Mol Microbiol. 1997 Sep;25(5):819–29. Ghosh S, Chakraborty K, Nagaraja T, Basak S, Koley H, Dutta S, et al. An adhesion protein of Salmonella enterica serovar Typhi is required for pathogenesis and potential target for vaccine development. Proc Natl Acad Sci. 2011 Feb 22;108(8):3348–53. Kim D, Kim JS, Bai X, Zhang J, Park M, Lee U, et al. Development of Miniprotein-Type Inhibitors of Biofilm Formation in Candida albicans and Candida auris. J Microbiol Biotechnol. 2025 Feb 25;35:e2411076. Aguirre García M, Gutiérrez-Kobeh L, López Vancell R. Entamoeba histolytica: Adhesins and Lectins in the Trophozoite Surface. Molecules. 2015 Feb 9;20(2):2802–15. Kumari A, Tripathi AH, Gautam P, Gahtori R, Pande A, Singh Y, et al. Adhesins in the virulence of opportunistic fungal pathogens of human. Mycology. 2021 Oct 2;12(4):296–324. Zielonka S, Krah S, editors. Genotype Phenotype Coupling: Methods and Protocols [Internet]. New York, NY: Springer US; 2020 [cited 2025 Jun 26]. (Methods in Molecular Biology; vol. 2070). Available from: http://link.springer.com/10.1007/978-1-4939-9853-1 Zeng Y, Woolley M, Chockalingam K, Thomas B, Arora S, Hook M, et al. Click display: a rapid and efficient in vitro protein display method for directed evolution. Nucleic Acids Res. 2023 Sep 8;51(16):e89–e89. Wu L, Wen C, Qin Y, Yin H, Tu Q, Van Nostrand JD, et al. Phasing amplicon sequencing on Illumina Miseq for robust environmental microbial community analysis. BMC Microbiol. 2015 Dec;15(1):125. Edman NI, Phal A, Redler RL, Schlichthaerle T, Srivatsan SR, Ehnes DD, et al. Modulation of FGF pathway signaling and vascular differentiation using designed oligomeric assemblies. Cell. 2024 Jul;187(14):3726-3740.e43. Johnson BA, Blevins RA. NMR View: A computer program for the visualization and analysis of NMR data. J Biomol NMR. 1994 Sep;4(5):603–14. Delaglio F, Grzesiek S, Vuister GeertenW, Zhu G, Pfeifer J, Bax A. NMRPipe: A multidimensional spectral processing system based on UNIX pipes. J Biomol NMR [Internet]. 1995 Nov [cited 2025 May 8];6(3). Available from: http://link.springer.com/10.1007/BF00197809 Alsoussi WB, Turner JS, Case JB, Zhao H, Schmitz AJ, Zhou JQ, et al. A Potently Neutralizing Antibody Protects Mice against SARS-CoV-2 Infection. J Immunol. 2020 Aug 15;205(4):915–22. Von Boehmer L, Liu C, Ackerman S, Gitlin AD, Wang Q, Gazumyan A, et al. Sequencing and cloning of antigen-specific antibodies from mouse memory B cells. Nat Protoc. 2016 Oct;11(10):1908–23. Sokurenko EV, Courtney HS, Maslow J, Siitonen A, Hasty DL. Quantitative differences in adhesiveness of type 1 fimbriated Escherichia coli due to structural differences in fimH genes. J Bacteriol. 1995 Jul;177(13):3680–6. Stahlhut SG, Tchesnokova V, Struve C, Weissman SJ, Chattopadhyay S, Yakovenko O, et al. Comparative Structure-Function Analysis of Mannose-Specific FimH Adhesins from Klebsiella pneumoniae and Escherichia coli. J Bacteriol. 2009 Nov;191(21):6592–601. Tchesnokova VL, Rechkina E, Larson L, Ferrier K, Weaver JL, Schroeder DW, et al. Rapid and Extensive Expansion in the United States of a New Multidrug-resistant Escherichia coli Clonal Group, Sequence Type 1193. Clin Infect Dis. 2019 Jan 7;68(2):334–7. Weissman SJ, Johnson JR, Tchesnokova V, Billig M, Dykhuizen D, Riddell K, et al. High-Resolution Two-Locus Clonal Typing of Extraintestinal Pathogenic Escherichia coli. Appl Environ Microbiol. 2012 Mar;78(5):1353–60. Andersen MJ, Fong C, La Bella AA, Molina JJ, Molesan A, Champion MM, et al. Inhibiting host-protein deposition on urinary catheters reduces associated urinary tract infections. eLife. 2022 Mar 29;11:e75798. Schneider CA, Rasband WS, Eliceiri KW. NIH Image to ImageJ: 25 years of image analysis. Nat Methods. 2012 Jul;9(7):671–5. Guiton PS, Hung CS, Hancock LE, Caparon MG, Hultgren SJ. Enterococcal Biofilm Formation and Virulence in an Optimized Murine Model of Foreign Body-Associated Urinary Tract Infections. Infect Immun. 2010 Oct;78(10):4166–75. Additional Declarations Yes there is potential Competing Interest. ACG, TRT, and DB are co-inventors on a provisional patent describing the adhesin minibinders (63/857,995). SJH is on the advisory board of Sequoia Vaccines Inc. SJH is a cofounder of Fimbrion Therapeutics that is developing FimH targeted therapies and may financially benefit if the company is successful. AHE received funding from Emergent BioSolutions, AbbVie, and Moderna that are unrelated to the data presented in the current study. AHE has received consulting and speaking fees from InBios International, Fimbrion Therapeutics, RGAX, Mubadala Investment Company, Moderna, Pfizer, GSK, Danaher, Third Rock Ventures, Goldman Sachs, and Morgan Stanley and is the founder of ImmuneBio Consulting. AJS and AHE are recipients of a licensing agreement with Abbvie that is unrelated to the data presented in the current study. MRT, KOT, JSP, KWD, AHE, and SJH are inventors on the US patent application US23/82426 regarding Abp2D monoclonal antibodies. Supplementary Files AdhesinpaperNatureMicroSubmissionSupplement.docx Adhesin paper - Nature Micro Submission - Supplement Cite Share Download PDF Status: Under Review Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7951484","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":538462580,"identity":"ed3d711e-7c1c-4591-88ca-6ff3374216f5","order_by":0,"name":"Adam Chazin-Gray","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA1klEQVRIiWNgGAWjYFACHiA2YGDgB7ETCkjRItkA0mJAtBaQrgMMEL0EAX/72YMffxTcsdt8fnXihwcGDPL8Ygfwa5E4k5csIWHwLHnbjbebJYAOM5w5O4GANTd4DCQMDA4nm904uwGkJcHgNgEt8jd4jH8kALUYzzi7+QdRWgxu8JhJHDA4bGfA37uNOFsMz+SYWTYYHE6QuMG7zSLBQIKwX+SOnzG++ePPYXv+/rObb/6osJHnlyagBQYSGyTAKiWIUw4C9gz8B4hXPQpGwSgYBSMLAADsY0XKLfqnsgAAAABJRU5ErkJggg==","orcid":"","institution":"University of Washington","correspondingAuthor":true,"prefix":"","firstName":"Adam","middleName":"","lastName":"Chazin-Gray","suffix":""},{"id":538462581,"identity":"7e4a4a55-3c2a-455a-b647-e35c5d53b893","order_by":1,"name":"Tuscan Thompson","email":"","orcid":"","institution":"University of Washington","correspondingAuthor":false,"prefix":"","firstName":"Tuscan","middleName":"","lastName":"Thompson","suffix":""},{"id":538462582,"identity":"55c8e32d-67cb-4c47-ad9c-bd5b0523fdff","order_by":2,"name":"Edward Lopatto","email":"","orcid":"https://orcid.org/0000-0002-6625-6143","institution":"Washington University in St. Louis","correspondingAuthor":false,"prefix":"","firstName":"Edward","middleName":"","lastName":"Lopatto","suffix":""},{"id":538462583,"identity":"03a0dd97-0a6b-4381-9ee9-13723e3c59d2","order_by":3,"name":"Pearl Magala","email":"","orcid":"","institution":"University of Washington","correspondingAuthor":false,"prefix":"","firstName":"Pearl","middleName":"","lastName":"Magala","suffix":""},{"id":538462584,"identity":"1bec17d4-d393-4dd8-aee1-2d4bd8074fc7","order_by":4,"name":"Patrick Erickson","email":"","orcid":"","institution":"University of Washington","correspondingAuthor":false,"prefix":"","firstName":"Patrick","middleName":"","lastName":"Erickson","suffix":""},{"id":538462585,"identity":"5ca5a9c4-424d-403f-a10a-98eb1c97e58e","order_by":5,"name":"Andrew Hunt","email":"","orcid":"","institution":"University of Washington","correspondingAuthor":false,"prefix":"","firstName":"Andrew","middleName":"","lastName":"Hunt","suffix":""},{"id":538462586,"identity":"5e7d372b-9995-47e9-b13e-06e391ae6334","order_by":6,"name":"Anna Manchenko","email":"","orcid":"","institution":"University of Washington","correspondingAuthor":false,"prefix":"","firstName":"Anna","middleName":"","lastName":"Manchenko","suffix":""},{"id":538462587,"identity":"c14f3cbd-9e6d-47ea-aa0d-2da6e09b4274","order_by":7,"name":"Pavel Aprikian","email":"","orcid":"","institution":"University of Washington","correspondingAuthor":false,"prefix":"","firstName":"Pavel","middleName":"","lastName":"Aprikian","suffix":""},{"id":538462588,"identity":"646ae404-9cc3-4fc4-a26e-a40b9b90c846","order_by":8,"name":"Veronika Tchesnokova","email":"","orcid":"","institution":"University of Washington","correspondingAuthor":false,"prefix":"","firstName":"Veronika","middleName":"","lastName":"Tchesnokova","suffix":""},{"id":538462589,"identity":"cca8dd27-1d1f-4ef0-8eee-bd7d6ece992c","order_by":9,"name":"Irina Basova","email":"","orcid":"","institution":"University of Washington","correspondingAuthor":false,"prefix":"","firstName":"Irina","middleName":"","lastName":"Basova","suffix":""},{"id":538462590,"identity":"e6fb023c-ac53-405d-a479-d1454fdce206","order_by":10,"name":"Denise Sanick","email":"","orcid":"","institution":"Washington University in St. Louis","correspondingAuthor":false,"prefix":"","firstName":"Denise","middleName":"","lastName":"Sanick","suffix":""},{"id":538462591,"identity":"e88002ac-e8e1-4fc0-abb2-a6c097228ed1","order_by":11,"name":"Kevin Tamadonfar","email":"","orcid":"","institution":"Washington University School of Medicine","correspondingAuthor":false,"prefix":"","firstName":"Kevin","middleName":"","lastName":"Tamadonfar","suffix":""},{"id":538462592,"identity":"a42d9232-38f9-41ee-8970-a260b8e58f79","order_by":12,"name":"Morgan Timm","email":"","orcid":"https://orcid.org/0000-0002-2947-9508","institution":"Washington University School of Medicine","correspondingAuthor":false,"prefix":"","firstName":"Morgan","middleName":"","lastName":"Timm","suffix":""},{"id":538462593,"identity":"572cb47d-393f-478d-b773-570d1672973e","order_by":13,"name":"Jerome Pinkner","email":"","orcid":"","institution":"Washington University School of Medicine","correspondingAuthor":false,"prefix":"","firstName":"Jerome","middleName":"","lastName":"Pinkner","suffix":""},{"id":538462594,"identity":"6d5e6336-4385-424d-bf52-567d1d395546","order_by":14,"name":"Karen Dodson","email":"","orcid":"","institution":"Washington University School of Medicine","correspondingAuthor":false,"prefix":"","firstName":"Karen","middleName":"","lastName":"Dodson","suffix":""},{"id":538462595,"identity":"fb547c78-02e1-4cd5-9ba6-62ab46daa7c8","order_by":15,"name":"Alex Kang","email":"","orcid":"","institution":"University of Washington","correspondingAuthor":false,"prefix":"","firstName":"Alex","middleName":"","lastName":"Kang","suffix":""},{"id":538462596,"identity":"8f9da772-3da5-4144-9ed9-4c60da5595be","order_by":16,"name":"Emily Joyce","email":"","orcid":"","institution":"University of Washington","correspondingAuthor":false,"prefix":"","firstName":"Emily","middleName":"","lastName":"Joyce","suffix":""},{"id":538462597,"identity":"2ea5c3af-2b0b-499c-8f86-9cb02b197e27","order_by":17,"name":"Asim Bera","email":"","orcid":"","institution":"University of Washington","correspondingAuthor":false,"prefix":"","firstName":"Asim","middleName":"","lastName":"Bera","suffix":""},{"id":538462598,"identity":"7075ffd9-b79a-47a9-9dde-b9b3d2046d6a","order_by":18,"name":"Aaron Schmitz","email":"","orcid":"","institution":"Washington University in St. Louis","correspondingAuthor":false,"prefix":"","firstName":"Aaron","middleName":"","lastName":"Schmitz","suffix":""},{"id":538462599,"identity":"76a71c6f-b238-4e68-941a-0094c3411394","order_by":19,"name":"Ali Ellebedy","email":"","orcid":"https://orcid.org/0000-0002-6129-2532","institution":"Washington University School of Medicine","correspondingAuthor":false,"prefix":"","firstName":"Ali","middleName":"","lastName":"Ellebedy","suffix":""},{"id":538462600,"identity":"be37e49d-7ccc-4d17-a396-2eee9549b946","order_by":20,"name":"Kelli Hvorecny","email":"","orcid":"https://orcid.org/0000-0002-4609-6170","institution":"University of Washington","correspondingAuthor":false,"prefix":"","firstName":"Kelli","middleName":"","lastName":"Hvorecny","suffix":""},{"id":538462601,"identity":"23dd9fde-dd81-45a3-a497-30f936929741","order_by":21,"name":"Gianluca Interlandi","email":"","orcid":"","institution":"University of Washington","correspondingAuthor":false,"prefix":"","firstName":"Gianluca","middleName":"","lastName":"Interlandi","suffix":""},{"id":538462602,"identity":"7520d7ce-d962-4c2d-bdbc-07965e3ecfcc","order_by":22,"name":"Mark Cartwright","email":"","orcid":"","institution":"Harvard University","correspondingAuthor":false,"prefix":"","firstName":"Mark","middleName":"","lastName":"Cartwright","suffix":""},{"id":538462603,"identity":"c9530c28-8302-4f58-85b4-f8d31415f726","order_by":23,"name":"Andyna Vernet","email":"","orcid":"","institution":"Harvard University","correspondingAuthor":false,"prefix":"","firstName":"Andyna","middleName":"","lastName":"Vernet","suffix":""},{"id":538462604,"identity":"b5b02613-f35a-4872-a010-5a279b7860c0","order_by":24,"name":"Sarai Bardales","email":"","orcid":"","institution":"Harvard University","correspondingAuthor":false,"prefix":"","firstName":"Sarai","middleName":"","lastName":"Bardales","suffix":""},{"id":538462605,"identity":"2a7fc080-fb9c-465e-9db6-883ecdbe0973","order_by":25,"name":"Desmond White","email":"","orcid":"","institution":"Harvard University","correspondingAuthor":false,"prefix":"","firstName":"Desmond","middleName":"","lastName":"White","suffix":""},{"id":538462606,"identity":"1c4e57a2-5f62-44bc-8e61-7dd651675fe4","order_by":26,"name":"Rachel Klevit","email":"","orcid":"https://orcid.org/0000-0002-3476-969X","institution":"University of Washington School of Medicine","correspondingAuthor":false,"prefix":"","firstName":"Rachel","middleName":"","lastName":"Klevit","suffix":""},{"id":538462607,"identity":"e2dd1464-e7b1-46c2-b1ef-57c5762adbe9","order_by":27,"name":"Evgeni Sokurenko","email":"","orcid":"","institution":"University of Washington","correspondingAuthor":false,"prefix":"","firstName":"Evgeni","middleName":"","lastName":"Sokurenko","suffix":""},{"id":538462608,"identity":"10bb9bbe-9057-4b88-be3d-e6b94e447abe","order_by":28,"name":"Scott Hultgren","email":"","orcid":"https://orcid.org/0000-0001-8785-564X","institution":"Department of Molecular Microbiology and Center for Women's Infectious Disease Research, Washington University School of Medicine","correspondingAuthor":false,"prefix":"","firstName":"Scott","middleName":"","lastName":"Hultgren","suffix":""},{"id":538462609,"identity":"06a8ff61-802b-4ba5-899c-9ee04f1d86e8","order_by":29,"name":"David Baker","email":"","orcid":"https://orcid.org/0000-0001-7896-6217","institution":"University of Washington","correspondingAuthor":false,"prefix":"","firstName":"David","middleName":"","lastName":"Baker","suffix":""}],"badges":[],"createdAt":"2025-10-27 11:15:58","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7951484/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7951484/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":95091424,"identity":"8bc24216-2361-4fea-9cf8-47e7b1f00c81","added_by":"auto","created_at":"2025-11-04 08:30:44","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":112898,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003e\u003cstrong\u003eDe novo\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e design of adhesin inhibitors. (A)\u003c/strong\u003e Schematic illustrating the structures on the gram-negative bacterial cell surface that are potentially accessible to protein-based inhibitors. Virulence factors that make up the bacterial pili, especially tip adhesins, are attractive targets due to their essential role in pathogenesis and the fact that they lie well off the membrane and beyond the protective lipopolysaccharide shell. Created with BioRender.com. \u003cstrong\u003e(B)\u003c/strong\u003e Example denoising trajectory for an Abp2D binder. “Hotspots” corresponding to the adhesin substrate binding pocket residues are highlighted in cyan.\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-7951484/v1/ca12784510b608bbc59e9c9a.png"},{"id":95091430,"identity":"ff6a9a94-abcf-46d8-9f77-d9d6f1857cba","added_by":"auto","created_at":"2025-11-04 08:30:44","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":200741,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003e\u003cstrong\u003eIn vitro\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e characterization of\u003c/strong\u003e\u003cem\u003e\u003cstrong\u003e \u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003eFimH minibinder F7.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(A)\u003c/strong\u003e Design model for FimH minibinder F7 (red) in complex with FimH LAS (gray).\u0026nbsp; \u003cstrong\u003e(B)\u003c/strong\u003e SEC trace of F7: sample was injected onto a Superdex 75 10/300 GL column. \u003cstrong\u003e(C) \u003c/strong\u003eThe binding affinity of F7 was determined with SPR. The SPR data indicate that F7 binds more tightly to the LAS (left; K\u003csub\u003ed\u003c/sub\u003e=119 nM) than HAS (right; K\u003csub\u003ed\u003c/sub\u003e=713 nM). \u003cstrong\u003e(D) \u003c/strong\u003eRaw CD data for F7. Full wavelength scans from 250 to 200 nm were performed at the indicated temperatures. \u003cstrong\u003e(E) \u003c/strong\u003eA\u003cstrong\u003e \u003c/strong\u003emelting curve for F7 indicates that it is thermostable up to 75°C. CD signal at 220 nm (helicity) was measured every 1°C.\u003cstrong\u003e (F) \u003c/strong\u003eCrystal structure of F7 (green) with FimH L34K (cyan) overlaid on the design model (red); inset: difference in FimH clamp loop between design and crystal structure showing an unexpected displacement of the clamp loop. The experimental structure exhibits a Ca RMSD of 0.69 Å to the design model.\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-7951484/v1/143ad7c6479de54d87905ac4.png"},{"id":95223389,"identity":"69c2fa03-28a7-4d55-ab19-207de91738bd","added_by":"auto","created_at":"2025-11-05 16:22:11","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":311099,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003e\u003cstrong\u003eIn cellulo \u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003echaracterization of\u003c/strong\u003e\u003cem\u003e\u003cstrong\u003e \u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003eFimH minibinder F7. (A)\u003c/strong\u003eInhibition ELISA results for minibinder F7 (pink; IC50=1.9 µM; 95% CI: 1.4-2.9 µM), noninhibitory minibinder A4 (teal), and mannose (black). Each experiment includes three replicates. Error bars show standard deviation. \u003cstrong\u003e(B) \u003c/strong\u003eDetachment ELISA results for F7 (pink; IC50=1.7 µM; 95% CI: 1.3-2.4 µM), A4 (teal), and mAb926 (black). Each experiment includes three replicates. Error bars show standard deviation. \u003cstrong\u003e(C)\u003c/strong\u003e Red blood cell aggregation inhibition assay with a serial dilution of F7 (MIC=69 nM).\u003cstrong\u003e (D)\u003c/strong\u003e Red blood cell disaggregation assay showing red blood cells incubated for one hour with bacteria (i), without bacteria (ii), with bacteria and 2% mannose (iii), incubated for 1 hour with bacteria and then for 15 minutes with minibinder F7 (iv).\u003cstrong\u003e (E) \u003c/strong\u003eEffect of F7 on bacterial adhesion to bladder epithelial cells. The addition of F7 decreases the number of bacteria present.\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-7951484/v1/83f7d4fa2c260660def62ec6.png"},{"id":95091426,"identity":"a88ab9f6-1bcd-4515-89c0-30eb92af97df","added_by":"auto","created_at":"2025-11-04 08:30:44","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":191853,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003e\u003cstrong\u003eIn vitro\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e characterization of\u003c/strong\u003e\u003cem\u003e\u003cstrong\u003e \u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003eAbp minibinder A7.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(A)\u003c/strong\u003e Design model for Abp minibinder A7 (red) in complex with Abp2D (gray).\u0026nbsp; \u003cstrong\u003e(B)\u003c/strong\u003e SEC trace of A7: sample was injected onto a Superdex 75 10/300 GL column. (\u003cstrong\u003eC) \u003c/strong\u003eThe binding affinity of A7 was determined with SPR. The SPR data indicate that A7 binds with high affinity to Abp1D (left; K\u003csub\u003ed\u003c/sub\u003e=50.4 nM) and Abp2D (right; K\u003csub\u003ed\u003c/sub\u003e=3.5 nM). \u003cstrong\u003e(D) \u003c/strong\u003eRaw CD data for A7. Full wavelength scans from 260 to 200 nm were performed at the indicated temperatures.\u0026nbsp; \u003cstrong\u003e(E) \u003c/strong\u003eA\u003cstrong\u003e \u003c/strong\u003emelting curve for A7 indicates that it is thermostable up to 80°C. CD signal at 220 nm (helicity) was measured every 1°.\u003cstrong\u003e (F) \u003c/strong\u003eCrystal structure of A7 (green) with Abp2D (cyan) overlaid on the design model (red); inset: beta-strand pairing interaction between A7 and an edge strand of Abp2D.\u0026nbsp; The experimental structure exhibits a Ca RMSD of 0.88 Å to the design model.\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-7951484/v1/83ec6ebf4d3d0dfb2f7dde94.png"},{"id":95224521,"identity":"15a53767-2187-415c-865b-7ecd31eca2f4","added_by":"auto","created_at":"2025-11-05 16:23:50","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":104985,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003e\u003cstrong\u003eIn cellulo\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e characterization of Abp minibinder A7.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(A) \u003c/strong\u003e\u0026nbsp;Inhibition ELISA results for minibinder A7 (IC50=2.6 nM; 95% CI: 1.8-3.0 nM) with \u003cem\u003eA. baumannii\u003c/em\u003e ACICU. Each experiment includes three biological replicates. Error bars show standard deviation. \u003cstrong\u003e(B) \u003c/strong\u003eDetachment ELISA results for A7 (IC50=40.0 nM; 95% CI: 25.5-52.1 nM) and \u003cem\u003eA. baumannii\u003c/em\u003e ACICU. Each experiment includes three replicates. Error bars show standard deviation. \u003cstrong\u003e(C)\u003c/strong\u003e A7 significantly inhibits the formation of biofilms (left) and disperses existing biofilms (right; n=3 independent replicates). One-way ANOVA test. ****P ≤ 0.0001, ***P ≤ 0.001, **P ≤ 0.01, *P ≤ 0.05. \u003cstrong\u003e\u0026nbsp;(D) \u003c/strong\u003eFluorescence imaging of fibrinogen-coated silicone tubing reveals differential adherence of \u003cem\u003eA. baumannii\u003c/em\u003e ACICU after pretreatment with or dispersion by A7 or a BSA control.\u003cstrong\u003e \u003c/strong\u003eThe catheters in this panel were captured in a single imaging session and uniform processing was applied to the entire image. Image representative of 4 independent replicates (except for 100 nM BSA detachment; n=3).\u003c/p\u003e","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-7951484/v1/2299be83fdb6f6226bfc3779.png"},{"id":95091425,"identity":"821fe2b2-8388-420e-bd1e-68a759ac272f","added_by":"auto","created_at":"2025-11-04 08:30:44","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":104607,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003e\u003cstrong\u003eIn vivo\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e characterization of\u003c/strong\u003e\u003cem\u003e\u003cstrong\u003e de novo\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e designed adhesin inhibitors. (A)\u003c/strong\u003e Schematic of UPEC UTI mouse model dosing regimen. Created with BioRender.com. \u003cstrong\u003e(B)\u003c/strong\u003e Titers of bacteria 48 hours post-infection (hpi) in the bladder after treatment with buffer (grey, n=34), minibinder F7 (pink, n=5), or monoclonal antibody (teal, n=21). Error bars show standard deviation. One-way ANOVA test. ****P ≤ 0.01, ***P ≤ 0.01, **P ≤ 0.01, *P ≤ 0.05.\u0026nbsp; \u003cstrong\u003e(C)\u003c/strong\u003e Schematic of ACICU CAUTI mouse model dosing regimen. Mice were infected with 10\u003csup\u003e8\u003c/sup\u003e cells of ACICU in a CAUTI model. Their catheters and bladders were processed for counting CFUs 3 hpi. Created with BioRender.com.\u003cstrong\u003e (D)\u003c/strong\u003e Bladder (left) and catheter (right) titers of ACICU and ACICU Δabp1 Δabp2 3 hpi. ACICU were treated with buffer control (gray, n=9) or A7 (pink, n=12). ACICU Δabp1 Δabp2 were treated with buffer (teal, n=11).\u0026nbsp; Error bars show standard deviation. One-way ANOVA test. **P ≤ 0.01, *P ≤ 0.05.\u003c/p\u003e","description":"","filename":"6.png","url":"https://assets-eu.researchsquare.com/files/rs-7951484/v1/f46b99031a0a2d8624316882.png"},{"id":104401750,"identity":"8bacbb97-aab3-4e93-a550-2f577e8c6c93","added_by":"auto","created_at":"2026-03-11 12:13:26","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1737797,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7951484/v1/78e6d727-4a81-4804-8129-4f6fc6560764.pdf"},{"id":95091431,"identity":"537c0814-b1df-41b6-92ae-a429c925162a","added_by":"auto","created_at":"2025-11-04 08:30:45","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":2853497,"visible":true,"origin":"","legend":"Adhesin paper - Nature Micro Submission - Supplement","description":"","filename":"AdhesinpaperNatureMicroSubmissionSupplement.docx","url":"https://assets-eu.researchsquare.com/files/rs-7951484/v1/466985f03750735d61411032.docx"}],"financialInterests":"\u003cb\u003eYes\u003c/b\u003e there is potential Competing Interest.\nACG, TRT, and DB are co-inventors on a provisional patent describing the adhesin minibinders (63/857,995). SJH is on the advisory board of Sequoia Vaccines Inc. SJH is a cofounder of Fimbrion Therapeutics that is developing FimH targeted therapies and may financially benefit if the company is successful. AHE received funding from Emergent BioSolutions, AbbVie, and Moderna that are unrelated to the data presented in the current study. AHE has received consulting and speaking fees from InBios International, Fimbrion Therapeutics, RGAX, Mubadala Investment Company, Moderna, Pfizer, GSK, Danaher, Third Rock Ventures, Goldman Sachs, and Morgan Stanley and is the founder of ImmuneBio Consulting. AJS and AHE are recipients of a licensing agreement with Abbvie that is unrelated to the data presented in the current study. MRT, KOT, JSP, KWD, AHE, and SJH are inventors on the US patent application US23/82426 regarding Abp2D monoclonal antibodies.","formattedTitle":"De Novo Design of Miniprotein Inhibitors of Bacterial Adhesins","fulltext":[{"header":"Main","content":"\u003cp\u003eMultidrug-resistant (MDR) bacterial pathogens constitute a significant and growing threat to public health (\u003cem\u003e1\u003c/em\u003e). To counter the alarming rise of these increasingly untreatable and deadly bacterial infections, accelerating the development of novel antimicrobial therapies is essential. Recent advances in protein design have opened the door to designing small (\u0026lt; 120 amino acids) \u003cem\u003ede novo\u003c/em\u003e protein binders (minibinders) that precisely target virtually any epitope (\u003cem\u003e2\u003c/em\u003e,\u003cem\u003e3\u003c/em\u003e), enabling therapeutics against viruses (\u003cem\u003e4\u003c/em\u003e), toxins (\u003cem\u003e5\u003c/em\u003e), and other disease targets (\u003cem\u003e6\u003c/em\u003e). However, a major challenge in applying these tools to combat bacteria is the accessibility of minbinders to intracellular and membrane-embedded target antigens due to their shielding by surface molecules such as lipopolysaccharides. Previous attempts have identified high-affinity protein binders to their integral outer membrane protein target antigens \u003cem\u003ein vitro\u003c/em\u003e, only to find that the binders cannot access their targets in a cellular context (7,\u003cem\u003e8\u003c/em\u003e).\u003c/p\u003e\n\u003cp\u003eWe reasoned that bacterial adhesins would constitute ideal targets for designed minibinder therapeutics targeting bacteria because they lie well beyond bacterial surface molecules and play a key role in the virulence of MDR bacteria by mediating their attachment to and invasion of host epithelial cells as well as biofilm formation (Fig. 1a). Among the diverse family of bacterial adhesins, chaperone-usher pathway (CUP) pili adhesins lie at the tips of extracellular fibers, and recognize receptors in a lectin-substrate interaction that mediates host tropisms. CUP pili are ubiquitous among gram-negative bacteria and are critical for the establishment and persistence of numerous bacterial infections, including both uncomplicated urinary tract infections (UTIs) and catheter-associated UTIs (CAUTIs; Fig. 1a). UTIs are one of the most common bacterial infections among women and their treatment accounts for approximately 15% of antibiotic usage in the United States (\u003cem\u003e9\u003c/em\u003e,\u003cem\u003e10\u003c/em\u003e). Over 50% of UTIs, especially those acquired in hospitals, are caused by MDR bacteria, limiting the effectiveness of a dwindling number of traditional antibiotic treatments (\u003cem\u003e11-13\u003c/em\u003e). Previous anti-adhesin therapeutic strategies have relied on small molecule ligand mimetic inhibitors (\u003cem\u003e14\u003c/em\u003e) or monoclonal antibodies (mAbs) (\u003cem\u003e15\u003c/em\u003e,\u003cem\u003e16\u003c/em\u003e). However, these strategies require identification of the adhesin substrate, which is often unknown, or labor-intensive screening for inhibitory mAbs. Designed minibinders could overcome this therapeutic bottleneck by enabling the rapid development of antibiotic-sparing therapies, relying solely on the adhesin structure and identification of its putative binding pocket. To investigate this, we sought to design minibinders against three CUP adhesins: FimH from \u003cem\u003eEscherichia coli\u003c/em\u003e, which accounts for 70-90% of UTIs, and Abp1D and Abp2D from \u003cem\u003eAcinetobacter baumannii\u003c/em\u003e, which is prevalent in hospital-acquired CAUTIs and is often multidrug resistant.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003e\u003cem\u003eDe novo\u003c/em\u003e design of minibinders targeting FimH\u003c/p\u003e\n\u003cp\u003eUropathogenic \u003cem\u003eE. coli\u0026nbsp;\u003c/em\u003e(UPEC) expresses type 1 pili, a prototypical CUP pilus, which are tipped with the mannose binding adhesin FimH (\u003cem\u003e17\u003c/em\u003e). UPEC FimH is essential for bacterial attachment to the bladder epithelium via mannosylated glycoproteins (\u003cem\u003e18\u003c/em\u003e). After attachment, UPEC forms intracellular bacterial communities that are recalcitrant to antibiotics and, if left untreated, can lead to recurrent UTIs and kidney infection (\u003cem\u003e18\u003c/em\u003e). FimH consists of an N-terminal mannose-binding lectin domain and a C-terminal pilin domain that attaches to the rest of the pilus (\u003cem\u003e18,19\u003c/em\u003e). The lectin domain is an allosteric protein that samples a conformational equilibrium between two well-characterized conformational states: a low-affinity state (LAS, also known as the \u0026ldquo;Tense\u0026rdquo; state) and a high-affinity state (HAS, also known as the \u0026ldquo;Relaxed\u0026rdquo; state) (\u003cem\u003e20\u003c/em\u003e). In the LAS, the body of the FimH lectin domain interacts with the pilin domain, fixing it in place, and allosterically disrupts the mannose binding pocket, resulting in a low affinity for mannose (\u003cem\u003e18,19\u003c/em\u003e). In the HAS, the lectin domain has little to no interaction with the pilin domain, allowing it to rotate freely relative to the pilin domain, and the mannose binding pocket loops are stably positioned to bind to mannose (\u003cem\u003e18\u003c/em\u003e).\u003c/p\u003e\n\u003cp\u003eWe used RFdiffusion (\u003cem\u003e3\u003c/em\u003e) to generate novel minibinder backbones against both states of FimH. We targeted designs to the binding pocket of published crystal structures (FimH in LAS: 3JWN; FimH in HAS: 1UWF) by specifying \u0026ldquo;hotspot\u0026rdquo; residues in the binding pocket as inputs to the model (Fig. 1b). We subsequently assigned sequences to these backbones using ProteinMPNN (\u003cem\u003e21\u003c/em\u003e) and scored designs with AlphaFold2 (AF2) Initial Guess (\u003cem\u003e22\u003c/em\u003e). The best sequences ranked by AF2 folding and interaction confidence metrics (predicted local distance difference test (pLDDT) and predicted aligned error (PAE) scores) were then iteratively refined by partially diffusing (\u003cem\u003e23\u003c/em\u003e) the AF2 predicted structures with RFdiffusion, reassigning sequences with ProteinMPNN, and filtering with AF2. We obtained an oligonucleotide array encoding the top 10,000 designs by AF2 and Rosetta interface metrics.\u003c/p\u003e\n\u003cp\u003eExperimental characterization of FimH minibinders\u003c/p\u003e\n\u003cp\u003eFimH minibinders were screened via cDNA display, rather than a cell-based screening approach like yeast surface display, because FimH binds to mannose on the cell surface. We panned against the wild-type FimH lectin domain, which adopts the HAS, and a mutant lectin domain stabilized in the LAS (\u003cem\u003e24\u003c/em\u003e). The top 96 designs enriched more than four-fold over the naive library and consisted of 78 binders identified through the HAS sorts and 39 binders identified through the LAS sorts. Twenty-one designs were enriched against both states. Eighty-one of the enriched sequences were designed against the LAS.\u003c/p\u003e\n\u003cp\u003eWe expressed these 96 FimH minibinders in \u003cem\u003eE. coli\u003c/em\u003e and found that 79 exhibited monodisperse size exclusion chromatography (SEC) traces and expressed sufficiently for downstream characterization (see Materials and Methods for details). In an initial surface plasmon resonance (SPR) binding screen, nine of these designs exhibited sub-micromolar binding against at least one of the FimH states (table S3). All nine designs were originally designed against the LAS. To directly assess the ability of FimH minibinders to inhibit and disrupt bacterial adhesion, we performed a series of red blood cell (RBC) aggregation assays using \u003cem\u003eE. coli\u0026nbsp;\u003c/em\u003estrain KB23 expressing type 1 fimbriae. \u003cem\u003eE. coli\u0026nbsp;\u003c/em\u003eaggregates RBCs by binding to mannosylated glycoproteins on their surface via FimH. Of the 96 minibinders we tested, eight significantly inhibited RBC aggregation (Fig. S1), seven of which also had submicromolar affinities in the SPR screen. Minibinder F7 had the lowest minimum inhibitory concentration (69 nM) in the RBC aggregation inhibition assay and was therefore selected for further characterization (Fig. S2). F7 was designed against the LAS, expressed well in \u003cem\u003eE. coli\u003c/em\u003e, and purified as a monodisperse species (Fig. 2a, 2b). Though it was designed against the LAS, F7 was identified by panning against the HAS. Further characterization by SPR confirmed that F7 binds to both conformations, but with a several-fold higher affinity to the LAS (K\u003csub\u003ed\u003c/sub\u003e=119 nM) than HAS (K\u003csub\u003ed\u003c/sub\u003e=713 nM) (Fig. 2c). F7 exhibited the expected circular dichroism (CD) spectra and proved to be highly thermostable, with a melting point above 75\u0026deg;C (Fig. 2d, 2e).\u003c/p\u003e\n\u003cp\u003eCo-crystallization of F7 with a mutant FimH lectin domain stabilized in the LAS (L34K) (\u003cem\u003e25\u003c/em\u003e) confirmed the designed binding interface with a C\u003cem\u003ea\u003c/em\u003e RMSD of 0.69 \u0026Aring; to the design model (PDB Code: 9Q1V; Fig. 2f). The experimentally determined structure of FimH bound to F7 was much closer to the canonical LAS conformation than the HAS conformation (C\u003cem\u003ea\u003c/em\u003e RMSD of 0.45 \u0026Aring; vs 2.26 \u0026Aring;). As expected, the minibinder occluded the binding pocket of FimH. Unexpectedly, it also wedged between one of the mannose-binding loops and the clamp loop, further displacing it from the binding pocket compared to the design model. The clamp-loop displacement induced by F7 closely resembles that of the recently determined structure of FimH bound to anti-FimH antibody mAb926, which was raised against the LAS and known to function through a parasteric mechanism (fig. S3) (\u003cem\u003e26\u003c/em\u003e).\u003c/p\u003e\n\u003cp\u003eGiven that F7 preferentially binds the LAS-stabilized mutant FimH and that the crystal structure of FimH in complex with the minibinder resembles the LAS, we suspected that F7 binding may induce a shift from the HAS to the LAS in wild-type FimH lectin domain. We compared F7\u0026rsquo;s interaction with the L34K variant of the lectin domain (LAS) and wild-type lectin domain (HAS) by (\u003csup\u003e15\u003c/sup\u003eN, \u003csup\u003e1\u003c/sup\u003eH)-HSQC NMR spectroscopy. The substantial structural differences between the LAS and HAS conformations of the isolated lectin domain are reflected in their distinct NMR spectra, which display very little peak overlap (fig. S4A, S4B). Addition of F7 to the LAS variant caused chemical shift perturbations localized at the binding interface, consistent with a binding event and/or local change in conformation, but not a global conformational change (fig. S5A). In contrast, the spectrum of the WT HAS domain showed widespread chemical shift changes, becoming similar to that of the LAS variant (fig. S5A). Notably, spectra of the F7-bound HAS and LAS variants now show extensive peak overlap, indicating a shared conformation (fig. S4C). Together, these data indicate that F7 binding to the isolated WT lectin domain induces a conformational change from the HAS to the LAS, shifting the conformational equilibrium in the opposite direction as binding the native ligand and existing mannoside therapeutics (\u003cem\u003e20\u003c/em\u003e).\u003c/p\u003e\n\u003cp\u003eWe quantitatively assessed the ability of minibinder F7 to block bacterial adhesion to surfaces coated with bovine RNAseB\u0026mdash;a model glycoprotein rich in Man\u003csub\u003e5\u003c/sub\u003e high-mannose type oligosaccharides that avidly binds FimH even in the LAS. We preincubated\u003cem\u003e\u0026nbsp;E. coli\u0026nbsp;\u003c/em\u003eKB23 with either F7, alpha-methyl-mannoside, or a minibinder that showed no inhibition in RBC assays (A4) and then measured its binding to RNaseB. F7 displayed an IC50 of 1.9 \u0026micro;M (95% CI: 1.4-2.9 \u0026micro;M), a more than two-thousand-fold improvement over mannose (Fig. 3a). We then performed a detachment ELISA by first adhering bacteria to RNAseB-coated plates and subsequently introducing either F7, noninhibitory minibinder A4, or mAb926, which is known to effectively dissolve preformed biofilms (27). F7 but not A4 successfully detached UPEC with an IC50 of 1.7 \u0026micro;M (95% CI: 1.3-2.4 \u0026micro;M), but fell short of mAb926 by two orders of magnitude (Fig. 3b). We also sought to characterize F7\u0026rsquo;s inhibitory ability in a series of more biologically relevant RBC assays. F7 successfully inhibited RBC aggregation by\u003cem\u003e\u0026nbsp;E. coli\u0026nbsp;\u003c/em\u003ewith a minimum inhibitory concentration of 69 nM (Fig. 3c), exceeding by more than a thousand-fold that of alpha-methyl-mannoside (fig. S6). We tested the ability of F7 to reverse bacterial aggregation by adding it to pre-aggregated mixtures of RBC and \u003cem\u003eE. coli\u003c/em\u003e. F7 successfully redispersed the RBCs within 15 minutes of introduction at 1 \u0026micro;g/ml (Fig. 3d). We confirmed the physiological relevance of F7\u0026apos;s inhibitory ability by comparing the adherence of bacteria to T24 urinary bladder tissue culture cells in the presence and absence of minibinder. F7-treated cells had considerably fewer bacteria present than either untreated cells (Fig. 3e) or cells treated with A4 (fig. S7).\u003c/p\u003e\n\u003cp\u003eGiven the sequence diversity of FimH across disease-causing strains, we tested whether F7 is inhibitory not only to the FimH variant it was designed against but also to different FimH variants expressed by the most common highly-virulent and/or multi-drug resistant clonal groups of UPEC. F7 successfully inhibited RBC aggregation by strains from the clonal groups ST73, ST95, ST69, and ST131-H30 (fig. S8, table S4). F7 also inhibited aggregation by \u003cem\u003eE. coli\u003c/em\u003e strain cas665 expressing \u003cem\u003eKlebsiella pneumoniae\u0026nbsp;\u003c/em\u003eFimH, which has approximately 85% sequence identity to\u003cem\u003e\u0026nbsp;E. coli\u0026nbsp;\u003c/em\u003eFimH, suggesting that F7 may be cross-reactive with FimH variants from other bacterial pathogens.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eDe novo\u003c/em\u003e design of minibinders targeting Abp adhesins\u003c/p\u003e\n\u003cp\u003eCAUTIs caused by MDR bacteria pose a significant economic toll on the healthcare system in the United States (\u003cem\u003e28\u003c/em\u003e). CAUTIs represent approximately 40% of hospital-acquired infections and have an increased risk of morbidity and mortality compared to uncomplicated UTIs (\u003cem\u003e28,29\u003c/em\u003e). CAUTIs result in a serious deterioration of quality of life through pain, discomfort, disruption of daily activities, and increased healthcare costs, exacerbated by the rapid spread of drug resistance in uropathogens (\u003cem\u003e28\u0026ndash;32\u003c/em\u003e).\u003c/p\u003e\n\u003cp\u003eOne particularly concerning CAUTI pathogen is \u003cem\u003eA. baumannii\u003c/em\u003e, which is classified by the CDC as a pathogen of urgent concern due to its high rates of multidrug resistance\u003cem\u003e\u0026nbsp;\u003c/em\u003eand increases in relative prevalence in CAUTIs compared to uncomplicated UTIs (\u003cem\u003e33\u003c/em\u003e). The \u003cem\u003eA. baumannii\u003c/em\u003e strain \u0026ldquo;ACICU\u0026rdquo; possesses two highly similar CUP pili, Abp1 and Abp2, that play critical roles in mediating CAUTI. The Abp1 and Abp2 pili are tipped with the Abp1D and Abp2D adhesins that allow these bacteria to bind to host-deposited fibrinogen on implanted catheters, which is a critical step for \u003cem\u003eA. baumannii\u0026nbsp;\u003c/em\u003eto establish a CAUTI (\u003cem\u003e34\u003c/em\u003e). While the receptor binding domains of these adhesins share only 70% sequence identity, they share a higher degree of structural homology as seen in previously solved crystal structures (Abp1D: 8DF0; Abp2D: 8DEZ).(34) Previous structure-function studies identified the putative fibrinogen-binding pockets of these adhesins; however, the exact structural binding interactions between fibrinogen and the adhesins remain unknown (\u003cem\u003e34\u003c/em\u003e). Similar to the dynamic conformational ensembles displayed by FimH, the binding affinity of the \u003cem\u003eA. baumannii\u003c/em\u003e Abp1D and Abp2D receptor binding domains (RBD) are controlled via dynamics of the flexible anterior binding loop, illustrated by comparing the loop\u0026rsquo;s conformation in the Abp2D crystal structure to an AlphaFold2 model for this protein (fig. S9). In addition, while vaccination with Abp2D protected mice from \u003cem\u003eA. baumannii\u003c/em\u003e CAUTI, these efforts to immunize mice with Abp2D failed to identify sufficiently inhibitory mAbs to effectively disrupt binding of ACICU to fibrinogen (IC50s \u0026gt; 100 \u0026micro;M) in a bacterial ELISA assay (fig. S10) (\u003cem\u003e35\u003c/em\u003e). Thus, we were motivated to apply protein design tools to develop more effective Abp inhibitors (\u003cem\u003e36\u003c/em\u003e).\u003c/p\u003e\n\u003cp\u003eWe used RFdiffusion (\u003cem\u003e3\u003c/em\u003e) with \u0026ldquo;hotspot\u0026rdquo; conditioning to generate novel minibinder backbones specifically targeting the putative fibrinogen-binding sites of both the Abp1D and Abp2D adhesins (Fig. 1b). Taking into account the anterior loop conformational differences between the high-affinity and low-affinity states, we designed backbones against the crystal structures of Abp1D and Abp2D receptor binding domains, which display the anterior binding loop in the open (low-affinity) conformation, exposing the putative binding pocket. We assigned sequences to these backbones using ProteinMPNN, predicted their structures with AF2 Initial Guess, and then iteratively optimized high-quality designs as described above. The top ~5000 scoring designs against Abp1D (~1300) and Abp2D (~3700), with a maximum of 15 sequences per backbone, were obtained as a DNA oligo chip.\u003c/p\u003e\n\u003cp\u003eExperimental\u003cem\u003e\u0026nbsp;\u003c/em\u003echaracterization of Abp minibinders\u003c/p\u003e\n\u003cp\u003eAbp adhesins do not bind mannose and are therefore amenable to screening minibinders using yeast surface display. Designed Abp binder sequences were amplified from the DNA oligo chip and cloned into a yeast display expression vector using yeast cloning. The resulting yeast display library was enriched for binding via two rounds of fluorescence-activated cell sorting (FACS) with 1 \u0026mu;M biotinylated Abp1D or Abp2D lectin domains using PE-conjugated streptavidin as a detection reagent. A third round of FACS sorting involved staining the enriched populations with 10-fold titrations of these adhesins from 1000 nM to 1 nM, revealing strong binding signal for the Abp2D-enriched population down to 1 nM, and binding signal down to 10 nM for the Abp1D-enriched population (fig. S11). Next-generation sequencing of sorted populations identified 31 highly enriched binder sequences across both populations, including some sequences that were enriched in both Abp1D and Abp2D sorts, suggesting that we identified cross-reactive binders. Interestingly, all 31 of these enriched binders were designed using the Abp2D crystal structure, which might be explained by the presence of a citrate molecule in the Abp2D crystal structure that opens its putative binding pocket and exposes more hydrophobic residues than in the Abp1D structure.\u003c/p\u003e\n\u003cp\u003eMinibinder sequences that were enriched during yeast surface display with Abp adhesins were expressed in and purified from \u003cem\u003eE. coli\u003c/em\u003e. Of the 31 enriched binders, twenty-four exhibited primarily monodisperse SEC traces and had sufficient yields for downstream characterization. An initial SPR binding screen of these 24 binders against both Abp1D and Abp2D identified 16 binders that exhibited sub-micromolar affinity to Abp2D, four of which also exhibited sub-micromolar affinity to Abp1D (table S5). A follow-up SPR experiment on the four cross-reactive binders revealed that Abp minibinder A7 (design model in Fig. 4a) exhibited high affinity to both Abp1D (K\u003csub\u003ed\u003c/sub\u003e=50.4 nM) and Abp2D (K\u003csub\u003ed\u003c/sub\u003e=3.5 nM; Fig. 4c). A7 expressed well (~50 mg/L of culture), behaved as a monodisperse species as assessed by SEC (Fig. 4b), exhibited the expected CD spectra (Fig. 4d), and was thermostable with a melting temperature greater than 85\u0026deg;C (Fig. 4e). Next, we confirmed the specificity of the designed adhesin binders using SPR. Minibinder A7 exhibited undetectable levels of binding to FimH LAS or HAS. Similarly, FimH minibinder F7 exhibited insignificant binding to Abp1D and Abp2D (fig. S12).\u003c/p\u003e\n\u003cp\u003eDetermining the structure of a co-crystal of A7 with Abp2D confirmed the designed interface, and exhibited a C\u003cem\u003ea\u003c/em\u003e RMSD of 0.88 \u0026Aring; to the design model (PDB Code: 9Q1H; Fig. 4f). This structure confirms the designed binding mode that utilizes a helix-turn-strand motif to simultaneously stabilize the core of the minibinder, insert a loop deep into the putative fibrinogen binding pocket of Abp2D, and strand pair with an edge strand of this adhesin. As designed, A7 stabilizes the anterior binding loop of Abp2D in an open conformation, thus neutralizing the adhesin in the low-affinity conformation and matching the loop conformation in the Abp2D crystal structure.\u003c/p\u003e\n\u003cp\u003eTo assess the ability of A7 to inhibit bacterial adhesion, we performed a bacterial ELISA assay with \u003cem\u003eA. baumannii\u003c/em\u003e ACICU. When preincubated with ACICU prior to being added to the fibrinogen-coated ELISA plate, A7 but not a BSA control significantly inhibited ACICU binding with an IC50 of 2.6 nM (95% CI: 1.8-3.0 nM; Fig. 5a). In a \u0026ldquo;detachment\u0026rdquo; ELISA, A7 successfully displaced adherent ACICU and exhibited an IC50 of 40.0 nM (95% CI: 25.5-52.1 nM; Fig. 5b). Since bacterial adhesins are known to contribute to biofilm formation (34), we also tested whether designed inhibitors could both prevent ACICU from forming biofilms and disperse preformed biofilms. Crystal violet staining of \u003cem\u003eA. baumannii\u0026nbsp;\u003c/em\u003ebiofilms grown on PVC plates demonstrated that 1 \u0026micro;M of A7 could prevent the formation of ACICU biofilms, unlike a BSA control (p \u0026lt; 0.0001, one-way ANOVA; Fig. 5c). In addition, 1 \u0026micro;M of A7 but not a BSA control could almost completely disperse preformed ACICU biofilms after 2 hours of treatment (p=0.0021, one-way ANOVA), mimicking the phenotype of an ACICU \u0026Delta;abp1 \u0026Delta;abp2 double knockout strain (Fig. 5c; p=0.0003, one-way ANOVA).\u003c/p\u003e\n\u003cp\u003eTo explore whether minibinder A7 could disrupt \u003cem\u003eA. baumannii\u003c/em\u003e ACICU biofilms in a more\u003cem\u003e\u0026nbsp;in vivo\u003c/em\u003e-like\u003cem\u003e\u0026nbsp;\u003c/em\u003econtext, inhibition and detachment studies were performed with fibrinogen-coated silicone catheters (\u003cem\u003e34,37\u003c/em\u003e). After coating silicone tubing with fibrinogen, ACICU was allowed to bind in the presence of A7 or BSA. We also tested the ability of A7 and BSA to disperse ACICU that had already bound to the silicone catheter. Immunostaining the tubing revealed that despite fibrinogen coating each piece of tubing, 100 nM of A7 but not the BSA control significantly prevented ACICU attachment to and dispersed prebound bacteria from the tubing, mimicking the phenotype of the ACICU \u0026Delta;abp1 \u0026Delta;abp2 strain (Fig. 5d, fig. S13, p=0.0052, one-way ANOVA).\u003c/p\u003e\n\u003cp\u003eClinical models of adhesin inhibitors\u003c/p\u003e\n\u003cp\u003eWe next explored whether these adhesin inhibitors would function \u003cem\u003ein vivo\u003c/em\u003e in clinical models of UTI. To test anti-FimH minibinder F7\u0026rsquo;s ability to combat uncomplicated UTI, mice were injected intraperitoneally with either 100 \u0026micro;g of the minibinder, mAb926, or a buffer-only control 24 hours prior to bacterial challenge. Mice were then challenged with model uropathogenic \u003cem\u003eE. coli\u003c/em\u003e strain CFT073 from the clonal group ST73 (10\u003csup\u003e6\u003c/sup\u003e CFU), and a second dose of protein was administered by intraperitoneal (IP) injection 24 hours after infection. Forty-eight hours post-infection, bacterial burden in the bladder was quantified (Fig. 6a). F7-treated mice showed a significant (~2 log) reduction in bacterial bladder titers compared to the mock control group (p = 0.034, one-way ANOVA) and reached similar levels of reduction as the antibody-treated group (Fig. 6b).\u003c/p\u003e\n\u003cp\u003eWe next tested anti-Abp minibinder A7 in a mouse model of CAUTI. Mice were catheterized and immediately infected with \u003cem\u003eA. baumannii\u003c/em\u003e ACICU (10\u003csup\u003e8\u003c/sup\u003e CFU) premixed with 0.25 mg/mL of minibinder or buffer. After 1 hour, mice were treated with IP injections of 100 \u0026micro;g of A7 or a buffer-only control. After 3 hours, bacterial burden in the bladder and catheter of each mouse was quantified (Fig. 6c). Mice that were treated with A7 exhibited significantly reduced bacterial loads in the bladder (p = 0.0035, one-way ANOVA) and catheter (p = 0.0014, one-way ANOVA) compared to mock-treated controls (Fig. 6d). This reduction in bacterial burden aligns with previous studies showing only modest (~1 log) attenuation of bacterial burden by ACICU \u0026Delta;abp1\u003cem\u003e\u0026nbsp;\u003c/em\u003e\u0026Delta;abp2, suggesting alternative adhesion mechanisms may be involved in this mouse model (\u003cem\u003e34\u003c/em\u003e). This proof of concept experiment lays the ground for future antibiotic-sparing strategies against pathogens of great concern.\u003c/p\u003e\n\u003cp\u003eOptimization of adhesin inhibitors\u003c/p\u003e\n\u003cp\u003eWe explored two routes to generating higher affinity designs. First, we used ProteinMPNN to resample sequences based on the designed backbone of minibinder F7, and selected variants with improved AF2 Initial Guess scores; this led to the identification of a new minibinder (C8) exhibiting ~10-fold higher affinity for FimH (LAS K\u003csub\u003ed\u003c/sub\u003e=15.0nM; HAS K\u003csub\u003ed\u003c/sub\u003e=243nM) and more than ~30-fold greater inhibitory activity in an inhibition ELISA with an IC50 of 37 nM (95% CI: 31-47 nM; Fig. S14, S15). Second, to enhance affinity via avidity effects, minibinder A7 was fused to de novo designed homo-oligomeric domains (38), generating multivalent binders. Oligomeric constructs such as oligomer C11 displayed significantly increased binding affinity to both adhesins, with K\u003csub\u003ed\u003c/sub\u003es in the picomolar range for Abp1D and Abp2D (fig. S16). In cellulo, oligomer C11 exhibits a significantly lower inhibition ELISA IC50 (0.36 nM) than A7 (95% CI: 0.15-0.61 nM; fig. S17). Such affinity and avidity-increasing strategies should be generally applicable for further improving the \u003cem\u003ein vivo\u003c/em\u003e efficacy of designed anti-adhesin minibinders.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003e\u003c/p\u003e\n\u003cp\u003eIn this study we show that \u003cem\u003ede novo\u003c/em\u003e protein design tools can be employed to design potent and specific minibinder inhibitors of bacterial adhesins, virulence factors critical for colonization and biofilm formation in a broad range of bacterial infections, including UPEC UTI and \u003cem\u003eA. baumannii\u003c/em\u003e CAUTI. The choice of target was key to \u003cem\u003ein vivo\u003c/em\u003e success, as the extracellular nature of these adhesins made them readily accessible to the designed proteins. Being able to specifically target minibinders to the substrate binding epitope using RFdiffusion with hotspot conditioning led to neutralizing binders without the need for antibody screening or knowledge of the native substrates. The conversion of FimH HAS to LAS by minibinder F7 shows that computational design allows not only the selection of target epitope but also target conformation. Targeting the substrate binding pocket of these adhesins should make it unlikely that resistance to these binders will arise quickly, as any mutation to the binding pocket may also disrupt adhesion. The specificity of these designed proteins for their targets should enable selective depletion of these pathogens without affecting commensal microbiota and limit pressure among the broader bacterial community to evolve resistance to the minibinders (\u003cem\u003e39\u003c/em\u003e). Future studies should explore whether resistance can be evolved to escape these minibinders and how they affect the broader microbiome.\u003c/p\u003e\n\u003cp\u003eFuture development of the adhesion-neutralizing therapeutics demonstrated here will require characterization and possibly additional optimization of the pharmacokinetics and bioavailability of these designed proteins. Blocking adhesin activity using minibinders has potential utility against non-UTI bacterial pathogens such as \u003cem\u003eStaphylococcus aureus\u003c/em\u003e, \u003cem\u003eStreptococcus pneumoniae,\u003c/em\u003e and \u003cem\u003eSalmonella\u003c/em\u003e (\u003cem\u003e40,41\u003c/em\u003e). The strategies developed here should be applicable to combat the adhesion and colonization of pathogens beyond bacteria, including fungi and amoebae (\u003cem\u003e42\u0026ndash;44\u003c/em\u003e).\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgments:\u0026nbsp;\u003c/strong\u003e Crystallographic diffraction data were collected at the Northeastern Collaborative Access Team beamlines at the Advanced Photon Source, which are funded by the National Institute of General Medical Sciences from the National Institutes of Health (P30 GM124165) and the ALS 4.2.2 beamline (P30 GM124169). This research used resources of the Advanced Photon Source, a US Department of Energy (DOE) Office of Science User Facility operated by Argonne National Laboratory under contract no. DE-AC02-06CH11357. This research used resources (FMX) of the National Synchrotron Light Source II, a U.S. Department of Energy (DOE) Office of Science User Facility operated for the DOE Office of Science by Brookhaven National Laboratory under Contract No. DE-SC0012704. The Center for BioMolecular Structure (CBMS) is primarily supported by the National Institutes of Health, National Institute of General Medical Sciences (NIGMS) through a Center Core P30 Grant (P30GM133893), and by the DOE Office of Biological and Environmental Research (KP1607011). Figures were made using Biorender.com. We would like to acknowledge Jacob M. Gershon, Hojae Choi, Yensi Flores Bueno, David S. Lee, Jeremiah Sims, Brian Coventry, and Jay Nix for materials, helpful discussions, and support; and L. Goldschmidt and K. Van Wormer for general operations.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding:\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eDefense Advanced Research Projects Agency grant HR0011-21-2-0012 (ACG, TRT)\u003c/p\u003e\n\u003cp\u003eNational Institutes of Health grant F30DK135390 (MRT)\u003c/p\u003e\n\u003cp\u003eNational Institutes of Health grant R01AI029549 (SJH)\u003c/p\u003e\n\u003cp\u003eNational Institutes of Health grant R37AI048689 (SJH)\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eNational Institutes of Health grant U19AI157797 (SJH, AHE)\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eNational Institutes of Health grant K99GM141364 (PM)\u003c/p\u003e\n\u003cp\u003eNational Institutes of Health grant R01AI171570 (REK, EVS)\u003c/p\u003e\n\u003cp\u003eThe Open Philanthropy Project Improving Protein Design Fund (ACG, TRT)\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eHoward Hughes Medical Institute (DB)\u003c/p\u003e\n\u003cp\u003eThe Schmidt Science Fellows, in partnership with the Rhodes Trust (ACH)\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor contributions:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eConceptualization: ACG, TRT, EDBL\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eMethodology: ACG, TRT, EDBL, PM, REK. PWE, KWD\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eInvestigation: ACG, TRT, EDBL, PM, DAS, EDBL, MC, AV, SB, DW, AH, KOT, AK, EJ, AKB, AM, PA, VT, IB, MRT, JSP, GI\u003c/p\u003e\n\u003cp\u003eVisualization: ACG, TRT, EDBL, PM, KLH\u003c/p\u003e\n\u003cp\u003eWriting: ACG, TRT, EDBL, KWD, KLH, DB, SJH, EVS\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eSupervision: DB, SJH, EVS, AHE\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthics declarations:\u0026nbsp;\u003c/strong\u003eACG, TRT, and DB are co-inventors on a provisional patent describing the adhesin minibinders (63/857,995). SJH is on the advisory board of Sequoia Vaccines Inc. SJH is a cofounder of Fimbrion Therapeutics that is developing FimH targeted therapies and may financially benefit if the company is successful. AHE received funding from Emergent BioSolutions, AbbVie, and Moderna that are unrelated to the data presented in the current study. AHE has received consulting and speaking fees from InBios International, Fimbrion Therapeutics, RGAX, Mubadala Investment Company, Moderna, Pfizer, GSK, Danaher, Third Rock Ventures, Goldman Sachs, and Morgan Stanley and is the founder of ImmuneBio Consulting. AJS and AHE are recipients of a licensing agreement with Abbvie that is unrelated to the data presented in the current study. MRT, KOT, JSP, KWD, AHE, and SJH are inventors on the US patent application US23/82426 regarding Abp2D monoclonal antibodies.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData and materials availability:\u003c/strong\u003e The crystal structures have been deposited to the Protein Data Bank (PDB) with accession codes 9Q1V (FimH-F7), 9Q1H (Abp2D-A7). All other data are available in the main text or the supplementary materials\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eVan Duin D, Paterson DL. Multidrug-Resistant Bacteria in the Community. Infect Dis Clin North Am. 2020 Dec;34(4):709\u0026ndash;22. \u003c/li\u003e\n\u003cli\u003eCao L, Coventry B, Goreshnik I, Huang B, Sheffler W, Park JS, et al. Design of protein-binding proteins from the target structure alone. Nature. 2022 May 19;605(7910):551\u0026ndash;60. \u003c/li\u003e\n\u003cli\u003eWatson JL, Juergens D, Bennett NR, Trippe BL, Yim J, Eisenach HE, et al. De novo design of protein structure and function with RFdiffusion. Nature. 2023 Aug 31;620(7976):1089\u0026ndash;100. \u003c/li\u003e\n\u003cli\u003eCao L, Goreshnik I, Coventry B, Case JB, Miller L, Kozodoy L, et al. De novo design of picomolar SARS-CoV-2 miniprotein inhibitors. \u003c/li\u003e\n\u003cli\u003eV\u0026aacute;zquez Torres S, Benard Valle M, Mackessy SP, Menzies SK, Casewell NR, Ahmadi S, et al. De novo designed proteins neutralize lethal snake venom toxins. Nature. 2025 Mar 6;639(8053):225\u0026ndash;31. \u003c/li\u003e\n\u003cli\u003eRoy A, Shi L, Chang A, Dong X, Fernandez A, Kraft JC, et al. De novo design of highly selective miniprotein inhibitors of integrins \u0026alpha;v\u0026beta;6 and \u0026alpha;v\u0026beta;8. Nat Commun. 2023 Sep 13;14(1):5660. \u003c/li\u003e\n\u003cli\u003eStorek KM, Auerbach MR, Shi H, Garcia NK, Sun D, Nickerson NN, et al. Monoclonal antibody targeting the \u0026beta;-barrel assembly machine of Escherichia coli is bactericidal. Proc Natl Acad Sci. 2018 Apr 3;115(14):3692\u0026ndash;7. \u003c/li\u003e\n\u003cli\u003eGokce-Alpkilic G, Huang B, Liu A, Kreuk LSM, Adebomi V, Bueso YF, et al. De novo design of high-affinity miniprotein binders targeting Francisella tularensis virulence factor. \u003c/li\u003e\n\u003cli\u003eChua KP, Fischer MA, Linder JA. Appropriateness of outpatient antibiotic prescribing among privately insured US patients: ICD-10-CM based cross sectional study. BMJ. 2019 Jan 16;k5092. \u003c/li\u003e\n\u003cli\u003eFleming-Dutra KE, Hersh AL, Shapiro DJ, Bartoces M, Enns EA, File TM, et al. Prevalence of Inappropriate Antibiotic Prescriptions Among US Ambulatory Care Visits, 2010-2011. JAMA. 2016 May 3;315(17):1864. \u003c/li\u003e\n\u003cli\u003eMadrazo M, Esparcia A, L\u0026oacute;pez-Cruz I, Alberola J, Piles L, Viana A, et al. Clinical impact of multidrug-resistant bacteria in older hospitalized patients with community-acquired urinary tract infection. BMC Infect Dis. 2021 Dec;21(1):1232. \u003c/li\u003e\n\u003cli\u003eWatkins RR, Thapaliya D, Lemonovich TL, Bonomo RA. Gepotidacin: a novel, oral, \u0026lsquo;first-in-class\u0026rsquo; triazaacenaphthylene antibiotic for the treatment of uncomplicated urinary tract infections and urogenital gonorrhoea. J Antimicrob Chemother. 2023 May 3;78(5):1137\u0026ndash;42. \u003c/li\u003e\n\u003cli\u003eWorby CJ, Schreiber HL, Straub TJ, Van Dijk LR, Bronson RA, Olson BS, et al. Longitudinal multi-omics analyses link gut microbiome dysbiosis with recurrent urinary tract infections in women. Nat Microbiol. 2022 May 2;7(5):630\u0026ndash;9. \u003c/li\u003e\n\u003cli\u003eCusumano CK, Pinkner JS, Han Z, Greene SE, Ford BA, Crowley JR, et al. Treatment and Prevention of Urinary Tract Infection with Orally Active FimH Inhibitors. Sci Transl Med [Internet]. 2011 Nov 16 [cited 2025 May 6];3(109). Available from: https://www.science.org/doi/10.1126/scitranslmed.3003021 \u003c/li\u003e\n\u003cli\u003eHvorecny KL, Interlandi G, Veth TS, Aprikian P, Manchenko A, Tchesnokova VL, et al. Antibodies disrupt bacterial adhesion by ligand mimicry and allosteric interference [Internet]. Immunology; 2024 [cited 2025 May 6]. Available from: http://biorxiv.org/lookup/doi/10.1101/2024.12.06.627246 \u003c/li\u003e\n\u003cli\u003eLopatto EDB, Santiago-Borges JM, Sanick DA, Malladi SK, Azimzadeh PN, Timm MW, et al. Monoclonal antibodies targeting the FimH adhesin protect against uropathogenic E. coli UTI [Internet]. Microbiology; 2024 [cited 2025 May 6]. Available from: http://biorxiv.org/lookup/doi/10.1101/2024.12.10.627638 \u003c/li\u003e\n\u003cli\u003eLe Trong I, Aprikian P, Kidd BA, Forero-Shelton M, Tchesnokova V, Rajagopal P, et al. Structural Basis for Mechanical Force Regulation of the Adhesin FimH via Finger Trap-like \u0026beta; Sheet Twisting. Cell. 2010 May;141(4):645\u0026ndash;55. \u003c/li\u003e\n\u003cli\u003eForoogh N, Rezvan M, Ahmad K, Mahmood S. Structural and functional characterization of the FimH adhesin of uropathogenic Escherichia coli and its novel applications. Microb Pathog. 2021 Dec;161:105288. \u003c/li\u003e\n\u003cli\u003eLe Trong I, Aprikian P, Kidd BA, Forero-Shelton M, Tchesnokova V, Rajagopal P, et al. Structural Basis for Mechanical Force Regulation of the Adhesin FimH via Finger Trap-like \u0026beta; Sheet Twisting. Cell. 2010 May;141(4):645\u0026ndash;55. \u003c/li\u003e\n\u003cli\u003eKalas V, Pinkner JS, Hannan TJ, Hibbing ME, Dodson KW, Holehouse AS, et al. Evolutionary fine-tuning of conformational ensembles in FimH during host-pathogen interactions. Sci Adv. \u003c/li\u003e\n\u003cli\u003eDauparas J, Anishchenko I, Bennett N, Bai H, Ragotte RJ, Milles LF, et al. Robust deep learning\u0026ndash;based protein sequence design using ProteinMPNN. \u003c/li\u003e\n\u003cli\u003eBennett NR, Coventry B, Goreshnik I, Huang B, Allen A, Vafeados D, et al. Improving de novo protein binder design with deep learning. Nat Commun. 2023 May 6;14(1):2625. \u003c/li\u003e\n\u003cli\u003eGl\u0026ouml;gl M, Krishnakumar A, Ragotte RJ, Goreshnik I, Coventry B, Bera AK, et al. Target-conditioned diffusion generates potent TNFR superfamily antagonists and agonists. Science. 2024 Dec 6;386(6726):1154\u0026ndash;61. \u003c/li\u003e\n\u003cli\u003eRodriguez VB, Kidd BA, Interlandi G, Tchesnokova V, Sokurenko EV, Thomas WE. Allosteric Coupling in the Bacterial Adhesive Protein FimH. J Biol Chem. 2013 Aug;288(33):24128\u0026ndash;39. \u003c/li\u003e\n\u003cli\u003eMagala P, Tuttle LM, Interlandi G, Carlucci LA, Mollica MY, Thomas WE, et al. Ultra-slow conformational dynamics and catch bond formation of a Bacterial Adhesin revealed by a single-domain variant of FimH. \u003c/li\u003e\n\u003cli\u003eHvorecny KL, Interlandi G, Veth TS, Aprikian P, Manchenko A, Tchesnokova VL, et al. Antibodies disrupt bacterial adhesion by ligand mimicry and allosteric interference [Internet]. Immunology; 2024 [cited 2025 Aug 4]. Available from: http://biorxiv.org/lookup/doi/10.1101/2024.12.06.627246 \u003c/li\u003e\n\u003cli\u003eKisiela DI, Avagyan H, Friend D, Jalan A, Gupta S, Interlandi G, et al. Inhibition and Reversal of Microbial Attachment by an Antibody with Parasteric Activity against the FimH Adhesin of Uropathogenic E. coli. Tsolis RM, editor. PLOS Pathog. 2015 May 14;11(5):e1004857. \u003c/li\u003e\n\u003cli\u003eMolina JJ, Flores-Mireles AL. CAUTIon \u0026mdash; not all UTIs are the same. Nat Rev Urol [Internet]. 2025 Aug 1 [cited 2025 Aug 15]; Available from: https://www.nature.com/articles/s41585-025-01065-z \u003c/li\u003e\n\u003cli\u003eYang X, Chen H, Zheng Y, Qu S, Wang H, Yi F. Disease burden and long-term trends of urinary tract infections: A worldwide report. Front Public Health. 2022 Jul 27;10:888205. \u003c/li\u003e\n\u003cli\u003eAndersen MJ, Flores-Mireles AL. Urinary Catheter Coating Modifications: The Race against Catheter-Associated Infections. Coatings. 2019 Dec 29;10(1):23. \u003c/li\u003e\n\u003cli\u003eFlores-Mireles A, Hreha TN, Hunstad DA. Pathophysiology, Treatment, and Prevention of Catheter-Associated Urinary Tract Infection. Top Spinal Cord Inj Rehabil. 2019 Jun;25(3):228\u0026ndash;40. \u003c/li\u003e\n\u003cli\u003eSaint S, Trautner BW, Fowler KE, Colozzi J, Ratz D, Lescinskas E, et al. A Multicenter Study of Patient-Reported Infectious and Noninfectious Complications Associated With Indwelling Urethral Catheters. JAMA Intern Med. 2018 Aug 1;178(8):1078. \u003c/li\u003e\n\u003cli\u003eHazen JE, Di Venanzio G, Hultgren SJ, Feldman MF. Catheterization of mice triggers resurgent urinary tract infection seeded by a bladder reservoir of Acinetobacter baumannii. Sci Transl Med. 2023 Jan 11;15(678):eabn8134. \u003c/li\u003e\n\u003cli\u003eTamadonfar KO, Di Venanzio G, Pinkner JS, Dodson KW, Kalas V, Zimmerman MI, et al. Structure\u0026ndash;function correlates of fibrinogen binding by Acinetobacter adhesins critical in catheter-associated urinary tract infections. Proc Natl Acad Sci. 2023 Jan 24;120(4):e2212694120. \u003c/li\u003e\n\u003cli\u003eHultgren S, Timm M, Tamadonfar K, Nye T, Pinkner J, Dodson K, et al. Vaccination with Acinetobacter baumannii adhesin Abp2D provides protection against catheter-associated urinary tract infection [Internet]. In Review; 2023 [cited 2025 Aug 15]. Available from: https://www.researchsquare.com/article/rs-3213777/v1 \u003c/li\u003e\n\u003cli\u003eHultgren S, Timm M, Tamadonfar K, Nye T, Pinkner J, Dodson K, et al. Vaccination with Acinetobacter baumannii adhesin Abp2D provides protection against catheter-associated urinary tract infection [Internet]. In Review; 2023 [cited 2025 May 31]. Available from: https://www.researchsquare.com/article/rs-3213777/v1 \u003c/li\u003e\n\u003cli\u003eDi Venanzio G, Flores-Mireles AL, Calix JJ, Haurat MF, Scott NE, Palmer LD, et al. Urinary tract colonization is enhanced by a plasmid that regulates uropathogenic Acinetobacter baumannii chromosomal genes. Nat Commun. 2019 Jun 24;10(1):2763. \u003c/li\u003e\n\u003cli\u003eEdman NI, Phal A, Redler RL, Schlichthaerle T, Srivatsan SR, Ehnes DD, et al. Modulation of FGF pathway signaling and vascular differentiation using designed oligomeric assemblies. Cell. 2024 Jul;187(14):3726-3740.e43. \u003c/li\u003e\n\u003cli\u003ePerez-Carrasco V, Soriano-Lerma A, Soriano M, Guti\u0026eacute;rrez-Fern\u0026aacute;ndez J, Garcia-Salcedo JA. Urinary Microbiome: Yin and Yang of the Urinary Tract. Front Cell Infect Microbiol. 2021 May 18;11:617002. \u003c/li\u003e\n\u003cli\u003eRosenow C, Ryan P, Weiser JN, Johnson S, Fontan P, Ortqvist A, et al. Contribution of novel choline‐binding proteins to adherence, colonization and immunogenicity of Streptococcus pneumoniae. Mol Microbiol. 1997 Sep;25(5):819\u0026ndash;29. \u003c/li\u003e\n\u003cli\u003eGhosh S, Chakraborty K, Nagaraja T, Basak S, Koley H, Dutta S, et al. An adhesion protein of Salmonella enterica serovar Typhi is required for pathogenesis and potential target for vaccine development. Proc Natl Acad Sci. 2011 Feb 22;108(8):3348\u0026ndash;53. \u003c/li\u003e\n\u003cli\u003eKim D, Kim JS, Bai X, Zhang J, Park M, Lee U, et al. Development of Miniprotein-Type Inhibitors of Biofilm Formation in Candida albicans and Candida auris. J Microbiol Biotechnol. 2025 Feb 25;35:e2411076. \u003c/li\u003e\n\u003cli\u003eAguirre Garc\u0026iacute;a M, Guti\u0026eacute;rrez-Kobeh L, L\u0026oacute;pez Vancell R. Entamoeba histolytica: Adhesins and Lectins in the Trophozoite Surface. Molecules. 2015 Feb 9;20(2):2802\u0026ndash;15. \u003c/li\u003e\n\u003cli\u003eKumari A, Tripathi AH, Gautam P, Gahtori R, Pande A, Singh Y, et al. Adhesins in the virulence of opportunistic fungal pathogens of human. Mycology. 2021 Oct 2;12(4):296\u0026ndash;324. \u003c/li\u003e\n\u003cli\u003eZielonka S, Krah S, editors. Genotype Phenotype Coupling: Methods and Protocols [Internet]. New York, NY: Springer US; 2020 [cited 2025 Jun 26]. (Methods in Molecular Biology; vol. 2070). Available from: http://link.springer.com/10.1007/978-1-4939-9853-1 \u003c/li\u003e\n\u003cli\u003eZeng Y, Woolley M, Chockalingam K, Thomas B, Arora S, Hook M, et al. Click display: a rapid and efficient in vitro protein display method for directed evolution. Nucleic Acids Res. 2023 Sep 8;51(16):e89\u0026ndash;e89. \u003c/li\u003e\n\u003cli\u003eWu L, Wen C, Qin Y, Yin H, Tu Q, Van Nostrand JD, et al. Phasing amplicon sequencing on Illumina Miseq for robust environmental microbial community analysis. BMC Microbiol. 2015 Dec;15(1):125. \u003c/li\u003e\n\u003cli\u003eEdman NI, Phal A, Redler RL, Schlichthaerle T, Srivatsan SR, Ehnes DD, et al. Modulation of FGF pathway signaling and vascular differentiation using designed oligomeric assemblies. Cell. 2024 Jul;187(14):3726-3740.e43. \u003c/li\u003e\n\u003cli\u003eJohnson BA, Blevins RA. NMR View: A computer program for the visualization and analysis of NMR data. J Biomol NMR. 1994 Sep;4(5):603\u0026ndash;14. \u003c/li\u003e\n\u003cli\u003eDelaglio F, Grzesiek S, Vuister GeertenW, Zhu G, Pfeifer J, Bax A. NMRPipe: A multidimensional spectral processing system based on UNIX pipes. J Biomol NMR [Internet]. 1995 Nov [cited 2025 May 8];6(3). Available from: http://link.springer.com/10.1007/BF00197809 \u003c/li\u003e\n\u003cli\u003eAlsoussi WB, Turner JS, Case JB, Zhao H, Schmitz AJ, Zhou JQ, et al. A Potently Neutralizing Antibody Protects Mice against SARS-CoV-2 Infection. J Immunol. 2020 Aug 15;205(4):915\u0026ndash;22. \u003c/li\u003e\n\u003cli\u003eVon Boehmer L, Liu C, Ackerman S, Gitlin AD, Wang Q, Gazumyan A, et al. Sequencing and cloning of antigen-specific antibodies from mouse memory B cells. Nat Protoc. 2016 Oct;11(10):1908\u0026ndash;23. \u003c/li\u003e\n\u003cli\u003eSokurenko EV, Courtney HS, Maslow J, Siitonen A, Hasty DL. Quantitative differences in adhesiveness of type 1 fimbriated Escherichia coli due to structural differences in fimH genes. J Bacteriol. 1995 Jul;177(13):3680\u0026ndash;6. \u003c/li\u003e\n\u003cli\u003eStahlhut SG, Tchesnokova V, Struve C, Weissman SJ, Chattopadhyay S, Yakovenko O, et al. Comparative Structure-Function Analysis of Mannose-Specific FimH Adhesins from Klebsiella pneumoniae and Escherichia coli. J Bacteriol. 2009 Nov;191(21):6592\u0026ndash;601. \u003c/li\u003e\n\u003cli\u003eTchesnokova VL, Rechkina E, Larson L, Ferrier K, Weaver JL, Schroeder DW, et al. Rapid and Extensive Expansion in the United States of a New Multidrug-resistant Escherichia coli Clonal Group, Sequence Type 1193. Clin Infect Dis. 2019 Jan 7;68(2):334\u0026ndash;7. \u003c/li\u003e\n\u003cli\u003eWeissman SJ, Johnson JR, Tchesnokova V, Billig M, Dykhuizen D, Riddell K, et al. High-Resolution Two-Locus Clonal Typing of Extraintestinal Pathogenic Escherichia coli. Appl Environ Microbiol. 2012 Mar;78(5):1353\u0026ndash;60. \u003c/li\u003e\n\u003cli\u003eAndersen MJ, Fong C, La Bella AA, Molina JJ, Molesan A, Champion MM, et al. Inhibiting host-protein deposition on urinary catheters reduces associated urinary tract infections. eLife. 2022 Mar 29;11:e75798. \u003c/li\u003e\n\u003cli\u003eSchneider CA, Rasband WS, Eliceiri KW. NIH Image to ImageJ: 25 years of image analysis. Nat Methods. 2012 Jul;9(7):671\u0026ndash;5. \u003c/li\u003e\n\u003cli\u003eGuiton PS, Hung CS, Hancock LE, Caparon MG, Hultgren SJ. Enterococcal Biofilm Formation and Virulence in an Optimized Murine Model of Foreign Body-Associated Urinary Tract Infections. Infect Immun. 2010 Oct;78(10):4166\u0026ndash;75. \u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"nature-portfolio","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"","title":"Nature Portfolio","twitterHandle":"","acdcEnabled":false,"dfaEnabled":false,"editorialSystem":"ejp","reportingPortfolio":"","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-7951484/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7951484/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"The rise of multidrug-resistant bacterial infections necessitates the discovery of novel antimicrobial strategies. Here, we show that protein design provides a generalizable means of generating new antimicrobials by neutralizing the function of bacterial adhesins, which are virulence factors critical in host-pathogen interactions. We de novo designed high-affinity miniprotein binders to FimH and Abp chaperone usher pili adhesins from uropathogenic Escherichia coli and Acinetobacter baumannii, respectively, which are implicated in mediating both uncomplicated and catheter-associated urinary tract infections (UTI) responsible for significant morbidity worldwide. The designed antagonists have high specificity and stability, disrupt bacterial recognition of host receptors, block biofilm formation, and are effective in treating and preventing murine models of uncomplicated and catheter-associated UTIs in vivo.","manuscriptTitle":"De Novo Design of Miniprotein Inhibitors of Bacterial Adhesins","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-11-04 08:30:39","doi":"10.21203/rs.3.rs-7951484/v1","editorialEvents":[],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"nature-communications","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"NCOMMS","sideBox":"Learn more about [Nature Communications](http://www.nature.com/ncomms/)","snPcode":"","submissionUrl":"https://mts-ncomms.nature.com/","title":"Nature Communications","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature Communications","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"aba384ec-2762-419b-b1bd-d7e77d482119","owner":[],"postedDate":"November 4th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":57273438,"name":"Biological sciences/Microbiology/Antimicrobials/Antimicrobial resistance"},{"id":57273439,"name":"Biological sciences/Biochemistry/Proteins/Lectins"},{"id":57273440,"name":"Biological sciences/Drug discovery/Biologics/Antibody fragment therapy"}],"tags":[],"updatedAt":"2026-04-28T10:56:13+00:00","versionOfRecord":[],"versionCreatedAt":"2025-11-04 08:30:39","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7951484","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7951484","identity":"rs-7951484","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}