A Cell‐Penetrating Peptide‐Derived Fluorescent Probe for Selective Isolation of Bacterial Spores Authors: Jaeyoung Yu, Tatsuru Okazaki, Yugo Kawabuchi, Chika Hayashida, Takahiro Muraoka, Tetsushi Mori Published: 2026 ABSTRACT ABSTRACTIn this study, we identify a peptide discovered through fluorescence‐based screening of a cell‐penetrating peptide (CPP) library that preferentially associates with bacterial spores and evaluate its utility as a fluorescence‐guided tool for spore isolation. Specifically, an eight‐ornithine peptide conjugated with 5(6)‐carboxyfluorescein (Orn8‐K‐FAM) generates a distinct and spatially confined fluorescence signal in spores. Across multiple spore‐forming bacterial species, more than 90% of spores were labelled by Orn8‐K‐FAM, whereas only a minor fraction of vegetative cells from non‐spore‐forming bacteria exhibited detectable fluorescence, indicating a high degree of preference at the cellular level under the conditions tested. To assess practical applicability, Orn8‐K‐FAM labeling was combined with flow‐cytometric sorting, enabling fluorescence‐guided isolation of spores from mixed microbial populations, including both a defined synthetic community and an environmentally derived bacterial fraction. Compared with conventional spore isolation approaches based on selective cultivation or heat‐mediated inactivation of vegetative cells, this workflow provides a cultivation‐independent means to separate spore‐associated cellular states from heterogeneous microbial samples. This work highlights the potential of CPP‐derived probes as analytical tools for spore‐focused studies in applied and environmental microbiology, while providing a foundation for further investigation into the scope and mechanistic basis of CPP‐spore interactions. Introduction Cell‐penetrating peptides (CPPs) are short (typically 8–30 amino acids), cationic or amphipathic peptides capable of crossing biological membranes and transporting a wide range of molecular cargoes, including nucleic acids, peptides, proteins and small molecules (Reveret et al. 2023). Since the discovery of the HIV‐1 TAT cell‐penetrating domain in the late 1980s, CPPs have become widely adopted as non‐viral molecular carriers because they enable access to the intracellular space by permeating biological membranes while avoiding the substantial physical or chemical perturbations associated with electroporation or chemical transformation (Okafor et al. 2025). Owing to these advantages, CPPs have been extensively used in mammalian systems as versatile tools for delivering macromolecules and enabling visualization and analysis of intracellular processes in living cells (Zhang et al. 2023). Their success in mammalian cell‐based research has positioned CPPs as a reliable delivery platform and has motivated attempts to expand their applicability to additional biological contexts. However, the success of CPPs in mammalian cells does not readily translate to bacteria. Whereas mammalian cells possess a single, fluid lipid bilayer that supports uptake via endocytic pathways as well as direct membrane permeation (Pancholi et al. 2025), bacterial envelopes lack vesicle‐mediated uptake mechanisms (Silhavy et al. 2010) and present major structural barriers, including a peptidoglycan layer and, in Gram‐negative species, an additional outer membrane (Pals et al. 2024). Consequently, CPP interactions with bacteria are primarily governed by physicochemical interactions with the cell envelope, including membrane penetration and membrane perturbation, rather than vesicular uptake, and overall CPP internalization is substantially less efficient than in mammalian systems (Mori et al. 2025). Peptides optimized for mammalian delivery, therefore, often show limited performance in bacterial contexts (Lee et al. 2021), underscoring the need to better understand how CPP properties intersect with the distinct architectures of bacterial cell envelopes. Despite these mechanistic constraints, accumulating evidence demonstrates that CPPs can function in bacterial systems. CPPs have been employed to deliver diverse molecular cargoes, among which CPP‐peptide nucleic acid (CPP‐PNA) conjugates represent a particularly successful example, enabling antisense‐mediated inhibition of essential genes and the development of species‐selective antimicrobial strategies (Vogel et al. 2025). CPPs have also been used to introduce plasmids for genetic recombination and to deliver fluorescent probes for the detection of specific microbial targets (Islam et al. 2019; Harris et al. 2022). Collectively, these advances indicate that CPP‐based strategies are gaining increasing importance in bacterial research and are emerging as versatile molecular platforms for probing and manipulating microbial systems. Our group has contributed to broadening CPP applications in bacterial systems by designing synthetic amino acid‐based CPPs with enhanced interactions in Gram‐negative bacteria (Inoue et al. 2021), optimizing CPP treatment conditions (Toyohara et al. 2019) and applying CPP‐PNA conjugates as molecular probes to investigate bacterial protein function (Yokoi et al. 2022; Kawabuchi et al. 2025). Building on these works, the present study was initiated as a screening effort to identify peptides with enhanced interactions towards Gram‐positive bacteria. During this screening, we unexpectedly observed that a CPP composed of eight‐ornithine residues (Orn8), when conjugated to 5(6)‐carboxyfluorescein (FAM), produced a distinct and spatially confined fluorescence pattern in a subset of Gram‐positive cells. This serendipitous observation prompted further investigation into the underlying biological structure associated with Orn8 interaction, leading to the identification of bacterial spores as the primary target. Motivated by this previously unreported CPP‐spore interaction, we evaluate Orn8‐K‐FAM as a recognition‐based fluorescent probe for spore detection and isolation. Specifically, we assess (i) the selectivity of Orn8‐K‐FAM across spore‐forming and non‐spore‐forming bacteria and (ii) its applicability for isolating spores from both defined mixed communities and environmentally derived microbiota. By reframing CPP‐bacteria interactions as recognition phenomena rather than exclusively as delivery events, this work highlights a potential expansion of CPP functional scope and suggests an approach for spore analysis in applied and environmental microbiology. Experimental Procedures Preparation of Laboratory‐Derived Spores and Environmental Microbial Fractions Laboratory‐Derived Spore Samples All bacterial strains used in this study—including spore‐forming and non‐spore‐forming species—were cultivated under species‐optimized media, temperature and oxygenation conditions. Detailed information on bacterial strains and growth parameters is provided in Data S1. To obtain purified spore preparations under controlled laboratory conditions, representative spore‐forming bacteria, Bacillus subtilis , Clostridium butyricum and Priestia megaterium were cultivated on organism‐appropriate agar plates (NBRC 702 agar for B. subtilis and P. megaterium, and NBRC 902 agar for C. butyricum ) and incubated for 1 week to induce sporulation (Abe et al. 2014). Colonies were harvested and suspended in 5 mL of 1 × phosphate‐buffered saline (PBS; Thermo Fisher Scientific, Japan), followed by heat treatment at 65°C for 30 min to inactivate vegetative cells while retaining heat‐resistant spores (Wen et al. 2022). Samples were washed three times with PBS by centrifugation (7500 × g, 3 min), and the final pellets were resuspended in PBS and adjusted to an optical density at 660 nm (OD660) of 1.0. These standardized spore suspensions were used for all subsequent CPP labeling and isolation experiments unless otherwise stated. Environmental Microbial Fractions To evaluate the applicability of CPP‐based spore isolation in a natural microbial background, bacterial fractions were prepared from the marine sponge Halichondria okadai as previously described (Yu et al. 2024). This host organism was selected as a representative environmental system known to harbour diverse spore‐forming bacteria (Yu et al. 2024). Briefly, sponge specimens collected at Mabori Beach (Kanagawa Prefecture, Japan) were rinsed twice with filtered seawater (FSW) to remove surface contaminants and cut into ~2 cm pieces. FSW was then added, and sponge tissues were mechanically disrupted by repeated vortexing and settling cycles to release associated bacteria. The resulting suspension was filtered through a 32 μm nylon mesh to remove large tissue debris and particulate material generated during sample disruption, followed by sequential low‐speed centrifugation (100 × g) to further remove residual particulates and high‐speed centrifugation (7500 × g) to collect bacterial cells. Final pellets were resuspended in FSW and adjusted to OD660 = 1.0. Design, Synthesis and Characterization of CPP‐FAM Probes The CPP library was designed to enable systematic evaluation of how peptide length and side‐chain structure influence CPP‐bacteria interactions. To this end, we designed and screened a panel of homopolymeric CPPs composed of four different cationic amino acids: the naturally occurring lysine (K) and the synthetic analogs ornithine (Orn), 2,4‐diaminobutyric acid (Dab) and 2,3‐diaminopropionic acid (Dap). Each CPP consisted of a single type of positively charged amino acid and was constructed in lengths ranging from six to twelve residues. These amino acids share a terminal amino group but differ in the number of methylene units in their aliphatic side chains—four in K, three in Orn, two in Dab and one in Dap—thereby providing a controlled framework to assess the combined effects of side‐chain length and peptide length on bacterial interactions. This library design was informed by our previous observations that CPP interactions with Gram‐negative bacteria are enhanced by increasing peptide length and incorporating amino acids with shorter side chains, which are thought to influence charge density and physicochemical interactions with the bacterial envelope (Inoue et al. 2021). Based on these findings, we examined whether similar sequence features would also modulate CPP interactions in Gram‐positive systems. All peptide‐based probes were synthesized by manual 9‐fluorenylmethoxycarbonyl (Fmoc) solid‐phase peptide synthesis (SPPS) as previously described (Inoue et al. 2021). Rink amide resin was used for peptide chain assembly from the C‐terminus to the N‐terminus, and completion of each coupling step was confirmed using a negative trinitrobenzene sulfonic acid (TNBS) test. For fluorescence‐based assays, CPPs were conjugated with 5(6)‐carboxyfluorescein (FAM) via the side chain of a C‐terminal lysine residue, allowing direct comparison with our previous study (Inoue et al. 2021). After synthesis, all probes were cleaved from the resin, purified by preparative reversed‐phase high‐performance liquid chromatography (RT‐HPLC), and verified by matrix‐assisted laser desorption/ionization time‐of‐flight mass spectrometry (MALDI‐TOF MS). Purified probes were lyophilized, dissolved in nuclease‐free water, quantified spectrophotometrically and stored at −20°C until use. Detailed synthesis, purification and analytical procedures are provided in the Data S1. Fluorescence‐Based CPP‐Bacteria Interaction Assay (CPP‐FAM Labeling) Fluorescence‐based CPP‐bacteria interaction assays were performed to assess the association of CPP‐FAM probes with bacterial cells or spores, based on a previously established protocol (Toyohara et al. 2019) with minor modifications. Bacterial cells or spore suspensions (100 μL, OD660 = 1.0) were collected by centrifugation (7500 × g, 5 min) and washed twice with 100 μL of PBS. Pellets were resuspended in 50 μL of CPP‐FAM working solution (2.0 μM in 0.5 × PBS) and incubated in the dark at 30°C for 1 h unless otherwise specified. The use of 0.5 × PBS was adopted from prior work (Toyohara et al. 2019), in which reduced buffer conditions, corresponding to lower osmotic strength, were suggested to enhance CPP‐bacteria interactions. Following incubation, samples were washed twice with PBS to remove excess free probes. To suppress extracellular fluorescence and minimize contributions from residual surface‐bound signals (Zhang et al. 2018), pellets were resuspended in 50 μL of 0.1% trypan blue (TB; Sigma‐Aldrich, Japan) prior to analysis. When visualization of cell membranes was required, FM4‐64 (1 μg/mL; Thermo Fisher Scientific) was used instead of TB (Cosi et al. 2025). FM4‐64‐treated samples were incubated at room temperature (RT; 25°C) for 15 min, washed twice with PBS and resuspended in 50 μL of PBS. Fluorescence signals were subsequently evaluated by fluorescence microscopy and/or flow cytometry, depending on the experimental objective. Fluorescence Microscopy Fluorescence microscopy was performed to visualize CPP‐FAM labeling patterns and spore‐associated fluorescence signals. Images were acquired using an Olympus BX53 upright fluorescence microscope equipped with a 100 × objective lens and a DP74 digital camera (Olympus, Japan). Green fluorescence from FAM was detected with a U‐FBWA mirror unit, and dual‐colour imaging of FAM and FM4‐64 was performed using a U‐FBW wideband mirror unit. All images were captured using identical exposure settings to enable direct qualitative comparison among samples. Flow Cytometry and Fluorescence‐Activated Cell Sorting (FACS) Quantitative analysis of CPP‐FAM labeling was conducted using a BD FACSAria II cell sorter (BD Biosciences, Japan) equipped with a 488 nm blue laser (nominal output ~13 mW) and a 530/30 nm bandpass filter for FITC detection. A 70 μm nozzle was used for all measurements. Prior to analysis, instrument performance and fluidics stability were routinely verified using CS&T research beads and Accudrop fluorescent beads (BD Biosciences). Forward scatter (FSC) and side scatter (SSC) parameters were calibrated using a flow cytometry size calibration kit (Thermo Fisher Scientific). Samples were diluted as necessary with 0.1% TB to maintain an event rate of 1000–2000 events/s; flow rate was adjusted accordingly. To minimize contributions from instrument‐ and buffer‐derived background noise, PBS and 0.1% TB controls were analysed independently to confirm background event rates below 10 events/s under all measurement conditions, and events corresponding to this background were excluded based on these control measurements. To reduce potential bias arising from arbitrary scatter gating—particularly given the small size and heterogeneity of bacterial cells and spores, as well as the compositional complexity of environmental samples—FSC and SSC parameters were not used to impose restrictive exclusion gates beyond this background‐based exclusion, and all samples were analysed under identical acquisition settings. Fluorescence thresholds defining CPP‐positive events were determined independently for each sample type based on the corresponding untreated control samples, and quantitative analysis was performed based on 20,000 acquired events per sample. For viability assessment, fluorescence‐positive events were sorted directly onto agar plates (240 cells per plate), incubated and colony‐forming units (CFUs) were quantified. CFU counts were normalized to untreated control samples that were processed in parallel under identical sorting conditions, with the same number of events deposited onto agar plates but without fluorescence‐based selection. For fluorescence‐based isolation experiments, approximately 1 × 106 CPP‐FAM‐positive events were sorted into 15 mL centrifuge tubes containing 5 mL of PBS, pelleted by centrifugation (7500 × g, 10 min), and resuspended in 100 μL of PBS for downstream analyses. Cultivation and Identification of Sorted Fractions To evaluate whether CPP‐FAM‐based fluorescence sorting enables the selective isolation of bacterial spores from environmental samples, CPP‐FAM fluorescence‐positive events (sorted fraction) and the untreated H. okadai‐derived bacterial fraction (unsorted) were subjected to cultivation‐based analysis, with consideration for potential downstream applications of the recovered cells. Samples were plated onto NBRC 902 agar after serial dilution (up to 104 fold when necessary) and incubated at 25°C for 2 weeks under anaerobic conditions to favour the germination and vegetative growth of spore‐forming anaerobic bacteria (Morvan et al. 2021; Song et al. 2022). After incubation, 30 colonies were randomly selected from each subsample and identified by 16S rRNA gene sequencing following a standard protocol (Yu et al. 2025). The nearly full‐length 16S rRNA gene was amplified using the universal primer pair 27F (5′–AKWGTTTGATCMTGGCTCAG) and 1492R (5′–GGHTACCTTGTTACGACTT). PCR reactions were prepared using AmpliTaq Gold 360 Master Mix (Thermo Fisher Scientific) and performed on C1000 Touch Thermal Cycler (Bio‐Rad, USA) under the manufacturer's recommended conditions: initial denaturation at 95°C for 10 min; 25 cycles of 95°C for 30 s, 54°C for 30 s and 72°C for 90 s; followed by a final elongation step at 72°C for 7 min. Amplicons were sequenced by FASMAC Co. Ltd. (Japan) using additional internal universal primers (591R: 5′–GWATTACCGCGGCKGCTG; 907F: 5′–AAACTYAAAKGAATTGACGG; 1100R: 5′–GGGTTGCGCTCGTTRYVGGAC) to obtain near full‐length 16S rRNA gene sequences. The resulting sequences were processed using Geneious Prime v2020.2.5 (Geneious, New Zealand), and taxonomic assignment was performed based on ≥ 98% sequence identity against reference 16S rRNA gene databases. Data Processing and Statistical Analysis All quantitative experiments were performed using three independent biological replicates, and results are presented as the mean ± standard deviation (SD). For fluorescence microscopy, representative images from at least three independent samples are shown. For qualitative analyses, such as colony identification following FACS, data from three biological replicates were pooled to illustrate the overall distribution of isolates, as genus‐level composition is not suitable for statistical comparison across individual samples. Image processing was limited to uniform adjustments in brightness and contrast and was performed using Adobe Photoshop CC 2018 v19.1.3 (Adobe Systems, Japan). Statistical analyses and graph generation were conducted using GraphPad Prism 8 (GraphPad Software Inc., USA). Results Screening Identifies Dab12‐K‐FAM as a CPP Showing Pronounced Fluorescence Association in Gram‐Positive Bacteria To compare CPP‐bacteria interaction profiles in Gram‐positive bacteria, a homopolymeric CPP‐FAM library differing in cationic amino acid type and peptide length was evaluated using Staphylococcus saprophyticus as a Gram‐positive model organism under fluorescence‐based assay conditions. Fluorescence microscopy revealed a clear difference among the tested variants: a pronounced fluorescence signal was observed only in cells treated with Dab12‐K‐FAM, whereas other CPP‐FAM variants showed weak or negligible signals (Figure 1A). Flow‐cytometric analysis corroborated the microscopy results. Dab12‐K‐FAM treatment resulted in a distinct fluorescent population relative to the negative control, with 99.6% ± 0.3% of events classified as fluorescent based on the defined threshold (Figure 1B and Figure S1). To assess whether Dab12‐K‐FAM labeling affected cell viability under the assay conditions, CFUs were quantified following treatment. CFU values remained comparable to those of the negative control, with no appreciable loss of viability observed within ±5% of the negative control (Figure 1C). The Dab12‐K‐FAM interaction profile was then examined across additional Gram‐positive bacterial species. In all species tested, Dab12‐K‐FAM treatment yielded a substantial fluorescent fraction, exceeding 67.5% in every case (Figure 1D and Figure S1). Together, these data indicate that Dab12‐K‐FAM consistently yields high fluorescence association across phylogenetically diverse Gram‐positive bacteria under the conditions tested. Fluorescence‐based screening of CPP‐bacteria interaction in Gram‐positive bacteria. (A) Fluorescence microscopy images of Staphylococcus saprophyticus treated with CPP‐FAM conjugates of varying peptide length and amino acid composition. Strong fluorescence was observed only in cells treated with Dab12‐K‐FAM. Scale bar, 5 μm. (B) Flow‐cytometric analysis of S. saprophyticus treated with Dab12‐K‐FAM, showing a clear shift towards a fluorescent population (green) relative to the negative control (NC; grey). The fluorescent fraction accounted for 99.6% of total events based on the defined fluorescence threshold. (C) Relative colony‐forming units (CFUs) of S. saprophyticus following Dab12‐K‐FAM treatment. CFU values are shown relative to the NC, with red dotted lines indicating ±5% of the NC. (D) Fraction of fluorescent cells across multiple Gram‐positive bacterial species after Dab12‐K‐FAM treatment. For all species examined, the fluorescent fraction exceeded 67.5%. Serendipitous Identification of Spore‐Associated Fluorescence by Orn8‐K‐FAM During fluorescence‐based screening of the CPP‐FAM library across multiple Gram‐positive bacteria, an unexpected fluorescence pattern was observed for Orn8‐K‐FAM in P. megaterium. Most other CPP variants evaluated, including Dab‐based peptides (Dab6–10) and Orn peptides of other lengths (Orn6 and Orn10), produced weak or non‐spatially confined fluorescence signals, whereas Orn8‐K‐FAM yielded a localized, circular fluorescence signal within a subset of cells (Figure 2A). In addition, this observation contrasted with Dab12‐K‐FAM, which exhibited strong fluorescence association at the population level in vegetative cells (Figure 1D). Because the screening was originally designed to assess CPP‐cell interaction profiles rather than to detect specific cellular structures, the localized Orn8‐K‐FAM fluorescence prompted further examination of the labelled region. Considering both the morphology of the fluorescence signal and the documented sporulation physiology of P. megaterium (Guha et al. 2025), we next examined whether this signal corresponded to bacterial spores. Although the culture conditions used in this study (30°C, NBRC 702 medium, aerobic shaking, overnight incubation) were not optimized for sporulation, sporulation has been reported to occur sporadically even under non‐inducing conditions in certain spore‐forming bacteria (Narula et al. 2015), supporting further examination of this possibility. Fluorescence‐based evaluation of Orn8‐K‐FAM association with bacterial spores. (A) Fluorescence microscopy images of Priestia megaterium vegetative cells incubated with different CPP variants. While several CPP variants produced weak or diffuse fluorescence signals, Orn8‐K‐FAM exhibited distinct, circular, and spatially confined fluorescence in a subset of cells (white arrows). FM4‐64 was used to visualize cell morphology. Scale bar, 5 μm. (B) Representative bright‐field and fluorescence images of Orn8‐K‐FAM‐treated P. megaterium. Fluorescence signals are selectively associated with spherical structures distinct from elongated vegetative cells, consistent with spore‐associated structures. Scale bar, 5 μm. (C) Flow‐cytometric analysis of Orn8‐K‐FAM fluorescence across multiple bacterial species. Strong fluorescence was detected in spores from spore‐forming bacteria, whereas non‐spore‐forming bacteria exhibited minimal fluorescence. Spores: A, P. megaterium; B, Bacillus subtilis ; C, Clostridium butyricum . Non‐spore‐forming Gram‐positive bacteria: D, Streptococcus thermophilus ; E, Enterococcus hirae ; F, S. saprophyticus . Non‐spore‐forming Gram‐negative bacteria: G, Escherichia coli ; H, Pantoea agglomerans . (D) Time‐course analysis of Orn8‐K‐FAM association with P. megaterium spores. Fluorescence was detectable within 3 min of incubation and remained detectable at all measured time points. Scale bars, 5 μm. Representative spore‐forming bacteria, including P. megaterium and B. subtilis , were first subjected to Wirtz spore staining to assess the presence of intracellular spore formation within vegetative cells. Under these conditions, green‐stained intracellular structures were observed within a subset of vegetative cells (Figure S2), consistent with sporulation occurring within vegetative cells. Based on these observations, we further evaluated spore‐associated fractions to examine the relationship between Orn8‐K‐FAM signals and spore‐like features. Spore formation and subsequent preparation procedures were first validated, where cells subjected to sporulation exhibited distinct morphological changes and produced structures positively stained by malachite green (Figure S2). When the prepared spores were subsequently evaluated using Orn8‐K‐FAM, bright fluorescence was observed in mature spores as well as in discrete, circular regions within vegetative cells (Figure 2B), supporting an association between Orn8‐K‐FAM signals and spore‐like features. The association between Orn8‐K‐FAM fluorescence and spores was further evaluated across multiple bacterial species. Spores prepared from spore‐forming bacteria (P. megaterium, B. subtilis and C. butyricum ) and vegetative cells from non‐spore‐forming bacteria (Gram‐positive: Streptococcus thermophilus, Enterococcus hirae, S. saprophyticus ; Gram‐negative: Escherichia coli, Pantoea agglomerans ) were analysed by flow cytometry following Orn8‐K‐FAM treatment. Vegetative cells of spore‐forming species were not included in this comparison to avoid potential misinterpretation arising from intracellular spore formation within vegetative cells, as observed in Wirtz‐stained preparations (Figure S2), which could confound attribution of the fluorescence signal to spores versus vegetative cellular structures. In all spore‐forming species examined, more than 90% of spore events exceeded the fluorescence threshold (Figure 2C and Figure S3). In contrast, approximately 10% or fewer events from non‐spore‐forming bacteria were classified as fluorescent. To characterize the temporal profile of this fluorescence association, P. megaterium spores were incubated with Orn8‐K‐FAM for varying durations. In addition to the spore‐selective fluorescence observed across multiple spore‐forming species, a detectable fluorescence signal was observed after 3 min of incubation and remained detectable at all subsequent time points examined, up to 60 min (Figure 2D). Together, these results indicate that Orn8‐K‐FAM exhibits preferential fluorescence association with bacterial spores across multiple species and that this association is established rapidly and maintained over the duration of the assay. Fluorescence‐Guided Isolation of Bacterial Spores From Mixed and Environmental Samples Following the identification of spore‐associated fluorescence by Orn8‐K‐FAM, we next examined whether this property could be applied to isolate spores from heterogeneous microbial samples. As an initial assessment, a defined model bacterial community was constructed that comprised spores from spore‐forming bacteria (P. megaterium, B. subtilis and C. butyricum ) and vegetative cells from non‐spore‐forming bacteria (Gram‐positive: S. thermophilus , E. hirae , S. saprophyticus ; Gram‐negative: E. coli , P. agglomerans ). After Orn8‐K‐FAM treatment, the fluorescent population was collected by flow‐cytometric sorting (Figure S4A) and analysed by species‐specific PCR (Table S1). DNA bands corresponding to P. megaterium, B. subtilis and C. butyricum were detected in the fluorescence‐positive fraction (Figure S4B), indicating that spores from the spore‐forming species were present in the sorted population. To evaluate applicability in a natural microbial background, Orn8‐K‐FAM‐based fluorescence sorting was next applied to a bacterial fraction derived from the marine sponge H. okadai. Prior to sorting, PCR analysis targeting the sporulation‐associated gene spo0A confirmed the presence of spore‐forming bacteria in the H. okadai‐derived fraction (Table S1 and Figure S4C) (Bueche et al. 2013). Flow‐cytometric analysis following Orn8‐K‐FAM treatment identified a fluorescence‐positive population corresponding to 22.3% of total events (Figure 3A and Figure S5A), which was subsequently collected by cell sorting. The sorted and unsorted fractions were examined by Wirtz spore staining, where the unsorted fraction showed numerous vegetative cells, whereas spore‐like structures were more clearly observed in the fluorescence‐sorted fraction (Figure S5B). To further characterize the sorted population, cells recovered from the fluorescence‐positive fraction and from the unsorted control were cultivated, and representative colonies were identified by 16S rRNA gene sequencing. Colonies obtained from the unsorted fraction included both spore‐forming and non‐spore‐forming genera (Figure 3B). In contrast, colonies recovered from the fluorescence‐sorted fraction were assigned to spore‐forming genera, including Bacillus, Clostridium and Paraclostridium (Figure 3B). Taken together, these results show that Orn8‐K‐FAM‐based fluorescent labeling, when combined with flow cytometry, supports the separation and recovery of spore‐associated cellular fractions from both defined mixed communities and environmentally derived microbial samples. Fluorescence‐guided isolation of bacterial spores from an environmental sample using Orn8‐K‐FAM. (A) Flow cytometry analysis of a bacterial fraction derived from the marine sponge Halichondria okadai following Orn8‐K‐FAM treatment. The fluorescence‐positive population (green) accounted for 22.3% of total events based on the defined fluorescence threshold. (B) Bacterial genera recovered by cultivation from control (unsorted) and fluorescence‐sorted fractions based on Orn8‐K‐FAM signals. Control samples yielded both spore‐forming and non‐spore‐forming genera, whereas sorted samples yielded spore‐forming genera preferentially under the culture conditions used. Discussion Interpreting CPP‐Cell Interactions in Fluorescence‐Based Screening Gram‐positive and Gram‐negative bacteria differ fundamentally in cell‐envelope architecture, a distinction that has important consequences for CPP‐cell interactions. In contrast to Gram‐negative bacteria, Gram‐positive species lack an outer membrane and instead possess a substantially thicker peptidoglycan layer (Yu et al. 2022), which often contains anionic polymers such as teichoic acids that contribute negative charge, the magnitude of which is partially neutralized through D‐alanylation (Zhou et al. 2022). Consequently, Gram‐positive cell surfaces may present both a robust physical barrier and a comparatively reduced anionic character. These features suggest that CPP interaction in Gram‐positive bacteria may depend more strongly on the cumulative positive charge and charge density of the peptide—often associated with peptide length—rather than on sequence features optimized for Gram‐negative envelopes. To systematically evaluate how side‐chain structure influences CPP‐bacteria interactions, we designed a homopolymeric CPP library composed of lysine and its structural analogs (Orn, Dab and Dap), which share a terminal primary amine but differ in methylene chain length. By restricting the library to primary amine‐containing residues, we established a controlled framework that isolates the effects of side‐chain length while minimizing additional variables such as amphipathicity or sequence patterning that would arise from mixed‐residue compositions. The peptide length range (6–12 residues) was selected to capture variations in charge density within the typical range of cationic CPPs (+4 to +12 net charge [Patel et al. 2019; Szabó et al. 2021]), thereby enabling comparison of length‐dependent interaction profiles. While this simplified library enabled interpretation of side‐chain length effects, extending this framework to include arginine‐containing peptides, which possess guanidinium groups known to mediate strong interactions with bacterial membranes (Frimodt‐Møller et al. 2021), as well as mixed‐residue sequences may be important for capturing a broader range of CPP interaction modalities. Consistent with this notion, fluorescence‐based screening in the present study revealed distinct CPP interaction profiles between Gram‐negative and Gram‐positive bacteria. In E. coli , strong fluorescence association was observed for Dab‐based CPPs even at shorter lengths (e.g., Dab6) (Inoue et al. 2021), whereas in the Gram‐positive bacterium S. saprophyticus , detectable fluorescence association required a longer Dab‐based sequence, with pronounced signals observed only for Dab12‐K‐FAM (Figure 1A). This length‐dependent asymmetry was not limited to Dab‐based peptides. CPPs composed of other cationic residues, such as Dap, showed a similar cell type dependence: while short Dap‐based peptides exhibited strong fluorescence association in E. coli (Inoue et al. 2021), no detectable fluorescence was observed in S. saprophyticus even at extended peptide lengths (Figure 1A). Together, these results suggest that CPP‐cell interaction properties may vary with cell envelope characteristics and bacterial physiological state, and that structural features effective in Gram‐negative bacteria may not translate directly to Gram‐positive envelope environments. The initial objective of this study was to identify CPPs with enhanced delivery potential towards Gram‐positive bacteria using fluorescence‐based screening as an experimental readout. Within this framework, Dab12 emerged as a candidate exhibiting strong fluorescence association across multiple Gram‐positive species (Figure 1D and Figure S1). However, subsequent CPP‐PNA delivery experiments demonstrated that fluorescence intensity alone was not predictive of functional antisense activity, as measured by growth inhibition resulting from target gene suppression. Although Dab12‐PNA conjugates targeting the essential gene acpP inhibited growth in several Gram‐positive species, the effective concentrations required for growth inhibition varied markedly among taxa (Figure S6). For example, P. megaterium exhibited pronounced growth inhibition at submicromolar concentrations, whereas L. plantarum showed only partial inhibition even at substantially higher concentrations (25 μM; Figure S6), despite displaying a high fluorescent fraction (98.8%; Figure 1D and Figure S1). Although CPP‐PNA activity can vary substantially depending on species‐specific physiology and cellular state (Vogel 2020), these observations indicate that fluorescence intensity alone does not necessarily correspond to efficient intracellular delivery or antisense‐mediated activity. Despite these limitations, a broadly applicable and standardized method for directly quantifying CPP‐mediated intracellular delivery in bacteria has not yet been established (Sarkar et al. 2025). Consequently, fluorescence‐based assays remain a practical and widely used approach for probing CPP‐bacteria interactions, particularly in exploratory screening contexts (Oikawa et al. 2018; Deprey et al. 2019). While such assays do not directly resolve peptide localization or functional uptake, they provide a sensitive means to detect reproducible fluorescence‐associated signal distributions across diverse bacterial systems (Deprey et al. 2019; Sarkar et al. 2025). Accordingly, fluorescence signals in the present study were interpreted conservatively as indicators of CPP‐cell interaction rather than as measures of delivery efficiency. Importantly, this interpretative framework enabled recognition of an unexpected interaction mode: compared to vegetative cells, Orn8‐K‐FAM showed a preferential fluorescence association with bacterial spores. This observation redirected the focus of the study from CPP‐mediated delivery towards recognition‐based analysis of bacterial spores. Reinterpretation of CPP Fluorescence Reveals Selective Recognition of Bacterial Spores by Orn8 Viewing fluorescence‐based CPP assays as indicators of CPP‐cell interaction rather than direct measures of intracellular delivery led to an unexpected finding that broadened the scope of this study. Compared to vegetative cells, Orn8‐K‐FAM generated strong and spatially confined fluorescence in bacterial spores, indicating preferential interaction with structural features characteristic of spores rather than with general cell envelopes. This selectivity is notable given the biological importance of spores, which represent one of the most resilient and persistent cellular forms in nature and enable bacteria to survive extreme environmental stresses, including heat, desiccation and chemical exposure (Setlow and Christie 2023). Such resilience underlies the long‐term persistence of spore‐forming pathogens, including Bacillus and Clostridium, in clinical, industrial and food‐associated environments, complicating disinfection and infection control (Maojin et al. 2025). Conversely, spores produced by nonpathogenic bacteria have been exploited in biotechnology as stable platforms for enzyme immobilization, probiotics and bioremediation (Hazan et al. 2025). Accordingly, robust and broadly applicable methods for detecting and isolating spores are of importance across both risk management and applied microbiology contexts. Conventional approaches for spore isolation typically rely either on selective cultivation following prior inactivation of vegetative cells—using antibiotics, defined carbon sources or tailored physicochemical conditions—or on heat treatment to eliminate vegetative‐cell viability while retaining heat‐resistant spores (Talahmeh et al. 2020; Wen et al. 2022). Although effective, these strategies have inherent limitations. Selective media require prior assumptions regarding target taxa and extensive optimization, increasing labour demands while potentially overlooking unexpected or low‐abundance spore formers (Talahmeh et al. 2020). Heat treatment, while operationally simple, can generate substantial cellular debris and extracellular polymers that co‐aggregate with spores, complicating downstream analyses such as microscopy, density‐based separation or colony isolation (Silvestri et al. 2015). In addition, flow cytometry‐based approaches using FSC/SSC parameters have been reported to enable discrimination of spores from vegetative cells in unstained samples (Cichocki et al. 2020). More recently, combinations of flow cytometry with nucleic acid stains (Chen et al. 2024) or metabolic dyes such as 5‐cyano‐2,3‐ditolyl‐2H‐tetrazolium chloride (Kawai and Nakano 2025) have been developed for rapid detection of spores in food and industrial contexts. However, to our knowledge, the application of these approaches to the physical isolation of spores from complex environmental samples remains limited, and their performance in highly heterogeneous microbial communities requires further evaluation. In this context, the combination of Orn8‐K‐FAM‐based fluorescent labeling with flow cytometry demonstrates the potential to isolate spores from complex environmental samples based on physicochemical properties, suggesting an alternative approach that does not rely on taxonomic or growth‐based assumptions. Nevertheless, careful interpretation of Orn8‐K‐FAM‐based isolation is warranted. Although Orn8‐K‐FAM exhibited a strong preference for spores, its selectivity was not absolute, as a small fluorescent fraction (< 10%) was detected among non‐spore‐forming bacteria (Figure 2C and Figure S3). Moreover, the fluorescent‐positive fraction observed in the H. okadai‐derived sample (22.3%; Figure 3A) exceeded previously reported estimates of spore‐forming bacteria in this microbiota (~6%) (Yu et al. 2024). While spore abundance may vary with environmental conditions and sampling period, this discrepancy suggests that limited labeling of non‐spore cells may occur. Consistent with this interpretation, Wirtz staining of the sorted population revealed a small number of red‐stained vegetative cells (Figure S5B). In addition, subsequent cultivation of the sorted fraction under anaerobic conditions favoured germination and growth of anaerobic or facultatively anaerobic spore‐forming bacteria, such as Clostridium and Paraclostridium, while suppressing obligate aerobes (Morvan et al. 2021; Song et al. 2022). Thus, the recovery of spore‐forming genera after cultivation reflects both the selectivity of Orn8‐K‐FAM labeling and the influence of post‐sorting culture conditions (Figure 3B). Taken together, while low‐level labeling of non‐spore cells may occur depending on sample composition and experimental context, the ability of Orn8‐K‐FAM to support recovery of spore fractions from complex environmental samples highlights its potential utility as a complementary tool for spore‐focused studies in applied and environmental microbiology. Mechanistic Considerations and Future Perspectives on Orn8‐Spore Interactions In the present study, we focused primarily on evaluating the applicability of Orn8‐K‐FAM as a fluorescence‐based tool for bacterial spore isolation, rather than on elucidating the detailed molecular mechanism underlying its spore selectivity. This emphasis reflects the broader context of CPP research, in which even the interaction and uptake mechanisms of CPPs in vegetative bacterial cells are not fully resolved (Moustafa et al. 2025). It is generally accepted that the initial association of cationic CPPs with bacterial cells is driven by electrostatic interactions between positively charged residues and negatively charged envelope components (Wojciechowska et al. 2020). Beyond this initial interaction, however, the downstream processes governing CPP translocation, access beyond the cell envelope, and functional intracellular delivery remain poorly understood and appear to be highly cell type dependent (Datta 2025; Moustafa et al. 2025). In some bacterial species, uptake of specific CPP‐cargo conjugates has been shown to involve dedicated transport systems, most notably the inner‐membrane protein SbmA, highlighting that active transport can contribute to CPP‐mediated delivery in certain contexts (Ghosal et al. 2013). However, the existence of CPP activity in bacteria lacking SbmA, as well as the identification of CPPs that do not rely on this transporter, indicates that CPP uptake cannot be explained by a single, universal mechanism (Puckett et al. 2012; Iubatti et al. 2022). As a result, the molecular basis of CPP‐mediated delivery in bacteria remains an open and actively debatable question. Bacterial spores represent an even more structurally complex and multilayered target than vegetative cells. From the exterior inward, spores are composed of an exosporium (when present), a proteinaceous spore coat, a peptidoglycan‐rich cortex and a core enclosed by the inner membrane and germ cell wall (Brunt et al. 2015). In this context, the observation that Orn8‐K‐FAM produced minimal fluorescence in both Gram‐positive and Gram‐negative vegetative cells, yet generated strong and selective signals in spores (Figure 2C and Figure S3), suggests preferential association with structural features unique to the spore rather than with general bacterial envelopes. At present, however, the molecular basis of this selectivity remains unresolved. It is unclear whether Orn8 interacts with specific components of the spore surface (e.g., coat‐associated structures) or whether its association reflects broader physicochemical properties characteristic of the spore. Further mechanistic insight could be gained by examining the temporal and spatial characteristics of Orn8‐K‐FAM labeling, including its association during defined stages of sporulation and its localization relative to individual spore layers. Such analyses would help clarify whether the observed fluorescence reflects limited access to outer spore layers or alternative interaction modes. Importantly, these questions arise directly from the reproducible and selective interaction documented here and will require dedicated mechanistic investigation beyond the scope of the present study. Accordingly, the current work focuses on defining the interaction profile and practical applicability of Orn8‐based fluorescence labeling, while establishing a foundation for future mechanistic studies. Finally, the scope of Orn8‐K‐FAM interactions beyond the bacterial systems examined here also warrants further exploration. While this study focused on spores produced by representative Gram‐positive bacteria, it will be informative to assess whether Orn8 exhibits comparable behavior towards other differentiated or dormant cell types, such as spores formed by actinobacteria (e.g., Streptomyces), myxobacteria or fungal species. Comparative studies across these systems may reveal whether Orn8 recognizes conserved features associated with spore formation or whether its selectivity is restricted to a narrower subset of bacterial spores. Such efforts would further refine both the mechanistic understanding and the potential applications of Orn8‐derived probes in microbial research. Conclusion This study shows that a peptide identified through CPP‐based fluorescence screening can exhibit preferential fluorescence association with bacterial spores. From a synthetic homopolymeric CPP library, we identified Orn8‐K‐FAM, which produced strong and spatially confined fluorescence in spores while showing minimal labeling in vegetative cells under the conditions tested. This spore‐associated fluorescence was observed across multiple spore‐forming species and, when combined with flow cytometry, supported fluorescence‐guided separation and recovery of spore‐associated fractions from both a defined mixed community and an environmentally derived microbial sample. Collectively, these findings support the potential of Orn8‐K‐FAM as a fluorescence‐based handle for spore isolation in heterogeneous microbial populations and provide a foundation for future work to define the scope and mechanistic basis of Orn8‐spore interactions. Author Contributions Jaeyoung Yu: conceptualization, data curation, formal analysis, investigation, methodology, validation, writing – original draft, writing – review and editing, visualization. Tatsuru Okazaki: data curation, investigation, methodology, writing – review and editing. Yugo Kawabuchi: investigation, writing – review and editing. Chika Hayashida: methodology, writing – review and editing. Takahiro Muraoka: project administration, resources, writing – review and editing. Tetsushi Mori: conceptualization, funding acquisition, project administration, resources, supervision, writing – review and editing. Funding This work was supported by Japan Society for the Promotion of Science, JP24H00519, P25100 and Japan Science and Technology Agency. Conflicts of Interest The authors declare no conflicts of interest. Supporting information TABLE S1: Primer sets used for PCR‐based verification of spore‐forming bacteria. Primer pairs targeted either species‐specific regions of the 16S rRNA gene or the sporulation‐associated gene spo0A. Species‐specific primers were used to confirm the presence of Priestia megaterium, Bacillus subtilis and Clostridium butyricum in Orn8‐K‐FAM fluorescence‐sorted fractions from the model bacterial community, whereas the spo0A primer set was used to assess the presence of spore‐forming bacteria in environmentally derived samples. FIGURE S1: Representative flow cytometry histograms showing fluorescence intensity distributions for untreated controls (NC) and Dab12‐K‐FAM‐treated samples (PC). The percentage of fluorescent cells was calculated based on the gating threshold indicated by the dashed line. These data correspond to the quantitative analyses presented in Figure 1B,D. FIGURE S2: Evaluation of spore preparation using Wirtz staining. Representative microscopy images of vegetative cells and prepared spore samples of P. megaterium and B. subtilis . Spore samples were obtained following sporulation and subsequent spore preparation procedures, including heat treatment to inactive vegetative cells. Spores are visualized as green‐stained structures, while vegetative cells appear red; red arrows indicate spores within vegetative cells. Scale bar, 5 μm. FIGURE S3: Representative flow cytometry histograms showing fluorescence intensity distributions for untreated controls (NC) and Orn8‐K‐FAM‐treated samples (PC). Spores: P. megaterium, Bacillus subtilis, C. butyricum . Non‐spore‐forming Gram‐positive bacteria: S. thermophilus , E. hirae , S. saprophyticus . Non‐spore‐forming Gram‐negative bacteria: E. coli , P. agglomerans . The percentage of fluorescent cells was calculated using a gating threshold defined from the NC population. These data correspond to the quantitative analyses presented in Figure 2C. FIGURE S4: PCR‐based confirmation of spore‐forming bacteria in Orn8‐K‐FAM fluorescence‐sorted fractions. (A) Representative flow cytometry histograms of the model bacterial community, showing fluorescence distributions of untreated control (NC) and Orn8‐K‐FAM‐treated samples (PC). (B) Species‐specific PCR analysis of Orn8‐K‐FAM fluorescence‐sorted fractions from the model bacterial community confirmed the presence of P. megaterium, B. subtilis and C. butyricum . (C) PCR analysis targeting the sporulation‐associated gene spo0A confirmed the presence of spore‐forming bacteria in bacterial fractions derived from Halichondria okadai. Lanes: M, 1 kb Plus DNA ladder (Thermo Fisher Scientific, Japan); 1–3, triplicate PCRs of fluorescence‐sorted fractions; P, positive control (PCR products from reference strains); NC, negative control (nuclease‐free water); ET, spo0A‐specific PCR of H. okadai‐derived bacterial fractions; EP, positive control for spo0A‐specific PCR (P. megaterium). FIGURE S5: (A) Flow cytometry analysis of a bacterial fraction derived from the marine sponge H. okadai without Orn8‐K‐FAM treatment (Negative control) (B) Wirtz spore staining of the H. okadai bacterial fraction before (Control) and after (Sorted) fluorescence‐activated cell sorting based on Orn8‐K‐FAM fluorescence. In sorted samples, spore structures (green; red arrows) were more readily identifiable, whereas the control fraction contained numerous vegetative cells. Scale bar, 5 μm. FIGURE S6: Growth inhibition of Gram‐positive bacteria by Dab12‐PNA conjugates targeting the essential gene acpP. Growth curves (OD660) illustrate concentration‐dependent suppression of bacterial growth, with the magnitude of inhibition varying among tested species. DATA S1: Supplemental materials and methods.