Traditional plant urologicals curb early uropathogenic E. coli infection and strengthen host innate defense

TL;DR

Herbal medicines reduce E. coli infection in the urinary tract by preventing bacterial adhesion and boosting the body's natural defenses.

Contribution

This study reveals that herbal extracts curb UPEC infection by inducing bacterial surface stress and enhancing host innate immunity.

Findings

Herbal extracts impair UPEC adhesion to bladder cells by reducing type 1 fimbriae expression.

HMP upregulate innate immunity genes in human bladder cells and Galleria mellonella larvae.

The extracts do not inhibit UPEC growth but reduce intracellular bacterial counts and improve survival after infection.

Abstract

Herbal medicinal products (HMP) are commonly used across Europe to treat urinary tract infections caused by uropathogenic Escherichia coli (UPEC), yet their mechanisms of action often remain unknown. We investigated the potential modes of action of several complex, government‑approved herbal preparations in clinical use. Four aqueous HMP extracts, fully profiled by LC‑MS, did not inhibit UPEC growth but altered early host‑UPEC interactions. A stress‑reporter assay revealed specific induction of surface stress in E. coli K-12 strain MG1655 via the BaeS‑BaeR two‑component system, while osmotic, pH, oxidative, SOS, and other stress pathways remained unchanged. This surface‑stress response impaired type 1 fimbriae expression and function, markedly decreasing adhesion of UPEC strain CFT073 to human T24 bladder cells and also lowering intracellular bacterial counts. The extracts did not negatively affect bacterial biofilm formation. To prove these in vitro effects in a more complex model, the HMP extracts were orally fed to Galleria mellonella larvae to investigate their impact on innate immunity and cellular host response. Oral feeding to uninfected larvae increased innate‑immunity gene (moricin, hemolin) expression in the midgut and in circulating hemocytes and markedly increased survival after UPEC infection. The HMP also caused an upregulation of the expression of innate‑immunity genes (IL‑6, IL‑8, and CXCL‑3) in uninfected human bladder epithelial cells, indicating an enhanced anti‑infective host response. These findings show that HMP can both lower uroepithelial susceptibility to UPEC, by impairing bacterial adhesion and invasion, and enhance host innate defenses. Consequently, such multi‑component herbal mixtures represent a promising alternative or complement to antibiotic therapy for UPEC infections through a dual‑target mechanism.

• Traditional herbal medicinal products (HMP) against urinary tract infections protect host cells from uropathogenic E. coli attack.

• Some HMP reduce UPEC adhesion to and invasion into bladder epithelial cells.

• Some HMP are immunostimulatory and reduce host cell susceptibility to infection.

The online version contains supplementary material available at 10.1007/s00253-026-13794-1.

Figures

Several traditional herbal urologicals induce envelope stress response in *E. coli* K-12 strain MG1655. The promoter activity of P*spy* in response to incubation with extract B was quantified as a biomarker for surface stress. The arrow indicates the time point when the extract was added to the bacterial culture. Shown are the results of representative experiments obtained for the different concentrations tested (A). Overview of the overall effect of the four extracts on the stress reporter systems (outcome of three biological replicates) (B)Fig. 2Regulation of envelope stress response induced

Impact of plant extracts on the expression of functional type 1 fimbriae. UPEC strain CFT073 was incubated with the individual plant extracts A to D to test for mannose-sensitive type 1 fimbriae-dependent agglutination of yeast cells. The extent of yeast agglutination by the treated bacteria or untreated control (UC) is indicated by + and − symbols below the photos. The agglutination was performed in the absence or presence of 3% D-Mannose in the yeast suspension. The addition of 3% D-Mannose to the yeast suspension served as a negative control for mannose-specific binding, thereby demonstrati

Treatment of bladder epithelial cells with plant extracts reduces bacterial adhesion and invasion. T24 human bladder epithelial cells were incubated with extracts A to D at 330 µg/ml concentration for 45 min followed by infection with UPEC strain CFT073 for 2.5 h, lysed and plated to determine bacterial adhesion (A). To determine the impact of the extracts on bacterial invasion, the T24 cells were incubated for 45 min (B) or 3 h (C) with the plant extracts (330 µg/ml) before they were treated for 1.5 h with gentamycin (200 µg/ml), lysed and plated on LB agar. The experiments were done with 3 b

Analysis of the immunostimulatory effects of plant extracts in the *G. mellonella* model of UPEC infection. The expression of the moricin, hemolin, apolipophorin III, and prophenoloxidase-encoding genes was analyzed in the midgut (A) and in circulating hemocytes (B) of larvae 24 h after oral administration of different plant extracts using quantitative real-time RT-PCR. Basal expression in the treated larvae was calculated as fold-change relative to untreated control larvae and normalized to the 18S rRNA housekeeping gene. Results represent the mean values of at least three independent determi

Full text

Introduction

Urinary tract infection (UTI) is one of the most common bacterial infections globally, affecting approximately 150 million people annually (Zeng et al. 2022). UTI refers to the infection in the urethra (urethritis), the bladder (cystitis), or the kidneys (pyelonephritis) (Mancuso et al. 2023). Mostly, UTI is caused by bacteria like Klebsiella pneumoniae,* Proteus mirabilis*,* Pseudomonas aeruginosa*, and Staphylococcus saprophyticus. However, uropathogenic Escherichia coli (UPEC) account for more than 80% of uncomplicated UTI cases (Frick-Cheng et al. 2020; Zhou et al. 2023). Antibiotic treatment for UTIs primarily functions by either killing (bactericidal) or inhibiting the growth of UPEC (bacteriostatic). However, the frequent use of antibiotics creates selective pressure, promoting the emergence of resistant strains, which makes the therapy more complicated (Kot 2019). Traditional herbal medicines are gaining increasing interest as they often hit multiple bacterial targets, unlike single-molecule antibiotics, making it harder for UPEC to evolve resistance (Marouf et al. 2022).

Plant extracts have a complex composition and can be regarded as multi-compound drugs, which can affect different targets. In the case of herbal extracts used against bacterial infections, they can reduce bacterial growth, motility, and biofilm formation; induce bacterial stress responses; interfere with the early pathogen-host interaction; or affect the overall fitness of the host (García-Heredia et al. 2016; Mulat et al. 2025; Shamim et al. 2023). For example, sub-inhibitory concentrations of allicin from Allium sativum (garlic) essential oil decrease UPEC biofilm formation, dispersal, adhesion ability, and swimming motility by affecting bacterial gene expression (Yang et al. 2016). Polyphenols described for Vaccinium macrocarpon (cranberry), Orthosiphon stamineus (java tea), or Equisetum arvense (horsetail) inhibit UPEC adhesion to uroepithelial cells by targeting bacterial adhesins (Deipenbrock et al. 2023; Deipenbrock and Hensel 2019; Mo et al. 2022; Sarshar et al. 2017). Similarly, extracts of Arctostaphylos uva ursi (bearberry), Agropyron repens (couch grass), and Apium graveolens (celery seeds) exert direct antiproliferative as well as antiadhesive effects against UPEC (Beydokthi et al. 2017; Rafsanjany et al. 2015). Recent studies indicated that the natural products resveratrol and caffeic acid phenethyl ester inhibit UPEC invasion into the host cells by suppressing the induction of focal adhesion kinases responsible for regulating actin-mediated cytoskeletal rearrangement in the host cell (Lewis et al. 2024).

Interestingly, traditional UTI treatment often relies on complex mixtures of various herbal materials, typically prepared as aqueous decoctions or teas. Notably, in 1976, the German government and its drug regulatory agency acknowledged this practice by introducing official “exempt standard formulations,” enabling the registration of traditionally used medicines in the formal pharmaceutical market for registered drug products (Bundesinstitut für Arzneimittel und Medizinprodukte 2024). The official German governmental list of exempt standard formulations includes seven traditional herbal medicinal product (HMP) mixtures for treating UTI and preparing bladder and kidney teas. Clinically, the respective efficacy is documented by the official authorization of these HMP as registered drug products on the European level by European Medical Agency EMA (European Medicines Agency 2017) or on the German level by the relevant drug registration authority BfArM (Bundesinstitut für Arzneimittel und Medizinprodukte 2024). On the other side, it remains unclear what the underlying mechanisms are for the observed positive effects against UTI. In most cases, direct antibacterial effects can be excluded for most of the respective plants from the complex mixtures. If bacterial virulence factors are affected or if the host cell innate defense is activated has not been investigated in large detail. Here, we systematically investigated the effects of four selected HMP formulations, containing Orthosiphon stamineus, Betulae sp., Ononis spinosa, Solidago sp., Agropyron repentis, and Equisetum arvense within complex mixtures, on UPEC adhesion, invasion into human bladder epithelial cells, alongside their impact on bacterial growth, stress response, as well as on innate host responses. Our results suggest that the type of herbal material in plant extracts determines their anti-infective properties, primarily by interfering with the early phases of UPEC-host interaction and by modulating the host cell response into a defense mode against the UPEC infection, even in the absence of strong direct antimicrobial activity.

Materials and methods

Plant material

The plant material used for the preparation of the extracts was obtained from traders from the pharmaceutical market as follows: leaves from Orthosiphon stamineus (batch 18235505 & 22000566003) and Betulae sp. (batch 18053215 & 22000996004), as well as roots from Ononis spinosa (batch 20003845007) were obtained from Caesar & Loretz (Hilden, Germany). Aerial material from Solidago gigantea (batch 41402) Solidago virgaureae (batch 49949), and rhizome of Agropyron repentis (batch 58002) were acquired from Alfred Galke (Bad Grund, Germany). Aerial material of Equisetum arvense (batch 3831B 200826-01) was obtained from Heinrich Klenk (Schwebheim, Germany). The herbal material was identified by AH and SB and was in accordance with the quality standards of the European Pharmacopeia. Voucher specimens for all used plant material are deposited at the Institute for Pharmaceutical Biology and Phytochemistry of the University of Münster (voucher numbers: IPBP-902, IPBP-903, IPBP-890, IPBP-891, IPBP-901, IPBP-906, IPBP-908, IPBP-889, and IPBP-895).

Preparation of extracts

The standard formulations A, B, and C were prepared as recently described (Boertz et al. 2025a). Briefly, 1.25 g of each of the relevant individual dried and cut plant materials were mixed with 100 ml of 90 °C warm water in a 250-ml round-bottom flask. The extraction was performed for 30 min under heating at 90 °C and gentle stirring. A reflux condenser was used to reduce evaporation of volatile compounds. After extraction, the suspension was filtered through a metal sieve and insoluble material was removed by centrifugation at 3.000×g for 10 min. The supernatant was collected and lyophilized. For extract D, 2.41 g of the aerial material from Solidago virgaurea, 1.32 g of leaves from Orthosiphon stamineus, and 1.27 g of roots from Ononis spinosa were combined and then extracted as mentioned above.

Analytical characterization of extracts

The chromatographic separation (LC-ESI-qTOF-MS) of the extracts was performed on a Dionex Ultimate 3000 RS Liquid Chromatography System (Thermo Fisher Scientific, Schwerte, Germany) on a Waters HSST3 column (2.1 × 100 mm, 1.7 µm) (Waters, Eschborn, Germany). A binary gradient of water + 0.1% formic acid (A) and acetonitrile + 0.1% formic acid (B) was used with the following profile: 0.0 min: 5% B, 0.4 min: 10% B, 6.1 min: 50% B, 8.1 min: 100% B, 15.0 min: 100% B, 15.1 min: 5% B, 20 min: 5% B. The injection volume was 2 µl. Detection of eluted compounds was achieved using a Dionex Ultimate DAD-3000 RS (Thermo Fisher Scientific, Schwerte, Germany) over a wavelength range of λ = 200 to 400 nm and a Daltonics micrOTOF-QII time-of-flight mass spectrometer (Bruker Daltonics, Bremen, Germany) equipped with an Apollo electrospray ionization source in positive mode at 3 Hz over a mass range of m/z = 50–1450. The following instrument setting was applied: nebulizer gas nitrogen at 4 bar; dry gas nitrogen, 9 l/min, 200 °C; capillary voltage 4500 V; end plate offset 500 V; transfer time 100 µs, prepulse storage 6 µs; collision energy 40 eV; and collision RF 130 Vpp. Internal calibration of the dataset (HPC method) was performed using a solution of sodium formate 50% in isopropanol that was infused during LC re-equilibration using a divert valve equipped with a 20-µl sample loop. Data were analyzed using Bruker DataAnalysis 4.1 SP1 (Bruker Daltonics).

Extracts were dissolved in water at a concentration of 5 mg/ml to allow the assignment of compounds to the used plants. The combined and separate extracts of the different herbal materials were dissolved in water at a concentration of 20 mg/ml.

Bacterial strains and cell cultures

For the bacterial stress reporter system, the E. coli K-12 strain MG1655 (Blattner et al. 1997) was used as the strain background. Individual reporter gene fusions consisting of the promoterless yellow fluorescent protein-encoding gene yfp and one of the promoters of the stress marker genes gadA, spy, otsAB, recA, and dps were inserted into the chromosomal attachment site of bacteriophage λ in E. coli strain MG1655 as described before (Chitto et al. 2025). The pyelonephritis model isolate E. coli CFT073 (Welch et al. 2002) or the gfp-labeled derivative CFT073-gfp was used for the experiments. The insertion of the gfpmut3.1-bla module of pGFPmut3.1 into the chromosomal attachment site of the bacteriophage lambda in E. coli CFT073 was carried out by amplifying the module from pGFPmut3.1 by PCR with primer pair (5′AAGCCAATGCCAGCGCCAGACGGGAAACTGAAAATGTGTTCACAGGTTGCTCCGGGCTATTGGCCGATTCATTAATGCAGC3′/5′AATGCCATCTGGTATCACTTAAAGGTATTAAAAACAACTTTTTGTCTTTTTACCTCTCACGTTAAGGGATTTTGGTCA-3′). This 2549-PCR product was afterwards used to insert gfpmut3.1-bla in the attB site of E. coli CFT073 using pKOBEG as the recombineering vector (Datsenko and Wanner 2000; Derbise et al. 2003). Correctness of the insertion was verified by PCR with primers UD1543 (5'-AGAATCGGGCGAGAAGAGG-3') and UD1544 (5'-TCGATGAAGGTGCCAGCG-3'). The UPEC strains were routinely grown statically from frozen glycerol stocks in lysogeny broth (LB) (10 g/l tryptone, 5 g/l yeast extract, and 5 g/l NaCl) for 24 h at 37 °C and on LB agar plates.

The human cell lines used were cultivated at a temperature of 37 °C in an incubator with a water-saturated and 5% CO_2_ atmosphere. The T24 cell line was derived from a urinary bladder transitional carcinoma patient (O’Toole et al. 1972). These cells were cultured in McCoy’s 5 A medium (Sigma-Aldrich, Taufkirchen, Germany) supplemented with 10% fetal bovine serum (Bio&Sell GmbH, Germany), 1 × non-essential amino acids (Sigma-Aldrich, Taufkirchen, Germany), 1 mM sodium pyruvate (Sigma-Aldrich, Taufkirchen, Germany), and 2 mM alanine-glutamine (Sigma-Aldrich, Taufkirchen, Germany). The human urinary bladder transitional carcinoma cell line RT-112 (Benham et al. 1977) was cultured in Waymouth’s medium (Thermo Fisher Scientific, Schwerte, Germany) supplemented with 10% fetal bovine serum (Bio&Sell GmbH, Germany), 1 × non-essential amino acids (Sigma-Aldrich, Taufkirchen, Germany), and 1 mM sodium pyruvate (Sigma-Aldrich, Taufkirchen, Germany). The 5637 cell line was isolated from a patient with urinary bladder transitional carcinoma (Fogh et al. 1977)and was cultured in RPMI 1640 medium with Glutamine (Sigma-Aldrich, Taufkirchen, Germany) supplemented with 10% fetal bovine serum (Bio&Sell GmbH, Germany), 1 × non-essential amino acids (Sigma-Aldrich, Taufkirchen, Germany), 1 mM sodium pyruvate (Sigma- Aldrich, Taufkirchen, Germany), and 2 mM alanine-glutamine (Sigma-Aldrich, Taufkirchen, Germany).

Stress reporter assay

To study the induction of bacterial stress response by the plant extracts, we quantified the intensity of yellow fluorescence over time in the different E. coli MG1655-based reporter strains as a measure of the promoter activity of the stress biomarkers in response to the various plant extracts. According to the recently published protocol (Chitto et al. 2025), overnight cultures of the reporter strains were grown in M9 medium (Sambrook et al. 1989) supplemented with 0.4% glucose and 0.4% casamino acids. The next day, the overnight cultures were diluted 1:200 in M9 medium supplemented with 0.4% glucose and 0.4% casamino acids, and 150 µl was pipetted per well in a black flat-bottom microtiter plate (Thermo Fisher Scientific, Schwerte, Germany). The microtiter plate was incubated at 37 °C with agitation in an Infinite F200 microplate reader (Tecan, Männedorf, Switzerland) until the bacterial cultures reached the logarithmic growth phase. Subsequently, the individual plant extracts were added in different concentrations (two-fold serial dilutions ranging from 1 to 0.03125 mg/ml final concentration; water was used as a solvent control) to the wells in a total volume of 200 µl, before the plate was further incubated in the microplate reader to read the absorbance and fluorescence values in ten-minute intervals. The optical density was measured at λ = 595 nm (± 10 nm). The fluorescence signal was measured using an excitation wavelength of λ = 485 nm (± 20 nm) and an emission wavelength of λ = 535 nm (± 25 nm).

Biofilm formation assay

The assessment of biofilm formation of UPEC strain CFT073 was based on the protocol described by Reisner and colleagues (Reisner et al. 2006). Briefly, overnight cultures of E. coli strain CFT073-gfp were prepared in LB either in the presence or absence of plant extracts (final concentration: 330 µg/ml) and kept in static conditions at 37 °C. These overnight cultures were diluted 1:200 in fresh LB either with or without the various plant extracts and at least six wells of a 96-well plate (clear U-bottom, PVC; Corning, USA) were filled with 180 μl per strain. The plate was then incubated for 48 h at 37 °C under static conditions. Subsequently, the plate was washed three times with 150 µl of PBS and dried for 30 min at 80 °C. 200 µl of 0.1% crystal violet solution was added to each well and stained for 20–30 min at room temperature. In the following, the wells were washed three times with PBS and dried for 10 min at 80 °C. After that, 200 µl destaining solution (80% ethanol, 20% acetone) was added to each well and incubated for 15 min. Finally, 100 µl solution was transferred from each well to a 96-well microtiter plate and optical density was measured at 570 nm in an Infinite F200 microplate reader (Tecan, Männedorf, Switzerland).

Yeast agglutination assay

UPEC strain CFT073-gfp was statically cultured at 37 °C overnight in LB with and without plant extracts. The overnight cultures were centrifuged at 6000 rpm for 2 min to pellet the bacteria. The bacterial pellets were then adjusted with PBS to an optical density at 600 nm of 5. Then, in a 24-well plate, 200 µl of bacterial suspension was mixed with 200 µl of a Saccharomyces cerevisiae suspension (10 mg/ml) in PBS with and without 3% D-mannose by swirling for approximately 5 min. The addition of 3% D-Mannose to the yeast suspension served as a negative control for mannose-specific binding, thereby demonstrating that any observed yeast agglutination was not due to non-specific interactions but mannose-specific adhesion mediated by the tested bacteria. The plate was gently mixed, and then the extent of yeast agglutination was observed in comparison to that caused by untreated bacteria. A high degree of agglutination indicates expression of functional type 1 fimbriae.

Bacterial adhesion assay

In each well of a 24-well plate (Sarstedt, Nümbrecht, Germany), 1 ml of a T24 bladder epithelial cell suspension with a density of 1 × 10^5^ cells/ml in McCoy’s 5 A medium was incubated at 37 °C and 5% CO_2_. After 48 h, the cells became confluent, and the old medium was exchanged for fresh McCoy’s 5 A medium. Afterwards, the individual plant extracts were added at different concentrations and the cells were incubated for 45 min. UPEC strain CFT073-gfp was previously cultured overnight in LB under static conditions at 37 °C. The optical density of the bacterial overnight culture was adjusted to OD at 600 nm of 1, followed by centrifugation at 5000 rpm for 10 min. The pellet was resuspended in 0.9% NaCl solution and washed twice. Then, the T24 cells were infected with 10^6^ bacteria (MOI of 10) and incubated for 2.5 h at 37 °C. After this, the cells were washed 5 times with HBSS (Hanks' balanced salt solution) buffer and subsequently lysed by the addition of 1 ml 0.025% Triton X-100 solution and 10 min shaking at room temperature on an orbital shaker. The lysed cells were serially diluted (1:100, 1:1000, 1:10,000), then plated on LB agar and incubated overnight at 37 °C to determine colony-forming units (CFU/ml).

Bacterial invasion assay

The invasion assay was performed as recently described in (Boertz et al. 2025). Briefly, UPEC strain CFT073-gfp was grown statically overnight in LB at 37 °C. 1 ml of T24 bladder epithelial cells (1 × 10^5^ cells/ml) was cultivated in McCoy’s 5 A medium at 37 °C and 5% CO_2_. After 48 h, when the cells became confluent, the old culture medium was replaced by fresh McCoy’s 5 A medium, plant extracts were added at different concentrations, and the cells were further incubated for 45 min. The optical density of the overnight culture of UPEC strain CFT073-gfp was adjusted to an OD at 600 nm of 1 and centrifuged at 5000 rpm for 10 min to pellet the bacteria. The bacterial pellet was then washed twice in 0.9% NaCl solution. After the 45-min preincubation time with the individual plant extracts, the T24 cells were infected with 10^6^ bacteria (MOI of 10) and further incubated for 2.5 h at 37 °C and 5% CO_2_. Then, the cells were washed at least three times with HBSS buffer followed by the addition of gentamycin (200 µg/ml final concentration) and incubation for 1.5 h to kill all the extracellular bacteria. Subsequently, the T24 cells were washed twice with HBSS and lysed in 1 ml 0.025% Triton X-100 solution by 10 min shaking at room temperature on an orbital shaker. The lysed cells were plated undiluted and diluted 1:10 on LB agar and incubated overnight at 37 °C to determine colony-forming units (CFU/ml).

RNA isolation and real-time PCR

T24 cells were incubated for 3.5 h with the plant extracts (long exposure time). Afterwards, RNA was isolated from the T24 cells with 1 ml Trizol (Thermo Fisher Scientific, Schwerte, Germany) according to the manufacturer’s protocol. In some cases, RNA was also isolated after a 45-min incubation with the extracts (short exposure time), followed by infection with E. coli strain CFT073-gfp for 2.5 h.

For complementary DNA (cDNA) synthesis, the RevertAid First-strand cDNA Synthesis kit was used (Thermo Fisher Scientific, Schwerte, Germany), and the cDNA was then quantified spectrophotometrically. 50 ng of cDNA per reaction was used for quantitative real-time RT-PCR using CFX 96 real-time PCR (Bio-Rad Laboratories, Munich, Germany) along with SsoAdvanced Universal IT SYBRGreen Smx (Bio-Rad Laboratories, Munich, Germany). The amplification parameters comprised an initial activation step (95–98 °C for 3–10 min), followed by 39–40 cycles of denaturation (95–98 °C for 15 s), annealing (56 °C for 15–30 s), and extension (65–72 °C for 15–30 s). The primers used for the target genes are listed in Supplementary Table 1. The relative expression levels of the target genes were determined using the ΔΔCT method, with 18S ribosomal RNA used as the reference gene for data normalization.

Oral application of plant extracts in G. mellonella, homogenization and RNA isolation, cDNA synthesis, and RT-PCR

The greater wax moth G. mellonella larvae were purchased from Fauna Topics Zoobedarf Zucht und Handels GmbH, Marbach am Neckar, Germany, and maintained as described previously (Mukherjee et al., 2020). The larvae were orally administered different plant extracts (330 µg/larva) by force-feeding, while those receiving water served as controls. RNA was isolated from midgut and hemocyte 24 h post-feeding from both treated and control larvae using TRIzol (Thermo Fisher Scientific, Schwerte, Germany) according to the manufacturer’s protocol. Larvae were surface-sterilized with 70% ethanol prior to hemolymph and midgut extraction. Hemolymph was obtained by piercing a proleg with a sterile needle and collected in Eppendorf tubes containing TRIzol. Midguts were dissected from surface-sterilized larvae, then homogenized in TRIzol using a sterile pestle. For RNA isolation, three biological replicates were prepared, each consisting of pooled midguts or hemocytes from at least five larvae. RT-PCR analysis of target genes, including the amplification parameters, target gene primers, and gene expression analysis was conducted as described above.

Impact of the plant extracts on the survival of G. mellonella larvae after UPEC infection

The UPEC strain CFT073, in its logarithmic phase, was washed and serially diluted in 0.9% NaCl. A 10-μl aliquot of cultures (1 × 10^5^ CFU/ml) was injected into larvae (1 × 10^3^ CFU/larva) 24 h after the oral administration of the plant extracts. Injections were performed through the left proleg using 1-ml disposable syringes with 0.4 by 20-mm needles mounted on a microapplicator. All experiments were conducted using larvae in their sixth developmental stage (weight ~ 250–300 mg). Larvae were considered dead if they exhibited no movement in response to touch following incubation at 37 °C post-injection.

Statistics

All experiments were performed a minimum of three times. All figures were created, and statistical analyses were conducted using GraphPad Prism 8.0.1 for Windows (GraphPad Software, San Diego, CA, USA). Figures were generated as mean ± standard error of the mean (SEM), and the unpaired two-tailed t-test was used for pairwise comparison of the treatment groups with the untreated control when the dataset showed normal distribution; otherwise, the groups were analyzed using the Mann–Whitney test. For Figs. 5A, B and Fig.S9 A-D comparisons with untreated uninfected controls and Fig. 5C, D comparisons with treated uninfected controls (relative expression normalized to 1) were evaluated using the unpaired two-sample t-test and Mann–Whitney test, as appropriate. For Fig. 6C, D, survival was analyzed using Kaplan–Meier curves, with group differences assessed by the log-rank test.

Results

Plant extracts induce bacterial envelope stress response and modulate type 1 fimbriae function without exhibiting detectable anti-UPEC activity

Traditionally used complex mixtures for UTI from different herbal materials, partly listed by the official German governmental list of exempt standard formulations (Bundesinstitut für Arzneimittel und Medizinprodukte (BfArM)), were selected for detailed investigations on potential anti-infective properties. Table 1 displays the qualitative and quantitative composition of the mixtures A to D. For the preparation of the test extracts, infusion with water for 30 min at 90 °C of the combined herbal material mixture was performed. After centrifugation, the respective extracts were lyophilized, and the dry extracts were subjected to analytical and functional investigations. The respective yields of the dry extracts A to D obtained are displayed in Table 1. All extracts were characterized by HPLC-(+)-ESI-qTOF-MS using the respective retention times, UV spectra, m/z values and MS2 fragmentation pattern in combination with data base search for peak alignment to known natural compounds (Table 2). To keep the complex data set from being too complex for each extract, only the main 20 peaks were analyzed in detail and used for the characterization of the respective extract. Table 1. Composition of standard formulation preparations A to D, included into the present study (based on traditional medicine)DesignationComponentsPercentage by mass [%]Yield lyophilized dry extract [%]^#^Plant speciesCommon name/plant nameABetula pendula, B. pubescensBirch leaves2513.7Agropyron repensCouch grass rhizome25Solidago gigantea, S. canadensis*(Giant) goldenrod aerial material25Ononis spinosaSpiny restharrow roots25BBetula pendula, B. pubescensBirch leaves2515.1Solidago gigantea, S. canadensis*(Giant) goldenrod aerial material25Ononis spinosaSpiny restharrow roots25Equisetum arvenseHorsetail aerial material25CBetula pendula, B. pubescensBirch leaves2513.6Solidago gigantea, S. canadensis*(Giant) goldenrod aerial material25Ononis spinosaSpiny restharrow roots25Orthosiphon aristatusJava tea leaves25DSolidago virgaureaGoldenrod aerial material48.214.0Orthosiphon aristatusJava tea leaves26.4Ononis spinosaSpiny restharrow roots25.4^*^Qualitative composition according the governmental standard composition of the German government^^Composition according herbal drug preparation from the German drug market^#^Related to dried herbal materialTable 2Characterized and tentatively identified compounds from extracts A, B, C, and DtR/minCmp.^†^MWIonIon formulaerr/mDamΣfragmentsUV max. λ nmIdentity/indistinguishable from^‡^Rel. peak area [%]Extract AExtract BExtract CExtractDOrigin #Ref1.401165.0790[M + H]^+^[C_9_H_12_NO_2_]^+^−1.65.6120200,256Phenylalanin1.51.6B, S, Or1.402288.0892[M + H]^+^[C_12_H_17_O_8_]^+^−4.418.9127200,256?On1.423327.1320[M + H]^+^[C_15_H_22_NO_7_]^+^2.816.5120,166,310200,260?2.0B2.844706.1780[M + H]^+^???515200,255,340?1.3S3.165328.1164[M + Na] + [C_15_H_20_NaO_8_]^+^−0.62.2167, 328, 329200, 216, 2841-O-(4-Hydroxyhydrocinnamoyl)-hexosid4.64.44.0B, On, S(Shu et al. 2021)3.406166.0624[M + H]^+^[C_9_H_11_O_3_]^+^0.81.4121, 149200, 220, 280?1.1B3.627354.0950[M + H]^+^[C_16_H_19_O_9_]^+^1.63163200, 216, 324(Neo)chlorogenic acid3.23.0B, S(Goswami et al. 1984; Ossipov et al. 1996)4.208432.1250[M + H]^+^[C_18_H_25_O_12_]^+^03.8127200, 260, 325Licoagrochalcone B2.82.62.22.6On(Li et al. 2000)4.289284.1780[M + H]^+^[C_19_H_25_O_2_]^+^7.734.1163, 241200, 260, 312?1.4On4.3910338.1015[M + H]^+^[C_16_H_19_O_8_]^+^−2.43.8147200, 312Cumaroyl quinic acid1.11.21.1B, E, S(Jin et al. 2007)4.6111386.1963[M + H]^+^[C_19_H_31_O_8_]^+^1.912207, 225200, 270, 320Roseosid1.31.1B, Or(Fuchino et al. 1998)4.7212480.0900[M + H]^+^[C_21_H_21_O_13_]^+^−123.9177, 209, 319, 369, 481, 495204, 265, 320Myricetinglycoside1.21.3B(Keinänen and Julkunen-Tiitto 1998)4.7313368.1101[M + H]^+^[C_17_H_21_O_9_]^+^??177200, 270, 320Feruoyl quinic acidE, S5.0214610.1530[M + H]^+^[C_27_H_31_O_16_]^+^−114.5200, 256, 332Quercetin-(6″-desoxyglycosyl)glycoside1.21.6S(Batyuk and Kol’tsova 1969)5.0915464.0950[M + H]^+^[C_21_H_21_O_12_]^+^−0.313.2303, 465. 479200, 256, 340Quercetin glycoside1.51.72.1B, E, S[(Ossipov et al. 1996; Pawłowska 2015; Phan et al. 2011) 10]5.3216594.1580[M + H]^+^[C_27_H_31_O_15_]^+^0.316.5200, 248, 336Kaempferol-disaccharide (hexoside + deoxyhexoside)1.9S5.3317434.0850[M + H]^+^[C_20_H_19_O_11_]^+^0.931.8303200, 256, 332Quercetin-arabinoside1.32.6B(Dallenbach-Toelke et al. 1986; Pietta et al. 1989)5.4718448.1000[M + H]^+^[C_21_H_21_O_11_]^+^1.618.2147, 303200, 248, 336Quercetin-rhamnoside3.02.82.54.0B, S(Bongartz and Hesse 1995; Pawłowska 2014)5.5819360.0828[M + H]^+^[C_18_H_17_O_8_]^+^219.1163200, 250, 328Rosmarinic acid1.71.7Or(Sumaryono et al. 1991)5.6920332.1814[M + Na]^+^[C_16_H_28_NaO_7_]^+^2.220.5153, 171204, 316?1.1B5.7921432.1060[M + H]^+^[C_21_H_21_O_10_]^+^0.125.9287200, 285, 330Kaempferol deoxyhexoside1.1?5.8622490.1110[M + H]^+^[C_23_H_23_O_12_]^+^−0.210?1.6S5.9523571.2040[M + H]^+^?−0.2?200, 280, 325?1.1On6.0124234.1018[M + H]^+^[C_10_H_20_O_4_P]^+^−0.655.3137Geranylphosphate1.1S6.0325500.1800[M + Na]^+^[C_23_H_32_NaO_12_]^+^0.917.1167204, 280, 320Glyasperin D1.11.3B6.0326244.1310[M + Na]^+^[C_12_H_20_NaO_5_]^+^−1.917.4209, 227204, 280?E6.0727532.1220[M + H]^+^[C_25_H_25_O_13_]^+^−0.48.6285204, 280, 320methylated flavone-5-O-(6″-malonylglucopyranoside)1.4E(Veit et al. 1990)6.0828557.2274[M + H]^+^[C_29_H_36_NO_10_]^+^−2.810.4204, 280, 320?1.4On7.5329180.1137[M + H]^+^[C_11_H_17_O_2_]^+^−1.78.5220, 280?1.2S, Or7.5330222.1618[M + H]^+^[C_14_H_23_O_2_]^+^0.420.9220, 280?S7.7931372.1210[M + H]^+^[C_39_H_51_O_25_]^+^0.48.8271Sinensetin1.52.4Or(Ohashi et al. 2000)8.1732342.1124[M + H]^+^[C_19_H_19_O_6_]^+^−0.96224, 320Tetramethoxyflavone1.61.6Or(Malterud et al. 1989; Ohashi et al. 2000)8.1833314.0790[M + H]^+^[C_17_H_15_O_6_]^+^−1.918.5224, 325DihydroxydimethoxyflavoneB8.2934234.1640[M + H]^+^[C_15_H_23_O_2_]^+^−1.712.8217224?S8.2935570.2489[M + NH4]^+^[C_31_H_42_NO_10_]^+^−326.7224Orthosiphol M1.5Or(Awale et al. 2001; Stampoulis et al. 1999)8.536330.1838[M + H]^+^[C_20_H_27_O_4_]^+^0.42.2224Diterpenbutenolid1.6S(Jurenitsch et al. 1988)9.5437676.2880[M + NH4]^+^[C_38_H_48_NO_11_]^+^1.918.9495, 555224Orthosiphol2.4Or(Masuda et al. 1992)9.6238218.1680[M + H]^+^[C_15_H_23_O]^+^−0.610.9224Beta-Betulenal2.21.81.53.6B, S(Demirci et al. 2000; Křepinský and Herout 1962; Weyerstahl et al. 1993)9.6839620.3948[M + Na]^+^[C_35_H_56_NaO_9_]^+^2.328.4439, 525, 543, 585224Acetylated/malonylated dammarane triterpene2.2B(Hilpisch et al. 1997)9.8140604.3672[M + NH4]^+^[C_34_H_56_NO_9_]^+^5.730.5423, 545224Acetylated/malonylated dammarane triterpene1.1B9.8341620.3932[M + Na]^+^[C_35_H_56_NaO_9_]^+^−0.79.8421, 439, 525, 543, 585224Acetylated/malonylated dammarane triterpene4.13.72.7B(Hilpisch et al. 1997)10.2842560.3360[M + H]^+^[C_32_H_49_O_8_]^+^2.912.2397, 501224Acetylated/malonylated dammarane triterpene1.3B(Hilpisch et al. 1997)10.3943604.4031[M + Na]^+^[C3_5_H_56_NaO_8_]^+^5.562.1341, 423, 527, 569224Acetylated/malonylated dammarane triterpene1.81.51.5B(Hilpisch et al. 1997; Reichardt 1981; Taipale et al. 1993)10.6444602.3872[M + NH4]^+^[C_35_H_58_NO_8_]^+^−5.39.7525, 543224Acetylated/malonylated dammarane triterpene1.41.4B10.9245620.3971[M + H]^+^[C_35_H_5_7O_9_]^+^−4.410.6439, 457, 543, 561224Acetylated/malonylated dammarane triterpene2.42.21.8B(Hilpisch et al. 1997)11.6246620.3964[M + Na]^+^[C_35_H_56_NaO_9_]^+^2.412.3543, 561, 621224Acetylated/malonylated dammarane triterpene3.73.52.8B(Hilpisch et al. 1997)11.7047604.4016[M + Na]^+^[C_35_H_56_NaO_8_]^+^–4.113.4423, 441, 509, 527, 545, 587224Acetylated/malonylated dammarane triterpene3.12.72.3B(Hilpisch et al. 1997; Reichardt 1981; Taipale et al. 1993)12.1148588.4088[M + H]^+^[C_35_H_56_NaO_7_]^+^6.27.6407, 511, 571224Acetylated/malonylated dammarane triterpene1.21.31.1B(Taipale et al. 1993)^‡^Specific names as used in reference; unspecific names without reference are based on interpretation of MS^1^- and where available MS^2^-data and UV spectra. The same compound is repeatedly assigned when non-distinguishable MS datasets occurred^#^B: Betula sp leaves., E: Equisetum arvense aerial parts, S: Solidago sp. aerial parts, On: Ononis spinosa roots, Or: Orthosiphon aristatus leaves

The minimum inhibitory concentration (MIC) of extracts A to D could not be determined. Compared to the untreated culture, cultivation of UPEC strain CFT073 in the presence of the four extracts did not result in a significant reduction in bacterial growth, even at a concentration of 1 mg/ml after 21 h of incubation (Supplementary Fig. 1). Furthermore, these plant extracts did not significantly inhibit the growth rate of UPEC compared to the untreated controls; on the contrary, at high concentrations, they tended to slightly improve bacterial growth (with the exception of extract C) (Supplementary Fig. 2). Interestingly, the complex extracts A, B, and C significantly increased biofilm formation at 330 μg/ml after 48 h (Supplementary Fig. 3).

To assess bacterial stress responses to the plant extracts, we used a stress reporter based on E. coli K-12 strain MG1655 (Chitto et al. 2025) that carries the promoterless yellow fluorescent protein-encoding yfp gene fused to the promoter of representative marker genes gadA (pH stress), spy (envelope stress), otsAB (osmotic stress), recA (SOS response), and dps (oxidative stress). We first evaluated the overall response of these stress reporters to the different plant extracts A to D. The Pspy module exhibited a notable increase in yellow fluorescence intensity following the addition of A, B, and C (Figs. 1A and Supplementary Fig. 4). This response was concentration-dependent, with the highest signal observed at 1 mg/ml, indicating a strong induction compared to the control. In contrast, extract D failed to induce Pspy activation. To determine whether the Pspy response in our reporter strain was specifically regulated or influenced by unspecific factors, we tested the plant extracts on the baeR^−^ and cpxR^−^ mutants of our reporter strain. We did not observe any Pspy induction in the baeR^−^ mutant strain for most of the extracts eliciting a positive spy response, but they led to a reduced Pspy induction in the cpxR^−^ mutant strain (Fig. 2A–D). This suggests that the observed surface stress response was indeed predominantly mediated via the BaeS-BaeR two-component system, with CpxA-CpxR acting as a modulator (Supplementary Fig. 4). However, other than Pspy, the plant extracts were unable to induce pH stress (gadA), organic solvent stress (pspA), osmotic stress (otsAB), SOS response (recA), and oxidative stress (dps) (Fig. 1B), suggesting specificity in bacterial surface stress response.Fig. 1. Several traditional herbal urologicals induce envelope stress response in E. coli K-12 strain MG1655. The promoter activity of Pspy in response to incubation with extract B was quantified as a biomarker for surface stress. The arrow indicates the time point when the extract was added to the bacterial culture. Shown are the results of representative experiments obtained for the different concentrations tested (A). Overview of the overall effect of the four extracts on the stress reporter systems (outcome of three biological replicates) (B)Fig. 2. Regulation of envelope stress response induced by extract B in E. coli K-12 strain MG1655. Fluorescence measurements to see the effect of extract B on the spy stress response in the A E. coli strain MG1655 background and its spy regulatory mutants baeR^-^ B and cpxR^−^ C. The arrow indicates the time point when the extract was added to the bacterial culture. D Comparison of total YFP expression between the wild type strain and the spy regulatory mutants. Depicted are the mean values and standard errors of the mean for three biological replicates and two technical replicates in comparison to the water control. *p < 0.05; **p < 0.005; ***p < 0.0005; ****p < 0.0001

We performed a yeast agglutination assay to determine whether elevated bacterial surface stress, as indicated by the Pspy induction, impairs the function of type 1 fimbriae in UPEC. We observed that bacterial treatment with the extracts B, C, and D resulted in yeast agglutination patterns similar to those of the untreated control (Fig. 3), indicating that the expression and functionality of type 1 fimbriae remained unaffected in response to these plant extracts. However, pre-treatment of UPEC with extract A caused less yeast agglutination. This result resembled the mannose-treated control, where type 1 fimbria-mediated agglutination of yeast cells by UPEC was blocked. This effect was specific to extract A (Fig. 3).Fig. 3. Impact of plant extracts on the expression of functional type 1 fimbriae. UPEC strain CFT073 was incubated with the individual plant extracts A to D to test for mannose-sensitive type 1 fimbriae-dependent agglutination of yeast cells. The extent of yeast agglutination by the treated bacteria or untreated control (UC) is indicated by + and − symbols below the photos. The agglutination was performed in the absence or presence of 3% D-Mannose in the yeast suspension. The addition of 3% D-Mannose to the yeast suspension served as a negative control for mannose-specific binding, thereby demonstrating that any observed yeast agglutination was not due to non-specific interactions, but mannose-specific adhesion mediated by the tested bacteria. The assay was conducted with 3 biological replicates. Similar agglutination patterns were observed; representative results are displayed

Plant extracts interfere with the early stages of UPEC infection of human urinary bladder epithelial cells

We assessed whether preincubating T24 bladder epithelial cells with the plant extracts (330 µg/ml) before infection with the UPEC strain CFT073 impacted UPEC adhesion. We observed that preincubation (45 min) of the host cells with extracts A to D significantly reduced the colony forming units (CFUs) of T24 cell associated UPEC bacteria compared to the untreated control (Fig. 4A), reflecting the bacterial adhesion to the host cells. We also determined that the viability of the T24 human bladder epithelial cell line was not affected by the incubation with the plant extracts at a concentration of 330 µg/ml (Supplementary Table 2, Supplementary Fig.5). This implies that the plant extracts substantially impaired the adhesion of UPEC to the epithelial cells.Fig. 4. Treatment of bladder epithelial cells with plant extracts reduces bacterial adhesion and invasion. T24 human bladder epithelial cells were incubated with extracts A to D at 330 µg/ml concentration for 45 min followed by infection with UPEC strain CFT073 for 2.5 h, lysed and plated to determine bacterial adhesion (A). To determine the impact of the extracts on bacterial invasion, the T24 cells were incubated for 45 min (B) or 3 h (C) with the plant extracts (330 µg/ml) before they were treated for 1.5 h with gentamycin (200 µg/ml), lysed and plated on LB agar. The experiments were done with 3 biological replicates with 2 technical replicates each and were compared to the untreated control, UC. Depicted are the mean values and standard errors of the mean for three biological replicates and two technical replicates. * p < 0.05; ** p < 0.005; *** p < 0.0005; **** p < 0.0001

We investigated how the duration of treatment with the plant extracts affects UPEC-urothelial cell interaction. A brief 45-min exposure of the T24 cells to the tested plant extracts—except for extract B—was sufficient to significantly reduce the number of intracellular UPEC (Fig. 4B). Prolonged exposure for 3 h resulted in a further decrease of intracellular UPEC in the case of extracts C and D and also significantly reduced intracellular bacterial counts in extract B-treated cells (Fig. 4C). Overall, all tested plant extracts caused a significant reduction in intracellular UPEC compared to the untreated control. The inhibitory effect displayed a clear concentration dependence, with the highest concentration (330 µg/ml) achieving the most pronounced reduction in intracellular UPEC replication (Supplementary Fig.6). We observed a similar significant reduction of intracellular UPEC in the human uroepithelial 5637 and RT-112 cell lines (Supplementary Fig. 7). Therefore, the four plant extracts studied effectively reduced the number of intracellular UPEC in three different human urinary bladder cell lines irrespective of cell line variations.

Furthermore, we observed that even after a 24-h period, the number of intracellular UPEC in the pretreated T24 cells was significantly reduced compared to the untreated control. However, upon evaluation of the number of intracellular bacteria after 24 h in relation to the value after 3 h of infection in untreated and pretreated T24 cells, no significant difference was observed (with the exception of extract D) (Supplementary Fig. 8A). We also examined whether extracts A-D influenced the replication of intracellular UPEC within the host cell after the bacteria had invaded it. However, the addition of the plant extracts after UPEC infection and gentamycin treatment during the 24-h UPEC infection had no impact (except a slight reduction caused by extract D) on the intracellular replication of the bacteria (Supplementary Fig. 8B).

Oral application of the plant extracts stimulates innate immune response and reduces UPEC infection in the surrogate insect host Galleria mellonella

To determine whether the anti-infective properties of the HMP correlate with an upregulation of the host innate immune response, we assessed their impact on the innate immunity of the surrogate host G. mellonella. This widely accepted in vivo screening model for anti-infectives and bacterial pathogenesis has already been successfully used in studying bacterial virulence mechanisms and host response to infection. We tested whether feeding plant extracts to uninfected G. mellonella larvae boosts the expression of innate immune genes in their midgut and hemocytes. To find out whether this immune activation helps the larvae resist later UPEC infection, RNA was isolated from the midgut and hemocytes of non-infected larvae that had been treated orally with the respective plant extracts (330 µg/larva) for 24 h. The expression of innate immunity genes was analyzed by RT-PCR.

Our results showed that genes coding antimicrobial peptides such as moricin and hemolin were strongly induced in the midgut by plant extracts A and D compared to naïve, unchallenged larvae (Fig. 5 A). Additionally, oral application of plant extracts A and D also resulted in enhanced moricin and hemolin expression in the hemocytes of larvae (Fig. 5B). In contrast, the other plant extracts (B and C) induced only low levels of innate immune gene expression in hemocytes.Fig. 5. Analysis of the immunostimulatory effects of plant extracts in the G. mellonella model of UPEC infection. The expression of the moricin, hemolin, apolipophorin III, and prophenoloxidase-encoding genes was analyzed in the midgut (A) and in circulating hemocytes (B) of larvae 24 h after oral administration of different plant extracts using quantitative real-time RT-PCR. Basal expression in the treated larvae was calculated as fold-change relative to untreated control larvae and normalized to the 18S rRNA housekeeping gene. Results represent the mean values of at least three independent determinations ± SE. Kaplan–Meier survival plots depict the survival of UPEC-infected larvae pre-exposed to plant extracts A and B (C) and extracts C and D (D), respectively. UPEC-infected larvae that received water instead of plant extracts served as controls. Results represent the means of at least three independent determinations for 10 animals per treatment. *p < 0.05; **p < 0.005; ***p < 0.0005

To assess whether the immunostimulatory plant extracts enhance host responses to UPEC infection, G. mellonella larvae were injected with the UPEC strain CFT073 24 h after oral administration of either the plant extracts (330 µg/larva) or water as a control. Our results showed that UPEC injection in control larvae resulted in high mortality (survival rate 50%) (Fig. 5C, D). However, larvae pre-treated with the various plant extracts exhibited improved survival rates: extract A (survival rate 82%), extract B (survival rate 80%), extract C (survival rate 90%) and extract D (survival rate 89%) (Fig. 5C, D).

Plant extracts enhance urothelial innate host cell responses to UPEC infection

To validate the relevance of our findings on the immunostimulatory effects of plant extracts in G. mellonella larvae, we also studied the innate immune response in different human bladder epithelial cell lines. For this, non-infected human uroepithelial cell lines (T24 and 5637) were incubated with 330 µg/ml of the plant extracts for 3.5 h, followed by RNA extraction and real-time PCR analysis of the expression of innate immune genes of the host cells. We observed a significant induction of the genes coding for IL-6, IL-8, and CXCL-3 in T24 cells, by most plant extracts, particularly A and D, compared to untreated cells (Fig. 6 A-B). Similar to T24 cells, we saw a significant increase in IL-6 expression at the transcriptional level in 5637 cells treated with the extracts A, B, and D (Supplementary Fig. 9 A). In addition, we observed a significant increase in transcript levels of the CXCL-3-encoding gene in these cells treated with the extracts A, B, and C compared to the control (Supplementary Fig. 9 B). Additionally, we examined the expression of these innate immune genes in treated vs. untreated RT-112 human bladder epithelial cells and observed that, similar to T24 and 5637 cells, treatment with extracts A and B led to increased IL-6 expression, although the fold changes were not as robust as those observed in the T24 and 5637 cell lines (Supplementary Fig. 9 C). Transcript levels of other innate immune genes, including those coding for IL-8, CXCL-3, and β-defensin, were not induced in RT-112 cells by the plant extracts (Supplementary Fig. 9B-D).Fig. 6. Plant extracts show an immunostimulatory effect in T24 cells. Fold-change of IL-6, IL-8, CXCL-3, and β-defensin transcript levels in T24 cells after preincubation with 330 µg/ml of extracts without UPEC CFT073 infection (A, B) and with UPEC CFT073 infection (C, D). Fold-change of selected innate immune-related genes in A, B is calculated relative to the untreated control and that in C, D relative to the treated uninfected control, both normalized to the 18S rRNA housekeeping gene. All data shown here result from 3 biological replicates and 2 technical replicates each and are compared with the UC. *p < 0.05; **p < 0.005; ***p < 0.0005; ****p < 0.0001

To explore whether the immunostimulatory plant extracts can further enhance immune gene expression upon UPEC infection, T24 cells were pre-treated with the extracts for 45 min and subsequently infected with the UPEC strain CFT073 for 3 h. The expression levels of a selected set of innate immune genes (coding for IL-6, IL-8 and CXCL-3) were then assessed and normalized to the treated but non-infected control, to evaluate the impact of UPEC infection on immune gene expression. Generally, infection of the treated T24 cells resulted in strongly increased immune gene expression. The relative expression level of IL-6 was significantly higher in treated, infected cells compared to treated, non-infected cells. In T24 cells treated with plant extracts B and C, infection also caused a strong and often significant increase in IL-8 and CXCL-3 expression (Fig. 6 C-D). Transcription of the gene coding for the anti-microbial peptide β-defensin was not affected by the treatment with the four plant extracts, but markedly induced upon UPEC infection (Fig. 6B and D, Supplementary Fig.9 B).

Discussion

Plant urologicals have a long tradition in the treatment of UTIs. Certain HMP formulations have demonstrated potential in interfering with the function of adhesive traits of the pathogens or preventing or alleviating UTI symptoms (Marouf et al. 2022; Mulat et al. 2025; Yang et al. 2016). We tested HMP which are widely used in European clinical practice and which are also government-registered drug products European Medicines Agency. The evidence for their effectiveness is largely based on traditional use, but mechanistic insights into their mode of action are often missing. Given the growing concern over bacterial resistance to antibiotics, there is an urgent need to search for additional or alternative treatment options to antibiotics. As HMP formulations may be promising alternatives or supplements to antibiotics in the treatment of UTIs, we aimed at elucidating the mechanisms of action of these complex HMP formulations at the pathogen-host interaction level, given the absence of obvious direct antibacterial effects.

To characterize the effects of four selected HMP formulations used in Germany for preparing bladder and kidney teas for UTI treatment, we sought to understand their impact on the bacterial stress response. This is because they may reduce the effectiveness of bacterial virulence factors and disrupt general cellular processes, thereby triggering stress in bacteria. UPEC and other E. coli strains are well adapted to survive adverse conditions during infection, such as nutrient scarcity, acid stress, oxidative environments, and DNA damage, due to their highly regulated stress response systems (Dawan and Ahn 2022; Fang et al. 2016). Biofilm formation was slightly increased upon bacterial cultivation in the presence of the herbal extracts (Suppl. Figure 3). As biofilm formation can also be seen as a response against external stress or unfavorable exogenous conditions (Rossi et al. 2018; Zhao et al. 2023), and since sub-inhibitory concentrations of antimicrobials can induce stress-related pathways (Brand et al. 2025), we decided to test whether or not watery extracts of the four HMP formulations induced well-characterized stress-response pathways in UPEC. Our set of bacterial stress biosensor strains enabled us to analyze the bacterial stress response to the plant extracts. We included the relevant key bacterial stress-related genes gadA (pH stress) (Castanie-Cornet and Foster 2001), spy (envelope stress) (He et al. 2021), otsAB (osmotic stress) (Purvis et al. 2005), recA (SOS response) (Diez et al. 2000), and dps (oxidative stress) (Calhoun and Kwon 2011) into our analysis. Interestingly, cultivation with extracts A to C increased the promoter activity of the spy gene (Fig. 1A), which is indicative of bacterial surface stress. However, other than Pspy, the plant extracts were unable to induce stress responses to pH, osmosis, the SOS response, or oxidative stress (Fig. 1B), suggesting specificity in the bacterial response to surface stress. As a periplasmic chaperone, Spy plays a role in maintaining the proper function of envelope-associated proteins, including adhesins that mediate UPEC attachment to receptors on the urothelium (Eto et al. 2007; Maalouf et al. 2022). The expression of the Spy-encoded periplasmic chaperone, which protects and refolds proteins, is activated in response to protein unfolding and other stress conditions. These include exposure to tannins, via the two-component regulatory systems BaeS-BaeR and CpxA-CpxR (Zoetendal et al. 2008). As extracts A to D contain various saponin- and tannin-like compounds with well-known surface activity (e.g. cmp 38 to 46 and cmp 19, respectively, as detected in the MS dereplication; see Table 2), these compounds could induce similar effects on the bacterial surface. From this point of view, the induction of the Spy chaperone is plausible. Our stress response test results, combined with our observations from the yeast agglutination assay (Fig. 3), demonstrate that despite surface stress, only extract A inhibits the function of type 1 fimbriae. The envelope stress response did not necessarily correlate with impaired expression of bacterial type 1 fimbriae mediating UPEC adhesion to bladder epithelial cells. This indicates that although various extracts induce surface stress in the bacteria, they can nevertheless have different effects on bacterial virulence properties. Such effects caused by extract A may reduce the ability of UPEC to adhere to epithelial cells, potentially offering a pathway for therapeutic intervention (Nielubowicz and Mobley 2010).

Adherence to uroepithelia is the prerequisite for subsequent stages of UPEC infection, including UPEC invasion into bladder epithelial cells (Dhakal and Mulvey 2009). Consequently, we investigated the impact of the four traditional herbal urologicals on UPEC adhesion and invasion. The plant extracts interfered with early stages of UPEC infection, i.e., bacterial adhesion to and invasion into the bladder epithelial cells. They effectively reduced the number of intracellular UPEC in three different human urinary bladder cell lines irrespective of cell line variations (Fig. 4, Suppl. Figures 6 and 7), but they did not interfere with the survival and replication of intracellular bacteria (Suppl. Figure 8). Pre-treatment of the host cells with the plant extracts could preventively reduce the interaction between UPEC and host cells as well as the bacterial invasion. Consequently, all tested plant extracts significantly reduced UPEC adhesion and invasion, but not bacterial replication in bladder epithelial cells. Our previous data on the effect of the aqueous extract of Orthosiphon stamineus, one of the plant components found in formulations C and D, confirm an anti-adhesive effect, as well as a reduced frequency of bladder and kidney infections, in a murine model of ascending urinary tract infections (Sarshar et al. 2017).

So far, our findings have shown that the plant extracts mainly inhibit UPEC infection by altering key properties of the host cells, rather than through antimicrobial activity. We are aware that the G. mellonella system cannot provide evidence of clinical efficacy and cannot replace animal infection models. Nevertheless, a wide range of literature, as well as toxicological databases, clearly shows that G. mellonella is a well-established and valid surrogate model screening platform in biomedical research and drug discovery. It is known to reproduce conserved aspects of innate immunity, reliably assess bacterial virulence and identify anti-infective activities that often translate to mammalian systems (de Matos Silva et al. 2024; Villani et al. 2025). To simulate the effects of bladder and kidney teas, and to study the effects of plant extracts after oral administration, as well as the combined action of different components of the host's innate and cellular responses, we fed the extracts to G. mellonella larvae prior to systemic UPEC infection and examined their survival. The midgut of G. mellonella larvae plays a critical role in initiating immune responses, including in hemocytes circulating in the hemolymph. Midgut-derived signals can trigger systemic immune responses, including hemocyte activation and priming (Grizanova et al. 2014). Insects like G. mellonella can express various classes of antimicrobial molecules, such as moricin (an antimicrobial peptide) and hemolin (a member of the immunoglobulin (Ig) superfamily) in response to bacterial infections (Asai et al. 2021; Xu et al. 2019). The midgut can affect hemocytes indirectly through humoral factors released into the hemolymph. For instance, damage to midgut epithelial cells can release immune activators (e.g., ROS, AMPs, or cytokines) that circulate and stimulate hemocytes (Grizanova et al. 2014). Moreover, when the midgut is exposed to sublethal concentrations of pathogens or their components (e.g., heat-killed bacteria), it can initiate a systemic alert that prepares hemocytes for future challenges. Our results indicate that oral feeding of individual plant extracts can enhance larval survival and immune gene expression in the larval midgut and hemocytes (Fig. 5), potentially enabling the host to eliminate UPEC more effectively, entering through the septic route.

We supported the relevance of our findings in the G. mellonella larval model, as an ethical and cost-effective alternative to mammalian infection models, by subsequent validation in different human bladder epithelial cell lines. Previous studies have shown that the expression of the cytokines IL-6 and IL-8, along with that of the chemokine CXCL-3, was induced early during UPEC infection in human cell lines and in a mouse UTI model as part of the host innate immune response against UPEC colonization (Armbruster et al. 2018; Koley and Mukherjee 2024). IL-6 can promote the host defense against UPEC by activating the immune system and triggering the expression of, e.g., antimicrobial peptides (Ching et al. 2018). Our results show that even in the absence of infection, preincubation with the tested plant extracts induces an immunostimulatory effect in T24 and 5637 cells, potentially preparing them to mount a more effective defense against subsequent infection. This also fits with our observation that treating T24 cells with extracts A and D, which induce very strong IL-6 expression (Fig. 6A), causes the most significant reduction in intracellular UPEC (Fig. 4B). The four tested plant extracts, especially extracts A and D, appear to trigger a baseline immune response, which can be further amplified by infection, thereby enhancing urothelial cell responses to infection. The differences in immune gene stimulation by the extracts may be attributed to inherent variations among the T24, 5637, and RT-112 cell lines. Although the induction of IL-6, IL-8, and CXCL-3 can have broader immunological relevance, our study does not allow conclusions about effects on other pathogens or cell types. At this stage, we cannot rule out that the HMP tested may exert similar immunostimulatory effects in other cell types and against other pathogens. The potentially broader immunostimulatory activity of the HMP beyond UPEC remains to be investigated in the future.

In summary, our study on the underlying mechanisms responsible for the mode of action of traditional aqueous plant extracts highlights their potential as agents that target multiple sites in the fight against UPEC infection. By acting on both bacterial and host factors, these extracts offer diverse mechanisms of host protection, supporting their use as preventive strategies against urinary tract infections. Several studies on plant extracts have mainly focused on the impact of the extracts on the expression and functionality of bacterial virulence factors, but very few studies have emphasized the role of natural compounds in regulating host responses during UTI (Lewis et al. 2024). Our findings demonstrate that the plant extracts interfere with the early stages of UPEC-host interaction on the host cell side rather than with bacterial pathogenicity traits. Interestingly, most of these extracts may function by preparing the host to fight against the infection. Hence, follow-up work needs to be conducted to understand the exact mechanisms by which these extracts exert the anti-invasive effect at the host cell level and to determine the contribution of individual herbal components and phytochemicals. At the present stage, it is not possible to attribute the observed effects to an individual herbal material or a single natural product of the complex mixtures, as not only the individual plants determine the composition of the extract, but also the mixture of the different plants affects the extractability of compounds. It is also essential to substantiate the physiological relevance and translational potential of the results. Our findings still need to be validated in an animal model that more accurately reflects the natural infection niche of UPEC. Yet, our findings provide the first mechanistic rationale for using aqueous plant extracts against UTIs at the pathogen-host interaction level, outlining their probable modes of action.

Supplementary Information

Below is the link to the electronic supplementary material.ESM 1(PDF 308 KB)