Agar-assisted biofilm cultivation method

Abstract

The invention relates to a microbial biofilm cultivation method comprising the use of a soft agar hydrogel substrate and one or more stains for detection and / or quantification of one or more microbial components and / or characteristics. The invention further relates to the use of the microbial biofilm cultivation method for screening of one or more antimicrobial compounds.

Description

[0001] AGAR-ASSISTED BIOFILM CULTIVATION METHOD

[0002] FIELD OF THE INVENTION

[0003] This invention relates to a microbial biofilm cultivation method comprising the use of a soft agar hydrogel substrate and one or more stains for detection and / or quantification of one or more microbial components and / or characteristics. The invention further relates to the use of the microbial biofilm cultivation method for screening of one or more antimicrobial compounds.

[0004] BACKGROUND TO THE INVENTION

[0005] When infecting hosts, many bacterial pathogens secrete extracellular polymers and aggregate to form biofilms. Biofilms are protective structures that improve the pathogen’s survival within the host.

[0006] Through biofilm formation, participating cells often show physiological states and behaviour extremely different to their dispersed forms. In particular, the chemoresistances of biofilm-associated cells tend to be much stronger, resulting in decreased susceptibility to antibiotics and enabling subsequent antibiotic resistances to form. This prevents or hinders effective medical treatment.

[0007] It is now widely recognised that biofilms are the more representative phenotype of bacterial cells found in natural systems (Branda et al., 2005), exhibiting similar resistances to antimicrobial exposure, compared to their natural counterparts i.e. planktonic / dispersed cells typically cultured in vitro.

[0008] Consequently, recent studies have emphasised screening of antimicrobial substances using in vitro biofilm systems as representative systems for infective behaviour, rather than the dispersed cell systems commonly used for screening.However, these studies report difficulties in producing consistent biofilms which hinders representative screening from taking place (Auger et al., 2017 and Jorge et al., 2015).

[0009] Furthermore, existing studies have predominantly emphasised the rheological properties of agar hydrogels, particularly stiffness, rather than their application as solid substrates for biofilm adherence, formation and maturation. For example, (Kolowe et al., 2015) showed that increasing the stiffness of the agar hydrogel (from 3% to 9%) correlates with enhanced bacterial adherence, which in turn promotes biofilm formation. However, a stiff agar hydrogel does not mimic the natural environment in which a biofilm would grow and it would be useful if a more physiologically relevant substrate could be developed for studying biofilm formation.

[0010] It would be useful if bacterial biofilms with higher numbers of cell viability could be generated consistently for use as a model to screen antibacterial compounds against a variety of biofilm-producing Gram-positive and Gram-negative bacteria. More particularly the development of standardised biofilms between cultivations and improved intra-batch consistency of biofilms would be beneficial.

[0011] It would also be useful if, using such a bacterial biofilm cultivation system, changes in cell viability and overall biomass could be monitored overtime during development of the biofilm.

[0012] SUMMARY OF THE INVENTION

[0013] ACCORDING TO A FIRST ASPECT OF THE INVENTION THERE IS PROVIDED a method of cultivating a microbial biofilm comprising the use of a soft agar hydrogel substrate and one or more stains for detection and / or quantification of one or more microbial components and / or characteristics. In particular, the soft agar hydrogel substrate has a concentration of about 1.2% (w / v). In a preferred embodiment of the invention, the soft agar hydrogel substrate has a diameter and height of about 8 mm and about 4 mm, respectively. For example, the soft agar hydrogel substrate may be comprised within a 96-well microtiter plate.The microbial biofilm may comprise any microorganism that has the capacity to produce a biofilm, for example bacteria, fungi and algae. In a preferred embodiment, the method is for cultivation of a bacterial biofilm, which may be comprised of Grampositive or Gram-negative bacteria. For example, the Gram-negative bacteria may be Escherichia coli (E. coli). Further for example, the Gram-positive bacteria may be Bacillus subtilis (B. subtil is).

[0014] In a possible embodiment, the agar may be bacteriological agar also known as agar agar.

[0015] The method may comprise or consist of the following steps:

[0016] a) providing a cultivation vessel having a solid substrate;

[0017] b) adding the agar into the cultivation vessel and allowing it to solidify into a hydrogel;

[0018] c) adding liquid microbiological media to the cultivation vessel;

[0019] d) inoculating microorganisms to be cultured into the cultivation vessel;

[0020] e) cultivating the microorganisms in the cultivation vessel to produce a biofilm of microorganisms; and

[0021] f) staining the microorganisms during cultivation with the one or more stain(s) for detection and / or quantification of one or more microbial components and / or characteristics.

[0022] Steps c) and d) may be performed separately or simultaneously. When performed simultaneously, the microorganisms are inoculated into the liquid microbiological media which is then added to the cultivation vessel.

[0023] For example, in step f) any one or more stain(s) suitable for identification of a microorganism, or microorganism viability, or both may be used. In particular, the microorganism may be a bacterium. Further for example, the one or more stain(s) may be a fluorometric or a colorimetric stain. The one or more stain(s) may comprise or consist of any one or more stain(s) as set out in Table 1.

[0024] Microbial cell viability may be quantified with the use of any one or more staining methods capable of detecting viable microbial cells or quantification of extracellularDNA such as DAPI (4',6-diamidino-2-phenylindole) or CV (crystal violet) to quantify general biomass / total cells, resazurin or tetrazolium dye or MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) for quantifying cell viability, using an organometallic ruthenium chelate such as SYPRO™ Ruby or CBB (Coomassie Brilliant Blue) for quantifying protein content, and / or using 5,5'-(6,22-Dioxo-11,14,17-trioxa-7,21-diazaheptacosane-1,27-diyl)bis(3,8-diamino-6-phenylphenanthridin-5-ium) diiodide (also known as Dye No. 35, ET-27 or GelRed®) for quantifying DNA concentration. The response may be measured using spectrophotometry and / or spectrofluorometry followed by microscopy including colorimetric or fluorescence microscopy.

[0025] In one preferred embodiment, two stains are used, the first, a resazurin fluorescent stain (as a proxy for cell viability) and the second, a (4’,6-diamidino-2-phenylindole) (also known as DAPI) fluorescent stain to determine general biomass I total cells.

[0026] It is to be appreciated that step e) may be performed for any period of time as appropriate for establishment of the microbial biofilm and is dependent on variables including the microorganism strain, the microorganism inoculation density / concentration, and the microorganism growth kinetics. A person skilled in the art when monitoring the microbial biofilm development would be capable of selecting the appropriate duration of cultivation in accordance with these variables.

[0027] However, in one possible example, step e) may be performed for between about 12 and 120 hours or any time period therebetween, preferably between about 24 and 120 hours or any time period therebetween, more preferably between about 48 and about 120 hours or any time period therebetween.

[0028] Step e) may further comprise a step of removing a volume of the liquid microbiological media from the cultivation vessel and replacing this volume with fresh liquid microbiological media to replenish nutrients for the microorganism growth. The step of replacement of the liquid microbiological media may be performed at about 24 hours from the start of cultivation, and about each 24 hours thereafter until the completion of cultivation of the biofilm. For example, 70 to 90% of the total liquid microbiological media, or any volume therebetween, more preferably 80% of the liquid microbiologicalmedia may be removed from the cultivation vessel and be replaced with an equivalent volume of fresh liquid microbiological media.

[0029] The cultivation vessel may be any laboratory-grade vessel in which liquid microbiological cultivation be performed. For example, the cultivation vessel may be one or more wells in a microtiter plate, more particularly, the cultivation vessel is a 96-well microtiter plate.

[0030] ACCORDING TO A SECOND ASPECT OF THE INVENTION THERE IS PROVIDED

[0031] a method of screening one or more antimicrobial agents for efficacy against a microbial biofilm with the use of a microbial biofilm cultivated according to the method of the invention.

[0032] The method may comprise the steps of:

[0033] i) providing a microbial biofilm of microorganisms cultivated according to the method of the invention upon which the one or more antimicrobial agents are to be screened;

[0034] ii) adding the one or more antimicrobial agents to the microbial biofilm;

[0035] iii) incubating the microbial biofilm with the one or more antimicrobial agents; and iv) detecting efficacy of the one or more antimicrobial agents against the microbial biofilm by measurement of one or more parameters such as microbial cell viability.

[0036] Microbial cell viability may be quantified with the use of any one or more staining methods capable of detecting viable microbial cells or quantification of extracellular DNA such as DAPI (4',6-diamidino-2-phenylindole) or CV (crystal violet) to quantify general biomass / total cells, resazurin or tetrazolium dye or MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) for quantifying cell viability, using an organometallic ruthenium chelate such as SYPRO™ Ruby or CBB (Coomassie Brilliant Blue) for quantifying protein content, and / or using 5,5'-(6,22-Dioxo-11,14,17-trioxa-7,21-diazaheptacosane-1,27-diyl)bis(3,8-diamino-6-phenylphenanthridin-5-ium) diiodide (also known as Dye No. 35, ET-27 or GelRed®) for quantifying DNA concentration. The response may be measured using spectrophotometry and / or spectrofluorometry followed by microscopy includingcolorimetric or fluorescence microscopy. In one preferred embodiment, two stains are used, the first, a resazurin fluorescent stain (as a proxy for cell viability) and the second, a (4’,6-diamidino-2-phenylindole) (also known as DAPI) fluorescent stain to determine general biomass / total cells.

[0037] The one or more antimicrobial agents may be selected to have one or more different concentrations and / or one or more different formulations.

[0038] The antimicrobial agent may be an antibacterial agent. For example, the antibacterial agent may be a broad-spectrum antibiotic agent such as ciprofloxacin or kanamycin.

[0039] ACCORDING TO A THIRD ASPECT OF THE INVENTION THERE IS PROVIDED a microbial biofilm cultivated according to the method of cultivation of the invention for use in a method of screening one or more antimicrobial agents for efficacy against the microbial biofilm.

[0040] The method of screening may be as hereinbefore described.

[0041] ACCORDING TO A FOURTH ASPECT OF THE INVENTION THERE IS PROVIDED

[0042] a kit for cultivating a microbial biofilm according to the method of cultivation of the invention, the kit comprising:

[0043] A. a cultivation vessel having a solid substrate;

[0044] B. agar;

[0045] C. liquid microbiological media;

[0046] D. optionally, one or more microorganisms to be cultivated;

[0047] E. one or more stain(s) for measurement of one or more microbial components and / or characteristics such as microbial cell viability; and

[0048] F. optionally, instructions for use of the kit.

[0049] The one or more stain(s) may be as hereinbefore described.

[0050] ACCORDING TO A FIFTH ASPECT OF THE INVENTION THERE IS PROVIDED a kit for screening one or more antimicrobial agents for efficacy against a microbial biofilm cultivated according to the method of the invention, the kit comprising:I. a cultivation vessel having a solid substrate;

[0051] II. agar;

[0052] III. liquid microbiological media;

[0053] IV. optionally, one or more microorganisms to be cultivated;

[0054] V. one or more stain(s) for measurement of one or more microbial components and / or characteristics such as microbial cell viability;

[0055] VI. optionally, instructions for use of the kit; and

[0056] VII. optionally, one or more antimicrobial agent to be tested.

[0057] The one or more stain(s) may be as hereinbefore described.

[0058] BRIEF DESCRIPTION OF THE DRAWINGS

[0059] The invention will now be further described, by way of example, with reference to the accompanying drawings.

[0060] Figure 1 shows a graph of fluorescence / absorbance spectra showing characteristic spectra and peak wavelengths achieved for each stain following sample pre-treatment and control-correction. Annotations indicate peak wavelengths selected for quantitative analysis between the samples.

[0061] Figure 2 shows a graph of the comparison of the fluorescence / absorbance response between suspended cells and cultivated biofilm samples, emphasizing the importance of including various controls and standards on the same plate. Dotplots to the left of the boxplots indicate the raw data (before control-correction), where grey dots indicate the response of unstained samples and dots in colour indicate the responses of stained samples. Annotations inset show the outcomes of multiple-sample comparison inferential statistics.

[0062] A - agar-layer

[0063] M - media

[0064] HC - high-density suspended cell sample

[0065] LC - low-density suspended cell sample

[0066] AB - agar-assisted biofilmUB- unassisted biofilm

[0067] * - significant increase between stained vs. unstained samples (t-test; p > 0.05 or U-test; p < 0.05). Sensitivity (in the case of samples and standards) and nonspecific staining (in the case of the controls).

[0068] t - significant difference compared to the agar control layer (Kruskal-Wallis test and Dunn’s Multiple Comparison Test, with BH correction; p < 0.05 or ANOVA (1 and 2 way) and Tukey's Honestly Significant Difference (HSD) post-hoc test; p > 0.05. (Specificity 1).

[0069] f - significant difference between the HC and LC sample (Specificity 2).

[0070] ¥ - significant difference compared to the assisted biofilm samples

[0071] 0 - significant difference compared to the unassisted biofilm samples

[0072] Figure 3 shows a graph of E. coli and B. subtilis biofilm kinetics over time using Resazurin, cultivated in the presence and absence of the agar hydrogel, compared to the low-density suspended cell sample. The boxplots depict the resorufin fluorescence response at 590 nm, normalised to the 560 nm emission baseline, which accounts for background autofluorescence. Annotations inset show the outcomes of multiple-sample comparison inferential statistics.

[0073] Levene’s test of variance, if p < 0.05 = Kruskal-Wallis and Dunn’s Multiple Comparison test. If p > 0.5 = 1 or 2-way ANOVA and Tukey HSD post-hoc test.

[0074] * - Significant difference between the agar-assisted and unassisted biofilm samples t - significant increase in cell viability from inoculation (t = 0 hours)

[0075] y - significantly lower cell viability compared to peak cell viability (t = 72 hours) for E. coli biofilms

[0076] £ - significant difference compared to peak cell viability (t = 48 hours) for B. subtilis biofilms.

[0077] Figure 4 shows a graph of the impact of Ciprofloxacin and Kanamycin on E. coli cell viability and DAPI response for agar-assisted and unassisted biofilms and the highland low-density suspended cell samples. A four-parameter logistic regression model in the form of an equation: Y = Min + (Max-Min / 1+(x / IC50)Hill coefficient) was utilised to determine the half-maximum inhibitory concentration (IC50 value). Uncertainties represent the standard error of the means.

[0078] AB - assisted biofilmUB- unassisted biofilm

[0079] HC - high-density suspended cell sample

[0080] LC - low-density suspended cell sample

[0081] Figure 5 shows a graph of the impact of ciprofloxacin and kanamycin on B. subtilis agar-assisted and unassisted biofilms and suspended cell samples. A four-parameter logistic regression model in the form of an equation: Y = Min + (Max-Min / 1 +(x / IC50)H il I coefficient) was utilised to determine the half-maximum inhibitory concentration (IC50 value). Uncertainties represent the standard error of the means.

[0082] N / A- IC50 value removed consequent of substantial outliers / variance in data

[0083] AB - assisted biofilm

[0084] UB- unassisted biofilm

[0085] HC - high-density suspended cell sample

[0086] LC - low-density suspended cell sample

[0087] Figure 6 shows a graph of the fluorescent spectra of the main components of the biofilm cultivation system (agar and media) using selected stains. Spectra represent the average ± standard error of n = 5 independent measurements. Samples were selected based on their reported propensity for autofluorescence and I or non-specific staining, highlighting the need for sample pre-treatment. The unstained controls (-) appear as dashed lines, while the stained samples (+) appear as solid lines.

[0088] Figure 7 shows a graph of resazurin fluorescent response and MTT absorbance response over time for the high- and low-density suspended cell sample, nutrient media and agar-only samples.

[0089] * - Overflow (RFU > 90000)

[0090] t - significantly higher response compared to the low-density suspended cell sample at the same incubation time

[0091] t - significantly higher response compared to the media control at the same incubation time

[0092] || - significantly higher response compared to the PBS control at the same incubation timeFigure 8 shows a graph of exemplar fluorescence I absorbance spectra of stains utilised to quantify general biomass I total cells using A) DAPI and B) CV, cell viability using C) Resazurin and D) MTT, protein content using E) SYPRO Ruby and F) CBB and lastly, DNA concentration using G) GelRed. Spectra represent the average ± standard error of n = 5 independent measurements. Samples were selected based on their propensity for autofluorescence and I or residual absorbance. The high and low-density suspended cell samples were centrifuged and resuspended in PBS to remove LB. The unstained controls (-) appear as dashed lines, while the stained samples (+) appear as solid lines.

[0093] DETAILED DESCRIPTION

[0094] Biofilms are protective structures composed of multiple bacterial cells which secrete an extracellular matrix (ECM), enabling cells to adhere to one-another and to a substrate.

[0095] E.coli cells were selected as a model Gram-negative species reportedly widely-capable of producing biofilms and B. subtilis was selected as the model Gramnegative species.

[0096] Biofilms are dynamic structures, initiated by the adherence of dispersed bacteria to a substrate. Adherence is facilitated through a combination of: nonspecific physicochemical interactions (Van der Waals forces, electrostatic and acid-base interactions) between cell membrane components and the surface; appendage- and flagella-mediated attachment processes; and a variety of specific interactions with the substrates and host-receptors (e.g. glycoproteins).

[0097] The attached cells aggregate and begin to generate ECM materials and to proliferate through binary fission or asymmetric division, forming microcolonies that eventually aggregate to one-another via shared ECM components. Individual microcolonies are subjected to distinct microenvironmental conditions (Stoodley et al., 2009 and Stewart and Franklin 2008) during this process.Consequently, each microcolony exhibits individualised growth-phases and physiological states (van Houdt and Michiels, 2005) and may maintain distinctions from one another as the biofilms merge together to form a mature biofilm. The biofilm can develop structural features such as fluid-filled channels or pores that enable nutrient and waste transport between microcolonies. Once the biofilm has matured, cells at the periphery of the biofilm and the internal channels may disperse, allowing the biofilm participants to colonise other regions of the greater environment.

[0098] Biofilm-associated cells and dispersed cells describe two different modes of bacterial growth: biofilms tend to be more complex, organised structures; dispersed cells behave as individual, free-floating microorganisms (Mikkelsen et al., 2007). The gene expression profiles - and therefore, physiological states and behaviour - of biofilm-associated cells are often extremely different to their dispersed forms (Bihter et al., 2017, Shemesh et al., 2007, Ren et al., 2004, Beloin et al., 2003, Schembri et al., 2003, Sun et al., 2020).

[0099] The production of reproducible biofilms is complicated by the temporal (Ren et al., 2004) and spatial (Floyd et al., 2015 and Hung et al., 2013) conditions under which the biofilm is cultivated (Lopez et al., 2010). This is largely because gene expression varies significantly between each stage of biofilm growth (Holden et al., 2021 and Beitelshees et al., 2018); between individual biofilms of the same strain / species; and even between different microcolonies (Stoodley et al., 2009; Stewart and Franklin 2008 and van Houdt and Michiels, 2005) formed within the same biofilm.

[0100] To counteract the aforementioned challenges, a method of producing a biofilm has been developed, which is, in a particular embodiment, based on microtiter plates, and was validated by the Applicant for the ability to grow consistent biofilms. Agar hydrogels cast within each well served as supportive substrates for biofilm culturing.

[0101] Existing research has predominantly emphasized the rheological properties of agar hydrogels, particularly stiffness, rather than their application as a support substrate for biofilm adherence, formation and maturation. (Kolowe at al., 2015) indicated that increasing the stiffness of the agar hydrogel - by increasing the concentration from 3%to 9% (Kolowe et al., 2015), or through chemical-crosslinking (Roizman et al., 2024) -correlates with enhanced initial bacterial adherence.

[0102] In this research, a soft (1.2% w / v) agar hydrogel was utilised as its reduced stiffness was postulated to more closely resemble that of biological surfaces, compared to intermediate or stiff agar formulations, providing a more physiologically-relevant substrate for studying bacterial adhesion and biofilm formation. Additionally, the reduced stiffness of softer agar hydrogels was hoped to facilitate more efficient diffusion of nutrients and oxygen to biofilm-associated cells, supporting enhanced metabolic activity and biofilm maturation.

[0103] The 1.2% (w / v) agar hydrogel is characterised by a comparatively low Young’s modulus, indicating a softer structural profile relative to the higher-concentration formulations previously reported by (Roizman et al., 2024) of 2% w / v agarose mixed with 2% w / v gelatine, and (Kolowe et al., 2015) with a 3% and 9% w / v molecular grade agar.

[0104] The agar hydrogel cast within the wells of the biofilm cultivation system was estimated to possess a diameter and height of approximately 8 mm and 4 mm, respectively. The volume of hydrogel (~200 pl) is substantially lower than those reported by (Kolowe et al. 2015), who utilised hydrogels manually cut to a diameter of 25 mm and a height of 2 mm (~982 pl). The reduced volumes of the agar hydrogel is advantageous as it allows for in situ casting within the microtitre plates, decreasing the number of processes required to produce the hydrogel in environments where hydrogel sterility is easier to maintain.

[0105] Furthermore, the smaller format enables high-throughput experimentation through enabling compatibility with 96-well microtitre plates, rather than being limited to 6- or 24- well plates. However, the agar hydrogel remains sufficiently thick (compared to agar-layers applied to microfluidic chips) to support and optimise bacterial adhesion, and in turn, subsequent biofilm formation and maturation.

[0106] The extracellular matrix is composed largely of a combination of self-produced polysaccharides, proteins and nucleic acids (Hung et al., 2013; Flemming andWingender, 2010 and Sauer et al., 2007). However, a remarkable diversity in ECM structure and components is observed between different bacterial stains or species (Wozniak et al., 2003 and Hentzer et al., 2001). This is attributed to varied cell structure, metabolic capabilities, lifestyle and adaptations to strain / species specific environmental niches. For example, Gram-negative and -positive bacteria interact differently with their surroundings or other species as a result of differences in the peptidoglycan layer or outer membrane, impacting the ECM structure and components.

[0107] Biofilm development is conventionally monitored by quantifying cell propagation, often using metabolic dyes such as Resazurin (Dalecki et al., 2016 and Van den Driessche etal., 2014) or MTT (de Carvalho et al., 2017 and Lin etal., 2017) and / or by quantifying the production of ECM materials onto a surface.

[0108] Consequent of the inherent variability in ECM components, a variety of fluorometric and colorimetric stains were utilised when monitoring and quantifying the ECM composition of samples investigated.

[0109] Colorimetric analysis allows quantification of a substance based on a change in photonic properties / colour as a complex is formed between the substrate and colorimetric stain (Xia et al., 2010). Conversely, fluorometric analysis measures the concentration of a substance using fluorescent responses, usually by staining the substance with fluorescent stains.

[0110] Fluorometric analysis is inexpensive and easy to perform (Rousar et al., 2012) and has been routinely used to monitor in vitro biofilm growth. Due to the unique properties of the fluorescent stains utilised and the use of characteristic emission and excitation wavelengths, a higher specificity (Christopoulos & Diamandis, 1996 and Diaz Garcia & Bad i a Laino, 2005) and sensitivity (Hawker et al., 2018) is achieved with this approach, compared to colorimetric analysis (Burton et al., 2006). Consequently, detection limits are lower for fluorometric analysis (Ortega Algar, Martos and Diaz, 2003 and Persson and Wedborg, 2001) and can be used to measure concentrations down to about one thousandth (Rinnan and Andersen, 2005) of that possible throughcolorimetric detection. However, fluorometry generally requires more complex and expensive equipment, compared to colorimetry.

[0111] As a broad-spectrum biocidal agent, ciprofloxacin was selected in this study as the baseline antibiofilm compound for subsequent antimicrobial screening comparing suspended cells and biofilm-associated cells.

[0112] Ciprofloxacin is a broad-spectrum fluoroquinolone drug (Castro et al., 2013), active against a wide array of Gram-positive and -negative bacteria (Shariati et al., 2022). Ciprofloxacin inhibits DNA gyrase - an enzyme which controls the unwinding of DNA (Wallace et al., 2018) during DNA replication and repair - preventing it from resealing the DNA after breaking and unwinding it (Shariati et al., 2022). Consequently, double-stranded breaks are formed in the DNA (Li et al., 2018), affecting cellular viability. As a broad-spectrum biocidal agent, ciprofloxacin was selected as the baseline antibiofilm compound.

[0113] Antibiotic resistance to kanamycin, which targets protein synthesis rather than DNA replication, was also investigated. This enabled evaluation of how biofilm formation influences cellular tolerance to antibiotics with distinct mechanisms of action.

[0114] In summary, the Applicant has devised a method for consistent cultivation of bacterial biofilms having sufficiently high numbers of viable cells for use as a model to screen antibacterial compounds against a variety of biofilm-producing bacteria over time during cultivation. More particularly, the Applicant has devised a method of cultivation of bacterial biofilms that allows standardised biofilms to be cultivated having improved intra-batch consistency.

[0115] As used herein, the term “about” refers to a value that may reasonably vary due to experimental error, measurement imprecision, or natural biological variability. Unless otherwise specified, “about” generally encompasses variations of ±10% of the stated value, or a range appropriate to the particular context, method of measurement, or field-specific precision.The presently disclosed subject matter will now be described more fully hereinafter with reference to an accompanying Example, in which representative embodiments are exemplified. The presently disclosed subject matter can, however, be embodied in different forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the embodiments to those skilled in the art.

[0116] EXAMPLE

[0117] 1. Methodology

[0118] Reagents and Apparatus

[0119] Unless otherwise stated, all reagents were of technical grade, or higher. All aqueous solutions were prepared using double-distilled water purified using a Milli-Q® system (MilliporeSigma); water with a minimum conductivity of 18.2 MQ.cm was used.

[0120] The following buffer reagents were purchased from Merck: Sodium chloride, NaCI (> 99.5% purity); potassium chloride, KCI (99% purity); sodium phosphate dibasic, Na2HPO4 (> 98% purity); potassium dihydrogen phosphate, KH2PO4 (> 99.5% purity); 32% hydrochloric acid, HCI; and lastly sodium hydroxide, NaOH (> 98% purity).

[0121] The following reagents, purchased from Merck, were used to produce culture media: Luria Broth, LB; glycerol ( 99% purity) and bacteriological agar. E. coli cells (strain DH5a) and B. subtilis cells were obtained from in-house cultures.

[0122] Pre-sterilised, 96-well microtitre plates made of clear polystyrene with a flat bottom and lids were purchased from either Merck or Lasec (South Africa).

[0123] The stains tested are set out in Tables 1 and 2 below.

[0124] TABLE 1 : Summary of stain specificity and sensitivity

[0125]

[0126] TABLE 2: Spectrophotometric characteristics of the various stains utilised in this study

[0127] >

[0128] >

[0129]

[0130]

[0131] Footnotes:

[0132] abs - peak absorbance for colorimetric stain (if applicable)

[0133] ex - peak excitation wavelength

[0134] em - peak emission wavelength

[0135] in - minor absorbance / excitation / emission peak

[0136] -> - indicates that peak wavelengths change upon binding to its target

[0137] Fluorescent and colorimetric stains utilised to evaluate biofilms were purchased from a variety of sources. From Merck: SYPRO™ Ruby protein gel stain, DAPI (4',6-diamidino2-phenylindole; 90% purity), GelRed®, Coomassie Brilliant Blue G-250 and Resazurin sodium salt. Crystal violet was purchased from Wallace’s Pharmacy (Makhanda, South Africa) and MTT (3-(4,5-Dimethylthiazol-2-yl)-2,5-Diphenyltetrazolium Bromide) was obtained from Thermo Fisher Scientific. Standards and positive controls for the stains: bovine serum albumin, BSA (> 98% purity) and deoxyribonucleic acid from salmon sperm, dsDNA (> 95% purity) were obtained from Merck. Other standards were constructed from cell cultures. A laboratory-based Gram staining kit was used for E. coli and B. subtilis Gram staining.

[0138] Fluorometry and colorimetry were performed using a Biotech Synergy MX microplate reader (between 0 - 4 absorbance units) using Gen5 1.10 for analytical apparatus. Centrifugation was performed using the Avanti® J-E centrifuge or 1.0 R Megafuge®. Fluorescence microscopy was executed using a Thermo Scientific ™ Invitrogen™ EVOS™ FL Auto 2. Culture manipulation - and the preparation of all relevant reagents- was conducted in a laminar flow hood (VividAir 4004V UD flow bench), previously sterilised using 70% ethanol, to avoid microbial contamination. Gram staining was visualised using an UOP biological microscope UB202i (Lasec, South Africa). The pH of buffer solutions was adjusted using a Laboratory pH / mV meter pH 8 Pro DHS (ACCSEN).

[0139] Preparation of the agar hydrogel

[0140] The 1.2% agar bacteriological solution was prepared by dissolving 1.2 g of agar bacteriological in 100 nnL of deionised water. This solution was then autoclaved, applying peak temperatures at 121 °C for 20 minutes.

[0141] The agar hydrogel was prepared by pipetting 100 pL of the 1.2% agar bacteriological solution into each well of a 96-well microtiter plate using a multichannel pipette. The agar was left to solidify at room temperature (21 ± 2°C) in the laminar flood hood overnight.

[0142] The agar hydrogel cast within the wells of the biofilm cultivation system was estimated to possess a diameter and height of approximately 8 mm and 4 mm, respectively.

[0143] Initial staining studies

[0144] Fluorometric and colorimetric assays were initially performed on exponential- growth phase cells using 96-well microtiter plates to allow optimisation of the biofilm cultivation system. Optimisation of aspects such as the concentration of the stain I controls, incubation time for each stain, excitation and emission wavelengths used and whether sample pre-treatment was required were investigated.

[0145] The high-density suspended cell sample was prepared by inoculating 10 mL of Luria broth / Nutrient broth no. 5 with 30 pL of E. coli / B. subtilis glycerol stock culture. The sample was incubated for 24 hours I until a change in turbidity of 0.8 was reached; dilution with sterile media was performed to ensure a turbidity of precisely 0.8 was achieved throughout the research.The low-density cell suspended sample (i.e. OD600 nm of 0.08) was prepared from the above, by mixing 1 mL of the high-density suspended cell sample with 9 mL of sterile media. Both the high and low-density suspended cell samples were centrifuged and resuspended in PBS, to eliminate LB autofluorescence and non-specific staining. Depending on the stain under investigation, several control samples were included on the plate. Generally, controls consisted of: PBS; uninoculated media; high- density suspended cells; low-density suspended cells. In addition, stain-specific standards were included: 1 mg / mL BSA solution (for SR I CBB staining); 2 pg / mL dsDNA (GelRed® / DAPI); high and low-density suspended cell samples (CV / MTT / RES).

[0146] A sample size of n = 5 was used for all samples to ensure that the results obtained are reliable and not consequent of errors in methodology I equipment used. Unstained controls were included for each sample to account for autofluorescence I non-specific staining. All samples, standards and controls were included on a single plate so that they could be accurately compared against the standard.

[0147] Each stain was added directly to the various samples and standards and then fluorimetric I colorimetric analysis could be performed. The stains were then measured at stain-specific emission and excitation wavelengths as detailed in Table 2. To capture colorimetric data, all samples were analysed by absorbance spectrophotometry after spectrofluorimetry, generating absorbance spectra of the samples between 300 and 700 nm in 10 nm increments.

[0148] Monitoring biofilm growth kinetics using selected stains

[0149] Biofilm formation and growth kinetics were monitored overtime using separate biofilms for each time interval. Spectrofluorometry followed by spectrophotometry was performed at 12, 24, 48, 72, 96 and 120 hours for individual stains, all samples were of size n = 5.

[0150] Into each well of a sterile 96-well microtiter plate (excluding wells utilised for the control biofilms), 100 pL of bacteriological agar was added using a multichannel pipette. The agar was left to solidify overnight at approximately 21 ± 2 °C in the laminar flow hood.The wells designated for the agar-assisted biofilm sample were supplemented with 230 pL of sterile liquid media and were individually inoculated using 20 pL of glycerol stock culture. The plate was incubated at 37°C on a shaker at 100 rpm.

[0151] Control biofilms (Unassisted biofilms) cultured directly onto the polystyrene surfaces of the plate’s wells were included to determine if the agar hydrogel supported biofilm development. Control biofilm samples were supplemented with 330 pl liquid media and inoculated with 20 pl glycerol stock culture.

[0152] A high and low-density suspended cell sample (prepared as described above), stainspecific standard and uninoculated media sample were also included on the plate. The uninoculated media sample was added to evaluate the non-specific binding I autofluorescence response of the specific media utilised and functioned as the negative control. Unstained controls of each sample, treated in the same manner as the relevant sample, were also included to account for autofluorescence / non-specific staining.

[0153] Unassisted biofilm samples grown onto the plastic substrate of the microwell were included in the plate as controls, both to account for autofluorescence I non- specific staining of the agar substrate and to determine whether the agar substrate successfully encouraged biofilm growth.

[0154] Every 24 hours, 80% of the total liquid well volume was removed from wells not being tested within 24 hours and was replaced with fresh media to prevent nutrient depletion. Wells not in use were filled with deionised water to prevent desiccation of the agar layer. Plates were closed with a sterile lid and placed in a large plastic bag containing a petri dish containing 10 mL of deionised water. The inside of the plastic bag was also lightly sprayed with deionised water. This increased the humidity, preventing desiccation of the agar / samples (Vogt and Harris, 1974).

[0155] Staining of biofilm samples

[0156] For quantification of biofilm-associated cell population using DAPI, 20 pL of DAPI (10 pg / mL) was added directly to the biofilm sample and left to incubate for 15 minutes at21 ± 2 °C. The plate was wrapped in tinfoil to prevent exposure to light. After this time, media and loosely-attached cells were removed from the wells and the biofilm was rinsed three times by the addition of PBS. PBS was added to a total well volume of 350 pL (including the agar layer) and spectrofluorometry was performed.

[0157] For the remaining stains, loosely-attached cells were discarded and the biofilms were rinsed three times by the addition of PBS. GR and SR (50 pL) were added and spectrofluorometry measured immediately.

[0158] For RES and MTT, spectrofluorometry I spectrophotometry was only performed after 90 and 60 minutes respectively. This was done to allow the biofilm-associated cells to react with the stains, converting it into a fluorescent product.

[0159] After the addition of CV or CBB, the plate was wrapped in tinfoil and incubated at 37°C on a shaker at 100 rpm. After 15 minutes, the biofilms were washed thrice by the addition of PBS. This was necessary as CBB binds with high affinity to the agar substrate, causing background autofluorescence. PBS was added to a total volume of 350 pL (including the agar layer).

[0160] For staining with resazurin, SR, DAPI and GR, it was not necessary to wash the wells. After staining with MTT, CV and CBB, the wells were washed thrice by the addition of PBS. PBS was then added to a total volume of 350 pl and spectrofluorometry I spectrophotometry was performed.

[0161] The suspended cell samples, stain-specific standard and media control were added to the plate immediately before staining of biofilm samples to avoid contamination. Staining was then performed as initially described, without additional post-staining rinsing.

[0162] Data processing and statistical treatment of data

[0163] All samples consist of (n = 5) independent measurements; unless specifically mentioned, the samples in each dataset are measured within the same microtiter plates at the same time.Exemplar fluorescence and absorbance spectra are presented as the mean ± standard error of replicates in spectra graphs. The majority of extracted absorbances and wavelength-specific fluorescent response (measured in RFUs) are control-corrected and / or normalised as percentages of their standards measured together on the same plate and the same Inferential statistics analysis was performed using BlueSky Statistics. Firstly, Levene’s test for equality of variances was performed; if insignificant (p > 0.05), then a means test using 1 or 2-way ANOVA (3 or more samples), followed by Tukey’s Honest Significant Different (HSD) post-hoc test and finally, a Student’s t-test (for two- sample comparisons) were conducted. Alternatively, if Levene’s tests’ outcome was p < 0.05, a Kruskal-Wallis one-way ANOVA was performed. This was followed by Dunn’s multiple comparison post-hoc test and the Mann-Whitney U Test. Unless otherwise indicated, significance, a, was set to 0.05 for all analyses. For IC50 determination, a four-parameter logistic regression model in the form of an equation: Y = Min + (Max- Min / 1+(x / IC50) Hill coefficient) was utilised to determine the half-maximum inhibitory concentration (IC50 value).

[0164] 2. Results

[0165] Initial staining studies reinforce the necessities of sample pre-treatment and for the inclusion of various standards, controls and unstained samples on each assay plate

[0166] The biofilm cultivation system employed a 96-well microtiter plate, with an agar substrate, that enabled high-throughput screening while ensuring adequate lateral space for biofilm growth and development - a requirement that could not be met with a 384-well microtiter plate or higher.

[0167] The ability of each selected stain to accurately detect the presence of biomass or specific extracellular matrix components without the need for sample pre-treatment (removal of media components and / or destaining of the sample after exposure) was assessed. When considering high-throughput screening, minimal sample pretreatment is desirable, in order to decrease the complexity and measurement variability of the assay.The various stains were measured at the specific excitation and emission wavelengths specific to that individual stain (Table 1). For example, resazurin studies were measured at an excitation of 500, emission start of 560 and emission stop of 660. Optical sensitivity was set at 100 initially and decreased by 10 if ‘overflow’ readings were observed. Bandpass widths were set at 9 and 20 nm for excitation and emission wavelengths, respectively.

[0168] All stains exhibited considerable autofluorescence or non-specific staining of the nutrient media and / or agar-only samples at the wavelengths used for fluorometric detection - presented in (Figure 6) - highlighting the need for sample pre-treatment and inclusion of appropriate standards and controls.

[0169] It was imperative to include the various standards / controls on a single plate to control instrument variation, path length consistency and plate edge effects, thereby, ensuring experimental consistency, reliable comparisons (Lilyanna et al., 2018) and promoting the formation of reproducible biofilms.

[0170] An agar-only and nutrient media sample were included in further studies to evaluate the autofluorescence / residual absorbance and / or non-specific binding response of agar and / or media - functioning as negative controls. For specific stains, including CV, CBB, and MTT; it was necessary to rinse the agar substrate / biofilm samples with PBS post-staining to remove excess stain.

[0171] Two separate suspended cell samples - one consisting of a high density of cells; the other of a low density of cells - were compared to the stain-specific standard in order to determine the stain’s ability to stain biomass in general and distinguish a difference between varied cell densities.

[0172] Pre-treatment of the suspended cell samples was necessary to account for the autofluorescence / non-specific staining of the nutrient media. This included centrifugation and resuspension of the suspended cell samples in PBS (as described previously in Methodology) to remove the nutrient media.For resazurin and MTT, spectrofluorometry / spectrophotometry was only performed after 90 and 60 minutes, respectively, to allow sufficient time for the biofilm-associated cells and stain to react. This is presented in (Figure 7).

[0173] To further eliminate the effects of autofluorescence / residual absorbance, spectra (Figure 1) are presented below as control-corrected i.e. subtracting the average of the unstained controls from the samples’ responses.

[0174] Following control-correction, the expected DAPI fluorescence peak at 447 nm was observed in both the suspended cell sample and the stain-specific standard, indicating successful staining of biomass within these samples (Figure 1 A). In addition, a shoulder initially detected at approximately 410 nm (Figure 8) is no longer present following control-correction. A significant fluorescence response is observed for the agar hydrogel and, to a smaller extent, the nutrient media sample.

[0175] For CV, the peak at 580 nm, consistent with the commonly reported peak at ~ 595 nm (Stiefel et al., 2016), is observed to become more prominent following controlcorrection (Figure 1 B). Additionally, an initial broad peak observed at 540 nm (Figure 8) is less evident after control-correction. A significant absorbance response is still observed for the agar hydrogel, and to a lower extent for the nutrient media sample, following control-correction.

[0176] For resazurin, it is evident that the peak at - 640 nm is more prominent following control-correction (Figure 1 C). The peak at 590 nm was taken to be indicative of cell viability. Minimal fluorescence of the agar hydrogel and nutrient media samples is observed.

[0177] For MTT, a broad absorbance plateau between 500 and 700 nm was observed, with three absorbance peaks visible after control-correction (Figure 1 D) at approximately 400 nm, 570 nm (Stockert et al., 2017) and 620 nm for the suspended cell sample. These peaks are more obvious after subtraction of the unstained controls. The absorbance response at 620 nm was used as an indicator of cell viability.For SYPRO Ruby, the expected spectra and peak at ~ 620 nm (Berggren et al., 2002) is observed for the stain-specific standard after control-correction (Figure 1 E). However, a prominent peak at 620 nm was not observed for the suspended cell samples. A slight shoulder between 500 and 550 nm was also observed, which was also present in the unstained suspended cell sample (Figure 8 E).

[0178] For CBB, a single peak is observed at ~ 600 nm for both the suspended cell sample and stain-specific standard (Figure 1 F), consistent with previous reports of CBB peak absorbance maxima (Grintzalis et al., 2015). A significant absorbance response is observed for the agar hydrogel.

[0179] After control-correction, the expected peak was observed at ~ 600 nm (Sayas et al., 2015) for the stain-specific standard when stained with GelRed (Figure 1 G). A prominent peak was not observed for the suspended cell sample. For the uninoculated media sample, the peak was observed to shift to ~ 620 nm.

[0180] After subtraction of the unstained controls, the data was further standardized by normalising responses against the high-density suspended cell sample. This was done by normalising the average of the high-density suspended cell sample at the peak wavelength specific to each stain and multiplying by 100 (Figure 2).

[0181] Standardised boxplots of suspended cell and cultivated biofilm studies

[0182] To ensure experimental consistency, reliable comparisons and to promote the formation of consistent biofilms, it was necessary to include the high- and low-density suspended cell samples, nutrient media and agar-only controls on a single plate to avoid variation caused by instrument variation, path length inconsistencies and plate edge effects (Lilyanna et al., 2018).

[0183] In further experiments, all samples were control-corrected (subtracted the average of the unstained controls from the samples’ responses) and standardised against the low-density suspended cell sample (normalised the average of the low-density suspended cell control at the stain-specific peak wavelength and multiplied by 100).Standardised boxplots comparing the suspended cell studies and cultivated biofilm studies are presented in Figure 2.

[0184] From (Figure 2 A, Study 1), it can be observed that the high-density suspended cell sample has a greater (but not significant) fluorescence response and DAPI response as compared to the low-density suspended cell sample. However, a significant difference is not observed between the suspended cell samples, stain-specific standard and negative controls (agar hydrogel and nutrient media). Significant nonspecific staining of the agar hydrogel (* annotation), and autofluorescence of the nutrient media sample, is observed.

[0185] As evident from (Figure 2 A, Study 2), the assisted biofilm sample shows a greater (but not statistically significant) fluorescence response and general biomass / total cells compared to the unassisted biofilm sample. A significant difference in DAPI response is observed between the high- and low-density suspended cell samples (J annotation). However, the high-density suspended cell sample is observed to have a significantly higher DAPI response compared to the assisted biofilm sample (0 annotation).

[0186] From (Figure 2 B, Study 2), a higher (but not significant) absorbance response and general biomass is observed for the assisted biofilm sample, as compared to the unassisted biofilm. However, the high-density suspended cell sample has a lower absorbance response (but not significant) than the low-density suspended cell sample. A similar trend is observed for the suspended cell samples in (Study 1).

[0187] A significant difference between the stained and unstained samples (* annotations) is observed for the low-density suspended cell samples (both Study 1 and 2), and the unassisted and assisted biofilm samples (Study 2). Significant non-specific staining is observed for the agar-only and nutrient media samples (* annotations). Significant residual absorbance of the unstained high-density suspended cell samples is observed at wavelengths utilised for CV-staining in both (Study 1 and 2.)

[0188] Although, CV is commonly used for quantification of general biomass / total cells; it presents challenges when applied to biofilms. The uptake of CV by cells varies depending on the bacterial growth phase (Latka & Drulis-Kawa, 2020), which isproblematic as biofilm-associated cells typically exhibit stationary-phase phenotypes58. Additionally, CV can form complexes (Latka & Drulis-Kawa, 2020) with negatively charged antibiotics (e.g., penicillin), potentially interfering with subsequent antibiotic testing.

[0189] As a result of the aforementioned challenges with CV staining, DAPI was utilised to quantify general biomass / total cells in further experiments. DAPI is advantageous as it can be multicomplexed (Magana et al., 2018) with several other stains of interest which is time efficient and reduces the amount of resources used.

[0190] Evident from (Figure 2 C Study 2), it can be observed that the assisted biofilm sample has a higher (but not significant) fluorescence response and cell viability as compared to the unassisted biofilm sample. This is supported by the significantly higher cell viability observed for the assisted biofilm compared to the low-density suspended cell sample (¥ annotations).

[0191] Resazurin-staining is capable of sensitively distinguishing a difference in biomass between the stained and unstained low-density suspended cell sample (Study 1), high-density suspended cell sample, assisted and unassisted biofilm samples (* annotations).

[0192] A significant difference in fluorescence response and cell viability can be observed between the high- and low-density suspended cell samples for both (Study 1 and 2, Figure 2 C).

[0193] For (Study 2), a significant difference between the high-density suspended cell sample and the negative controls (agar hydrogel and nutrient media) can also be observed. Significant non-specific staining of the agar hydrogel and nutrient media sample is observed (* annotations).

[0194] As evident from (Figure 2 D, Study 2), a greater absorbance response (but nonsignificant) and cell viability is observed for the assisted biofilm sample, as compared to the unassisted biofilm sample. However, the high-density suspended cell sample had greater cell viability compared to the assisted biofilm sample ( annotation).An increased absorbance response was observed for the high-density suspended cell sample, as compared to the low-density suspended cell sample for both (Study 1 (significant) and 2 (non-significant)). From (Study 1), it was observed that the high-density suspended cell sample had a greater fluorescence response and cell viability as compared to the negative controls (agar layer and nutrient media samples; t and m annotations). Significant residual absorbance of the unstained high-density suspended cell sample was also evident at wavelengths utilised for MTT-staining in both (Study 1 and 2; * annotations).

[0195] As evident from (Figure 2 E Study 2), a significantly higher protein concentration is observed for the unassisted biofilm and high- and low-density suspended cell samples as compared to the assisted biofilm sample (0 annotations). Significant autofluorescence of most of the unstained samples is observed, resulting in an overall negative fluorescence response following control-correction and standardization of the data.

[0196] The high-density suspended cell sample has a greater (but not significant) protein content compared to the low-density suspended cell sample for both Study 1 and 2. Autofluorescence of the negative controls (agar hydrogel and nutrient media samples) was also observed in (Study 1).

[0197] Similarly, for CBB (Figure 2 F, Study 2), a much lower protein concentration is observed for the assisted biofilm sample, compared to the unassisted biofilm sample and the high- and low-density suspended cell samples. Significant residual absorbance is also observed for the assisted biofilm sample and the low-density suspended cell sample, resulting in an overall negative absorbance response following control correction and standardization of the data. An increased protein content (but not-significant) is observed for the high-density suspended cell sample compared to the low-density suspended cell sample for both (Study 1 and 2). Significant nonspecific staining of the agar hydrogel is still evident after rinsing of excess stain (* annotation).SYPRO Ruby is reportedly accurate, simple and easy to use and has greater sensitivity compared to CBB (Anaya et al., 2007). As a result of the reported increased sensitivity and significant non-specific staining of the agar hydrogel by CBB, SR was utilised for further experiments.

[0198] From (Figure 2 G, Study 2), it can be observed that the assisted biofilm has a lower fluorescence response and eDNA concentration as compared to the unassisted biofilms sample. The low-density suspended cell sample is also observed to have a greater eDNA concentration as compared to the high-density suspended cell sample in both (Study 1 and 2). Significant non-specific staining and autofluorescence of the nutrient media sample is also observed.

[0199] Monitoring biofilm development using selected stains: E. coli and B. subtilis

[0200] Existing methods of culturing biofilms largely fail to cultivate consistent biofilms in a manner compatible with technologies used routinely for high-throughput screening. Having identified the stains to be used for subsequent analyses (Figure 2), biofilm growth and development was monitored over time to determine if the biofilm cultivation system encouraged the growth of consistent biofilms and to study the influence of the agar layer on overall biofilm growth.

[0201] The biofilm kinetics presented in (Figure 3) depicts the resorufin fluorescence response at 590 nm, normalised to the 560 nm emission baseline, which accounts for background autofluorescence.

[0202] Multiple independent replicates (n = 5) of the normalised suspended cell sample were included on each plate to normalise the fluorescent readings between timeframes. This was necessary as significant variation in fluorescent signal is observed when using multiple plates.

[0203] The use of the normalised suspended cell sample ensured that the capability of the biofilm cultivation system to produce consistent biofilms was compared between the same batch of cells. The extent of biofilm growth is therefore also normalised to their distinctive metabolic properties when cultivating in the same growth media.As evident from (Figure 3 A), the E. coli agar-assisted biofilm samples exhibited greater, though not statistically significant, cell viability compared to the unassisted biofilms samples. For the assisted-biofilm, cell viability increased over time, reaching a peak at 72 hours and declining thereafter. Significantly lower cell viability was observed at t = 24, 48 and 120 hours, compared to the peak cell viability observed at 72 hours (y annotations, Figure 3 A). In comparison, the unassisted biofilms maintained relatively stable cell viability over the time tested.

[0204] A statistically significant increase in cell viability was observed from inoculation (t = 0 hours) at t = 48, 72 and 96 (f annotations) for the assisted-biofilm sample. In contrast, a statistically significant increase in cell viability from inoculation was not observed for the unassisted biofilm sample.

[0205] From (Figure 3 B), the agar-assisted biofilm samples exhibited higher cell viability at 0, 24, 72 and 120 hours. The viability of the B. subtilis assisted biofilms remained relatively constant across the tested time points, with a reduction observed at 96 hours, followed by a subsequent increase reaching peak viability at 120 hours.

[0206] Conversely, viability of the unassisted biofilm samples fluctuates significantly with peaks observed at 48 and 96 hours. A significant reduction in cell viability was observed at 0, 24, 72, 96 and 120 hours compared to the peak at 48 hours (J annotations).

[0207] As evident from the results, the presence of the soft agar hydrogel resulted in markedly higher cell viability, enhanced biofilm formation and maturation for both E. coli and B. subtilis. This was achieved using a soft, 1.2% agar hydrogel, without the addition of additional components or chemical cross-linking.

[0208] Antimicrobial susceptibility of E. coli suspended cell and cultivated biofilms using Resazurin and DAPI assays.The impact of ciprofloxacin and kanamycin on the viability and DAPI responses of assisted and unassisted E. coli biofilms and the high and low-density suspended cell samples was assessed using resazurin and DAPI, and is presented in (Figure 4).

[0209] For DAPI responses, the fluorescence response was normalised against the average of the low-density suspended cell sample. This was performed using the suspended cell sample, rather than the stain-specific standard (dsDNA), to reduce costs, experimental complexity and to ensure sufficient space was available on the microtitre plate.

[0210] Conversely, cell viability was assessed using the resorufin fluorescence response at 590 nm, normalised to the 560 nm emission baseline to correct for background signal. Although the resazurin-based antimicrobial screening data could be normalised to the low-density suspended cell sample, as done for DAPI, normalisation using the emission baseline yielded more consistent and reliable model outputs.

[0211] As evident from (Figure 4 A), a decline in cell viability became apparent at ciprofloxacin concentrations above 61 pg / mL for both agar-assisted and unassisted biofilm samples. Notably, the agar-assisted biofilm exhibited a significantly higher halfmaximum inhibitory concentration (IC50) - 138.64 ± 24.16 pg / mL - compared to the unassisted biofilm - 88.45 ± 18.33 pg / mL.

[0212] Comparable IC50values were observed for the high-density suspended cell sample (Figure 4 B) and the unassisted biofilm (84.41 ± 3.76 pg / mL and 88.45 ± 18.33 pg / mL, respectively). In contrast, the low-density suspended cell sample exhibited a substantially lower IC50value of 33.17 ± 3.62 pg / mL (Figure 4 B).

[0213] As evident from (Figure 4 C), cell viability was observed to decline at kanamycin concentrations above 3.13 pg / mL for both the assisted and unassisted biofilm samples. The assisted biofilm exhibited a higher, but not statistically significant, IC50value (3.81 ± 0.58) compared to the unassisted biofilm (3.78 ± 0.23 pg / mL). However, a significantly higher IC50value is observed for the suspended cell samples, compared to the biofilm samples, with values of 14.4 ± 20.2 and 11.8 ± 3.79 pg / mL (Figure 4 D).From (Figure 4 E), it can be observed that DAPI responses declines at ciprofloxacin concentrations above 61 pg / mL for both the agar-assisted and unassisted biofilm samples. However, a greater DAPI response and IC50value is observed for the assisted biofilm sample, as compared to the unassisted biofilm sample, with values of 81.53 ± 25.62 and 70.16 ± 15.20 pg / mL, respectively.

[0214] As evident from (Figure 4 F), a significant increase in DAPI response is observed at ciprofloxacin concentrations above 30.5 pg / mL for the high and low-density suspended cell samples.

[0215] As evident from (Figure 4 G), the DAPI response observed for the agar-assisted and unassisted biofilms remained largely consistent as the kanamycin concentration increased, resulting in IC50values at the maximum concentration (100 ± 30 and 100 ± 12 pg / mL for the assisted and unassisted biofilm samples, respectively).

[0216] A similar trend is observed for the high and low-density suspended cell samples, with a decline in DAPI response is only observed at the higher kanamycin concentrations (Figure 4 H). This is supported by the IC50values observed for the high and low-density suspended cell samples, with values of 94.75 ± 93.01 and 91.97 ± 51.00 pg / mL, respectively.

[0217] From these results, it is evident that the soft agar hydrogel resulted in increased cell viability, enhanced biofilm formation, and significantly higher antibiotic tolerance for E. coli cells.

[0218] Antimicrobial susceptibility of B. subtilis suspended cell and cultivated biofilms: Resazurin and DAPI assays.

[0219] The impact of ciprofloxacin and kanamycin on the viability and DAPI responses of agar-assisted and non-assisted B. subtilis biofilms and suspended cells was assessed using resazurin and DAPI, and is presented in (Figure 5).As evident from (Figure 5 A), a significant decline in cell viability was observed at ciprofloxacin concentrations exceeding 122 pg / mL for both the B. subtilis assisted and unassisted biofilm samples. However, a higher IC50value and ciprofloxacin tolerance was observed for the assisted biofilms (170.3 ± 90.92), compared to the unassisted biofilms (144.88 ± 15.99 pg / mL).

[0220] In comparison, the B. subtilis suspended cell samples have a lower tolerance to ciprofloxacin, with a significant decline in cell viability observed at concentrations above 61 pg / mL (Figure 5 B). This is supported by the significantly lower IC50values obtained for the high and low-density suspended cell samples with values of 111.26 ± 25.75 and 98.21 ± 46.41 pg / mL, respectively.

[0221] From (Figure 5 C), a significant reduction in cell viability was observed at kanamycin concentrations above 3.13 pg / mL for the agar-assisted biofilm sample. In contrast, cell viability of the unassisted biofilm sample remained largely consistent over the tested kanamycin concentrations. However, a greater IC50value of 7.49 ± 2.70 pg / mL was observed for the assisted biofilm sample, compared to the unassisted biofilm sample, with an IC50value of 5.10 ± 1.00 pg / mL.

[0222] In comparison, a lower tolerance to kanamycin was observed for the suspended cell samples, with a significant decline in cell viability evident from concentrations above 0.78 pg / mL for both the high and low-density suspended cell sample (Figure 5 D). This is supported by the lower IC50values observed for the high and low-density suspended cell samples with values of 1.1 ± 0.1 and 0.72 ± 0.04 pg / mL, respectively.

[0223] As evident from (Figure 5 E), DAPI responses were observed to remain relatively constant until biofilms were exposed to ciprofloxacin concentrations above 122 pg / mL, at which point a significant increase in response was observed. A significantly higher tolerance to ciprofloxacin was observed for the agar-assisted biofilm, compared to the unassisted biofilm, with IC50values of 240 ± 183.4 and 156.8 ± 103.1 pg / mL, respectively.A similar trend was observed for the high and low-density suspended cell samples with an increase in DAPI responses observed at ciprofloxacin concentrations above 122 pg / mL (Figure 5 F). A significantly lower ciprofloxacin tolerance was observed for the high and low-density suspended cell samples, compared to the assisted biofilm sample, with IC50values of 132.9 ± 51.56 and 159 ± 51.15 pg / mL.

[0224] Conversely, for the unassisted biofilm sample, the DAPI responses remained relatively stable across the tested ciprofloxacin concentrations until 244 pg / mL, at which point a reduction in response was observed.

[0225] From (Figure 5 G), a slight decline in DAPI responses were observed at kanamycin concentrations above 122 pg / mL for the agar-assisted biofilm sample. A similar trend is observed for the high and low-density suspended cell samples (Figure 5 H).

[0226] Conversely, DAPI responses remained stable over the tested kanamycin concentrations for the unassisted biofilm sample, with no evident decline in DAPI response. The IC50 values for Figure G and H were not removed (indicated as N / A) owing to substantial outliers and excessive variance which prevented reliable IC50 estimates.

[0227] These results largely supported that the presence of the soft agar hydrogel enhanced bacterial adhesion, in turn increasing cell viability and biofilm formation and maturation. This was achieved using a soft agar hydrogel (1.2%) without the need for cross-linking or addition of other components like gelatine.

[0228] 3. Discussion

[0229] Unstained controls of all samples and standards were included on a single plate to account for autofluorescence / residual absorbance and / or non-specific staining. The unstained controls were subtracted from the corresponding stained samples’ responses. Following control-correction (Figure 1), the expected spectra and peak wavelengths were achieved for each of the stains.Following control-correction, the prominent peak at 447 nm confirms that DAPI effectively bound to biomass components in the suspended cell sample and stainspecific standard (Figure 1 A), in line with previous reports (Saby et aL, 1997). The absence of the previously observed shoulder at 410 nm (Figure 8) suggests that autofluorescence / non-specific staining of common bacterial components - Flavins (Mihalcescu et aL, 2015), NAD(P)H (Bao et aL, 2008), aromatic amino acids and some proteins (Mullerova et aL, 2022) - which are excited at similar wavelengths used for DAPI staining (450 nm; Lakowicz, 1983), was eliminated through proper controlcorrection. DAPI was shown to interact with components of the agar substrate and nutrient media samples, resulting in an overestimation of general biomass. However, through the inclusion of appropriate controls, control-correction and standardization, this background signal can be mitigated.

[0230] The peak at 580 nm, similar to the commonly reported peak at 595 nm (Stiefel et aL, 2016), was observed to become more prominent after control-correction (Figure 1 B), supporting effective binding of CV to biomass. The reduction of the broad peak initially observed 540 nm suggests removal of residual absorbance / non-specific staining associated with eDNAor polysaccharides (Li et aL, 2003). This indicates that controlcorrection successfully minimised background interference, allowing for more accurate measurements. However, CV was shown to interact with components of the agar hydrogel and nutrient media, causing significant background signal.

[0231] The presence of the peak at 590 nm (Figure 1 C) confirms the reduction of resazurin into fluorescent resorufin (Chen et aL, 2017) and was taken to be indicative of cell viability. The enhanced peak at 640 nm supports that control-correction improved detection of unreacted resazurin (Zrimsek et aL, 2004), providing a clear distinction between background and signal.

[0232] The broad absorbance plateau and distinct peaks confirm the reduction of MTT to formazan (Benov, 2021; Figure 1 D). The peak at 570 nm represents formazan formation, while the peak at 620 nm likely results from formazan aggregation caused by the lack of DMSO solubilisation (Septisetyani et aL, 2014), which was purposefully excluded to allow subsequent fluorescence microscopy of the same samples. The persistence of the - 400 nm peak suggests incomplete reduction and the presence ofresidual MTT (Mao et al., 2013). Given that the 620 nm absorbance response consistently exceeded that at 570 nm, this wavelength was used as an indicator of cell viability.

[0233] The presence of the peak at 620 nm for the stain-specific standard (Figure 1 E) confirms the stain’s expected spectral response (Berggren et al., 2002). The absence of this peak for the suspended cell sample was expected as SYPRO Ruby’s effectiveness in staining intracellular proteins in intact bacterial cells has not been demonstrated in previous studies. Rather, SYPRO Ruby’s applications have been largely limited to staining of surface proteins (Anaya et al., 2007) or proteins immobilised on membranes and gels (Demina et al., 2009). The slight shoulder between 500 and 550 nm may indicate minimal staining of extracellular proteins. Consequent of the significant autofluorescence observed for the unstained samples (Figure 8 E), it was anticipated that negative fluorescence responses would be achieved following control-correction.

[0234] For CBB, the peak observed at 600 nm for both the suspended cell sample and stainspecific standard (Figure 1 F) confirms successful binding of CBB and biomass components and supports the established CBB absorbance spectrum characteristics (Grintzalis et al., 2015). CBB interacts with components of the agar hydrogel, requiring further data standardization.

[0235] After control-correction, the detection of the expected peak at 600 nm for the stainspecific standard (Figure 1 G) confirms that GR was performing as expected (Sayas et al., 2015). The shift to 620 nm for the uninoculated media sample is attributed to distinct turbidity of each sample, which inhibits light passing directly through the sample to reach the detector. The absence of a clear peak in the suspended cell sample is attributed to the inability of GR to penetrate the bacterial cell membrane (Doroshenko et al., 2014) and stain intracellular DNA. This result highlights the stains suitability for detection of extracellular DNA rather than intracellular DNA.

[0236] After subtraction of the unstained controls, the data was further standardized by normalising responses against the high-density suspended cell sample, representing cells in stationary-growth phase. This was done by normalising the average of thehigh-density suspended cell sample at the peak wavelength specific to each stain and multiplying by 100 (Figure 2).

[0237] Using DAPI-staining, the agar hydrogel was observed to successfully encourage biofilm formation and growth, resulting in an increased response (Figure 2 A, Study 2). This is attributed to the presence of actively metabolising cells which produce increased amounts of intracellular and extracellular DNA which can then be stained.

[0238] The greater eDNA concentration observed for the high-density suspended cell sample (Figure 2 A, Study 2, annotation) is attributed to the presence of largely stationary-phase / dying cells which release intracellular DNA through cell death and lysis of internal components. This is supported by the observed difference in DNA concentration between the low- and high-density suspended cell samples samples which are largely composed of log- and stationary-phase cells, respectively for both Study 1 (not-significant) and Study 2 (significant; J annotation).

[0239] Significant non-specific staining of the agar hydrogel and nutrient media was observed, showing that DAPI interacts with components of the agar and nutrient media (Study 1). Autofluorescence of components within the nutrient media (LB broth) is observed at wavelengths utilised for DAPI-staining.

[0240] The presence of the agar-layer was observed to encourage biofilm formation and growth, resulting in increased general biomass / total cells using CV-staining (Figure 2 B, Study 2). The greater biomass of the low-density suspended cell sample observed in both (Study 1 and 2), is attributed to CV relying on a peak detection close to 600 nm - the wavelength typically employed for monitoring turbidity (Hecht et al., 2016). Crystal violet has been shown to interact with components within the agar substrate and nutrient media samples, resulting in an overestimation of signal.

[0241] The presence of the soft agar hydrogel successfully encouraged biofilm growth and formation, shown by the increased cell viability of the assisted biofilm sample (Figure 2 C, Study 2) compared to the unassisted biofilm sample and low-density suspended cell sample ( annotation).Resazurin-staining is shown to be a sensitive stain capable of distinguishing a difference in cell densities between the suspended cell samples (t annotations, Study 1 and 2) and between the negative controls (f and m annotations, Study 1).

[0242] The presence of the agar hydrogel was shown to encourage biofilm formation and growth and consequently, cell viability (Figure 2 D, Study 2). MTT-staining was also capable of sensitively distinguishing a difference in cell densities between the highland low-density suspended cell samples in both Study 1 (significant; t annotation) and Study 2 (not-significant), supporting the observed differences between the assisted and unassisted biofilm samples.

[0243] For (Study 1), MTT-staining was also capable of sensitively distinguishing a difference in cell density between the high-density suspended cell sample and the negative controls (t and m annotations). Significant residual absorbance of the unstained high-density suspended cell sample was also evident at wavelengths utilised for MTT-staining in both (Study 1 and 2; * annotations) - close to 600 nm which is the wavelength typically employed for monitoring cell turbidity.

[0244] Present research largely supports that resazurin has greater sensitivity than MTT and is a fast and simple method of measuring cell viability (Martin et al., 2005; Riss et al., 2004). In comparison, MTT requires solubilisation of formazan crystals before cell viability can be reliably assessed (Martin at al., 2005), which is labour intensive and time consuming. Consequently, it was determined that resazurin would be used for further cell viability measurements.

[0245] A low protein content was observed for the assisted biofilm sample, showing that the agar layer successfully encouraged biofilm formation and growth, resulting in increased cell viability and consequently, minimal release of intracellular proteins through cell death and lysis (Figure 2 E Study 2). A greater protein content was observed for the high-density suspended cell sample, comapred to the low-density suspended cell sample. This is attributed to the presence of largely stationary-phase I dying cells which release intracellular proteins through cell death and lysis of internal components.Significant autofluorescence of the unstained low-density suspended cell sample and assisted biofilm sample was observed at wavelengths utilised for SR-staining. This was attributed to the reliance on peak wavelength close to 600 nm - the wavelength typically used for monitoring cell turbidity. As a result, an overall negative fluorescence response was observed for these samples following control-correction. SR-staining was observed to interact with components of the agar hydrogel and nutrient media samples. This was expected as previous research (Volgenant et al., 201) indicates significant non-specific staining of uninoculated media by SR, and of Luria broth (Zhang et al., 2019), specifically.

[0246] Similarly, for CBB (Figure 2 F, Study 2), the agar layer increased cell viability of biofilm-associated cells, decreasing cell death, lysis and release of intracellular protein into the surrounding media. Residual absorbance was also observed for the low-density suspended cell sample and assisted biofilm sample, resulting in an overall negative absorbance response following control-correction. Significant non-specific staining of the agar hydrogel was still observed after rinsing of excess stain. This was expected as existing research shows that CBB can penetrate and bind to components within the agarose gel matrix (Arakawa et al., 2022).

[0247] SYPRO Ruby is reportedly accurate, simple and easy to use and has greater sensitivity compared to CBB (Anaya et al., 2007). As a result of the reported increased sensitivity and significant non-specific staining of the agar hydrogel by CBB, SR was utilised for further experiments.

[0248] Using GR-staining (Figure 2 G, Study 2), the agar layer was observed to increase biofilm formation and growth, increasing cell viability and resulting in decreased release of intracellular DNA through prevention of cell death and lysis. As GR in unable to penetrate the bacterial cell membrane, intracellular DNA cannot be stained and quantified (Doroshenko et al., 2014).

[0249] The low-density suspended cell sample has a greater (non-significant) DNA concentration compared to the high-density suspended cell sample (Study 1 and 2). This is attributed to the presence of largely log-phase / actively-metabolising cells,which produce and release large amounts of intracellular DNA into the surrounding environment which can then be stained with GR.

[0250] GR is observed to interact with components within the agar hydrogel and nutrient media sample (Study 1), resulting in significant signal and an overestimation of DNA concentration. Autofluorescence of the negative controls is also evident at the peak wavelength utilised for GR-staining (620 nm) - close to the wavelength typically used to monitor cell turbidity.

[0251] The formation and development of E. coli and B. subtilis agar-assisted and unassisted biofilms was monitored over time using resazurin (Figure 3 A and B). This was compared to the low-density suspended cell sample.

[0252] Variability in the metabolic state of the suspended cell controls was minimised by using (n=5) independent samples that were diluted and applied as controls. However, experimental differences between different time points may still have influenced the overall measurements. It is proposed that variations in turbidity between individual biofilm samples have also affected the observed fluorescent signal by proportionally absorbing excitation and emission photons.

[0253] Higher estimates of viable cells was observed for the agar-assisted biofilm (Figure 3 A) and was attributed to difference in the number of viable cells, supported by Figure 1 C, which shows minimal interaction between resazurin and the agar hydrogel. Thus, supporting that the presence of the agar layer encourages biofilm formation and development, increasing cell viability.

[0254] Cell viability of the assisted biofilm increased over time, reaching a peak at 72 hours and declining thereafter. The decline in cell viability observed after 72 hours is attributed to the shift of biofilm population from actively metabolising to an inner cohort of stationary / stressed / non-viable cells.

[0255] In comparison, cell viability remained relatively constant over the times tested for the unassisted biofilm, highlighting the role of the agar layer in promoting increased cell viability.Cell viability increased significantly between inoculation and 48, 72 and 96 hours for the assisted biofilms.

[0256] In contrast, the absence of significant viability changes from inoculation for the unassisted biofilm suggests that, in the absence of the agar layer, biofilm formation and development is limited. Collectively, these results highlight the importance of the agar layer in promoting biofilm formation and development and enhanced cellular activity of E. coli biofilms.

[0257] Although cell viability remained relatively constant over time for the B. subtilis assisted biofilm, higher cell viability was observed at 0, 24, 72 and 120 hours (Figure 3 B), compared to the unassisted biofilm. Thus, supporting that the presence of the agar layer encourages biofilm formation and development.

[0258] In contrast, cell viability of the unassisted biofilms fluctuated significantly over time, with viable cell estimates peaking at 48 and 96 hours with a significant reduction in cell viability observed at 0, 24, 72, 96 and 120 hours (J annotations). This pattern could reflect a biphasic growth behaviour of the unassisted B. subtilis biofilms.

[0259] These results largely support the hypothesis that the presence of the soft agar hydrogel enhanced cell viability, biofilm formation and maturation in both E. coli and B. subtilis. Importantly, this was achieved using an agar hydrogel considered as soft (1.2%), which has previously been reported to be less conducive for bacterial cell adhesion. This finding therefore challenges existing assumptions regarding the limitations of soft agar hydrogels in encouraging bacterial attachment and biofilm formation.

[0260] Antimicrobial screening of E. coli and B. subtilis biofilm-associated and suspended cells was then performed over varied antibiotic concentrations using ciprofloxacin and kanamycin (Figure 4 and 5).

[0261] The half-maximum inhibitory concentration of antibiotic necessary to decrease cell viability by 50% of the initial cell viability (ICso) is often used as an important metric to compare antibiotic efficacy.The higher IC50value observed for the agar-assisted biofilm indicates an enhances ciprofloxacin tolerance relative to the unassisted biofilm, consequent of the presence of the agar layer (Figure 4 A). The similarity between the IC50values obtained for the high-density suspended cell sample and unassisted biofilm suggests the absence of the agar layer yields an antibiotic tolerance profile similar to that of dispersed cells (Figure 4 B). As anticipated, the substantially lower IC50value observed for the low-density suspended cell sample reflects the increased susceptibility of less crowded cells to ciprofloxacin exposure.

[0262] A half-maximum inhibitory concentration for ciprofloxacin ranging from 0.5 to 32 pg / mL (Naves et al., 2010) has been previously reported for E. coli biofilm-associated cells. The significantly higher IC50values obtained for the assisted and unassisted biofilm samples largely supports that the biofilm cultivation system, in the presence or absence of the agar layer, was capable of producing more consistent biofilms with increased tolerance to ciprofloxacin.

[0263] A minimum inhibitory concentration (MIC) for kanamycin of 0.74 ±0.01 (Sudha and Aranganathan, 2024) and 4.5 pg / mL (Pereira et al., 2012) have been reported for E. coli suspended and biofilm-associated cells, respectively. Similar IC50values were observed for the assisted and unassisted biofilm samples.

[0264] The slightly elevated IC50value observed for the assisted biofilm (Figure 4 C), compared to the unassisted biofilms, supports that the presence of the agar layer enhance the biofilms’ tolerance to kanamycin by encouraging growth or providing a more protective microenvironment that supports biofilm development. However, significantly higher IC50values were observed for the suspended cell samples (Figure 4 D). This likely reflects the more efficient uptake of kanamycin by metabolically active biofilm-associated cells, compared to the dense suspended cell cultures where oxygen depletion limits antibiotic uptake.

[0265] The higher IC50value observed for the assisted biofilm sample, compared to the unassisted biofilm sample (Figure 4 E), supports that the agar layer increasestolerance of biofilm-associated cells to ciprofloxacin as the associated-cells are able to survive higher ciprofloxacin concentrations.

[0266] In comparison, an increase in DAPI response is observed at higher ciprofloxacin concentrations for the suspended cell samples (Figure 4 F). This is attributed to increased cell death and lysis of the suspended cells at higher ciprofloxacin concentrations, resulting in the release of intracellular DNA into the surrounding medium, which is subsequently stained by DAPI. It is anticipated that at higher ciprofloxacin concentrations, the assisted and unassisted biofilm samples would exhibit a similar trend.

[0267] DAPI responses remained largely consistent as the kanamycin concentration increased for both biofilm samples (Figure 4 G). This suggests that the kanamycin concentration used was insufficient to effectively inhibit both the agar-assisted and unassisted E. coli biofilm samples. This is supported by the IC50values approaching the upper limit of the kanamycin concentrations tested. However, it is expected that at higher kanamycin concentrations, a significant decline in DAPI response would be observed. Conversely, the suspended cell samples show decreased tolerance to kanamycin (Figure 4 H), supported by the observed decline in DAPI responses and the IC50values obtained.

[0268] Presently, there is insufficient published information regarding the MIC or IC50values for B. subtilis biofilm-associated cells for ciprofloxacin and kanamycin, inhibiting comparison with the IC50values obtained in this research.

[0269] The higher IC50values observed for the B. subtilis assisted biofilm (Figure 5 A) indicate that the presence of the agar layer promotes biofilm formation and development, increased cell viability and resilience to antibiotic exposure. The reduced IC50values for the suspended cell samples (Figure 5 B) support that dispersed cells are more susceptible to ciprofloxacin compared to their biofilm-associated counterparts.

[0270] The higher IC50value observed for the assisted biofilm sample (Figure 5 C) supports that the presence of the agar layer encourages increased cell viability and toleranceto kanamycin. However, the unassisted biofilm maintained relatively stable viability across the tested concentration range, suggesting that its kanamycin tolerance may extend beyond the upper limits of the concentrations evaluated. The lower IC50values observed for the suspended cell samples (Figure 5 D) further supports the increased tolerance of biofilm-associated cells.

[0271] The observed increase in DAPI responses at higher ciprofloxacin concentrations (Figure 5 E) indicates substantial biofilm disruption and cell lysis consequent of antibiotic induced damage. As a result, intracellular DNA is released into the surrounding medium which is subsequently DAPI-stained (as previously described for Figure 4 F). The higher IC50value of the assisted biofilm sample suggests that the agar-layer enhances tolerance to ciprofloxacin. This is supported by the significantly lower IC50values observed for the suspended cell samples (Figure 5 F).

[0272] Although the assisted biofilm exhibited a higher IC50value, the sustained stability of DAPI responses across the tested ciprofloxacin range (Figure 5 E) indicates that the unassisted biofilm may possess greater overall tolerance. This is potentially due to a larger proportion of cells experiencing nutrient-limited or slower growth conditions that stimulate stress response mechanisms prior to antibiotic exposure. Additionally, ciprofloxacin demonstrates reduced efficacy against nutrient-limited or slow-growing cells because the activity of its molecular targets - DNAgyrase and Topoisomerase IV (Wallace et al., 2018; Li et al., 2018) - are diminished under these conditions.

[0273] In comparison to (Figure 5 E), a decline in DAPI response was observed at higher kanamycin concentrations for the assisted biofilm sample (Figure 5 G). This may suggest that the kanamycin concentrations used were insufficient to substantially disrupt the biofilm and cause subsequent release of intracellular DNA. Rather, a decrease in extracellular DNA levels was observed, reflecting kanamycin-mediated disruption of translation and synthesis of faulty or non-functional proteins (Thomas et al., 2017; Fosso et al., 2014), limiting DNA production and secretion.

[0274] The sustained stability of DAPI response across the tested kanamycin range (Figure 5 G) indicates that the unassisted biofilm may possess greater overall tolerance. Thismay be attributed to stress-induced antibiotic tolerance, as previously described for (Figure 5 E).

[0275] These results confirm that the presence of the soft agar hydrogel significantly decreased antibiotic susceptibility of both E. coli and B. subtilis against both ciprofloxacin and kanamycin - antibiotics with differing mechanisms of action.

[0276] 4. Conclusions

[0277] The biofilm cultivation system allowed high-throughput screening while providing sufficient lateral space to support biofilm formation. Each stain was capable of detecting biomass / specific extracellular matrix components at stain-specific excitation and emission parameters.

[0278] However, the results demonstrated that most stains exhibited substantial autofluorescence or non-specific staining of the growth media and agar layer. Consequently, sample pre-treatment and inclusion of appropriate controls and standards on a single plate was necessary. The negative controls (agar only and growth media) were included to account for autofluorescence and non-specific staining, while, the stain-specific standard and suspended cell samples allowed us to distinguish a difference between varied cell densities.

[0279] To further eliminate the effects of autofluorescence / non-specific staining, spectra were presented as control-corrected (Figure 1) through subtracting the average of the unstained controls from the samples’ responses.

[0280] Following control-correction, the expected spectra and characteristic peak wavelengths were observed for all stains. Minimal autofluorescence and / or nonspecific staining of the negative controls was observed for some of the stains.

[0281] The data was then further standardised by normalising responses against the high-density suspended cell sample. This was done by normalising the average of the high-density suspended cell sample at the peak wavelength specific to each stain and multiplying by 100 (Figures 2).From Figure 2 A, a slightly higher DNA concentration was observed for the assisted biofilm sample, suggesting that the agar layer encouraged biofilm development and subsequent DNA production. Although a significant difference in DNA concentration was observed between the high and low-density suspended cell samples, the high-density sample displayed a substantially greater DNA concentration to the assisted biofilm. This is likely attributable to the reduced metabolic rate of biofilm-associated cells which typically produce lower concentrations of DNA compared to rapidly dividing suspended cells.

[0282] The greater cell viability observed for the assisted biofilm, relative to the unassisted biofilm and low-density suspended cell sample (¥ annotations), supports that the presence of the agar layer encourages biofilm formation and development and increased cell viability. Resazurin staining effectively differentiated a difference in biomass between the stained and unstained assisted and unassisted biofilm samples and the high and low-density suspended cell samples (* annotations). Moreover, resazurin was capable of sensitively distinguishing a difference in biomass between the high and low-density suspended cell samples.

[0283] Based on results from Figure 2, resazurin and DAPI were used for subsequent analyses. The resazurin assay was employed to assess if the biofilm cultivation system supported the growth of consistent biofilms and to evaluate the influence of the agar layer on biofilm growth. This analysis was conducted using E. coli and B. subtilis biofilm-associated cells and suspended cells.

[0284] The biofilm kinetics presented depicts the resorufin fluorescence response at 590 nm, normalised to the 560 nm emission baseline, which accounts for background autofluorescence.

[0285] For the assisted E. coli biofilm, cell viability increased over time, peaking at 72 hours before declining. A significant increase in viability relative to inoculation was observed at numerous of the times tested, supporting that the agar layer encourages biofilm growth and increased cellular activity. In contrast, cell viability remained relatively constant over the times tested for the unassisted biofilm.For B. subtilis, the assisted biofilm maintained a more stable viability overtime, relative to the unassisted biofilm. However, greater cell viability was evident at multiple time points, indicating the agar layer supported biofilm growth. Conversely, cell viability fluctuated largely over time for the unassisted biofilm.

[0286] The biofilm kinetics largely demonstrated that the presence of the agar layer promoted more consistent and higher cell viability for both E. coli and B. subtilis biofilms. However, varied growth patterns were observed for assisted and unassisted biofilms.

[0287] The impact of ciprofloxacin and kanamycin on the viability and DAPI responses of assisted and unassisted E. coli (Figure 4) and B. subtilis (Figure 5) biofilms and the high and low-density suspended cell samples was assessed using resazurin and DAPI.

[0288] Antimicrobial screening using DAPI-staining reflects the fluorescence response at 447 nm normalised to the low-density suspended cell sample. Although resazurin-based antimicrobial screening data could be normalised to the low-density suspended cell sample, as done for DAPI, normalisation using the emission baseline yielded more consistent and reliable model outputs.

[0289] Antimicrobial testing largely indicated that the E. coli assisted biofilm exhibited greater tolerance to ciprofloxacin than the unassisted biofilms, with significantly greater IC50values (Figure 4). The unassisted biofilms exhibited similar sensitivity to the high-density suspended cell sample and the low-density suspended cell sample was much more susceptible.

[0290] Conversely, antimicrobial screening using kanamycin yielded similar tolerances for the assisted and unassisted biofilms and were substantially lower than those of the suspended cell samples. This is attributed to more efficient uptake of kanamycin by metabolically active biofilm-associated cells, compared to the dense suspended cell cultures were oxygen depletion limits kanamycin uptake.Higher ciprofloxacin concentrations caused a reduction in DAPI responses, with the assisted biofilm maintaining greater response relative to the unassisted biofilm. These results largely support that the presence of the agar layer increased tolerance to ciprofloxacin. Conversely, kanamycin had minimal impact on the DAPI responses of both the assisted and unassisted biofilms at the concentrations tested. However, lower tolerance to kanamycin was observed for the high and low-density suspended cell samples. Thus, suggesting that kanamycin may not be sufficient to inhibit the biofilms at the concentrations tested.

[0291] For B. subtilis, assisted biofilms exhibited greater tolerance to ciprofloxacin compared to unassisted biofilms with higher IC50values (Figure 5). The suspended cells were more susceptible, showing significant decline in cell viability at lower ciprofloxacin concentrations and lower IC50values relative to the assisted and unassisted biofilms. Thus, supporting that the biofilm cultivation system was capable of increasing tolerance to ciprofloxacin in the presence and absence of the agar layer.

[0292] Higher DNA levels were observed for the assisted biofilm samples at higher ciprofloxacin concentrations, while unassisted biofilms showed more stable DNA levels. The increase in DAPI responses at higher ciprofloxacin concentrations indicates substantial biofilm disruption and cell lysis consequent of antibiotic induced damage. As a result, intracellular DNA is released into the surrounding medium which is subsequently DAPI-stained, resulting in an increased fluorescence response.

[0293] For kanamycin, slightly reduced cell viability was observed at higher concentrations for the assisted biofilms, while the unassisted biofilms remained relatively stable over time. Thus, suggesting that kanamycin was not sufficient to inhibit biofilm-associated cells at the concentrations tested. This is supported by the lower tolerance and IC50values observed for the suspended cell samples.

[0294] These results largely support the hypothesis that the presence of the soft agar hydrogel enhanced cell viability, biofilm formation and maturation, and antibiotic resistance in both E. coli and B. subtilis. Importantly, this was achieved using an agar hydrogel considered as soft (1.2%), which has previously been reported to be lessconducive for bacterial cell adhesion (Kolowe et al., 2015). This finding therefore challenges existing assumptions regarding the limitations of soft agar hydrogels in encouraging bacterial attachment and biofilm formation (Kolowe et al., 2015).

[0295] REFERENCES

[0296] 1. Branda, S., Vik, A., Friedman, L. and Kolter, R. (2005). Biofilms: the matrix revisited. Trends Microbiol., 13(1), pp.20-26. https: / / doi.orq / 10.1016Zi.tim.2004.11.006

[0297] 2. Koo, H., Raymond, A., Howlin, R., Stoodley, R and Hall-Stoodley, L.(2017).

[0298] Targeting microbial biofilms: current and prospective therapeutic strategies. Nat. Rev. Microbiol., 15 (12), pp. 740-755. https: / / doi.Org / 10.1038 / nrmicro.2017.99

[0299] 3. Mikkelsen, H., Duck, Z., Lilley, K.S. and Welch, M. (2007). Interrelationships between colonies, biofilms, and planktonic cells of Pseudomonas aeruginosa. J. Bacterio!., 189 (6), pp. 2411-2416. https: / / doi.org / 10.1128 / JB.01687-06 4. Ren, D„ Bedzyk, L.A., Thomas, S.M., Ye, R.W. and Wood, T.K. (2004). Gene expression in Escherichia coli biofilms. Appl. Microbiol. Biotechnol., 64, pp. 515- 524. https: / / doi.org / 10.1007 / s00253-003-1517-v

[0300] 5. Sharma, D., Misba, L. and Khan, A. (2019). Antibiotics versus biofilm: an emerging battleground in microbial communities. Antimicrob. Resist. Infect. Control., 8 (1). https: / / doi.Org / 10.1186 / s13756-019-0533-3

[0301] 6. Auger, J.P., Chuzeville, S., Roy, D., Mathieu-Denoncourt, A., Xu, J., Grenier, D.

[0302] and Gottschalk, M. (2017). The bias of experimental design, including strain background, in the determination of critical Streptococcus suis serotype 2 virulence factors. PLOS ONE, 12.

[0303] https: / / doi.org / 10.1371 / iournal.pone.0181920.

[0304] 7. Jorge, R, Lourenco, A. and Pereira, M.O. (2015). Data quality in biofilm high- throughput routine analysis: intralaboratory protocol adaptation and experiment reproducibility. J. AOAC Int., 98 (6), pp. 1721-1727. https: / / doi.Org / 10.5740 / iaoacint.15-066

[0305] 8. Floyd, K.A., Moore, J.L., Eberly, A.R., Good, J.A., Shaffer, C.L. et al. (2015).

[0306] Adhesive fibre stratification in uropathogenic Escherichia coli biofilms unveilsoxygen mediated control of type 1 pili. PLOS Pathog., 11.

[0307] https: / / doi.org / 10.1371 / iournal.ppat1004697

[0308] Holden, E.R., Yasir, M., Turner, A.K., Wain, J., Charles, I.G. et al. (2021). Massively parallel transposon mutagenesis identifies temporally essential genes for biofilm formation in Escherichia coli. Microb. Genom., 7(11), p. 673. https: / / doi.Org / 10.1099 / mqen.0.000673

[0309] Stoodley, P., Sauer, K., Davies, D.G. and Costerton, J.W. (2009). Biofilms as complex differentiated communities. Annu. Rev. Microbiol., 56, pp.187-209. https: / / doi.Org / 10.1146 / annurev.micro.56.012302.160705

[0310] Beitelshees, M., Hill, A., Jones, C.H. and Pfeifer, B.A. (2018). Phenotypic variation during biofilm formation: Implications for anti-biofilm therapeutic design. Materials, 11(7), p. 1086. https: / / doi.org / 10.3390 / ma11071086 Domka, J., Lee, J., Bansal, T. and Wood, T.K. (2007), Temporal geneexpression in Escherichia coli K-12 biofilms. Environmental Microbiology, 9, pp.

[0311] 332-346. htps: / / doi.Org / 10.1111 / i.1462-2920.2006.01143.x

[0312] Stewart, P.S. and Franklin, M.J. (2008). Physiological heterogeneity in biofilms. Nature Reviews Microbiology, 6, pp. 199 -210. https: / / doi.org / 10.1038 / nrmicro1838

[0313] Hall-Stoodley, L., Costerton, J.W. and Stoodley, P. (2004). Bacterial biofilms: from the natural environment to infectious diseases. Nature Reviews Microbiology, 2, pp. 95-108. htps: / / doi.Org / 10.1038 / nrmicro821

[0314] Prigent-Combaret, C., Vidal, O., Dorel, C., and Lejuene, P. (1999). Abiotic surface sensing and biofilm-dependent regulation of gene expression in Escherichia coli. Journal of Bacteriology, 181, pp. 5993-6002. httDs: / / doi.orq / 10.1128 / ib.181.19.5993-6002.1999

[0315] Houdt, R. and Michiels, C.W (2005). Role of bacterial cell surface structures in Escherichia coli biofilm formation. Research in Microbiology, 156 (5-6), pp.626-633. htps: / / doi.Org / 10.1016 / i.resmic.2005.02.005

[0316] Besharova, O., Suchanek, V.M., Hartmann, R., Drescher, K. and Sourjik, V. (2016) Diversification of Gene Expression during Formation of Static Submerged Biofilms b Escherichia coli. Front. Microbiol. 7, p. 1568. doi: 10.3389 / fmicb.2016.01568Hussain, H.L, Iqbal, Z., Seleem, M.N., Huang, D., Sattar, A. et al. (2017). Virulence and transcriptome profile of multidrug- resistant Escherichia coli from chicken. Sci. Rep., 7, p. 8335. https: / / doi.orq / 10.1038 / s41598-017-07798-1 Hung, C., Zhou, Y, Pinkner, J.S., Dodson, K.W., Crowley, J.R. et al. (2013). Escherichia coli biofilms have an organized and complex extracellular matrix structure. MBio, 4 (5). https: / / doi.Org / 10.1128 / mBio.00645-13

[0317] Ageorges, V., Monteiro, R., Leroy, S., Burgess, C.M., Mariagrazia, P. et al. (2020). Molecular determinants of surface colonisation in diarrhoeagenic Escherichia coli (DEC): from bacterial adhesion to biofilm formation. Microbiol. Rev., 44 (3), pp. 314-350. https: / / doi.Org / 10.1093 / femsre / fuaa008

[0318] Chen, J.L., Steele, T.W.J. and Stuckey, D.C. (2017). Metabolic reduction of resazurin: location within the cell for cytotoxicity assays. Biotechnol. Bioeng., 115 (2), pp. 351-358. https: / / doi.Org / 10.1002 / bit.26475

[0319] Wozniak, D.J., Wyckoff, T.J., Starkey, M., Keyser, R., Azadi, P. et al. (2003). Alginate is not a significant component of the extracellular polysaccharide matrix of PA 14 and PAO1 Pseudomonas aeruginosa biofilms. PNAS, 100, pp.

[0320] 7907-7912. https: / / doi.org / 10.1073 / pnas.1231792100

[0321] Zhang, X., Li, Z., Pang, S., Jiang, B., Yang, Y. et al. (2021). The impact of cell structure, metabolism and group behaviour for the survival of bacteria under stress conditions. Archives of Microbiology, 203, pp. 431-441. https: / / doi.org / 10.1007 / s00203-020-02050-3

[0322] Dalecki, A.G., Crawford, C.L. and Wolschendorf, F. (2016). Targeting biofilm associated Staphylococcus aureus using resazurin based drug-susceptibility assay. J. Vis. Exp., 111, e53925. https: / / doi.Org / 10.3791 / 53925

[0323] de Carvalho Dias, K., Barbugli, P.A., De Patto, F., Lordello, V.B., de Aquino Penteado, L. et al. (2017). Soluble factors from biofilm of Candida albicans and Staphylococcus aureus promote cell death and inflammatory response. Bmc Microbiol., 17(1), pp. 1-9. htps: / / doi.org / 10.1186 / s12866-017-1031-5 Ravaioli, S., Campoccia, D., Speziale, P., Pietrocola, G., Zatorska, B. et al. (2020). Various biofilm matrices of the emerging pathogen Staphylococcus lugdunensis: exopolysaccharides, proteins, eDNA and their correlation with biofilm mass. Biofouling, 36(1), pp.86-100. https: / / doi.org / 10.1080 / 08927014.2020.1716217Sedmak, J. J. and Grossberg, S. E. (1977). A rapid, sensitive, and versatile assay for protein using Coomassie brilliant blue G250. Analytical biochemistry, 79 (1-2), pp. 544-552. https: / / doi.Org / 10.1016 / 0003- 2697(77)90428-6

[0324] Nosyk, O., Haseborg, E., Metzger, U. and Frimmel, F. (2008). A standardized pre-treatment method of biofilm flocs for fluorescence microscopic characterization. Journal of Microbiological Methods, 75(3), pp.449-456. https: / / doi.Org / 10.1016 / i.mimet.2008.07.024

[0325] Stiefel, P., Rosenberg, U., Schneider, J., Mauerhofer, S., Maniura-Weber, K. et al. (2016). Is biofilm removal properly assessed? Comparison of different quantification methods in a 96-well plate system. Applied Microbiology and Biotechnology, 100(9), pp. 4135-4145. https: / / doi.Org / 10, 1007 / s00253-016-7396-9

[0326] Grintzalis, K., Georgiou, C.D. and Schneider, Y.J. (2015). An accurate and sensitive Coomassie Brilliant Blue G-250-based assay for protein determination. Anal. Biochem., 1 (480), pp. 28-30. https: / / doi.Org / 10.1016 / i.ab.2015.03.024

[0327] Song, Y., Yang, Y, Lin, X., Li, X., Zhang, X. et al. (2020). In-situ and sensitive stability study of emulsion and aluminium adjuvanted inactivated foot-and-mouth disease virus vaccine by differential scanning fluorimetry analysis. Vaccine, 38(14), pp. 2904-2912. https: / / doi.Org / 10.1016 / i.vaccine.2020.02.068 Castro, W., Navarro, M. and Biot, C. (2013). Medicinal potential of ciprofloxacin and its derivatives. Future med. chem., 5(1), pp. 81-96. https: / / doi.org / 10.4155 / fmc.12.181

[0328] Ahmed, M.N., Porse, A., Sommer, M.O.A., Hoiby, N. and Ciofu, O. (2018). Evolution of antibiotic resistance in biofilm and planktonic Pseudomonas aeruginosa populations exposed to subinhibitory levels of ciprofloxacin. Anti microb. Agents Chemother., 62. https: / / doi.org / 10.1128 / aac.00320-18 Abdi-Ali, A., Hendiani, S., Mohammadi, P. and Gharavi, S. (2014). Assessment of biofilm formation and resistance to imipenem and ciprofloxacin among clinical isolates of Acinetobacter baumannii in Tehran. Jundishapur Journal of Microbiology, 7(1), e8606. https: / / doi.org / 10.5812 / iim.8606

[0329] Anderl, J.N., Franklin, M.J. and Stewart, P.S. (2000). Role of antibiotic penetration limitation in Klebsiella pneumoniae biofilm resistance to ampicillinand ciprofloxacin. Antimicrob. Agents Chemother., 44. https: / / doi.Org / 10,1128 / aac.44.7.1818-18242000

[0330] Stockert, J.C., Horobin, R.W., Colombo, L.L. and Blazquez-Castro, A. (2017). Tetrazolium salts and formazan products in Cell Biology: viability assessment, fluorescence imaging, and labelling perspectives. Acta Histochemica, 120 (3), pp. 159-167. https: / / doi.Org / 10.1016 / i.acthis.2O18.02.005.

[0331] Saby, S. Sibille, I., Mathieu, L., Paquin, J.L. and Block, J.C. (1997). Influence of water chlorination on the counting of bacteria with DAPI (4',6-diamidino-2-phenylindole). Applied and Environmental Microbiology, 63(4), pp. 1564- 1569. https: / / doi.org / 10.1128 / aem.63.4.1564-1569.1997

[0332] Sayas, E., Garcia-Lopez, F. and Serrano, R. (2015). Toxicity, mutagenicity, and transport in Saccharomyces cerevisiae of three popular DNA intercalating fluorescent dyes. Yeast, 32(9), pp. 595-606. https: / / doi.Org / 10.1002 / vea.3081 Lilyanna, S., Ng, E.M.W., Moriguchi, S., Chan, S.P., Kokawa, R. et al. (2018). Variability in microplate surface properties and its impact on ELISA. The Journal of Applied Laboratory Medicine, 2 (5), pp. 687 699. https: / / doi.org / 10.1373 / ialm.2O17.023952

[0333] Mihalcescu, I., Van-Melle Gateau, M., Chelli, B., Pinel, C. and Ravanat, J.L. (2015). Green autofluorescence, a double-edged monitoring tool for bacterial growth and activity in micro-plates. Phys. Biol., 12(6), p. 066016. https: / / doi.org / 10.1088 / 1478-3975 / 12 / 6 / 066016

[0334] Bao, N., Jagadeesan, B., Bhunia, B.K., Yao, Y. and Lu, C. (2008). Quantification of bacterial cells based on autofluorescence on a microfluidic platform. J. Chromatogr., 1181(1-2), pp. 153-158. https: / / doi.Org / 10.1016 / i.chroma.2007.12.048

[0335] Mullerova, L., Markova, K., Obruca, S. and Mravec, F. (2022). Use of flavin-related cellular autofluorescence to monitor processes in microbial biotechnology. Microorganisms, 10 (6), p. 1179. https: / / doi.org / 10.3390 / microorganisms10061179

[0336] Lakowicz, J.R. (2006) Principles of fluorescence spectroscopy, 3. Springer New York, NY. https: / / doi.Org / 10.1007 / 978-0-387-46312-4

[0337] Zrimsek, R, Kune, J., Kosec, M. and Mrkun, J. (2004), Spectrophotometric application of resazurin reduction assay to evaluate boar semen quality. Int. J. Androl., 27, pp. 57-62. https: / / doi.Org / 10.1046 / i.0105-6263.2003.00448.xBenov, L. (2021). Improved formazan dissolution for bacterial MTT assay. Microbiol. Spectr., 9 (3). https: / / doi.Org / 10,1128 / spectrum,01637-21 Mao, Z., Liu, Z., Chen, L., Yang, J., Zhao, B. et al. (2013). Predictive value of the surface enhanced resonance raman scattering-based MTT assay: a rapid and ultrasensitive method for cell viability in situ. Anal. Chem., 85 (15), pp.

[0338] 7361-7368. https: / / doi.Org / 10.1021 / ac401254s

[0339] Berggren, K.N., Schulenberg, B., Lopez, M.F., Steinberg, T.H., Bogdanova, A. et al. (2002). An improved formulation of SYPRO Ruby protein gel stain: comparison with the original formulation and with a ruthenium II tris (bathophenanthroline disulfonate) formulation. Proteom, 2, pp. 486-498. http: / / dx.doi.org / 10.1002 / 1615-9861(200205)2:5%3C486::AID-PROT486%3E3.Q.CO;2-X

[0340] Anaya, C., Church, N. and Lewis, J. P. (2007). Detection and identification of bacterial cell surface proteins by fluorescent labelling. Proteomics, 7(2), pp.

[0341] 215-219. htps: / / doi.Org / 10.1002 / pmic.200600175

[0342] Demina, LA., Serebryakova, M.V., Ladygina, V.G. etal. (2009). Proteome of the bacterium Mycoplasma gallisepticum. Biochemistry Moscow, 74, pp. 165-174. https: / / doi.org / 10.1134 / S0006297909020072

[0343] Doroshenko, N., Tseng, B.S., Howlin, R.P., Deacon, J., Wharton, J.A. et al. (2014). Extracellular DNA impedes the transport of vancomycin in Staphylococcus epidermidis biofilms pre-exposed to subinhibitory concentrations of vancomycin. Antimicrob. Agents Chemother., 58(12), pp.

[0344] 7273-7282. htDs: / / doi.org / 10.1128 / AAC.03132-14

[0345] Hecht, A., Endy, D., Salit, M. and Munson, M.S. (2016). When wavelengths collide: bias in cell abundance measurements due to expressed fluorescent proteins. SynBio, 5(9), pp. 1024-1027. https: / / doi.Org / 10.1021 / acssvnbio.6b00072

[0346] Latka, A. and Drulis-Kawa, Z. (2020). Advantages and limitations of microtiter biofilm assays in the model of antibiofilm activity of Klebsiella phage KP34 and its depolymerase. Sci. Rep., 10 (1). htDs: / / doi.org / 10.1038 / s41598-020-77198-5

[0347] Magana, M., Sereti, C., loannidis, A., Mitchell, C.A., Ball, A.R. et al. (2018). Options and limitations in clinical investigation of bacterial biofilms. Clin. Microbiol. Rev., 31 (3). htDs: / / doi.org / 10.1128 / CMR.00084-16Riss, T., Moravec, R.A., Niles, A.L., Duellman, S., Benink, H.A. et al. (2004). The assay guidance manual: cell viability assays. NLM, pp. 1-39. https: / / www.ncbi.nlm.nih.qov / books / NBK144065 /

[0348] Martin, A., Morcillo, ., Lemus, D., Montoro, E., da Silva Telles, M.A. et al. (2005). Multicentre study of MTT and resazurin assays for testing susceptibility to first-line anti-tuberculosis drugs. Int. J. Tuberc. Lung. Dis., 9(8), pp. 901-906. https: / / pubmed.ncbi.nlm.nih.gov / 16104638 /

[0349] Arakawa, T., Nakagawa, M., Tomioka, Y., Sakuma, C., Li, C. et al. (2022). Chapter 4 - Gel-electrophoresis based method for biomolecular interaction. Meth. Cell Biol., 169, pp. 67-95. https: / / doi.Org / 10.1016 / bs.mcb.2O21.12.030. Volgenant, C., van der Veen, M.H., de Soet, J. J. and ten Cate, J.M. (2013). Effect of metalloporphyrins on red autofluorescence from oral bacteria. Eur. J. Oral Sci., 121, pp. 156-161. https: / / doi.org / 10.1111 / eos.12045

[0350] Zhang, X.Y., Sun, K„ Abulimiti, A., Xu, P.R and Li, Z.Y (2019). Microfluidic system for observation of bacterial culture and effects on biofilm formation at microscale. Micromachines, 10(9), p. 606. https: / / doi.Org / 10.3390 / mi10090606 Fosso, M.Y, Li, Y. and Garneau-Tsodikova, S. (2014). New trends in the use of aminoglycosides. Med Chem Comm, 5 (8), pp. 1075-1091. https: / / doi.org / 10.1039 / C4MD00163J

[0351] Naves, P., Del Prado, G., Ponte, C. and Soriano, F. (2010). Differences in the in vitro susceptibility of planktonic and biofilm-associated Escherichia coli strains to antimicrobial agents. Journal of Chemotherapy, 22(5), pp. 312-317. https: / / doi.org / 10.1179 / ioc.2010.22.5.312

[0352] Pereira, A. M., Abreu, A. C. and Simoes, M. (2012). Action of kanamycin against single and dual species biofilms of Escherichia coli and Staphylococcus aureus. Journal of Microbiology Research, 2 (4), pp. 84-88. https: / / doi.Org / 10.5923 / i.microbiology.20120204.04

[0353] Sudha, S. S. and Aranganathan, V. (2024). Antibiofilm analysis, synergistic potential and biocompatibility evaluation of a bacteriocin from Bacillus subtilis (MK733983). Indian Journal of Microbiology, 64(4), pp. 1646-1663. https: / / doi.Org / 10.1007 / s12088-024-01206-9

[0354] Guegan, C., Garderes, J., Le Pennec, G., Gaillard, F., Fay, F. et al. (2014). Alteration of bacterial adhesion induced by the substrate stiffness. Colloids andSurfaces B: Biointerfaces, 114, pp. 193-200. https: / / doi.Org / 10.1016 / i.colsurfb.2013.10.010

[0355] Roizman, D., Herzog, M., Nath, A., Pachaimuthu, N., Hujeirat, A. et al. (2024). Crosslinked agarose-gelatine beads as a substrate for investigating biofilms of bacterial pathogens. bioRxiv, pp. 1-38. https: / / doi.org / 10.1101 / 2024.08.16.608334

[0356] Kolowe, K.W., Peyton, S.R., Schiffman, J.D. (2015). Fewer Bacteria Adhere to Softer Hydrogels. ACS Appl Mater Interfaces, 7(35), pp. 19562-19569. doi: 10.1021 / acsami.5b04269.

Claims

CLAIMS1. A method of cultivating a microbial biofilm comprising the use of a soft agar hydrogel substrate and one or more stain(s) for detection and / or quantification of one or more microbial component(s) and / or characteristic(s), the method comprising or consisting of the following steps:a. providing a cultivation vessel having a solid substrate;b. adding liquid agar having a concentration of about 1.2% (w / v) into the cultivation vessel and allowing it to solidify on the solid substrate into a soft agar hydrogel;c. inoculating microorganisms into the cultivation vessel and cultivating the microorganisms on the soft agar hydrogel to form a biofilm; and d. staining the microorganisms during cultivation with the one or more stain(s) for detection and / or quantification of one or more microbial components and / or characteristics.

2. The method according to claim 1, wherein the microbial biofilm comprises a microorganism with the capacity to produce a biofilm selected from the group consisting of bacteria, fungi and algae.

3. The method according to either claim 1 or claim 2, wherein the microbial biofilm is a bacterial biofilm selected from a Gram-positive or Gram-negative bacteria.

4. The method according to claim 3, wherein the bacteria is Escherichia coli E.coli) or Bacillus subtilis (B. subtilis).

5. The method according to any one of claims 1 to 4, comprising or consisting of the following steps:a) providing a cultivation vessel having a solid substrate;b) adding the agar into the cultivation vessel and allowing it to solidify into a hydrogel;c) adding liquid microbiological media to the cultivation vessel;d) inoculating microorganisms to be cultured into the cultivation vessel;e) cultivating the microorganisms in the cultivation vessel to produce a biofilm of microorganisms; andf) staining the microorganisms during cultivation with the one or more stain(s) for detection and / or quantification of one or more microbial components and / or characteristics.

6. The method according to claim 5, wherein steps c) and d) are performed separately or simultaneously.

7. The method according to claim 5, wherein step f) comprises staining the microorganisms during cultivation with one or more stain(s) suitable for identification of a microorganism, or microorganism viability, or both.

8. The method according to claim 7, wherein the stain is a fluorometric or a colorimetric stain comprising or consisting of any one or more stain(s) as set out in Table 1 and 2.

9. The method according to either claim 7 or 8, wherein the fluorometric or a colorimetric stain is for detecting viable microbial cells or quantification of extracellular DNA comprising or consisting of DAPI (4',6-diamidino-2- phenylindole), CV (crystal violet), resazurin, tetrazolium dye or MTT (3-(4,5- dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide), an organometallic ruthenium chelate, CBB (Coomassie Brilliant Blue), or 5,5'-(6,22-Dioxo- 11,14,17-trioxa-7,21 -diazaheptacosane- 1 ,27-diyl)bis(3,8-diamino-6- phenylphenanthridin-5-ium) diiodide; or is one or both of a resazurin fluorescent stain and a DAPI fluorescent stain.

10. A method of screening one or more antimicrobial agent(s) for efficacy against a microbial biofilm with the use of a microbial biofilm cultivated according to the method of any one of claims 1 to 9.

11. The method according to claim 10, further comprising the steps of:i) providing a microbial biofilm of microorganisms cultivated as described in any one of claims 1 to 9 upon which one or more antimicrobial agent(s) are to be screened;ii) adding the one or more antimicrobial agent(s) to the microbial biofilm; iii) incubating the microbial biofilm with the one or more antimicrobial agent(s);iv) detecting efficacy of the one or more antimicrobial agent(s) against the microbial biofilm by measurement of one or more parameter(s) including microbial cell viability.

12. The method according to claim 11, wherein the microbial cell viability is quantified with the use of any one or more staining method(s) capable of detecting viable microbial cells or quantification of extracellular DNA comprising or consisting of DAPI (4',6-diamidino-2-phenylindole), CV (crystal violet), resazurin, tetrazolium dye or MTT (3-(4,5-dimethylthiazol-2-yl)-2,5- diphenyltetrazolium bromide), an organometallic ruthenium chelate, CBB (Coomassie Brilliant Blue), or 5,5'-(6,22-Dioxo-11,14,17-trioxa-7,21- diazaheptacosane-1,27-diyl)bis(3,8-diamino-6-phenylphenanthridin-5-ium) diiodide; or is quantified with the use of resazurin, or DAPI or both.

13. The method according to claim 11 or 12, wherein step iv) is measured using microscopy including colorimetric or fluorescence microscopy, optionally including measurement of absorbance and fluorescence response by absorbance spectrophotometry and fluorometry prior to microscopy.

14. The method according to any one of claim 10 to 13, wherein the antimicrobial agent is an antibacterial agent or an antibiotic agent, or is ciprofloxacin, or kanamycin.

15. A microbial biofilm cultivated according to any one of claims 1 to 9 for use in a method of screening one or more antimicrobial agent(s) for efficacy against the microbial biofilm.

16. The microbial biofilm according to claim 15, wherein the method of screening is as described in any one of claims 10 to 14.

17. A kit for cultivating a microbial biofilm as described in any one of claims 1 to 9, the kit comprising:A. a cultivation vessel having a solid substrate;B. agar;C. liquid microbiological media;D. optionally, one or more microorganism(s) to be cultivated;E. one or more stain(s) for measurement of one or more microbial component(s) and / or characteristic(s) including microbial cell viability; and F. optionally, instructions for use of the kit.

18. A kit for screening one or more antimicrobial agent(s) for efficacy against a microbial biofilm cultivated according to any one of claims 1 to 9, the kit comprising:I. a cultivation vessel having a solid substrate;II. agar;III. liquid microbiological media;IV. optionally, one or more microorganism(s) to be cultivated;V. one or more stain(s) for measurement of one or more microbial component(s) and / or characteristic(s) including microbial cell viability;VI. optionally, one or more antimicrobial agent(s); andVII. optionally, instructions for use of the kit.

19. The kit according to either claim 17 or 18, wherein the one or more stain(s) are selected from the group comprising or consisting of DAPI (4',6-diamidino- 2- phenylindole), CV (crystal violet), resazurin, tetrazolium dye or MTT (3-(4,5- dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide), an organometallic ruthenium chelate, CBB (Coomassie Brilliant Blue), or 5,5'-(6,22-Dioxo- 11,14,17-trioxa-7,21-diazaheptacosane-1,27-diyl)bis(3,8-diamino-6- phenylphenanthridin-5-ium) diiodide; or consists of resazurin, or DAPI or both.

Images