Comparative In Vitro Staphylococcus aureus Biofilm Evaluation on 3D-Printed Polylactic Acid and Polyethylene Terephthalate Glycol-modified Surfaces

Summary

This study evaluates biofilm formation on 3D-printed polylactic acid (PLA) and polyethylene terephthalate glycol (PETG) using static, semi-static, and dynamic in vitro models. The methodology supports safer medical device development and advances biofilm research in 3D-printed biomaterials.

Abstract

Three-dimensional (3D) printing has accelerated the development of customized medical devices, but biofilm formation on printed materials remains a major threat to implant safety and performance. Because material chemistry and print-dependent surface features can influence bacterial attachment and antibiotic tolerance, standardized in vitro approaches that enable meaningful comparisons across commonly used 3D-printing polymers are needed. Here, we compare biofilm development on polylactic acid (PLA) and polyethylene terephthalate glycol (PETG) using a multi-model in vitro evaluation framework that captures complementary aspects of biofilm biology, including early adhesion, maturation, and antimicrobial tolerance, under both static and flow conditions. S. aureus biofilms were established and assessed using quantitative (viable bacterial load and minimal biofilm eradication concentration [MBEC]) and qualitative (scanning electron microscopy) endpoints. Both PLA and PETG supported biofilm formation across models; however, PLA tended to show higher early adhesion and greater biofilm density. In static assays, PLA demonstrated higher CFU values than PETG, whereas vancomycin MBEC values were similar between materials. Assay-dependent differences in MBEC were observed across platforms, underscoring how model structure can influence apparent antimicrobial susceptibility. Under dynamic flow, biofilm burden increased relative to static conditions, with minimal material-dependent differences. Collectively, these results highlight the susceptibility of both polymers to biofilm formation and demonstrate the value of a multi-model framework for evaluating material-associated biofilm behavior and benchmarking antimicrobial performance on 3D-printed device-relevant substrates.

Introduction

Recent advances in 3D printing have led to the growing adoption of 3D-printed medical devices due to their cost-effectiveness, high precision, complex structural design, efficient material utilization, and patient-specific design^1,2. These advancements are driving significant clinical progress in the development of customized orthopedic^3,4,5,6, craniofacial^7,8, dental^1,9, and spinal^10,11 implants.

Numerous techniques and materials now allow precise control over design parameters, making it possible to optimize surface characteristics and enhance overall device functionality^2,12,13. The implant’s surface is critically important, as it serves as the main interface with the host environment and influences immune responses, bacterial adhesion, and tissue integration—factors essential for the device’s long-term performance^14,15. However, this same interface also presents a major challenge, as infections associated with implantable medical devices can lead to severe complications and are often difficult to treat due to the ability of pathogens to form surface-attached biofilms^16,17,18,19.

In this context, in vitro testing of biofilm models on materials intended for the manufacture of implantable medical devices represents a fundamental step in the risk assessment and product development process²⁰. These assays enable the investigation of the material’s susceptibility to microbial adhesion, biofilm formation, and maturation, as well as the evaluation of the effectiveness of antimicrobial strategies or surface modifications aimed at preventing microbial growth.

A further limitation of many in vitro biofilm assays is that they are frequently performed on a single, standardized substrate (often one polymer or a generic laboratory material), implicitly assuming that results are transferable across device materials²¹. This is problematic for 3D-printed medical devices, where polymer chemistry and manufacturing-dependent surface features can vary substantially and may directly influence bacterial attachment, biofilm maturation, and antimicrobial tolerance²². Consequently, evaluating biofilm behavior on the actual polymers used in fabrication, rather than extrapolating from a single reference substrate, addresses an important gap in current testing practice.

Because the implant surface is the key interface with the host, it strongly influences bacterial adhesion and biofilm formation, which underpin many device-associated infections and treatment failures. We hypothesized that commonly used 3D-printed polymers differ in their ability to support S. aureus biofilm development and in their susceptibility to vancomycin-mediated biofilm eradication. Accordingly, we performed a qualitative and quantitative evaluation of biofilm formation on 3D-printed material models. In the present study, a qualitative and quantitative evaluation of biofilm formation on 3D-printed material models was performed.

Protocol

All hazardous chemicals (e.g., glutaraldehyde, sodium cacodylate buffer, Hexamethyldisilazane [HMDS], and ethanol) were handled in accordance with institutional Environmental Health and Safety (EHS) policies. Preparation and use of volatile/toxic reagents were performed in a certified chemical fume hood with appropriate personal protective equipment (lab coat, eye protection, and chemically resistant gloves), and flammable solvents were kept away from ignition sources. Chemical wastes were segregated and collected in properly labeled containers (including dedicated arsenic-containing waste for cacodylate and solvent waste for HMDS/ethanol) for disposal through the institutional hazardous-waste program. Biohazardous materials (bacterial cultures, contaminated plates, and disposables) were decontaminated according to biosafety procedures (e.g., autoclaving or approved disinfectants) prior to disposal, and any ethylene oxide sterilization was conducted in accordance with certified facility protocols and required aeration practices.

3D printed models
The models were designed using free computer-aided design software and printed using a 3D printer equipped with fused deposition modeling technology, with 1.75 mm (diameter) filaments of polylactic acid (PLA) and polyethylene terephthalate glycol (PETG) (Table of materials). The printing parameters used for fabricating the specimens were as follows: nozzle (0.4 mm) movement speed of 1800 mm/min; first layer speed of 300 mm/min; nozzle temperature of extrusion of 220 °C (PLA) and 255 °C (PETG); bed temperature of 60 °C; layer height of 0.08 mm; extrusion width of 0.48 mm and; infill density of 100% for all samples. None of the samples required the use of support structures. 3D models consisted of discs, flat-bottom plates, the Calgary biofilm device (CBD), and the modified Robbins device (MRD). The discs were designed as a cylinder with a 6 mm (diameter) x 6 mm (height).

For the CBD, a model was created (Table of materials). The CBD consisted of two parts: a base plate, a standard 96-well microtiter plate (ANSI/SLAS format), and a lid (peg plate): a polymer (PLA or PETG) lid with 96 cylindrical pegs with 2 mm (diameter) and 5 mm (height). The MRD was developed with 4 mm (diameter and height) pins fixed in a screw-thread model (Table of materials). After completing the printing process, the models were carefully removed from the build plate, visually inspected for any deformations, and subsequently sterilized with ethylene oxide.

Disc model of static biofilm
This method was used to induce biofilm formation on PLA using the previously described 6 mm × 6 mm cylindrical discs²³ (Figure 1).

Figure 1: PLA and PETG discs and a 24-well plate for biofilm formation on these discs’ models. (A) Representative 3D-printed PLA and PETG discs prior to bacterial inoculation and biofilm formation. (B) PLA and PETG discs positioned in a 24-well plate during biofilm induction under in vitro culture conditions. Abbreviations; PLA = polylactic acid; PETG = polyethylene terephthalate glycol modified. Please click here to view a larger version of this figure.

S. aureus was stored at −80 °C in tryptic soy broth (TSB) supplemented with 15% glycerol and, prior to each experiment, was cultured on blood agar and incubated for 24 h at 37 °C. An isolated colony was resuspended in 10 mL of TSB supplemented with 50 mM glucose, added to a 50 mL conical tube, and incubated for 24 h at 37 °C in an orbital shaker at 120 rpm. The resulting culture was centrifuged for 5 min at 200 x g, and the supernatant was discarded. The pellet was washed three times with 10 mL of sterile 0.9% NaCl. In the final washing step, the pellet was resuspended in 10 mL of fresh TSB to obtain a bacterial suspension with a turbidity equivalent to the 0.5 McFarland standard (approximately 1.5 × 10⁸ CFU/mL), as determined by nephelometry. The suspension was then diluted 1:10 in 10 mL of TSB in a 15 mL conical tube to achieve a final bacterial concentration of 10⁷ CFU/mL. For experiments with discs, 2 mL of the previous bacterial suspension was added to each well of sterile 24-well plates to cover each group of discs (PLA and PETG) for 2 h to allow cell adhesion. The discs were then aseptically transferred using sterile tweezers to a new sterile 24-well plate containing 2 mL of 0.9% NaCl to remove planktonic cells. Fresh 10 mL of TSB was added to the wells, and the plates were incubated at 37 °C for 24 h. During this step, cells adhering to the disc surfaces formed a biofilm. Following incubation, the culture medium was aspirated, and the discs were washed with 2 mL of 0.9% NaCl to remove residual medium and non-adherent cells. A 24-well plate was prepared with 1 mL of vancomycin solution at serial dilutions (2–4,096 mg/L), and the discs were submerged for 24 h. A control group without vancomycin (NaCl 0.9%) was included. After 24 h, the discs were washed with 2 mL NaCl 0.9%.

To quantify adhered bacteria on the discs, each disc was transferred using sterile tweezers to a microtube containing 1 mL of sterile 0.9% NaCl and sonicated in an ultrasonic bath for 15 min at 40 kHz and 37 °C to detach the adhered cells. The microtubes were vortexed for 60 s to ensure complete disaggregation of cell clusters. Aliquots of 100 µL of the undiluted sample (10°) and 10⁻² and 10⁻⁴ dilutions in sterile 0.9% NaCl were plated on TSA and incubated for 24 h at 37 °C. Bacterial growth was quantified by counting CFU on a plate and expressed as CFU/mL.

24-well flat-bottom plates for minimal biofilm eradication concentration
For experiments in 24-well plates, 3 mL of bacterial suspension (prepared as described above) was added to each well for 2 h to allow adequate cell adhesion. Afterwards, the bacterial suspension was aspirated, and 3 mL of sterile 0.9% NaCl was added to remove planktonic cells. Fresh 3 mL of TSB was then added, and the plates were incubated at 37 °C for 24 h. During this step, cells adhering to the well surfaces formed a biofilm. Following incubation, the TSB was aspirated, and the wells were washed with 3 mL of 0.9% NaCl to remove residual medium and non-adherent cells. 3 mL of vancomycin solution at serial dilutions (2–4,096 mg/L) was added to each well. A control group without vancomycin (NaCl 0.9%) was included.

After 24 h, the solution was removed, and each well was gently washed twice with 3 mL of 0.9% NaCl. Three milliliters of 0.9% NaCl were added to each well, the plate was capped and sealed with parafilm, and then sonicated in an ultrasonic bath for 15 min at 40 kHz and 37 °C to detach the adhered biofilm. Aliquots of 100 µL were collected from the wells, and the undiluted sample (10°) and 10⁻² and 10⁻⁴ dilutions prepared in sterile 0.9% NaCl were plated on TSA and incubated for 24 h at 37 °C. Bacterial growth was then quantified by counting CFU/mL. All experiments were performed in triplicate. Minimal biofilm eradication concentration (MBEC) was defined to be a reduction of at least 3 log₁₀ in comparison with the control group (Figure 2).

Figure 2: Printed 24-well plates. (A) 24-well plates printed with PLA. (B) 24-well plates printed with PETG. Abbreviations; PLA = polylactic acid; PETG = Polyethylene terephthalate glycol. Please click here to view a larger version of this figure.

The Calgary biofilm device
The CBD consists of a 96-well microtiter plate and a lid with 96 pegs, each designed to fit precisely into a well when the lid is placed on the plate. To induce biofilm formation, each well of a sterile 96-well plate was filled with 200 µL of the bacterial suspension, and sterile peg lids were then placed onto the plate. The assembled plates were incubated at 37 °C for 24 h under static conditions to allow biofilm development on the pegs. Following incubation, the peg lids were gently rinsed in a 96-well plate containing 200 µL of 0.9% NaCl to remove planktonic cells. The plate with the pegs was submerged in another 96-well plate containing vancomycin at progressive concentrations (2–4,096 mg/L) and incubated at 37 °C for 24 h under static conditions.

The plate with pegs was rinsed in a 96-well plate containing 200 µL of 0.9% NaCl, and the pegs were carefully stamped in a 15 cm TSA plate and incubated at 37 °C for 24 h. Bacterial growth was observed, and MBEC was defined as the lowest concentration at which no growth occurred. All experiments were performed in triplicate (Figure 3).

Figure 3: Calgary device and 96-well flat-bottom plates printed with PLA and PETG. (A) Calgary device printed with PLA. (B) 96-well flat-bottom plates printed with PLA. (C) Calgary device printed with PETG. (D) 96-well flat-bottom plates printed with PETG. Abbreviations; PLA = polylactic acid; PETG = polyethylene terephthalate glycol. Please click here to view a larger version of this figure.

A dynamic system under continuous conditions of flow
Each MRD contains five individual pegs fixed in a screw-thread model and arranged linearly along a channel with a rectangular cross-section. The dynamic system consisted of a flask containing 1 L of solution with the antibiotic to be tested connected to a peristaltic pump via a silicone tube coupled to the MRD. Assembly was done in a laminar air flow cabinet to prevent contamination. The medium flow rate was maintained at 1 mL/min and circulated through the system for 24 h. The fluid passing through the MRD is discarded into another glass flask (without recirculation).

S. aureus biofilms were induced on the pegs of an MRD, dropping the pegs on 24-well plates with TSB with 3 mL at a concentration of 107 CFU/mL. The pegs were gently rinsed by immersion in a 24-well plate filled with 3 mL of 0.9% NaCl solution. The pegs were carefully screwed into the MRD (n = 5). Vancomycin solution (40 mg/L) diluted in 0.9% NaCl was infused through the MDR using the pump as described above for 24 h. A control group without antibiotics (0.9% NaCl) was included. This concentration was chosen based on the serum concentration obtained during high doses of intravenous vancomycin, as previously described²⁴. The pins were aseptically removed from the MRD using sterile tweezers, placed in microtubes containing 1 mL of sterile 0.9% NaCl after washing with 0.9% NaCl, and sonicated in an ultrasonic bath for 15 min at 40 kHz and 37 °C to ensure complete detachment of the biofilm. The microtubes were vortexed for 60 s, and the procedure was repeated three times to disrupt residual biofilm aggregates. Aliquots of 100 µL were plated on TSA, and the plates were incubated for 24 h at 37 °C. Bacterial growth was quantified and expressed as CFU/mL (Figure 4).

Figure 4: Modified Robbins device with PETG and PLA and a photo of the dynamic system. (A) modified Robbins device printed with PETG or PLA. (B) Photo of the dynamic system. Abbreviations; PLA = polylactic acid; PETG = polyethylene terephthalate glycol. Please click here to view a larger version of this figure.

Scanning electron microscopy
To visualize the morphology and structure of the biofilms, scanning electron microscopy (SEM) was performed on the discs. Biofilm was induced on discs as described in the protocols. The discs were carefully removed from the plate using sterile tweezers and were fixed with a solution (0.68 g sucrose, 0.42 g sodium cacodylate, 0.6 mL 30% glutaraldehyde), with 19.4 mL of deionized water, for 45 min. Subsequently, the specimens were transferred to a buffer solution (0.68 g sucrose and 0.42 g sodium cacodylate) for 10 min. The samples were then dehydrated in a graded ethanol series (35%, 50%, 70%, and 100%) for 10 min each. Afterwards, the specimens were transferred to another Petri dish and covered with 100% hexamethyl disilazane for 10 min. Finally, the discs were sputter-coated with gold using a rotary pump and mounted on a metal stub for observation under different magnifications on a scanning electron microscope (Figure 5).

Figure 5: Scanning electron microscopy of PLA and PETG 3D printed models with and without biofilm induction. (A) Scanning electron microscopy photograph of PLA without biofilm. (B) Scanning electron microscopy photograph of PETG without biofilm. (C) Scanning electron microscopy photograph of biofilm-forming bacteria on PLA discs. (D) Scanning electron microscopy photograph of biofilm-forming bacteria on PETG discs. C and D show typical features of biofilms: bacterial organization into a 3D structure, adherence to surfaces, and the presence of an extracellular matrix. White scalebar represents 10 µm. Abbreviations; PLA = polylactic acid; PETG = Polyethylene Terephthalate Glycol. Please click here to view a larger version of this figure.

Statistical analysis
Some analyses were descriptive. For quantitative data, the sample size justified the use of the median with the interquartile range, with the Mann-Whitney test for statistical comparison. A p-value < 0.05 was considered statistically significant.

Results

Disc model of static biofilm
This assay is used to determine the MBEC on individual specimens. Progressive two-fold dilutions of vancomycin demonstrated baseline biofilm formation of 7.48 × 10⁵ CFU/mL [IQR 6.34–8.72] for PLA and 3.60 × 10⁵ CFU/mL [IQR 3.08–5.52] for PETG. Although both materials showed biofilm densities within the 10⁵ range, a statistically significant difference in biofilm production was observed between them (p = 0.015). When exposed to vancomycin, ...

Discussion

This study proposes a comprehensive in vitro evaluation of biofilm formation on 3D-printed medical-grade polymers, specifically PLA and PETG, using four complementary experimental platforms: static disc assays, 24-well plate models, the CBD, and a dynamic flow system based on an MRD. By integrating these models, the study aims to characterize how distinct printing materials behave when exposed to S. aureus biofilm induction, and to determine the MBEC of vancomycin across different test conditions. This ...

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
0.9% NaCl	Fresenius Kabi Brasil, Ceara, Brazil	17811
3D printer with Fused Deposition Modeling technology (Bambu X1-Carbon)	Bambu Lab, Shenzhen, China	PF001-P
Blood agar	NewProv, Paraná, Brazil	PA31
Calgary device for biofilm downloadable	No company	See link (without catalogue numbers)	https://www.printables.com/model/1464751-calgary-model-biofilm
Cell culture multiwell plate, 24 well	Greiner Bio-One, Kremsmünster Österreich, Austria	662160
Centrifuge (5910 Ri)	Eppendorf, Hamburg, Germany	5943000511
Centrifugue tube sterile 15 mL	Corning, Missouri, USA	430790
Centrifugue tube sterile 50 mL	Corning, Missouri, USA	430291
Electronic Pipette Controller (Easypet® 3)	Eppendorf, Hamburg, Germany	4430000018
Ethanol (≥99.5%, ACS reagent)	Sigma-Aldrich, Missouri, USA	459844
Glutaraldehyde	Merck, Darmstadt, Germany	8143931000
Hexamethyldisilazane	Merck, Darmstadt, Germany	440191
Incubator (502/4-C)	Fanem, São Paulo, Brazil	502007700
Laminar Air Flow cabinet (BioSEG 9)	Veco, São Paulo, Brazil	1EAC3770
Mechanical pipette (Eppendorf Reference® 2)	Eppendorf, Hamburg, Germany	4924000061
Modified Robbins Device downloadable	No company	See link (without catalogue numbers)	https://www.printables.com/model/1481700-robbins-device-modified
Nephelometer	Alfakit, Florianópolis, Brazil	7718
Orbital shaker (INNOVA 44R)	Eppendorf, Hamburg, Germany	M1282-0314
Parafilm M	Amcor, Illinois, USA	PM-996
Peristaltic pump (Minipuls® 3)	Gilson, Villiers-le-Bel, France	F117604
Petri dish (90 x 15 mm)	Greiner Bio-One, São Paulo, Brazil	630164
Pipette tip 1–200μL	Corning, Missouri, USA	CLS4142
Polyethylene terephthalate glycol (PETG)	Creality, China	44759
Polylactic acid (PLA) Ultrafuse® PLA PRO1	BASF, Maarssen, Netherlands	PR1-7501A075
Rotary pump (Q150R ES Plus)	Quorum Technologies, Lewes, UK	13034
Scanning electron microscope (Vega3 LMU)	Tescan, Brno, Czech Republic	See link (without catalogue numbers)	https://tescan.com/product-portfolio
Serological pipette 10 mL	Corning, Missouri, USA	CLS4488
Sodium cacodylate	Sigma-Aldrich, Missouri, USA	C0250
Staphylococcus aureus ATCC® 25923 strain	Laborclin, Pinhais, Brazil	660674
Sucrose	Sigma-Aldrich, Missouri, USA	59378
The 24-well flat-bottom downloadable	No company	See link (without catalogue numbers)	www.printables.com/model/104339-24-well-plates
Tryptic soy agar	Kasvi, Madrid, Spain	K25-1068
Tryptic soy broth	Sigma-Aldrich, Missouri, USA	22092
Ultra Low Freezer ( IULT 335 D)	Indrel, Paraná, Brazil	12038
Ultrasonic bath (Soniclean 15)	Sanders Medical, Minas Gerais, Brazil	See link (without catalogue numbers)	https://sandersdobrasil.com.br/produto/lavadora-ultrassonica-soniclean-15d/
Vancomycin	ABL, São Paulo, Brazil	15,56,20,041
Vortex (FLEXVORTEX)	Loccus, São Paulo, Brazil	4500