Into the well-A close look at the complex structures of a microtiter biofilm and the crystal violet assay

Abstract

The microtiter assay is one of the most widely used methods for assessing biofilm formation. Though it has high throughput, this assay is known for its substantial deviation from experiment to experiment, and even from well to well. Since the assay constitutes one of the pillars of biofilm research, it was decided to examine the wells of a microtiter plate directly during growth, treatment, and the steps involved in crystal violet (CV) measurements. An inverted Zeiss LSM 880 confocal laser scanning microscope was used to visualize and quantify biomass directly in the wells of the microtiter plate. Green fluorescent protein-tagged Pseudomonas aeruginosa, PAO1, and live/dead stains were used to assess the structure, state, and position of biomass build-up. Microscopic observations were compared with colony-forming unit (CFU) and CV measurements. The development and the structured architecture of biomass was observed in real-time in the wells. Three-dimensional images of biomass were obtained from all of the microtiter wells; these showed variations from well to well. CV staining showed large variations in remaining biomass, depending on the method selected to remove the supernatant prior to CV staining (i.e. pipetting or manually discarding the fluid by inversion, washed or unwashed wells). Colony-forming unit counts or live/dead staining used to evaluate biomass with or without antibiotic treatment proved imprecise due to aggregation, limited removal of biomass, and overestimation of dead staining. The highly structured microenvironment of biomass in microtiter wells needs to be considered when designing and analyzing experiments. When using microtiter plates, stochastic variation due to growth and handling may lead to flawed conclusions. It is therefore recommended that this assay be used as a screening tool rather than as a stand-alone experimental tool.

Keywords: Biofilm; Confocal laser scanning microscopy; Crystal violet; In vitro validation; Microtiter assay; Pseudomonas aeruginosa.

Fig. 1

Representative 3D projections of undisturbed P. aeruginosa biomass at the bottom of microtiter wells grown for 24, 48, and 72 h. Images consist of 7 × 7 stitched z-stack images with 100x magnification. A–C) A 24-h old biomass. D–F) A 48-h old biomass. G–I) A 72-h old biomass. J-K) Zoom in on structures present in 72 h old biomass.

Fig. 2

A-C) 3D projections of three neighboring wells each containing 72-h old biofilm. All three wells were inoculated, grown, and imaged in the same manner.

Fig. 3

A) Quantitative measurements of undisturbed P. aeruginosa biomass in 24-, 48-, and 72-h old wells. There was significantly more biomass (μm [3]) in 72-h old wells than in 24- and 48-h old wells. B) CFU count from undisturbed wells at the three timepoints. C–E) 3D projections of undisturbed P. aeruginosa biomass stained with live/dead stain. F–H) Side view of P. aeruginosa biomass stained with live/dead stain.

Fig. 4

White columns represent CFU counts from removed biomass in microtiter wells grown for 24, 48, or 72 h. The wells were scraped and ultra-sonicated to remove the biomass. The removed biomass was then degassed and ultra-sonicated before serial dilution was conducted prior to CFU determination. Prior to serial dilution, each sample was evaluated for the amount of biomass bound in aggregates larger than 100 μm³. The fraction of the total biomass bound in aggregates is represented in the gray columns.

Fig. 5

Removal of total biomass for the CFU count by means of scraping or both scraping and ultra-sonication. A) 3D projections of remaining biomass after thorough scraping. B) 3D projections of remaining biomass after thorough scraping followed by ultra-sonication. Images consist of 7 × 7 stitched z-stack images with a 100x magnification. C) Quantitative measurements of biomass left in the wells after scraping or both scraping and ultra-sonication, compared to the undisturbed control. D) The number of CFUs removed by scraping alone, ultra-sonication alone, or by the two methods combined.

Fig. 6

Representative 3D projections of 24-h old biomass remaining at the bottom of microtiter wells after the supernatant had been removed as part of the crystal violet staining procedure, by means of inversion, pipetting, or pipetting followed by a rinse with saline. Images consist of 7 × 7 stitched z-stack images with a 100x magnification. A–B) Biomass remaining after the removal of the supernatant by means of inversion. C–D) Biomass remaining after removal of the supernatant by means of pipetting. E–F) Biomass remaining after rinsing with saline. G) Quantitative measurements of 24-h old biomass either undisturbed or remaining at the bottom of microtiter wells after the supernatant had been removed as part of the crystal violet staining procedure, by means of inversion, pipetting, or pipetting followed by a rinse with saline. H) OD₅₉₅ measurement of crystal violet staining of remaining biomass after the supernatant was removed either by inversion or pipetting followed by a rinse with saline. I) CFU count for the biomass in undisturbed wells or remaining in the wells after inversion or pipetting and rinse. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

Fig. 7

3D projections of 24-h old biomass treated with 10 μg mL⁻¹ tobramycin and stained with live/dead stains. A) The biofilm 1 h after the beginning of treatment. B) The biofilm 3 h after the beginning of treatment. C) The biofilm 5 h after the beginning of treatment. D) The biofilm 17 h after the beginning of treatment. E) Side view of 17-h old treated biofilm, also shown in 5D. Zoom showing the top layer of PI-stained biomass on top of Syto9-stained biomass. F) Fraction of biomass stained green with Syto9. Biomass of both PI- and Syto9-stained cells were quantified, and the part stained with Syto9 was taken as a fraction of the total. G) CFU after 17 h of Tobramycin (10 μg mL⁻¹) treatment or untreated biomass in microtiter wells grown 24, 48, or 72 h before treatment. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

Fig. 8

Schematic drawing of scraping procedure. A) Firstly, all surfaces of the walls were scraped in an up-and-down motion while pumping with the pipet. B) Secondly, the bottom were scraped in a back-and-forth motion while pumping with the pipet. C) Finally, the edge between the walls and bottom and the corners, were scraped while pumping the pipet, before the supernatant was removed, followed by another suction to insure removal of all fluid., followed by ultra-sonication (230 VAC, Branson, USA) of the whole plate in a regime of 5 min degassing followed by 5 min of ultra-sonication. The degassing step was used to remove excess gas build-up and bubbles in the samples, which could otherwise inhibits the effect of the ultra-sonication treatment. Biomass was subsequently pipetted out, before it was 10-fold serial diluted and plated on LB plates (1.5% agar). The effectiveness of biomass removal by means of both scraping and ultra-sonication was evaluated using CLSM, as described above.