Relaxation-weighted MRI analysis of biofilm EPS: Differentiating biopolymers, cells, and water

Abstract

Biofilms are a highly complex community of microorganisms embedded in a protective extracellular polymeric substance (EPS). Successful biofilm control requires a variety of approaches to better understand the structure-function relationship of the EPS matrix. Magnetic resonance imaging (MRI) is a versatile tool which can measure spatial structure, diffusion, and flow velocities in three dimensions and in situ. It is well-suited to characterize biofilms under natural conditions and at different length scales. MRI contrast is dictated by T ₁ and T ₂ relaxation times which vary spatially depending on the local chemical and physical environment of the sample. Previous studies have demonstrated that MRI can provide important insights into the internal structure of biofilms, but the contribution of major biofilm components-such as proteins, polysaccharides, and cells-to MRI contrast is not fully understood. This study explores how these components affect contrast in T ₁ -and T ₂ -weighted MRI by analyzing artificial biofilms with well-defined properties modeled after aerobic granular sludge (AGS), compact spherical biofilm aggregates used in wastewater treatment. MRI of these biofilm models showed that certain gel-forming polysaccharides are a major source of T ₂ contrast, while other polysaccharides show minimal contrast. Proteins were found to reduce T ₂ contrast slightly when combined with polysaccharides, while cells had a negligible impact on T ₂ but showed T ₁ contrast. Patterns observed in the model biofilms served as a reference for examining T ₂ and T ₁ -weighted contrast in the void spaces of two distinct AGS granules, allowing for a qualitative evaluation of the EPS components which may be present. Further insights provided by MRI may help improve understanding of the biofilm matrix and guide how to better manage biofilms in wastewater, clinical, and industrial settings.

Keywords: AGS; Biofilm; EPS; MRI; NMR.

Fig. 1

Cross-sectional T_2,eff maps of alginate model biofilms with increasing concentration of polysaccharide (1, 2 and 3% from A-C, respectively). The calculated T_2,eff times are color-coded, with the shorter relaxation times, corresponding to the gel phase, in blue and longer relaxation times, corresponding to bulk water outside the gel, in yellow/green. Alginate concentration is strongly correlated with T₂ relaxation in the gel phase but does not influence bulk water relaxation. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

Fig. 2

T_2,eff distribution profile of alginate model biofilms from Fig. 1.

Fig. 3

T_2,eff maps of mixed polysaccharide model biofilms consisting of gellan (G) and alginate (A) (2% gellan, G/A = 3, G/A = 1, G/A = 1/3, and from A-D, respectively). Total biopolymer concentration is fixed at 2% w/w in all samples. From A-D, as the concentration of gellan decreases, there is more T₂ contrast. The images highlight the sensitivity of MRI to alginate concentrations in mixed gels.

Fig. 4

T_2,eff distribution profile of mixed polysaccharide model biofilms from Fig. 3.

Fig. 5

T_2,eff maps of alginate (PS, i.e. polysaccharide) model biofilms with increasing concentration of protein (PN) (2% alginate, PN/PS = 1, and PN/PS = 2 from A-C, respectively). Alginate concentration is fixed at 2% w/w in all samples. PN did not increase T₂ contrast as expected, instead, PS-PN interactions decreased T₂ MRI contrast.

Fig. 6

T_2,eff distribution profile of alginate/protein model biofilms from Fig. 5.

Fig. 7

LHS: T_2,eff maps of alginate with and without 20% v/v cells (A and B, respectively). RHS: Corresponding T₁-weighted images of alginate samples with and without cells (C–F and G-J, respectively), with each image collected at different echo times, i.e. 8, 24, 48, and 72 ms from left to right. The addition of cells shows clear T₁ contrast, but T_2,eff relaxation times are unaffected.

Fig. 8

T_2,eff distribution profiles of alginate samples with and without cells from Fig. 7 (A-B). Image of 10% v/v cells encapsulated in alginate is not shown in Fig. 7, though included in this distribution profile.

Fig. 9

T₁-weighted signal intensity distribution profiles of alginate samples with and without cells from Fig. 7 (C & G), taken from the first echo at 8 ms. Image of 10% v/v cells encapsulated in alginate is not shown in Fig. 7, though included in this distribution profile.

Fig. 10

LHS: T_2,eff map of a full-scale Vroomshoop AGS granule. The regions of interest (ROI) compare a void space inside the granule (ROI 1) with bulk water outside of the granule (ROI 2). RHS: Corresponding T₁-weighted image showing the same ROI.

Fig. 11

T_2,eff distribution profiles of Vroomshoop AGS granule from Fig. 10 (A), showing enhanced T₂ contrast inside the void (ROI 1). This relaxation behavior could be attributed to lightly crosslinked polymer.

Fig. 12

T₁-weighted signal intensity distribution profiles of Vroomshoop AGS granule from Fig. 10 (B), showing enhanced T₁ contrast inside the void (ROI 1). This signature could be attributed to the presence of cells.

Fig. 13

LHS: T_2,eff map of a full-scale Utrecht AGS granule. The regions of interest (ROI) compare a void space inside the granule (ROI 1) with bulk water outside of the granule (ROI 2). RHS: Corresponding T₁-weighted image showing the same ROI.

Fig. 14

T_2,eff distribution profiles of Utrecht AGS granule from Fig. 13 (A), showing similar T₂ contrast between the void and bulk water (ROI 1 and 2, respectively). This relaxation behavior suggests the void is empty or contains very dilute polymer.

Fig. 15

T₁-weighted signal intensity distribution profiles of Utrecht AGS granule from Fig. 13 (B) showing enhanced T₁ contrast inside the void (ROI 1). This signature could be attributed to the presence of cells.