Results | UCLouvain

Controlled aggregation

1. Fluorescent bacteria

To distinguish genetically different strains during controlled aggregation assays, we generated fluorescently tagged variants. While GFP is used as the adhesion “glue,” we constructed additional strains expressing either mTurquoise (blue) or mScarlet (red) fluorescent proteins. This allows us to track strain identity and monitor mixed populations in aggregation experiments.

Schematic of fluorescence constructs — **Figure 1.** Fluorescence construction (mTurquoise was also replaced by mScarlet and mNeonGreen).

Fluorescent strain 1 — **Figure 2.** LMD9 transformed with mTurquoise (left) and mScarlet (right).

Fluorescent strain 2 — **Figure 2.** LMD9 transformed with mTurquoise (left) and mScarlet (right).

2. Aggregation assay

To harness metabolic interdependence between genetically distinct strains, it is essential to control their spatial organization. By minimizing the distance between complementary strains, we can ensure efficient metabolite exchange and functional cooperation. For this purpose, we use an αRep and a nanobody recognizing GFP as adhesion modules, enabling controlled aggregation. In addition, strains were engineered to express distinct fluorescent markers: mTurquoise (blue) or mScarlet (red), which allow us to distinguish between different populations and confirm their spatial proximity.

The αRep and nanobody constructs were first integrated via natural transformation in LMD9 (see natural transformation protocol). To know if our construct was integrated in the genome we used antibiotic resistance. Our construct was composed of the gene coding the protein of interest but also the gene conferring the antibiotic resistance (here to chloramphenicol).

Integration of the constructions verification — **Figure 3.** Colonies after overnight growth on M17g-chloramphenicol (20 µg/mL). In the center: serial dilutions 10by 10, from up (10⁰) to bottom (10^-7) of each transformation performed, which are insertion of an alpharep, nanobody, fucanase in a WT LMD-9 background, from left to right. On the left: controls. A-rep, Nano, Fuca are LMD-9 which received sXIP to induce natural competence but not their respective DNA fragment. Medium is only the medium (milk) in which we performed the transformation.

At this point, we knew that our constructs were well integrated in the genome of the bacteria. What we also wanted to verify was the spot of integration but also their length. To do so, we did a DNA extraction after all the transformation steps. From this DNA extraction, we did a polymerase chain reaction with primers that anneal themselves on the specific region that we wanted to verify.

PCR verification 1 — **Figure 4.** Agarose gel as a proof that the constructions were inserted at the right spot in the genome and that they were completely inserted. The left band being our Nanobody construction and the right band being our Alpha-rep construction. Both band are at 5000bp.

After that, we performed a visual aggregation test, which consists of looking at our cells under epifluorescence microscopy after incubating our two subpopulations for 30 minutes in the presence or absence of GFP. The results were quite convincing. We observed aggregation between red and blue strains, but we also observed aggregation between strains that had the same colour. With an increase of GFP concentration, we observed more aggregation. We also need to be critical about our results because we couldn’t see any ring of GFP (which is the phenotype that we were expecting).

The reason why we didn’t see the GFP could also be the fact that CFP and GFP signals overlap. Thus, it might be that with the epifluorescence microscopy, the resolution isn’t high enough to distinguish CFP and GFP. For this issue, we will probably change the signal that we use from EGFP to YFP. This will allow us to potentially avoid any overlap between signals.

Due to this, we are not sure that our constructs are displayed at the surface. Thus, further controls and tests (e.g. Ri03, aggregation test with LMD9 where HtrA is deleted) will have to be done to confirm that the increase of aggregation is due to our tripartite system.

Aggregation trial without GFP. — **Figure 5.** Epifluorescence microscopy of, in red, LMD9 nano-body and, in cyan, LMD9 α-rep. This sample does not contain any GFP.

Aggregation trial with 10 nM GFP. — **Figure 6.** Epifluorescence microscopy of, in red, LMD9 nano-body and, in cyan, LMD9 α-rep. This sample contains 10 nM of GFP.

Aggregation trial with 100 nM GFP. — **Figure 7.** Epifluorescence microscopy of, in red, LMD9 nano-body and, in cyan, LMD9 α-rep. This sample contains 100 nM of GFP.

After converting our microscopy images to a map and extract the coordinates of every cell, we looked after the closest cell to every cell and extracted the distance separating them. We obtained data on 12977 cells in total. By grouping our data by GFP concentration in different samples, we obtained a significant decrease in the mean distance between neighboring cells upon the increase of GFP concentration.

Mean distance between neighboring cells grouped by GFP concentration — **Figure 8.** Mean distance between neighboring cells from 11 samples, grouped by the GFP concentration used during co-culture.
** = p ≤ 0.01; ******** = p ≤ 0.0001; ns = not significant between 0 and 10 nM of GFP.

These results indicate that upon GFP concentration increase, we observe a decrease in mean distance between neighboring cells that we can assimilate to our construction. But this does not ensure that our assumed aggregation is heterologous as expected. To further confirm this aggregation, we plot the same dataset by the identity of the origin cell (color) and the neighbor cell. This led to this plot:

**Figure 9.** Mean distance between neighboring cells depending on the pair. “From” (columns) indicates the color of the reference cell, and “to” the color of the neighboring cell.
n = 12 977; ns = no significant groups.

Even if this barplot shows you equivalent means, we are quite interested in its analysis. Ideally, we would want to see significant decrease in the mean distance in heterologous couples (mScarlet to mTurquoise and mTurquoise to mScarlet) or this is not the case here. The fact is that our statistics are biased by our own chassis. Indeed, “strepto” in Streptococcus thermophilus means chain. And like a chain, S. thermophilus cells are making a small thread of 3 to 4 cells (as observed in our microscopy results), genetically the same, implying that the neighboring cell of a cell inside a chain is, except correct aggregation, the neighbor of a cell of the same color. These data are so biased towards homogenous couples (mScarlet to mScarlet and mTurquoise to mTurquoise). Taking that into account, this non-significative barpot seems more hyping for our project.

Division of labour

For the division of labour, the first part was to transform our strain with the fucanase construct. This transformation was successful (see figure 3). After that, we verified the insertion site and the length of our construct.

PCR verification 2 — **Figure 10.** Agarose gel as a proof that the fucanase construction was inserted at the right spot with the right length. The fourth band at the right end of the gel is at 5000 bp which is the awaited length.

After that, we tested whether the fucanase was functional, i.e. whether it could degrade fucoidans into fucose. Unfortunately, this test did not yield conclusive results (see Engineering ).

Metabolic interdependence

1. Production of an LMD9 strain that can grow on galactose

To maintain balanced co-cultures with approximately 50% of each strain, we plan to exploit metabolic interdependence as a stabilizing strategy. The long-term goal is to engineer a Gal⁺ LacZ⁻ strain that depends on its partner for growth. At present, our construct is still Gal⁺ LacZ⁺, and future work will focus on eliminating LacZ to enforce the desired interdependence. This approach should ensure stable coexistence of the strains and allow robust cooperative behaviour in mixed cultures.

By plating LMD9 on M17 supplemented with galactose, we saw a colony that grew after 72 hours. We used Streptococcus salivarius as a first control since this bacterium is able to grow on galactose.

**Figure 11.** LMD9 on M17 galactose (left) and HSISS4 on M17 galactose (right).

**Figure 11.** LMD9 on M17 galactose (left) and HSISS4 on M17 galactose (right).

In the future, we will perform a PCR targeting the PrtS locus as a second control, since PrtS is quite specific to Streptococcus thermophilus. As a third control, could be a PCR and a sequencing of the gal locus of our single colony that we will compare to the gal locus of wild-type LMD9.

A fourth control, we will plate a fluorescent LMD9 strain; if colonies appear, we will confirm their fluorescence using an epifluorescence microscope.

Perspectives

In the future, for the division of labour using the fucanase and for the controlled aggregation using the Alpha-rep and nanobody, we are planning to use an inhibitor of the production of exopolysaccharides. This compound is Ri03 also known as 5-(4-Chlorophenyl)-2-furoic acid. It inhibits dTDP-L-rhamnose formation. This compound will allow us to have strains with less exopolysaccharides and having less of those is potentially required for our proteins to be well displayed. Before using this compound on our strain transformed with our constructs, we will have to determine the bacterial growth in presence of Ri03. This test will allow us to determine the right concentration and conditions to have the optimal effect on our strains. For the next steps, it would be interesting to use a confocal microscope to obtain higher-quality images and thus better visualize the GFP. We would also like to perform a sedimentation assay to gain another insight into cell aggregation. This test is carried out using a spectrophotometer — the faster the sample sediments, the more aggregated it is. Finally, for the metabolic interdependence module, we plan to delete the β-galactosidase gene in the gal⁺ strain so that it becomes lacZ⁻.