Detection and Quantification of Mono-Rhamnolipids and Di-Rhamnolipids Produced by Pseudomonas aeruginosa

Summary

Pseudomonas aeruginosa produces the rhamnolipid biosurfactants. Thin-layer chromatography detects and determines the proportion of mono- and di-rhamnolipids produced by each strain. Quantification of total rhamnolipids involves assessing rhamnose equivalents present in these biosurfactants extracted from the culture supernatants using the orcinol method.

Abstract

The environmental bacterium Pseudomonas aeruginosa is an opportunistic pathogen with high antibiotic resistance that represents a health hazard. This bacterium produces high levels of biosurfactants known as rhamnolipids (RL), which are molecules with significant biotechnological value but are also associated with virulence traits. In this respect, the detection and quantification of RL may be useful for both biotechnology applications and biomedical research projects. In this article, we demonstrate step-by-step the technique to detect the production of the two forms of RL produced by P. aeruginosa using thin-layer chromatography (TLC): mono-rhamnolipids (mRL), molecules constituted by a dimer of fatty acids (mainly C10-C10) linked to one rhamnose moiety, and di-rhamnolipids (dRL), molecules constituted by a similar fatty acid dimer linked to two rhamnose moieties. Additionally, we present a method to measure the total amount of RL based on the acid hydrolysis of these biosurfactants extracted from a P. aeruginosa culture supernatant and the subsequent detection of the concentration of rhamnose that reacts with orcinol. The combination of both techniques can be used to estimate the approximate concentration of mRL and dRL produced by a specific strain, as exemplified here with the type strains PAO1 (phylogroup 1), PA14 (phylogroup 2), and PA7 (phylogroup 3).

Introduction

Pseudomonas aeruginosa is an environmental bacterium and an opportunistic pathogen of great concern due to its production of virulence-associated traits and its high antibiotic resistance^1,2. A characteristic secondary metabolite produced by this bacterium is the biosurfactant RL, which is produced in a coordinated manner with several virulence-associated traits such as the phenazine pyocyanin, an antibiotic with redox activity, and the protease elastase³. The tensio-active and emulsification properties of RL have been exploited in different industrial applications and are currently commercialized⁴.

Most P. aeruginosa strains, belonging to phylogroups 1 and 2, produce two types of RL: mRL, which consists of one rhamnose moiety linked to a fatty acid dimer mainly of 10 carbons, and dRL, which contains an additional rhamnose moiety linked to the first rhamnose⁴ (see Figure 1). However, it has been reported that two minor P. aeruginosa phylogroups (groups 3 and 5) only produce mRL^5,6. The two types of RL contain a mixture of fatty acid dimers, which, as mentioned, are mainly C10-C10, but smaller proportions of molecules containing C12-C10, C12-C12, and C10-C12:1 dimers are also produced. The characterization of the RL congeners produced by different strains using HPLC MS/MS has been reported^7,8. The methods described in this work can only differentiate between mRL and dRL but cannot be used for the characterization of the RL congeners.

P. aeruginosa and some Burkholderia species are natural producers of RL⁹, but the former bacterium is the most efficient producer. However, commercially used RL is currently produced in Pseudomonas putida KT2440 derivatives expressing P. aeruginosa genes to avoid the use of this opportunistic pathogen^10,11. The detection and quantification of RL produced by P. aeruginosa are of great importance for studying the molecular mechanisms involved in the expression of virulence-related traits¹², in the characterization of strains belonging to clades 3 or 5¹³, and for constructing P. aeruginosa derivatives that overproduce these biosurfactants while having reduced virulence¹⁴. The production of biosurfactants by different microorganisms has been detected based on some general characteristics of these compounds, such as the collapse drop method or emulsification index¹⁵, but these methods are neither accurate nor specific¹⁶.

Here, we describe the protocol to detect mRL and dRL using the liquid extraction of total RL from the culture supernatants of different P. aeruginosa-type strains and the separation of both types of RL using TLC. In this method, the RL extracted from the culture supernatant is separated by their differential solubility in the solvents used for TLC, causing differential migration on the silica gel plate. Thus, mRL have a more rapid migration than dRL, and they can be detected as separate spots when the plates are dried and stained with α-naphthol.

The method described here for detecting mRL and dRL by TLC is based on a previously published article¹⁷, which is easy to perform and does not require expensive equipment. This method has been useful for detecting RL in various P. aeruginosa isolates¹³ using appropriate controls, such as a P. aeruginosa-derived mutant unable to produce RL. However, it is not the preferred method for characterizing novel biosurfactants produced by bacteria other than Pseudomonas aeruginosa due to its lack of specificity.

Additionally, a method for quantifying the rhamnose equivalents of total RL extracted from a P. aeruginosa culture supernatant is presented. This method quantifies these biosurfactants based on the reaction of orcinol with reductive sugars, resulting in a product that can be measured spectrophotometrically at 421 nm, as previously described¹⁸. Since the reaction with orcinol is not specific to rhamnose, it is important to perform this method with RL extracted from the culture supernatant that does not contain significant amounts of other sugar-containing molecules, such as lipopolysaccharides (LPS). An acidified chloroform/methanol mixture is used here for liquid extraction of RL¹⁸, but ethyl acetate can also be used, and solid-phase extraction (SPE) yields very good results¹⁹. The orcinol method described here does not require sophisticated equipment and can provide reliable results if performed with special care in preparing the analyzed samples, as discussed. To ensure proper sample preparation, it is important to include a Pseudomonas aeruginosa rhlA mutant unable to produce RL²⁰ and to perform three biological and three technical replicates for each determination.

There has been significant controversy in the literature^16,21regarding RL determination by the orcinol method, with some studies suggesting that RL production is overestimated and that the assay lacks specificity for rhamnose, potentially detecting other sugars. However, we demonstrate here that the methods described can be accurate and specific under the appropriate conditions. Furthermore, for comparison with the procedures outlined in this article, we utilize UPLC-MS/MS detection of a dRL standard and demonstrate that similar results are obtained with the orcinol method. The detailed protocol for quantifying RL using this method is included in Supplementary File 1.

These protocols are exemplified using the type strains PAO1 (phylogroup 1), PA14 (phylogroup 2), and PA7 (phylogroup 3). These strains were chosen because they are well-characterized and produce different RL profiles.

Protocol

This procedure is schematized in Supplementary Figure 1. The reagents and equipment used for the study are listed in the Table of Materials.

1. Detection of mRL and dRL in culture supernatants of P. aeruginosa using TLC

1. Start with the centrifuged broth of the P. aeruginosa strain of interest, cultivated in liquid medium for 24 h (to reach the stationary phase of growth where RL is produced). Typically, these cultures contain 1 x 10⁹ bacteria per mL.
2. Adjust the pH of the culture to 2 using concentrated HCl.
3. Put 5 mL of the acidified culture supernatant in a 50 mL polypropylene tube and add 5 mL of a 2:1 chloroform: methanol mixture.
4. Agitate the tube by inversion three times, each time for 10 s, and leave the tube without inversion for 2 min between each agitation.
5. Leave the tube without agitation for approximately 3 h until the two phases are separated, or centrifuge the tube for 10 min at 3,000 x g at 4 °C.
6. Transfer the organic phase (bottom layer) to a clean tube and leave the tube in an extraction hood until dryness occurs.
7. Repeat the process starting from step 1.3, putting the organic phase from the second chloroform: methanol extraction, into the tube that was used in the first extraction.
8. Evaporate the organic phase until 1 mL or 1.5 mL are left. Transfer this volume to a 1.5 mL centrifuge tube and evaporate the solvent to dryness overnight.
9. Add 50 µL of methanol to the dried tube to resuspend RL.
10. Perform thin-layer chromatography on silica gel plates.
    NOTE: The size of the TLC plate should be cut according to the number of samples that will be analyzed. Each sample should be applied at 1.5 cm, and the point of application should be 1 cm from the edge of the plate (draw a horizontal line with a pencil).
11. Apply 5 µL of each sample using a 10 µL micropipette.
12. The TLC liquid phase consists of a 65:15:2 mixture of chloroform: methanol: acetic acid. To prepare 35 mL of this mixture, mix 26 mL of chloroform, 6 mL of methanol, and 0.8 mL of a 20% stock solution of acetic acid. Mix these solvents and place them in the closed TLC chamber for at least 10 min before beginning the chromatography, so the atmosphere is saturated with the volatile solvents.
13. Place the TLC plate into the chamber, avoiding contact with the point where the samples were applied.
14. Close the chamber and leave the TLC until the solvent reaches 1 cm before the edge of the plate. At this point, remove the plate and let it dry (a flow of air can be applied to accelerate the process).
15. Prepare a solution of α-naphthol by dissolving 5 g of this compound in 33 mL of ethanol. Once dissolved, add 127.5 mL of ethanol, 12.6 mL of water, and 20.5 mL of cold H₂SO₄.
16. To detect the presence of mRL and dRL, spray the α-naphthol solution onto the plate in the extraction hood and place the sprayed plate in an oven at 85 °C for several minutes until the pinkish mark of RL is apparent.
17. Use a picture of the TLC to calculate the proportion of mono- and di-RL present in each sample using software that detects the density of each spot.

2. Quantification of the total amount of RL measuring the rhamnose equivalents present in the biosurfactant

1. Place 1.2 mL of a stationary phase culture (grown for 24 h) in a 1.5 mL centrifuge tube and centrifuge for 3 min at 3,000 x g at 4 °C. Collect the supernatant in a clean tube.
2. Transfer 333 µL to a clean tube (perform this step in triplicate) and add 1 mL of ether.
3. Vigorously mix in a vortex for 30 s. Repeat this procedure with one tube at a time.
4. Centrifuge for 2 min at 3,000 x g at 4 °C. Collect the organic phase (upper phase), transfer it to a clean tube, and leave it open in the extraction hood until dryness occurs.
5. Repeat the extraction with ether as described in step 2.2. Repeat steps 2.3 and 2.4.
6. Once completely dried, add 1 mL of water to each tube. Leave the tubes for 12 h to allow the RL to hydrate, then agitate vigorously in the vortex.
7. Place a clean flask in ice for 5 min (as it is an exothermic reaction). Prepare a solution of 60% H₂SO₄ (protect the flask from light).
8. In a clean tube, prepare a 1.6% orcinol solution in distilled water. To prepare the orcinol reagent, mix 4.4 mL of the 60% H₂SO₄ solution with 0.6 mL of the 1.6% orcinol solution.
9. Calculate the final volume of the orcinol reagent needed. Add 900 µL of this reagent to each sample and to each concentration of the calibration curve, using different rhamnose concentrations (typically, 9 concentrations of rhamnose are used in the range of 1 µg/mL to 9 µg/mL), and one with 100 µL of water, ensuring that all are performed in triplicate.
10. Add 100 µL of each RL sample to 900 µL of the orcinol reagent in a glass tube and mix the two solutions.
11. Incubate for 30 min in a water bath preheated to 80 °C.
12. Allow the tubes to cool to room temperature.
13. Read the absorbance of the samples and calibration curve at 421 nm using a quartz cell.
14. Calculate the concentration of rhamnose in each sample by interpolating the absorbance on the calibration curve and considering the volume used for the determination.
15. To calculate the µM concentration, divide the concentration obtained in µg/mL by 182.2 (the molecular weight of rhamnose) and multiply by 1000.

Results

In this article, three different P. aeruginosa type strains were utilized to represent three phylogroups, each with varying RL production levels and proportions of mRL and dRL. These strains include PAO1 (a wound isolate from Australia, 1955²²), PA14 (a plant isolate from the USA, 1977²³), and PA7 (a clinical isolate from Argentina, 2010²⁴). As a negative control, the PAO1 rhlA mutant was employed, which is incapable of RL production. All strains were cultivated for 24 h in PPGAS medium, specifically designed to promote high RL levels²⁵, at 37 °C. The PPGAS cultures were inoculated at an optical density of 0.05 at 600 nm using an LB medium²⁶ overnight culture. Typically, P. aeruginosa strains achieve an optical density of 2 measured at 600 nm in PPGAS medium after 24 h, which roughly corresponds to 1 x 10⁹ bacteria per mL. To collect the culture supernatants for RL detection and quantification, the cultures were centrifuged at 3,000 x g at 4 °C for 15 min, and the cell pellet was discarded. The results obtained using these two methods (Figure 2 and Figure 3) illustrate that both the types of RL produced by each strain and the quantity of their production differ among the three strains analyzed. As depicted in Figure 2, PAO1 produces approximately 30% mRL and 70% dRL, while PA14 produces an equal ratio of 50% mRL and 50% dRL, and PA7 exclusively produces mRL. This RL production profile is consistent with previous reports^12,14.

Figure 3A illustrates the amount of rhamnose equivalents present in RL produced by each of these three types of strains. From the detected rhamnose amounts, it is evident that strain PA14 produces the highest amount of RL, while strain PA7 produces the lowest amount. However, these results alone cannot be used to estimate the µM concentration of RL produced by each strain. Additionally, the orcinol method alone cannot provide an approximate RL concentration, as it relies on characterizing the proportions of mRL and dRL produced by each strain of interest. Therefore, to obtain an estimate of the µM concentration of RL, the proportion of each type of RL determined in the TLC must be considered, and a standard curve with different concentrations of rhamnose should be included (Figure 3B).

The orcinol method can indeed be utilized to quantify the concentration of RL when only one type of this biosurfactant is produced. In such cases, the μM concentration of the detected rhamnose corresponds directly to the μM concentration of mRL. For dRL, since two rhamnoses are produced by the hydrolysis of each dRL molecule, the μM concentration of rhamnose detected in RL needs to be divided by 2 to obtain their μM concentration (refer to Figure 4).

However, the majority of P. aeruginosa strains produce both types of RL at varying proportions (as shown in Figure 2), making it possible to determine only an approximate concentration of the total RL. Taking this into account, we estimated RL production by each strain as follows: For the PA7 strain, since it only produces mRL, the rhamnose forming part of RL (44.6 µg/mL + 4.5 µg/mL) corresponds directly to the µM concentration (244.78), representing the concentration of this biosurfactant in the culture supernatant, as each RL molecule contains one rhamnose moiety. However, for the PAO1 strain, although the detected rhamnose concentration was 111.55 µg/mL + 11.41 µg/mL, only 30% corresponds to mRL. Therefore, 70% of the RL molecules contain two rhamnoses per molecule. To estimate the concentration of RL, the µM rhamnose concentration (612.24) was divided by 5, considering that 1/5 corresponds to the µM concentration of mRL (122.49) and 2/5 (244.89) to dRL. Thus, the approximate µM concentration of RL produced by this strain is 367.39.

For the PA14 strain, the detected rhamnose concentration was 194.39 µg/mL + 11.5 µg/mL, which was converted to the µM concentration (1,066.9) and divided by 3. The result represents the concentration of each RL type, considering that each type was produced at 50% and that dRL contains 2 rhamnoses per molecule. Thus, this strain produces 355.63 µM of mRL and the same concentration of dRL, resulting in an approximate µM concentration of RL of 711.27, almost twice as much as strain PAO1 and almost three times the concentration produced by the PA7 strain.

Figure 1: Chemical composition of main congeners of mono-rhamnolipids and di-rhamnolipids. (A) Mono-rhamnolipids. (B) Di-rhamnolipids. Please click here to view a larger version of this figure.

Figure 2: Detection of mono-rhamnolipids and di-rhamnolipids by thin-layer chromatography. (A) Picture of a TLC plate showing RL standards, and the RL produced by strains PAO1, PA14, PA7, and the PAO1 derived ΔrhlA mutant. (B) Estimation of the proportion of each RL type (mRL and dRL) in each of the samples tested in (A), using ImageJ software. Please click here to view a larger version of this figure.

Figure 3: Quantification of RL concentration using the orcinol method. (A) Rhamnose concentration (µg/mL) contained in RL extracted from the culture supernatants of PAO1 (black bar), PA14 (light gray bar), and PA7 (dark grey bar) strains. The bars denote the standard deviation. (B) The rhamnose calibration curve for the experiment shown in (A). Please click here to view a larger version of this figure.

Figure 4: Validation of the orcinol method compared with UPLC-MS/MS for dRL quantification. (A) The dRL standard was quantified using UPLC-MS/MS. (B) The orcinol method compared with rhamnose concentration expressed in mM. (C) The same mM concentration of dRL as rhamnose gives approximately twice the absorbance at 421 nm, as expected. Please click here to view a larger version of this figure.

Supplementary Figure 1: Schematic representation of the protocol. Please click here to download this File.

Supplementary File 1: Detail protocol to quantify dRL by UPLC-MS/MS. Please click here to download this File.

Discussion

The most accurate method for detecting and quantifying RL is HPLC coupled with mass spectrometry (MS)^7,8,27; however, it requires specialized and expensive equipment that may not be accessible to many researchers. The methods described here can be routinely performed with basic laboratory materials and equipment to detect and estimate RL concentrations, but they have some limitations, particularly their inaccuracy in determining mRL and dRL mixtures. Moreover, the preparation of the sample used for RL quantification should be carefully executed. Cultures used for RL detection and quantification should not be in the late stationary phase, as cells might be aggregated or lysed, resulting in non-homogeneous cell suspensions that could cause variability in the results obtained. Additionally, cellular debris might interfere with the orcinol method.

Accurate results can be obtained when the described protocol is performed with appropriate controls, such as a Pseudomonas aeruginosa strain unable to produce RL (like the PAO1 ΔrhlA mutant²⁰ used in this work, Figure 2). These results are useful for characterizing different P. aeruginosa isolates^12,13, genetically modifying strains to increase RL production¹⁴, or studying the molecular genetics of RL production by various P. aeruginosa isolates¹².

Thus, the methods described here can be utilized to characterize Pseudomonas aeruginosa strains for their production of virulence factors and assess their biotechnological potential, especially in the early stages of investigation. It's important to note that these methods are not intended as substitutes for techniques like HPLC/MS but rather as accessible techniques for initial strain characterization and genetic manipulation of P. aeruginosa strains.

Furthermore, if the research focus is on RL production for biomedical purposes, it's crucial to acknowledge that RL production is just one of several factors produced by Pseudomonas aeruginosa. Additionally, the type of RL produced or the amount of their production is not strictly correlated with the virulence of a particular strain, as virulence is influenced by multiple factors²⁸.

Regarding the reproducibility of the methods described here, special attention should be paid when obtaining the organic phases in both the detection of RL by TLC (step 1.6) and the measurement of RL using the orcinol method (step 2.4). Even a small amount of the other phase taken during extraction can lead to considerable variation in the results obtained. The extraction of RL from the culture supernatant is a critical step in achieving reliable and reproducible results with the orcinol method. It's crucial that when collecting the organic phase, the interphase and the liquid phase are excluded. If an ether phase is not completely dried within 3 h, it likely contains water, necessitating a new extraction.

The use of the orcinol method for RL quantification has been debated, with arguments against its accuracy and specificity. It has been claimed that RL are often overestimated, and the method is not specific as orcinol can react with different sugars, not just rhamnose. The current data demonstrate that if RL are efficiently extracted from the culture supernatant, this technique is specific, as shown by the lack of reaction when the rhlA mutant is used as a negative control. However, to provide additional evidence of the method's accuracy in detection, independently of the extraction procedure, we detected a dRL standard by UPLC-MS/MS as described previously²⁷, and compared the results obtained with the orcinol method¹⁸ (Supplementary File 1, Figure 4A). We found, as reported¹⁸, a very good correlation between the concentration of RL measured using this method, where each μM concentration of dRL detected corresponds to approximately twice the absorbance at 421 nm as the same rhamnose μM concentration (Figure 4B,C). This control experiment clearly demonstrates the accuracy of this method for RL quantification.

However, due to the challenge of measuring the concentration of mRL and dRL mixtures with the orcinol method, most publications report the results as rhamnose equivalents in RL. This value may not have a direct correlation with the actual RL concentration, as illustrated in the examples presented here, but nonetheless, it serves as a useful means of comparing RL quantities under different conditions, especially when using the same P. aeruginosa strain.

An alternative method for mRL and dRL quantification has been reported, which is based on the solubilization of the hydrophobic dye Victoria Pure Blue BO, commonly used in ballpoint pens²⁹. This method is quick and inexpensive. However, this protocol was developed to quantify the heterologous production of RL by P. putida and is not feasible for use with RL produced by P. aeruginosa, as the expression of the blue toxin pyocyanin interferes with this method.

In summary, the methods described here offer a viable alternative for the detection and quantification of RL produced by P. aeruginosa that do not require sophisticated and expensive equipment. They provide a practical approach for researchers to assess RL production in a variety of experimental settings.

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
1-NAPHTHOL	SIGMA-ALDRICH	70442
ACETIC ACID	J.T. BAKER	9508-02
CENTRIFUGE			For centrifuging tubes 1.5 mL and  50 mL
CHLOROFORM	J.T. BAKER	9180-02
DRYING OVEN
ETHER	J.T. BAKER	9244-02
GLASS PIPETTE	SIGMA-ALDRICH	CLS706510
HYDROCHLORIC ACID	J.T. BAKER	5622-02
LB
L-RHAMNOSE MONOHYDRATE	SIGMA-ALDRICH	R-3875
METHANOL	J.T. BAKER	9049-02
ORCINOL MONOHYDRATE	SIGMA-ALDRICH	O1875
PPGAS Broth			Tris HCL (0.12M), Potassium Chloride ( 0.02M) Ammonium Chloride (0.02M),  Peptone (1%), pH 7.4   Autoclaved. Add  Glucose (5%) and Magnesium Sulfate (0.0016M)
QUARTZ CELL (CUVETTE)	SIGMA-ALDRICH	Z276669
RECTANGULAR TLC DEVELOPING TANK	FISHER SCIENTIFIC	K4161801020
RHAMNOLIPIDS	SIGMA-ALDRICH	R-90
SPECTROPHOTOMETER			VIS
SPRAYER	SIGMA-ALDRICH	Z529710-1EA
SULFURIC ACID	J.T. BAKER	9681-02
TES TUBES 5mL	CORNING	352002
TLC SILICA GEL 60 F254	MERCK	1.05554.0001
WATER BATH			> 80 °C