Rhamnolipids are conserved biosurfactants molecules - Springer Link

Appl Microbiol Biotechnol (2013) 97:7297–7306 DOI 10.1007/s00253-013-4876-z

APPLIED GENETICS AND MOLECULAR BIOTECHNOLOGY

Rhamnolipids are conserved biosurfactants molecules: implications for their biotechnological potential Amedea Perfumo & Michelle Rudden & Thomas J. P. Smyth & Roger Marchant & Paul S. Stevenson & Neil J. Parry & Ibrahim M. Banat

Received: 27 February 2013 / Revised: 20 March 2013 / Accepted: 21 March 2013 / Published online: 7 April 2013 # Springer-Verlag Berlin Heidelberg 2013

Abstract A range of isolates of Pseudomonas aeruginosa from widely different environmental sources were examined for their ability to synthesise rhamnolipid biosurfactants. No significant differences in the quantity or composition of the rhamnolipid congeners could be produced by manipulating the growth conditions. Sequences for the rhamnolipid genes indicated low levels of strain variation, and the majority of polymorphisms did lead to amino acid sequence changes that had no evident phenotypic effect. Expression of the rhlB and rhlC rhamnosyltransferase genes showed a fixed sequential expression pattern during growth, and no significant up-regulation could be induced by varying producer strains or growth media. The results indicated that rhamnolipids are highly conserved molecules and that their gene expression has a rather stringent control. This leaves little opportunity to manipulate and greatly increase the yield of rhamnolipids from strains of P. aeruginosa for biotechnological applications. Keywords Rhamnolipids . Biosurfactants . Pseudomonas aeruginosa . Bioreactor . Comparative gene analysis . Gene expression

A. Perfumo : M. Rudden : T. J. P. Smyth : R. Marchant : I. M. Banat (*) School of Biomedical Sciences, University of Ulster, Coleraine BT52 1SA, UK e-mail: [email protected] P. S. Stevenson : N. J. Parry Research and Development, Unilever, Port Sunlight, Wirral, Liverpool CH63 3JW, UK

Introduction Rhamnolipids are natural surfactants discovered in 1949 (Jarvis and Johnson 1949) but only in recent times have they begun to attract scientific and commercial interest (Marchant and Banat 2012a, b). Rhamnolipids are typically produced by Pseudomonas aeruginosa. In this bacterium, they play a variety of physiological roles that include solubilisation and up-take of hydrocarbons and hydrophobic compounds, adhesion to surfaces, formation and maintenance of biofilms and cell motility (Van Hamme et al. 2006). The numerous physicochemical properties that these biosurfactants display (e.g. surface/interface tension reduction, de-/emulsification and wetting) combined with activity at low concentrations (in milligrams per litre) and under extreme conditions (temperature, pH and salinity) have stimulated their potential industrial applications. Rhamnolipids can be used as ingredients in soaps, detergents, cosmetics, household and even pharmaceutical products (Banat et al. 2000, 2010; Maier and Soberón-Chavez 2000; Fracchia et al. 2012). They are also effective in environmental bioremediation, microbially enhanced oil recovery and sludge removal technologies (Perfumo et al. 2010; Franzetti et al. 2010). However, the successful commercialisation of rhamnolipids is hampered mainly by the low yield of these molecules in P. aeruginosa strains, which does not exceed in general 20 g/L. Although there are reports of much higher product yields, up to 100 g/L (Lang and Wullbrandt 1999), concerns persist about the accuracy of some of the quantification procedures used. Increasing the production rates is therefore necessary for rhamnolipids to become commercially competitive. Efforts in this direction focused at first on manipulating the production conditions through the selection of substrates, for example inexpensive vegetable or

7298

waste oils that could also enhance the synthesis of rhamnolipids (Nitschke et al. 2005a; Rahman et al. 2002; Makkar et al. 2011). Genetic engineering was subsequently attempted but in general with only modest results such as 2–3-fold increase obtained by random mutagenesis (Mukherjee et al. 2006). Moreover, the possibility, suggested by some studies in the past (Déziel et al. 2000; Mata-Sandoval et al. 2001), to obtain molecular variants of rhamnolipids by simply selecting the producer strains and/or the cultivation conditions further stimulated interest from industry. Déziel et al. (2000) reported for example that P. aeruginosa 57RP produced rhamnolipids mainly of the type Rha-Rha-C10-C10 when growing on mannitol, whereas Rha-Rha-C10 only was produced on naphthalene (Déziel et al. 2000). Genetically modified strains, however, needed to be created in the laboratory to achieve the synthesis of only the monorhamnolipid Rha-C10-C10 (Cabrera-Valladares et al. 2006; Cha et al. 2008). Despite a quite large body of literature, our scientific knowledge on rhamnolipids is still inadequate to fully support their exploitation at the industrial scale. We still lack defined experimental evidence to address one of the main questions: to what extent can rhamnolipid biosynthesis be manipulated for commercial purposes? The study we report here was part of the Susclean (sustainable and clean) project supported by Unilever UKCR Ltd. to assess the use of microbial surfactants in detergent formulations. We first followed a traditional approach consisting of screening a number of P. aeruginosa strains from various ecological niches (water, hydrocarboncontaminated soil and hospitals, including infected cystic fibrosis patients) and testing rhamnolipid production at the bioreactor scale. To better understand the results indicating limited productivity and a low degree of structural variation, we investigated the genetic diversity of the rhamnolipidencoding genes (rhl genes), rhlABRI operon and rhlC, by comparative sequence analysis. We also studied the expression of the rhamnosyltransferase genes, rhlB and rhlC, by RT-qPCR to assess the role of gene regulation over the genetic constraints. With this study, we aim at providing new, critical, insights into the biotechnological potential as well as the ecological significance of rhamnolipid biosurfactants.

Appl Microbiol Biotechnol (2013) 97:7297–7306

used as a reference strain (Stover et al. 2000). All strains were maintained at −70 °C in cryovials and cultivated on either Luria–Bertani broth (Difco) or Pseudomonas isolation agar plates (Difco). To maintain unaltered bacterial genomic features, subcultivation steps were minimised. Phenotypic characterisation for motility (swimming and twitching) and biofilm formation was carried out as previously reported (Deligianni et al. 2010). Rhamnolipids were produced on proteose peptone glucose ammonium salt (PPGAS) medium (Zhang and Miller 1994) and mineral salt medium (MSM) supplemented with different carbon substrates (Perfumo et al. 2006) in either shaken flasks (200 rpm) or bioreactor. A 5-L benchtop BIOSTAT B bioreactor (B. Braun Biotech International, Melsungen, Germany) was equipped with sensors for temperature, dissolved oxygen, pH and foam. Foam formation was controlled by using both foam-breaker impellers and a silicone-type antifoam Y-30 solution (Sigma-Aldrich) diluted to 5 % (v/v) final concentration. The antifoam system was adjusted to medium–high sensitivity (level 3). Aeration was maintained constant at 2 L/min, and the agitation was adjusted from 200 to 700 rpm according to need. Samples of the fermentation broth were taken from the sampling port and used for optical density measurement and rhamnolipid analysis. Bacterial growth was determined as optical density, OD(600 nm), using a Novaspec II, Pharmacia Biotech, spectrophotomer (Cambridge, UK). Analysis of rhamnolipids Synthesis of rhamnolipids was monitored by measuring the reduction of the surface tension using a digital tensiometer (K10ST, Krüss, Hamburg, Germany) equipped with a deNouy platinum ring. Rhamnolipids were analysed after thorough extraction and purification of the fermentation broth as described previously (Smyth et al. 2010). Quantification was determined in terms of dry weight (in grams per litre) of the pure, honey-like, rhamnolipid material, and the chemical characterisation was obtained by electrospray ionization tandem mass spectrometry (ESI–MS) and high–performance liquid chromatography (HPLC) (Smyth et al. 2010). For comparison with most data available in the literature, rhamnolipid concentration was additionally determined using the orcinol assay (Chandrasekaran and Bemiller 1980).

Materials and methods DNA and phylogenetic analysis of rhl genes Bacterial strains and cultivation conditions Nine P. aeruginosa isolates were used in this study (AP02-1, AD7, DS10, ST5, 1, 17, 26, 55, 80). They all were partially characterised as described elsewhere (Deligianni et al. 2010; Perfumo et al. 2006; Rahman et al. 2002). Strain PAO1 was

Genomic DNA was extracted using the FastPrep™ FP120 cell disrupter and a commercial Fast DNA® Spin® Kit for Soil (BIO101, Carlsbad, CA, USA). PCR amplification of full length rhl genes was carried out using newly designed sets of primers (Table 1). PCR reactions were performed

Appl Microbiol Biotechnol (2013) 97:7297–7306 Table 1 Oligonucleotide primers used in this study for PCR amplification, RT-qPCR and sequencing of rhl genes

Gene

Forward primer

Sequence (5′→3′)

Reverse primer

Sequence(5′→3′)

rhlA

rhlA2-f

ATGGCCGCTGAGTTACTTGT

rhlA2-r

GACGGTCTCGTTGAGCAGAT

rhlA3-f rhlAB-fa rhlB2-f rhlB3-f rhlB4-f rhlR-fa rhlR2-f rhlR3-f rhlI-fa rhlI2-f rhlI3-f rhlC2-f rhlC3-f rhlC4-f rhlC33-fb

CATTTCAACGTGGTGCTGTT TCATGGAATTGTCACAACCGC CATTTCCTCGACCTGGAGTC GGTACACCCCAAGTTCAACG CACGCCATCCTCATCGCC TGCATTTTATCGATCAGGGC CGGTGCTGGCATAACAGATA CTGGGCTTCGATTACTACGC TTCATCCTCCTTTAGTCTTCCC TGCCGTTCATCCTCCTTTAG CTCTCTGAATCGCTGGAAGG CTCGTCATTCTGGCTGGTCT ATCCATCTCGACGGACTGAC AACTGGCGGCGGCGTTTCC ACCGGATAGACATGGGCGT

rhlA3-r rhlAB-ra rhlB2-r rhlB3-r rhlB4-r rhlR-ra rhlR2-r rhlR3-r rhlI-ra rhlI2-r rhlI3-r rhlC2-r rhlC3-r rhlC4-r rhlC34-rb

CTACTCCGTCGCTTATGCAA ATACGGCAAAATCATGGCAAC ATCGAGAAAGCGTTGCAGTT AAAAAGCCTCCGTCATTCCT GGTCAGTTCGTCGCTCAGC CACTTCCTTTTCCAGGACG GCTCGAAGCTGGAGATGTTC CCTTCCAGCGATTCAGAGAG TTCCAGCGATTCAGAGAGC GCAGGCTGGACCAGAATATC GCAGAGAGACTACGCAAGTCG GCAGGCTGTATTCGGTGTC GGGCGATTCGTTCTACTTCC AGTCCTGGTCGAGCAGCAGCA GATCGCTGTGCGGTGAGTT

rhlB

rhlR

rhlI

rhlC

a

7299

Zhu et al. (2004)

b

Bazire et al. (2005)

using the GC-RICH PCR system-dNTPack (Roche) with MgCl2 concentration adjusted to 2.5 mM. A PCR cycle suitable for amplifying all rhl genes consisted of initial denaturation at 95 °C for 3 min, 30 cycles of denaturation at 95 °C for 60 s, annealing at 56 °C for 60 s, extension at 72 °C for 90 s and final extension at 72 °C for 10 min. PCR products were gel-purified by band excision and DNA quantified by NanoDrop™ ND-1000 UV/VIS spectrophotometer (Thermo Fisher Scientific, Waltham, MD, USA). Sequencing reactions were prepared using an ABI Prism® BigDye™ Terminator reaction kit (Applied Biosystems, Foster City, CA, USA) with addition of 5 % (v/v) dimethyl sulfoxide and run on an ABI Prism® 3100 Genetic Analyzer (Applied Biosystems). Phylogenetic analyses were performed using the Molecular Evolutionary Genetics Analysis (MEGA) program suite version 4.0 (Tamura et al. 2007). Genetic distances were estimated as the number of nucleotide differences from pairwise comparisons using the Jukes–Cantor method, and phylogenetic trees were constructed by the neighbourjoining method with 1,000 bootstrap replicates. RNA isolation and cDNA synthesis RNA was extracted from both PPGAS and MSM cultures at time points along the growth curve using 1-ml cell pellet. RNA was immediately stabilised with RNAprotect Bacteria Reagent (Qiagen) and extracted using the RNeasy Mini Plus Kit (Qiagen) following manufacturer’s instructions.

Absence of genomic DNA contamination was confirmed by standard PCR with RNA as template. RNA was quantified with a NanoDrop™ ND-1000 UV/VIS spectrophotometer and quality verified by gel electrophoresis. Two protocols were tested to synthesise cDNA. Random hexamer primers (Promega, Madison, WI, USA) with Superscript II Reverse Transcriptase (Invitrogen, Carlsbad, CA, USA) were used according to manufacturer’s instructions in a total volume of 20 μl. cDNA for rhlB and rhlC was also synthesised using gene-specific primers rhlB3-f/rhlB2-r (325 bp amplicon length) and rhlC33-f/rhlC34-r (133 bp amplicon length) (Table 1). The housekeeping gene rpoD was used as control and amplified with primers rpoD1-f (5′AGCAATCTCGTCTGAAAGAG-3′) and rpoD1-r (5′GCTGTCTCGAATACGTTGAT-3′) (163 bp amplicon length). Absence of unspecifically synthesised products was confirmed by standard PCR using cDNA as template. cDNAs were synthesised in biological triplicates for each time point. Real-time PCR (qPCR) and data analysis Real-time PCR was performed using the Roche LightCycler LC480 system (Roche) according to the manufacturer’s instructions. Each 10 μl PCR reaction contained 1 μl cDNA template, 0.5 μM each primer, 5 μl LightCycler 480 SYBR Green I Master (Roche) and 2 μl nuclease-free water. Cycling parameters were as follows: 95 °C for 5 min, 50 cycles of 95 °C for 10 s, 59 °C for 10 s and 72 °C for 10 s. Product specificity for each gene was confirmed by melt curve

7300

analysis, and only artefact free-reactions were deemed valid. For all cDNA samples, negative control reactions were run on the same plate, acceptable Cq for residual gDNA amplified were set at >36 cycles and for no template controls >40. Positive gDNA samples were also run on each plate to assess inter plate variability. Relative gene expression data analysis was determined using calibrator-normalised quantification with PCR efficiency correction. Primer efficiencies were determined from cDNA calibration curve slopes for each gene generated by the LightCycler® 480 system (software version 1.5) using the equation E=10[−1/slope] (Pfaffl 2001). cDNA standards were prepared from triplicate 5-fold serial dilutions of pooled cDNA samples covering over 5 orders of magnitude. cDNA templates were diluted to within the standard curve linear range for target runs. For relative expression, each cDNA target sample was normalised to the reference gene rpoD, and relative expression ratios were calculated from a calibrator sample which was time point zero along the growth curve using the following equation: Conc: target gene Conc: target gene : Conc: reference gene ðsampleÞ Conc: reference gene ðcalibratorÞ All qPCR samples were performed with triplicate technical replicates with SD 95 %) rhamnolipids was obtained and used for testing in detergent formulations. All strains produced rhamnolipids with very similar chemical composition regardless of the differences in the production media/conditions used. HPLC analyses on purified samples showed that all rhamnolipids were synthesised consistently as a mixture of six to seven congeners with negligible differences in their ratios (Fig. 1). Direct analysis of the cell-free supernatant by ESI-MS revealed a degree of error in determining the chemical profile of rhamnolipids; however, this procedure is widespread since it is quick and easy to perform. First, the high background from media residues (particularly if oily substrates were used) caused weakening or even suppression of the signal from individual rhamnolipids, especially if present in low concentrations. Second, a number of extra ions could be detected. They represented in-source fragmentations but in the past have been erroneously assigned to additional rhamnolipid structures, for example Rha-C8 (m/z 305), Rha-C10 (m/z 333) and Rha-C12 (m/z 361) (Déziel et al. 2000). Our results indicated that P. aeruginosa synthesises rhamnolipids with highly conserved chemical characteristics. No major changes in the molecular profile could be induced via manipulation of the production media and conditions. These findings are in agreement with many reports which have looked at variability in rhamnolipids produced from a range of substrates (Nitschke et al. 2005b; Li et al. 2011). Comparative sequence analysis of the rhl genes in different P. aeruginosa strains To address the question of the conservative over variable nature of rhamnolipid biosurfactants, we investigated the


7301

Fig. 1 HPLC spectrum showing relative absorption against time for rhamnolipids typically synthesised by P. aeruginosa. Rhamnolipids are produced as highly conserved mixture of six to seven molecular

congeners with two predominant species, the di-rhamnolipid RhaRha-C10-C10 (produced by RhlC) and the mono-rhamnolipid RhaC10-C10 (produced by RhlB)

range and the extent of the genetic variation of the rhl genes among P. aeruginosa strains isolated from highly diverse environments and showing different phenotypic traits. This would also represent a first step towards assessing the possibility to exploit natural gene variation for industrially tailored biosurfactant products. The genes of the rhlABRI operon showed a quite low sequence variation of 1.5 to 3.0 % (Table 2). The total number of single nucleotide polymorphisms (SNPs) was comparable amongst environmental and clinical isolates, and the singleton sites (at least two types of nucleotides with, at most, one occurring multiple times) were more frequent than the informative sites (at least two types of nucleotides and at least two of them occurring with a minimum frequency of two). The nucleotide substitutions leading to non-synonymous mutations, i.e. changes in amino acid sequence in the encoded enzyme, were only few (Table 2) and were predicted to have a neutral effect because replaced amino acids had similar properties (Betts and Russel 2003). The gene rhlC showed a sequence variation of 4.7 %, with a total number of 46 SNPs detected, most of which occurred in the clinical strains of P. aeruginosa (Table 2). We also observed a higher number of non-synonymous mutations that resulted in 3.7 % of the amino acids being substituted within the enzyme rhamnosyltransferase-2 (Table 3). However, there was no obvious phenotypic effect induced, e.g. higher proportion of di-rhamnolipids synthesised by these strains. Compared to

the rhl genes, other genes constituting the core genome of P. aeruginosa assessed in previous studies showed a much lower degree of variation of about ≤1 % (Finnan et al. 2004; Kiewitz and Tümmler 2000). We then investigated the potential phylogenetic relationship between the rhl genes and strain/environment. All rhl gene sequences available in the public database were included as outliners to make the analysis more accurate. Some strains showed a genetic relatedness irrespective of their environmental source. For example, water isolates AP02-1 and AD7 and cystic fibrosis isolates 17 and 26 clustered together for the rhlA gene and were very close for the genes rhlBRI. Another recurrent cluster was composed of cystic fibrosis strain 55 and the highly virulent PA14 (Mikkelsen et al. 2011) which carries a mutation in the ladS gene impairing biofilm formation, as well as by the rhamnolipid-deficient strain 80 and the hypervirulent Liverpool epidemic strain LESB58. The most diverse strains were soil isolate DS10 and cystic fibrosis isolate 1, which was further confirmed by the genetic distance computation (Fig. 2). Overall, our results showed that rhl genes have a limited sequence variation and that mutations that are rather strain-specific do not clearly correlate to any particular biological function of rhamnolipid molecules or environmental source of the producer organism (e.g. hydrophobic substrates solubilisation in contaminated environments, biofilm formation in infected cystic fibrosis patients etc.).

7302 Table 2 Sequence diversity and analysis of the polymorphisms in rhl genes among P. aeruginosa strains computed with MEGA 4.0


Gene locus

rhlA

rhlB

rhlR

rhlI

rhlC Strain PAO1 has been used as reference strain

Strains

All strains Environ strains Clinical strains All strains Environ strains Clinical strains All strains Environ strains Clinical strains All strains Environ strains Clinical strains All strains Environ strains Clinical strains

Sequence divergence (%)

No. of polymorphisms Total

Informative

Singleton

Synonymous

Nonsynonymous

1.46

13 8 7 27 19 17 13 11 8 19 17 14 46 15 45

3 2 1 15 8 11 8 4 6 12 11 8 23 5 19

10 6 6 12 11 6 5 7 2 7 6 6 23 10 26

12 7 6 22 16 13 13 11 8 16 15 11 34 10 33

1 1 1 5 3 4 0 0 0 3 2 3 12 5 12

2.1

1.79

3.13

4.7

phase of the culture, with rhlB which is responsible for the formation of the mono-rhamnolipid, expressed strongly and increasingly from 6 to 12 h and then dropping to low levels subsequently. The rhlC gene on the other hand, which is responsible for the addition of the second rhamnose moiety to form the di-rhamnolipid, is only expressed to any extent after 12 h when the expression of rhlB is already reduced. The regulation of the two genes is most likely co-ordinated in a sequential fashion. We also studied the expression of the rhlB and rhlC genes in some of the strains with the highest degree of

Expression of the rhamnosyltransferase genes rhlB and rhlC during synthesis of rhamnolipids To establish the time course for the rhamnolipid gene expression in relation to the batch growth cycle of P. aeruginosa, the reference strain PAO1 was grown under phosphate-limited conditions in PPGAS medium. Cell growth and surface tension were measured and the relative expression of the rhlB and rhlC genes determined at time intervals up to 48 h. The results (Fig. 3) show very low levels of expression of either gene until the early stationary Table 3 Non-synonymous polymorphisms at rhlC gene in P. aeruginosa isolates

SNPs

AA replacement

Position

Strains Environ AP02-1

Strain PAO1 has been used as reference strain

GAA→GGA

Glu→Gly

21

CAC→TAC GCG→ACG GTG→CTG CGC→TGC ACG→GCG ACC→ATC ACC→CGG TCC→CCC GTA→ACA GCT→GTT TCC→CCC

His→Tyr Ala→Thr Val→Leu Arg→Cys Thr→Ala Thr→Ile Thr→Arg Ser→Pro Val→Thr Ala→Val Ser→Pro

61 78 84 111 140 151 301 317 318 319 320

Clinical AD7

DS10

ST5

1

17

26

55

80

• • •

•

• • • • • • •

• • • • •

• • • • •

• • • •


7303

Fig. 2 Phylogenetic tree constructed with the sequences of the rhamnosyltransferase-1 (rhlB) and 2 (rhlC) from the P. aeruginosa strains used in this study and expanded with some outliers. The distances were calculated according to the Jukes-cantor model, and 1,000 replicates (bootstrap percentages are shown at the nodes) were

computed for each analysis. The trees were drawn to scale with branch lengths in the same units as those of the evolutionary distances used to infer the phylogenetic tree. Strains are labelled with their origin source (in brackets). The scale bar indicates the number of substitutions per site

physiological and genetic divergence (AP02-1, DS10, 1, 17 and 55) and in the reference strain PAO1. For a correct data interpretation, we standardised on previous studies (Waite et al. 2005) considering the genes as differentially upregulated when showing an average change of expression (n-fold) of ≥2.5 compared to the reference gene rpoD. In order to assess also differences in response to the cultivation conditions, P. aeruginosa strains were cultivated using two well established and rather different media for rhamnolipid production PPGAS and MSM. The PPGAS medium (Zhang and Miller 1994) contained glucose as carbon substrate and was limited in phosphate content, while nitrogen was

supplied both in organic (e.g. proteose peptone) and inorganic (ammonia) form. On the other hand, MSM was a mineral medium rich in salts and containing glycerol as carbon source. Growth on PPGAS was faster due to the

Fig. 3 Sequential expression of genes rhlB and rhlC during growth of P. aeruginosa. rhlB is expressed at an early stage of the growth, while rhlC is switched on only subsequently when rhlB is switched off. This gene regulation mechanisms controls the flow of rhamnolipids at a molecular level, where mono-rhamnolipids are synthesised first and then converted to di-rhamnolipids

Fig. 4 Up-regulation of genes rhlB and rhlC in different rhamnolipidproducing strains (AP02-1, DS101, 17 and 55) and varying production media (PPGAS and MSM). Manipulation of the rhamnolipid production conditions resulted in changes of the expression of rhlB and rhlC below the threshold level of 2.5-fold to be considered as preferential expression

7304

readily available glucose, and once the stationary phase was reached, all the cultures showed the distinctive characteristics of biosurfactant production, e.g. considerable foam and reduction of ST ≤30 mN/m. HPLC spectra showed that rhamnolipids were produced in the typical mixture. An increase of the gene expression of rhlB and rhlC of about one to three times greater than the expression of rpoD was observed and strain DS10 showed the highest up-regulation degree, especially when growing on mineral medium (Fig. 4).

Discussion Biosurfactant producing microorganisms have been known for many years and have been the subject of on-going research. It is only in the last few years, however, when a sustainability agenda gained high profile in manufacturing industry that the goal of replacing chemical surfactants, derived from non-renewable hydrocarbon sources, with sustainable biosurfactants became a priority. Not surprisingly yield and cost of production together with fitness for purpose became key issues. Rhamnolipids produced by P. aeruginosa have been shown to possess suitable characteristics for use in a number of applications (Chen et al. 2010a, b); however, sufficient yields have never been achieved which would make full commercialisation viable. In addition, one of the standard analytical methods, the orcinol method, almost certainly gives over inflated yield values when compared to a more rigorous extraction and HPLC quantification (Smyth et al. 2010). The main purpose of the present study was to try to determine the extent of natural genetic variation in P. aeruginosa strains isolated from a range of different environmental situations. The hypothesis that some strains might be hyperproducers as a result of selective pressures exerted by environmental conditions was tested. The gene sequencing results were not encouraging and showed that the rhamnolipid gene cluster is remarkably non-variable, although slightly more variable than the core genome of the organism. The sequence variation that could be detected largely consisted of SNPs that had no effect on the amino acid sequence of the protein produced. These results clearly indicate that a search for hyperproducer strains in the environment is unlikely to be successful. Similarly improbable seems the possibility to retrieve strains producing rhamnolipids with unique chemical profiles. The other standard approach which involves manipulating the medium composition and growth conditions also had little effect in the strains we examined from widely different environmental situations. We were also unable to confirm


that rhamnolipid production is greater on water-insoluble substrates rather than water-soluble ones, as often claimed in the past. Rhamnolipid production is known to be under tight regulatory control as part of the quorum sensing system (Reis et al. 2011), and it is therefore not surprising that disturbing the system is difficult. In addition to simply looking at the end product of the synthetic process, i.e. the rhamnolipids excreted into the medium, we also followed the gene expression of the two rhamnosyltransferase genes rhlB and rhlC. Where rhamnolipids have been produced commercially or for specific applications (Banat et al. 1991), long-batch cultivation times have been used in the past based on the assumption that since rhamnolipids are produced during the stationary phase prolonged cultivation will give higher yields. The time course experiments following gene expression have shown that the rhlB gene expression is effectively switched off before the second rhamnosyltransferase is produced. These results are also indicative that prolonging cultivation times in the presence of excess carbon substrate is unlikely to lead to greater yields. Examination of the relative expression of the rhlB and rhlC genes in the range of environmental strains used in the study confirmed the earlier results that there is little variation in either the genetic structure of the system or its operation in strains from widely different sources. Results from this present study are not encouraging for the use of P. aeruginosa for large-scale production of rhamnolipids, and perhaps a future, more productive, approach would be to examine the recently identified other rhamnolipid producers (e.g. Burkholderia strains) which may also have the benefit of being non-pathogenic organisms. Using a non-pathogen for production would greatly ease production problems and also perhaps make the product more easily incorporated into home and personal care products. Acknowledgments We thank Prof James Dooley at the University of Ulster for providing the clinical strains used in this study. A.P. is grateful to Dr. Urs Ochsner for his valuable advice at the beginning of this study. This work was supported by Unilever and the Department of Trade and Industry technology programme and a CAST award from the Department of Education and Learning Northern Ireland and Unilever to Michelle Rudden.

References Banat IM, Franzetti A, Gandolfi I, Bestetti G, Martinotti MG, Fracchia L, Smyth TJ, Marchant R (2010) Microbial biosurfactants production, applications and future potential. Appl Microbiol Biotechnol 87:427–444

Appl Microbiol Biotechnol (2013) 97:7297–7306 Banat IM, Makkar RS, Cameotra SS (2000) Potential commercial applications of microbial surfactants. Appl Microbiol Biotechnol 53:495–508 Banat IM, Samarah N, Murad M, Horne R, Banerjee S (1991) Biosurfactant production and use in oil tank cleanup. World J Microbiol Biotechnol 7:80–88 Bazire A, Dheilly A, Diab F, Morin D, Jebbar M, Haras D, Dufour A (2005) Osmotic stress and phosphate limitation alter production of cell-to-cell signal molecules and rhamnolipid biosurfactant by Pseudomonas aeruginosa. FEMS Microbiol Lett 253:125–131 Betts MJ, Russel RB (2003) Amino acid properties and consequences of substitutions. In: Barnes MR, Gray IC (eds) Bioinformatics for geneticists. Wiley-Interscience, New York Cabrera-Valladares N, Richardson AP, Olvera C, Treviňo LG, Déziel E, Lépine F, Soberón-Chavez G (2006) Monorhamnolipids and 3(3-hydroxyalkanoyloxy)alkanoic acids (HAAs) production using Escherichia coli as a heterologous host. Appl Microbiol Biotechnol 73:187–194 Cha M, Lee N, Kim M, Lee S (2008) Heterologous production of Pseudomonas aeruginosa EMS1 biosurfactant in Pseudomonas putida. Bioresour Technol 99:2192–2199 Chen ML, Penfold J, Thomas RK, Smyth TJP, Perfumo A, Marchant R, Banat IM, Stevenson P, Parry A, Tucker I, Grillo I (2010a) Mixing behavior of the biosurfactant, rhamnolipid, with a conventional anionic surfactant, sodium dodecyl benzene sulfonate. Langmuir 26:17958–17968 Chen ML, Penfold J, Thomas RK, Smyth TJP, Perfumo A, Marchant R, Banat IM, Stevenson P, Parry A, Tucker I, Grillo I (2010b) Solution self-assembly and adsorption at the air-water interface of the monorhamnose and dirhamnose rhamnolipids and their mixtures. Langmuir 26:18281–18292 Chandrasekaran EV, Bemiller JN (1980) Constituent analysis of glycosaminoglycans. In: Whistler RL (ed) Methods in carbohydrate chemistry. Academic, New York, pp 89–96 Deligianni E, Pattison S, Berrar D, Ternan NG, Haylock RW, Moore JE, Elborn SJ, Dooley JSG (2010) Pseudomonas aeruginosa cystic fibrosis isolates of similar RAPD genotype exhibit diversity in biofilm forming ability in vitro. BMC Microbiol 10:38–51 Déziel E, Lépine F, Milot S, Villemur R (2000) Mass spectrometry monitoring of rhamnolipids from a growing culture of Pseudomonas aeruginosa 57RP. Biochim Biophys Acta 31:145–152 Finnan S, Morrissey JP, O’Gara F, Boyd EF (2004) Genome diversity of Pseudomonas aeruginosa isolates from cystic fibrosis patients and the hospital environment. J Clin Microbiol 42:5783–5792 Fracchia L, Cavallo M, Martinotti MG, Banat IM (2012) Biosurfactants and bioemulsifiers biomedical and related applications—present status and future potentials. In: Ghista DN (ed) Biomedical science, engineering and technology, pp 325–370 Franzetti A, Tamburini E, Banat IM (2010) Applications of biological surface active compounds in remediation technologies. Adv Exp Med Biol 672:121–134 Jarvis FG, Johnson MJ (1949) A glycol-lipide produced by Pseudomonas aeruginosa. J Am Chem Soc 71:4124–4126 Kiewitz C, Tümmler B (2000) Sequence diversity of Pseudomonas aeruginosa: impact on population structure and genome evolution. J Bacteriol 182:3125–3135 Lang S, Wullbrandt D (1999) Rhamnose lipids—biosynthesis, microbial production and application potential. Appl Microbiol Biotechnol 51:22–32

7305 Li AH, Xu MY, Sun W, Sun GP (2011) Rhamnolipid production by Pseudomonas aeruginosa GIM 32 using different substrates including molasses distillery wastewater. Appl Biochem Biotechnol 163:600–611 Maier RM, Soberón-Chavez G (2000) Pseudomonas aeruginosa rhamnolipids: biosynthesis and potential applications. Appl Microbiol Biotechnol 54:625–633 Makkar RS, Cameotra SS, Banat IM (2011) Advances in utilization of renewable substrates for biosurfactant production. Appl Microbiol Biotechnol Express 1(1):1–5 Marchant R, Banat IM (2012a) Biosurfactants: a sustainable replacement for chemical surfactants? Biotechnol Lett 34:1597– 1605 Marchant R, Banat IM (2012b) Microbial biosurfactants: challenges and opportunities for future exploitation. Trends Biotechnol 11:558–565 Mata-Sandoval JC, Karns J, Torrents A (2001) Effect of nutritional and environmental conditions on the production and composition of rhamnolipids by P. aeruginosa UG2. Microbiol Res 155:249–256 Mikkelsen H, McMullan R, Filloux A (2011) The Pseudomonas aeruginosa reference strain PA14 displays increased virulence due to a mutation in ladS. PLoS One 6:e29113. doi:10.1371/ journal.pone.0029113 Mukherjee S, Das P, Sen R (2006) Towards commercial production of microbial surfactants. Trends Biotechnol 24:509– 515 Nitschke M, Costa SG, Contiero J (2005a) Rhamnolipid surfactants: an update on the general aspects of these remarkable biomolecules. Biotechnol Prog 21:1593–1600 Nitschke M, Costa SG, Haddad R, Gon alves LA, Eberlin MN, Contiero J (2005b) Oil wastes as unconventional substrates for rhamnolipid biosurfactant production by Pseudomonas aeruginosa B1. Biotechnol Prog 21:1562–1566 Perfumo A, Banat IM, Manganella F, Marchant R (2006) Rhamnolipid production by a novel thermophilic hydrocarbon-degrading Pseudomonas aeruginosa AP02-1. Appl Microbiol Biotechnol 71:132–138 Perfumo A, Rancich I, Banat IM (2010) Possibilities and challenges of biosurfactant uses in petroleum industry. In: Sen R (ed) Biosurfactants. Landes Bioscience, Austin, pp 135–145 Pfaffl MW (2001) A new mathematical model for relative quantification in real-time RT-PCR. Nucleic Acid Res 29(9): e45 Rahman KSM, Rahman TJ, McClean S, Marchant R, Banat IM (2002) Rhamnolipid biosurfactant production by strains of Pseudomonas aeruginosa using low-cost raw materials. Biotechnol Progress 18:1277–1281 Reis RS, Pereira AG, Neves BC, Freire DMG (2011) Gene regulation of rhamnolipid production in Pseudomonas aeruginosa—a review. Bioresource Technol 102:6377–6384 Smyth TJP, Perfumo A, Marchant R, Banat IM (2010) Isolation and analysis of low molecular weight microbial glycolipids, vol. 5, part 2. In: KN Timmis (ed) Handbook of hydrocarbon and lipid microbiology. Springer, Berlin, pp 3705– 3723 Stover CK, Pham XQ, Erwin AL, Mizoguchi SD, Warrener P, Hickey MJ, Brinkman FS, Hufnagle WO, Kowalik DJ, Lagrou M, Garber RL, Goltry L, Tolentino E, Westbrock-Wadman S, Yuan Y, Brody LL, Coulter SN, Folger KR, Kas A, Larbig K, Lim R, Smith K, Spencer D, Wong GK, Wu Z, Paulsen IT, Reizer J, Saier MH, Hancock RE, Lory S, Olson MV (2000) Complete genome

7306 sequence of Pseudomonas aeruginosa PAO1, an opportunistic pathogen. Nature 406:959–964 Tamura K, Dudley J, Nei M, Kumar S (2007) MEGA4: Molecular Evolutionary Genetics Analysis (MEGA) software version 4.0. Mol Biol Evol 24:1596–1599 Van Hamme JD, Singh A, Ward OP (2006) Physiological aspects, part 1 in a series of papers devoted to surfactants in microbiology and biotechnology. Biotechnol Adv 24:604–620 Waite RD, Papakonstantinopoulou A, Littler E, Curtis MA (2005) Transcriptome analysis of Pseudomonas aeruginosa

Appl Microbiol Biotechnol (2013) 97:7297–7306 growth: comparison of gene expression in planktonic cultures and developing and mature biofilms. J Bacteriol 187:6571– 6576 Zhang Y, Miller RM (1994) Effect of a Pseudomonas biosurfactant on cell hydrophobicity and biodegradation of octadecane. Appl Environ Microbiol 60:2101–2106 Zhu H, Bandara R, Conibear TCR, Thuruthyil SJ, Rice SA, Kjelleberg S, Givskov M, Willcox MDP (2004) Pseudomonas aeruginosa with LasI quorum-sensing deficiency during corneal infection. IOVS 45:1897–1903