Rhamnolipids are one of the most effective biosurfactants that are of great interest in industrial applications such as enhancing oil recovery, health care, cosmetics, pharmaceutical processes, food processing, detergents for protein folding, and bioremediation due to their unique characteristics such as low toxicity, surface active property to reduce surface/interfacial tensions, and excellent biodegradability. The genes and metabolic pathways for rhamnolipid synthesis have been well elucidated, but its cost-effective production is still challenging. Pseudomonas aeruginosa, the most powerful rhamnolipid producer, is an opportunistic pathogen, which limits its large scale production and applications. Rhamnolipid production using engineered strains other than Pseudomonas aeruginosa such as E. coli and Pseudomonas putida has received much attention. The highest yield of rhamnolipids is achieved when oil-type carbon sources are used, but using cheaper and renewable carbon sources such as lignocellulose would be an attractive strategy to reduce the production cost of rhamnolipids for various industrial applications.
The environmental concerns and risks of reduced energy and food resources arisen from daily life and industrial applications have become important topics. Chemicals are indispensable elements in the society because they are widely used in food, pharmaceutical industry, and many other industrial applications. How to produce such chemicals and how to eliminate their effect on environment are of great concern due to their potential side effects on the environment and human health. Surfactants are a class of chemicals consisted of both a hydrophilic head and a hydrophobic moiety. Surfactants are able to reduce surface and interfacial tension and form microemulsion when they are dissolved in different solvents. Surfactants have been widely used in both cosmetic and chemical industries. Although most surfactants are produced from plant oils, animal fat, and petro-chemicals, they are difficult to be degraded in nature, which may be a potential threat to the environment [1]. To reduce the effect on the environment and make full use of renewable resources, biosurfactants have great potential to be widely used due to their characteristics such as similar structures to chemically synthesized surfactants and nontoxicity to environment. Biosurfactants are either produced by organisms or synthesized from renewable resources [1].
The structures of biosurfactants produced by organisms are relatively simple compared with surfactants. The hydrophilic moiety includes amino acids, peptides, cations, and sugar groups [2]. The hydrophobic moiety is normally formed by saturated and unsaturated fatty acids. Based on these moieties, biosurfactants are classified as glycolipids, lipopeptides, lipoproteins, phospholipids, fatty acids, polymeric surfactants, and particulate surfactants [2]. Many bacteria have been shown to produce biosurfactants and detail information can be obtained from several reviews [2–5]. This review herein focuses on rhamnolipids that belong to glycolipids formed by carbohydrate and aliphatic acids. In addition to genetic regulation of rhamnolipid production, the variety and suitability of renewable resources used for rhamnolipid production are summarized in the review.
Rhamnolipids are comprised of one or two rhamnose units linked to one or two β-hydroxy fatty acids (Fig. 1). The length of the fatty acids is diverse from 8 to 16 [6]. Studies have shown that the most frequently produced rhamnolipids contains β-hydroxy fatty acid unit with ten carbons and there were approximately 60 kinds of rhamnolipids identified so far [7]. In addition to Pseudomonas species [8–10], several microorganisms from other species such as Acinetobacter calcoaceticus [11–14] have been shown to be able to produce rhamnolipids [7]. Pseudomonas aeruginosa are the main rhamnolipid producer and the yield of rhamnolipids varies when the strain was cultured in different medium or under different fermentation conditions [7].
Fig.1 Structures of rhamnolipids. The length of the carbon chain (m) may vary and the most common rhamnolipids contain 10-carbon chains
Rhamnolipids are one of well-characterized biosurfactants in terms of both function in industries and microbial production at genetic levels. Microbial production of rhamnolipids may be important for organism to obtain hydrophobic nutrient, bacterial swarming motility, and pathogenesis. Rhamnolipids are shown to be able to reduce surface tension of water, tension between water and oil, and form micelles with critical micelle concentration (CMC) ranging from 50 to 200 mg/L [15]. These characteristics make rhamnolipids useful for some industrial applications, ranging from detergents in cosmetics industry to emulsification reagent for enhancing oil recovery in oil industry [16]. This type of biosurfactants is thermal stable and can sustain activity within a pH range from 5 to 10 [17], which makes it an attractive biosurfactant that can be used in combination with other chemically synthesized surfactants and used under a condition with high temperature such as an oil field where oil reservoir temperature can be over 45 °C [18].
Rhamnolipids are composed of a unit of hydrophilic moiety-rhamnose and a hydrophobic moiety-β-hydroxy fatty acid. Using Pseudomonas aeruginosa (P. aeruginosa) as a model strain, genes that are essential for rhamnolipid production were identified and characterized. For the sugar moiety synthesis, deoxythymidine diphospho(dTDP)-L-rhamnose is the precursor for rhamnolipid synthesis (Fig. 2). This dTDP-L-rhamnose also serves as a substrate for lipopolysaccharide (LPS) synthesis in different bacteria as the O-antigen structure of LPS contains L-rhamnose [19,20]. Converting D-glucose-1-P to dTDP-L-rhamnose is achieved through four enzymes encoded by the rmlBDAC operon [19]. These four genes are conserved among bacteria. As dTDP-L-rhamnose is critical for both LPS and rhamnolipid synthesis, these enzymes can be served as good antibacterial targets [21]. D-glucose-1-P can be produced by phosphoglucomutase using D-glucose as a substrate and further converted into dTDP-L-rhamnose via four intermediates including dTDP-glucose, dTDP-6-deoxy-D-xylo-4-hexulose, and dTDP-6-deoxy-L-xylo-4-hexulose [12,22]. Expression of these genes is also regulated by the stationary phase sigma factor and quorum-sensing regulator [21].
Fig.2 Simplified metabolic pathways for rhamnolipid production. Abbreviations: Acyl-CoA, Acetyl- coenzyme; HAA, β-3-(3-hydroxyalkanoyloxy) alkanoic acid; PHA, polyhydroxyalkanoates; ACP, acyl carrier protein; LPS, lipopolysaccharide; RhlA, β-3-(3-Hydroxyalkanoyloxy) alkanoic acid synthetase; RhlB, Rhamnosyltransferase 1; RhlC, Rhamnosyltransferase 2. Glucose and fatty acids are highlighted in blue boxes, rhamnolipids and their precursors are highlighted in green, and other compounds are shown in black
The hydrophobic moiety of rhamnolipids is synthesized through the classical pathway of fatty acid synthesis starting from two-carbon units [12]. RhlA encoded by RhlA gene is indispensible for the synthesis of β-3-(3-hydroxyalkanoyloxy) alkanoic acid (HAA) that is the precursor of mono-rhamnolipids (23). Although there may be some debates on the substrate of RhlA, the lipid portion of rhamnolipids is from the pathway of fatty acid synthesis. To produce HAA-a moiety with ten carbons, β-hydroxyacyl-ACP (acyl carrier protein) needs to be synthesized through the fatty acid synthesis pathway which is similar in bacteria and different from mammals (12). Acetyl- coenzyme (CoA) derived from glucose or β-oxidation can be converted to malonyl-CoA by acetyl-CoA carboxylase. Malonyl-CoA and ACP will be converted to malonyl-ACP by malonyl-CoA:ACP transacylase-product of fabD gene (Fig. 2). There are two stages including initiation and elongation of fatty acid synthesis in E. coli. Three β-ketoacyl-ACP syntheses are critical for these steps. Some of the enzymes from P. aeruginosa have been also characterized [24–26].
To catalyze sugar unit and lipid portion to form rhamnolipids, other genes are needed. The rhlR gene was first identified to be important for rhamnolipids synthesis [27,28]. RhlR protein is suggested to be a transcription activator for the activation of rhlAB operon. In addition, the rhlIgene was identified to be responsible for the synthesis of butanol-L-homoserine lactone (C4-HSL) that is an autoinducer binding to RhlR [29]. The RhlR in complex with C4-HSL can recognize conserved “las box” to activate RhlAB operon. RhlB is encoded by rhlB gene and contains rhamnosyltransferase 1 activity that can produce mono-rhamnolipids using HAA (ten carbons) and dTDP-L-rhamnose as substrates. Although the RhlAB operon is regulated by RhlR and C4-HSL, other environmental factors such as nutritional status may also affect its expression because a study has shown that this operon was not expressed at the logarithmic phase even in the presence of these activators [30]. The rhlC gene also encodes a protein with rhamnosyltransferase 2 activity which can produce di-rhamnolipids when mono-rhamnolipids are synthesized. All related genes of rhamnolipids synthesis are also regulated by quorum sensing that is a step for production of autoinducers. These compounds are critical for cell-cell communication and activation of transcription regulators [12,31]. The regulation of rhamnolipid production has been described in recent reviews [12,32,33]. The two quorum sensing systems of P. aeruginosa are activated under certain conditions such as limited phosphate or nitrogen [34,35]. Therefore, it is not surprising that the precursors for rhamnolipid production were found to be affected by growth conditions and the composition of the cultural medium such as carbon sources [12]. Many studies have demonstrated the fermentation conditions and selection of carbon sources can affect the yield of rhamnolipids dramatically.
The yield of rhamnolipids by P. aeruginosa is the highest compared with other bacterial strains. As P. aeruginosa is an opportunistic pathogen, it limits the large-scale production of rhamnolipids. In addition to the pathogenic concern, the highest yield of rhamnolipids can be reached when P. aeruginosa was grown using plant oil as carbon source, which requires additional steps in downstream purification [36,37]. Based on the knowledge of genes that regulate rhamnolipid synthesis, metabolic engineering has been conducted to enhance the yield and reduce the cost in downstream purification steps, which has been reviewed in detail [32]. The yield of rhamnolipids from the engineered strains is still low, which may not fit for the industrial expectation. A recent study demonstrated that an engineered strain-P. putida KT2440 was able to produce rhamnolipids using glucose as a substrate and the production of rhamnolipids was not coupled with biomass accumulation [38].
It is well known that rhamnolipids are biodegradable biosurfactants and have a wide range of industrial applications. It has been noted that this type of biosurfactants are still not widely used mainly due to the high production cost [12]. To reduce the cost, two strategies have been commonly used. One is, as aforementioned, to conduct metabolic engineering on some strains to increase the yield. Although it is well understood for the genes that are critical for rhamnolipids synthesis, the yield of rhamnolipids from modified strains is still not high enough for industrial application [32]. The other one is to produce rhamnolipids using cheaper resources. The latter strategy seems to be promising if some wastes can be used as nutrient for the production of rhamnolipids and other biosurfactants [39,40].
Many carbon sources have been used for rhamnolipid production (Fig. 3) [41,42]. The yield of rhamnolipids varied when different types of carbon sources were used in fermentation [38]. The well characterized carbon sources can be summarized as sugars, glycerol [43], n-alkanes [44], oils [14] and others sources such as polycyclic aromatic hydrocarbons (PAHs) [45]. The highest yield of rhamnolipids can be obtained when oil-type carbon sources were used during fermentation, which may arise from the fact that oil-type carbon sources can be degraded through the β-oxidation pathway which can be energetically favorable to supply the hydrophobic moiety of rhamnolipids. For example, the yield of rhamnolipids can be reached from 10 g/L to 45 g/L when oil-type substrates were used as carbon sources [42]. Trummler et al. used resting cells to produce a mixture of rhamnolipids with a yield up to 45 g/L when rapeseed oil was used as a carbon source [46]. When glucose was used as a carbon source, the yield of rhamnolipids could reach 2 g/L to 6 g/L under different conditions [42,47]. Any wastes that can be used as aforementioned carbon sources should therefore have great potential in rhamnolipid production [40] (Table 1).
Fig.3 Some carbon sources used for rhamnolipid production. The commonly used carbon sources for rhamnolipid production are shown in black boxes. Wastes containing useful carbon sources are highlighted in gray. Wastes during oil production or refinement, or containing oil are not shown. Some of them are listed in Table 1
Tab.1 Renewable carbon sources that can be used for rhamnolipid production
Glycerol is a common substrate that can be utilized by many bacteria. Glycerol was demonstrated to serve as a carbon source for rhamnolipid production [69,70]. Therefore, any waste that can be converted into glycerol can also serve as a substrate for rhamnolipid production (Fig. 3). Glycerol can be obtained from several resources such as animal fat, oil, and waste of biodiesel production. Glycerin from biodiesel production is rich in glycerol and was demonstrated to serve as a carbon source for rhamnolipid production, but the price of glycerol from biodiesel production is considered to be high [49].
Oils are usually good substrates for rhamnolipid production. Soapstock is a waste during oil refinement and can be used as animal feed or converted into biodiesel [71]. A study showed that soapstock from sunflower oil process can be used for rhamnolipid production with a yield of 15.9 g/L. P. aeruginosa LBI was able to produce rhamnolipids by using unconventional carbon sources such as oily wastes from soybean, corn, babassu, cottonseed, and palm oil refinery [50,54]. The yield of rhamnolipids between 6.8 to 11.72 g/L was obtained [50]. It was shown that the consumption of fatty acids was at different rates for different oil wastes. Soybean oil soapstock was shown to be an attractive carbon source for rhamnolipid production [50]. The production of rhamnolipids using cottonseed may arise from consumption of glycerol component [50]. By varying carbon/nitrogen ratios in the cultural medium, Benincasa et al. was able to obtain a high yield of rhamnolipids using wastewater and soapstock from sunflower oil process [53]. The produced rhamnolipids in the soybean waste were able to remove crude oil from sand and were demonstrated to have antimicrobial activity against several bacteria [51,52,55].
Sugar and sugar-containing wastes from food-processing industries are also a promising substrate for rhamnolipid production. The yield of rhamnolipids is low in a suger-containing medium [12], but the cost of sugar-containing waste is lower than that of other wastes containing oil or glycerol. Whey is a byproduct from dairy industry and is rich in lactose, protein, and minerals. Whey can be converted into value added products such as food ingredient [1,72]. In addition to whey, molasses from sugar industry can also be utilized to produce rhamnolipids. Molasses are rich in sugars (~50% sucrose) and can be obtained at a lower price. Pseudomonas strains were able to produce rhamnolipids with a yield of 1.5 g/L to 3.2 g/L [64–66].
Another waste that can be converted to sugars is lignocellulose-which is the most abundant renewable substrate. Lignocellulose is composed of cellulose, hemicellulose and lignin. Cellulose can be enzymatically converted to glucose because it is formed by glucose units through β-1-4 glycosidic bonds [73]. Hemicellulose is formed by a mixture of xylose, mannose, galactose, xylose, and rhamnose. According to the source of materials, lignocellulose is classified into three groups: primary cellulosics-derived from special plants containing high content cellulose, agricultural waste cellulosics-byproducts from agricultural product/food processing, and municipal waste cellulosics-paper or paper related products derived from municipal solid wastes [1]. Converting lignocellulose to sugars and value-added chemicals has been studied for decades. Sugars from cellulose can be a valuable source for rhamnolipid production. Studies have shown that three main classes of cellulases are critical for converting cellulose to sugars. They are endoglucanase, exocellobiohydrolase, and β-glucosidase [74]. As cellulose and lignin form a tight complex, pretreatment of the lignocellulose is required to release cellulose. Hemicellulose can also be degraded in the pretreatment step [75]. A recent study showed that wheat straw can be pretreated with acids followed by enzymatic hydrolysis using enzymes produced by Trichoderma reesei NCIM 1186. The product was then used as a nutrient source for rhamnolipid production with a yield of 9.38 g/L [68]. This result clearly demonstrated that lignocellose can be used as a substrate for rhamnolipid production. Some byproducts such as furfurals from pretreatment of ligonocellose could inhibit bacterial growth [76]. Strains from Pseudomonas also have an advantage to use treated lignocellose as a substrate because they might be able to degrade furfurals [77].
Accumulated studies have demonstrated the wide application of rhamnolipids in many fields such as food process, bioremediation, and detergent industry [7,55,78–82]. Therefore, the needs for rhamnolipids will be increased. How to obtain rhamnolipids with a low cost is of great interest [1,39,72]. The highest yield (112 g/L) of rhamnolipids can be obtained using P. aeruginosa according to a patent. However, concerns still remain for P. aeruginosa because it is an opportunistic pathogen. Large scale fermentation of such a strain may not be allowed. To obtain rhamnolipids with lower cost, other alternative ways may be considered.
As P. aeruginosa might not be suitable for rhamnolipid production in large scales, some researchers are screening different strains from various environments. As many assay methods are available and well set up [42], several non-pathogenic strains have been screened. Among these strains, most of them are Pseudomonas species. P. chlororaphis was shown to be able to produce rhamnolipids using glucose as a carbon source [9]. A strain- Thermus thermophilus HB8 was shown to be able to produce both polyhydroxyalkanoates and rhamnolipids using glucose or sodium gluonate as the sole carbon source [13]. Although the rhamnolipid yield of these isolates was lower than that of P. aeruginosa, further modification of the strains may improve the yield dramatically.
Metabolic pathway engineering was also used to produce rhamnolipids using non-pathogenic strains (see reviews [32,42] ) because the genes involved in rhamnolipid production are well known. Several engineered strains have proven to produce rhamnolipids in bacteria such as E. coli and P. putida [8,38,83,84]. Although the yield of rhamnolipids was still lower than that in P. aeruginosa, this is a promising direction to reduce the cost for rhamnolipids production.
Studies have shown that fermentation conditions affect the yield of rhamnolipids and product-substrate conversion factor. Many factors such as carbon/nitrogen ratio and fermentation time are critical for rhamnolipid production. For example, the yield of rhamnolipids was found to vary at different cultural duration [55]. During 17 h incubation, rhamnolipids were detected (0.2 g/L). Additional study showed that rhamnolipids accumulated at different rates in two major stages. In the first stage, 3.54 g/L rhamnolipids were detected with a volumetric rhamnolipid productivity of 0.06 (g/L·h). In the second stage (72‒96 h), the rate was increased to 0.22 (g/L·h) and rhamnolipid concentration reached 5.64 g/L [55]. Batch and fed-batch fermentation strategies give various rhamnolipid yields. A kinetics study needs to be conducted when a suitable strain is obtained, which can reduce the cost dramatically.
Using renewable sources for rhamnolipid production is a direct way to reduce the production cost. The cost of nitrogen source is low compared with carbon sources [85]. Choosing a suitable carbon source has great impact on the production cost. Corn steep liquor is the commonly used nitrogen source for rhamnolipid production [86]. Based on the accumulated studies, the yield of rhamnolipids using a carbon source containing oil is higher than that using glucose [42]. Wastes from oil processes were shown to be a promising substrate for rhamnolipid production due to the high yield of rhamnolipids (Table 1). This type of wastes is relatively expensive and can be easily re-used as other resources such as animal feed. The lignocellulose waste will be a promising resource for rhamnolipid production as it is one of the most abundant resources [87]. It is noted that pre-treatment of lignocellulose is required for sugar generation. This type of waste is of great concern for their environmental effect. Utilization of such waste will be a good choice for rhamnolipid production. Different area/countries may have various types of lignocellulose wastes to be utilized. For example, empty fruit bunch (EFB) is a waste from palm oil production [88]. It is a common waste in Southeast Asia and its treatment is of important for the environment. EFB is a promising substrate for rhamnolipid production because its annual production is high and it can be converted into fermentable sugars [75,89]. In addition, mixing lignocellulose hydrolysates with wastes or wastewater from oil-processing industry or crude oil processing companies may also be attractive for rhamnolipid production.
In summary, rhamnolipids are important biosurfactants having potential application in industries. Rhamnolipids can be produced in low cost by using a suitable strain and a low-cost carbon source under optimized fermentation conditions.
This research is supported by the Science and Engineering Research Council (SERC) of the Agency for Science, Technology and Research (A*STAR) of Singapore (SERC grant number: 1526004161).
The authors have declared that no competing interests exist.
