- Email: [email protected]
Growth and lovastatin production by Aspergillus terreus under different carbohyrates as carbon sources
Author’s Accepted Manuscript Growth And Lovastatin Production by Aspergillus Terreus Under Different Carbohyrates as Carbon Sources Muhamad Hafiz Abd Rahim, Hanis H. Harith, Alejandro Montoya, Ali Abbas www.elsevier.com/locate/bab
PII: DOI: Reference:
S1878-8181(17)30045-2 http://dx.doi.org/10.1016/j.bcab.2017.04.011 BCAB547
To appear in: Biocatalysis and Agricultural Biotechnology Received date: 23 January 2017 Revised date: 22 February 2017 Accepted date: 20 April 2017 Cite this article as: Muhamad Hafiz Abd Rahim, Hanis H. Harith, Alejandro Montoya and Ali Abbas, Growth And Lovastatin Production by Aspergillus Terreus Under Different Carbohyrates as Carbon Sources, Biocatalysis and Agricultural Biotechnology, http://dx.doi.org/10.1016/j.bcab.2017.04.011 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Growth And Lovastatin Production by Aspergillus Terreus Under Different Carbohyrates as Carbon Sources Muhamad Hafiz Abd Rahim1,2, Hanis H Harith3, Alejandro Montoya1, Ali Abbas1* 1
School of Chemical and Biomolecular Engineering, The University of Sydney, Sydney, Australia
2
The Department of Food Science, Faculty of Food Science and Technology, Universiti Putra Malaysia,
Serdang, Malaysia 3
Department of Biomedical Science, Faculty of Medicine and Health Sciences, Universiti Putra Malaysia,
Serdang, Malaysia *Corresponding author: Associate Professor Ali Abbas, Director of the Laboratory for Multiscale Systems, School of Chemical and Biomolecular Engineering, J01 - Chemical Engineering Building, The University of Sydney. Ph: +61 2 9351 3002; E-mail: [email protected]
ABSTRACT Carbon source is a key component of metabolites synthesis in microorganisms. This work examined the effects of selected carbon sources in the form of carbohydrates, on the growth of Aspergillus terreus ATCC 20542 and the production of lovastatin. Slowly metabolised carbohydrates, such as D-galactose (consumption rate, r=3.11), produced a high microbial biomass, XFINAL (9.44 g/L) compared to other carbohydrates, but with a low biomass yield coefficient (YLOV/X=1.68). In contrast, D-ribose (YLOV/X=) which showed moderate biomass growth (XFINAL=8.78 g/L) and consumption rate (r=5.44 g/day), produced the highest lovastatin amount (51.81 mg/L, day 6). These indicate little correlation between biomass growth and lovastatin production. Notably, culture consisting of pellets with short hairy surface feature is associated with enhanced lovastatin production. Our findings suggest that the production of lovastatin by Aspergillus terreus is highly influenced by the choice of carbohydrates that will shape the pellet morphology rather than the rate of carbohydrates metabolism. Keywords: Aspergillus terreus; carbon source; lovastatin ; fermentation, filamentous fungi
Abbreviation(s) r, consumption rate; YLOV/X, biomass yield coefficient; YLOV/S, substrate yield coefficient; XFINAL, biomass growth; umax, specific growth rate 1.0 INTRODUCTION Lovastatin (C24H36O5) is a commonly prescribed cholesterol-lowering drug belonging to the statins group that acts by competitively inhibiting the formation of 3-hydroxy-3-methyl glutaryl CoA (HGM-CoA) reductase, a rate limiting enzyme of cholesterol biosynthesis in the liver. This compound is produced naturally by a few fungus species including the commonly known Aspergillus terreus ATCC 20542 (Manzoni et al., 1998). Although unclear, lovastatin is thought to have a protective role in fungus on the basis that it is effective at reducing sterols important for the growth of other microbes (Debakey and Endo, 2008). Recent studies also suggest that lovastatin may have pharmacological potential beyond cholesterol-lowering ability, such as for the treatment of cancer (Hindler et al., 2006), Alzheimer’s disease (Sparks, 2011) and osteoporosis (Gonyeau, 2005). The factors that influence the production of lovastatin include the types of carbon and nitrogen source (Hajjaj et al., 2001), vitamin B content (Bizukojc et al., 2007) and physical factors such as aeration rate (Bizukojc and Ledakowicz, 2008), dissolved oxygen concentration (Casas López et al., 2004), pH (Lai et al., 2005), and types of fermentation (Barrios-González and Miranda, 2010). One of the most important aspects, carbon source, has been widely studied in terms of its impact on the growth and metabolism of A. terreus (Casas López et al., 2003; Hajjaj et al., 2001; Jia et al., 2009a, 2009b; Kumar et al., 2000). A higher ratio of carbon to nitrogen (also known as ‘nitrogen starvation’) is thought to be an important factor in lovastatin production (Casas López et al., 2003). Similarly, slowly-metabolised carbon sources, such as α-lactose and glycerol, are better carbon choices which favour higher lovastatin production (Casas López et al., 2003). The use of rapidly-metabolised carbon sources, such as as D-glucose, may lead to reduced lovastatin production via catabolite repression mechanism (Lai et al., 2007). It also favours the production of primary metabolite such as itaconic acid (Lai et al., 2007). Despite this, some studies have not reported repressed lovastatin yield when these carbons (D-glucose and fructose) were used (Casas López et al., 2003; Hajjaj et al., 2001). The present study investigated the effects of different sources of carbon in the form of carbohydrate on the growth and metabolite production of A. terreus under similar culture conditions. Here, we evaluated the effects of different carbohydrates comprised of monosaccharides (D-glucose, D-xylose, D-fructose, D-galactose, Dmannose, D-ribose, D-arabinose) and disaccharides (α-lactose, maltose, D-sucrose) on lovastatin production and the morphology of A. terreus.
2.0 MATERIALS AND METHODS 2.1 Culture condition The culture conditions used in this study have been described previously with slight modifications (Abd Rahim et al., 2015). The basal media contained minimal salt media (containing KH2PO4, 0.4 g/L, MgSO4•7H2O, 0.2 g/L, and NaCl2 0.4 g/L) to ensure that the effect on the yield and morphology is due to the carbon sources. 1 mL aliquot of spore suspension (107 spores/mL) was inoculated into 125 mL conical flask containing 50 mL of media at 30 ˚C, 180 rpm in a shaking incubator. Seed culture was performed 24 to 30 hours prior to the actual experiment, using 8 g/L of yeast extract and 10 g/L of α-lactose in the basal salt media to allow the fungus to achieve exponential growth before cultivation. Ten different carbohydrates comprised of monosaccharides (D-glucose, D-xylose, D-galactose, D-mannose, Darabinose, D-fructose, D-ribose) and disaccharides (D-maltose, α-lactose, D-sucrose) were used in this study at 20 g/L (Fig. 1). Glycerol was used in control experiment. Initial pH was kept at 6.5. No further pH control was applied during the duration of the experiment. 2.2 Dry cell weight determination The biomass yield was determined gravimetrically. The recovery of fungus biomass was carried out through filtration on No. 2 Whatman filter paper. The biomass was washed twice with distilled water, followed by drying at 80 ºC for 24 hours or until constant weight is achieved. 2.3 Analytical method Lovastatin (Sigma Aldrich, Sydney, Australia) standard was prepared according to manufacturer’s instruction. The main analysis was performed using High Performance Liquid Chromatography (HPLC), Agilent 1200, using a UV detector at a wavelength of 238 nm, with a reference wavelength of 360 nm. The column used was XDB Eclipse Zorbax C-18. Parameters used for HPLC were as follows: Column temperature at 30˚C, sample chamber temperature at 4˚C, flow rate at 1 ml/min, injection volume at 10 µL and 95% acetonitrile and 0.1% phosphoric acid solution as mobile phases. The resultant HPLC peaks are shown in Fig. 2. Glycerol concentration was determined using free glycerol calorimetric detection kit (Sigma-Aldrich), which is based on fluorescence reading at 575 nm. The sugar content in the samples was determined using phenolsulfuric acid method (Masuko et al., 2005) with slight modification. 2.4 Image capture procedure The images of the fungal pellet were taken using Digital Blue QX7 Computer Microscope (Digital Blue, Atlanta, USA), a light microscope equipped with camera. The fungal pellets were harvested from the fermentation system at day 4 (where the fully formed morphology was observed). The pellets were placed on a
microscope slide and the excess liquid around the pellet was dried using Kimtech Wipes. The images were taken immediately following the drying procedure. 2.5 Kinetic calculations The common growth kinetic stoichiometric parameters were used to describe the growth of A. terreus. These include: The substrate consumption rate, r is calculated in days r= Si is the initial concentration; while So is the final concentration of substrate (s).
The yields, Y are both measured in relation to the biomass production and total of substrate consumed in the reaction YLOV/X = YLOV/S = Pmax in this equation equals to maximum lovastatin concentration, and Pi is the initial lovastatin concentration.
μmax is specific growth rate obtained by plotting the specific biomass weight versus time μmax = Xt is the production of biomass at specific time while Xi is the initial biomass weight.
XFINAL is the final biomass weight at day 6, respectively. 2.6 Statistical analysis All experiments were conducted at least in quadruplicates. Data obtained for metabolite production involving lovastatin was analysed using one-way ANOVA with Tukey post hoc test. All statistical analysis was performed using Graphpad Prism, version 7.0. Error bars represent 95% confidence interval.
3.0 RESULTS AND DISCUSSION 3.1 The relationship between kinetic parameters, morphology and lovastatin production of A. terreus Morphology of fungus is one of the key factors that influence the production of metabolites. The pellet form is known to produce better metabolites yield in submerged fermentation compared to filamentous fungus, therefore the pellets were used for seed culture (Casas López et al., 2005; Rodríguez Porcel et al., 2005). In general, we observed spherical, round pellets with all treatments. However, there were three distinct morphologies of A. terreus, mainly grouped based on their hair characteristics. Fig. 3 showed the images of these fungal pellets under the light microscope, which are classified as short-haired, long-haired or no-hair characteristics. These hair characteristics will serve as our point of reference throughout the discussion. Generally, we observed that the production of lovastatin is mainly mixed-growth associated with all the substrates tested, as lovastatin was detected in the early and late phase of biomass production. This was also observed by Bizukojc et al., where similar pattern of production was observed in the presence of α-lactose (Bizukojc and Ledakowicz, 2007). The kinetic parameters for the production of lovastatin by A. terreus using different carbohydrates are shown in Table 1. Both lowest and highest yield coefficients (YLOV/X and YLOV/S) were detected within monosaccharides group. Lowest yield coefficients were detected in the presence of Dgalactose, D-arabinose and D-glucose; while the highest was observed in the presence of monosaccharides, specifically D-xylose and D-fructose. Disaccharides, in general, gave a satisfactory production of lovastatin relative to the biomass and substrate consumption, as their YLOV/X and YLOV/S values were high, with maltose giving the highest yield coefficient. This observation is also in contrasts with the theory that slowly metabolised carbon is better for lovastatin production (Bizukojc and Ledakowicz, 2009; Casas López et al., 2003). It may be due to the limited range of carbons tested in this study (only a maximum of three carbons were tested) or the different compositions of media which may affect the production of lovastatin. The carbohydrates with high lovastatin yield, namely D-xylose, D-fructose, D-ribose, D-maltose and D-sucrose exhibited similar features throughout the investigation. These include short-haired pellet morphology, moderate specific growth rate (umax), and moderate final biomass (Xfinal), with pH approaching 7. Previous studies have shown that short-haired pellets (Casas López et al., 2005) and moderate to high ØFINAL (Casas López et al., 2005; Rodríguez Porcel et al., 2006, 2005) are associated with good lovastatin production. This may be due to the balance between growth and oxygen transfer, since short hair and higher diameter allow better gas absorption. Conversely, we observed pellets with long hair feature, displaying gelatinous-like morphology in Darabinose (Fig. 3c), which corresponded with lowest X FINAL. This is consistent with previous reports suggesting that long hair morphology may reduce the nutrient transfer ability (Jia et al., 2009a). Others with low yield coefficient, namely D-glucose, D-galactose, and α-lactose exhibited no-hair and compact morphologies with pH away from neutral. The absence of hair may limit the uptake of nutrients into the fungal pellet, which could lead to lower lovastatin production. Nevertheless, there is still no plausible explanation on why or how hair characteristics contribute to lovastatin production.
3.2 Lovastatin production and carbohydrates consumption by A. terreus Slowly-metabolised carbons, particularly α-lactose and glycerol, have been described as the optimal carbon source for lovastatin production (Bizukojc and Ledakowicz, 2009; Casas López et al., 2003). Table 2 demonstrated several lovastatin yields from previous findings using sole carbon and nitrogen sources. From the table, it is clear that the production of lovastatin vary in several studies, most likely due to different experimental conditions. In addition, only certain carbon sources were used regularly in previous studies (glucose, glycerol, lactose, fructose) while other carbon sources (maltose, xylose, arabinose, galactose) were hardly used during the investigations. Our findings show that there are other carbohydrates substrates with higher efficiency for lovastatin production such as D-xylose and D-fructose. Table 3 shows the production of lovastatin over the span of 4 days, while Fig. 4 demontrates the carbohydrates consumption rate over the span of 6 days. Unlike some fungal system (Bettiga et al., 2009; Klaubauf et al., 2013), we observed that the group or structure of sugars (mono- and disaccharides, or pentoses and hexoses) have no effects on the production of biomass and lovastatin. In general, these carbohydrates can be characterised according to the lovastatin production, low (5 – 11 mg/L), medium (18 – 32 mg/L) and high (43 – 57 mg/L). Carbohydrates which produced significantly lower lovastatin yield compared to the control are D-arabinose (5.78 mg/L), D-galactose (11.16 mg/L) and D-glucose (17.28 mg/L). Of note, D-arabinose and D-galactose had low uptake rate and both were detectable in the media even after 6 days of fermentation. These correlated with a halt in lovastatin production at day 4, suggesting that both sources may not be suitable for the production of metabolites. Like other microorganisms, fungi are known for their ability to develop a mechanism to readily uptake available carbon sources in nature for survival. Given that L-arabinose is more common and readily available than D-arabinose, this may explain why A. terreus was unable to efficiently use the latter as the carbohydrate source. In fact, the most established pathway known for D-arabinose catabolism in fungi is the oxidoreductase pathway, which involves L-arabinose and not Darabinose (Seiboth and Metz, 2011). D-galactose, on the other hand, has been shown to be poorly metabolised by certain types of fungi, such as A. niger (Edgecombe, 1938). The inhibitory effect of D-galactose on lovastatin production is most likely due to the lack of both galR and galX regulators that control the utilisation of Dgalactose for growth in A. terreus (Christensen et al., 2011). Given that only metabolite production was inhibited but not its biomass growth, this indicates that additional factors or pathways may be involved, and thus reflects the complex nature of the fungi. Although D-glucose produced significantly lower lovastatin than glycerol, the amount (17.28 mg/L) is significantly higher than both D-arabinose and D-galactose. Apart from the unfavourable morphology as discussed previously, the lower production of lovastatin may also be attributed to the well-known catabolite repression mechanism of glucose on metabolite production (Hajjaj et al., 2001). Our data also showed that this carbohydrate was metabolised rapidly, perhaps due to its abundance in nature (Lehninger et al., 2005) using established pathways such as the classical Embden-Meyerhof or pentose phosphate pathway (Bonnarme et al., 1995; Greene Lilly and Barnett, 1956).
The medium group consist of glycerol (25.58 mg/L), D-mannose (28.14 mg/L), α-lactose (30.53 mg/L) and Dmaltose (31.94 mg/L) and were grouped together as they showed no significance when compared to glycerol (control). Except for D-mannose, all these carbohydrates were slowly consumed by A. terreus. Although the production of lovastatin in α-lactose and glycerol started off quite slow, the production continued up to day 6. In contrast, lovastatin production in D-mannose peaked at day 2 with little change observed until day 6. This may suggest that the production in D-mannose was not limited by carbohydrate availability, but rather, the production was halted after certain amount of lovastatin was produced. While this may indicate that D-mannose is unfavourable for the fermentation of A. terreus, the fungus appeared to tolerate D-mannose better than Dgalactose and D-arabinose (the low-production group) as it was consumed until exhausted, accompanied with higher production of lovastatin. The consumption of D-mannose may be aided by mstA, a high-affinity transporter present in A. terreus which efficiently transports D-glucose, D-mannose and D-xylose into cells (Vankuyk et al., 2004). Nevertheless, it is quite surprising to find that α-lactose and glycerol did not produce as much lovastatin, compared to previous reports (Casas López et al., 2003; Pecyna and Bizukojc, 2011; Szakács et al., 1998). Apart from the limited types of carbon comparison in those studies, these inconsistencies may be attributed to the variation in the media composition. For example, complete media consisting of trace metals and extra nutrients (for example, biotin, manganese chloride and sodium molybdate) was used in previous studies which is likely to affect the interactions and relationships between carbons and fungus utilisation (Casas López et al., 2003). More importantly, both α-lactose and glycerol also produced high biomass. This suggests that these carbohydrate types may favour fatty acid biosynthetic pathway (refer to Fig. 5 for metabolite production pathway) instead of metabolite production. As such, this may compete with the lovastatin production pathway. For α-lactose in particular, the limited lovastatin production may also be explained by the catabolism of α-lactose into its respective monomers which usually involve two pathways - extracellular or intracellular hydrolysis (Seiboth and Metz, 2011). Perhaps, given that α-lactose can be hydrolysed to D-galactose and D-glucose, this may explain the limited production of lovastatin as both D-galactose and D-glucose are not the optimal carbohydrates for lovastatin production. Similarly, D-maltose can be hydrolysed into two units of glucose, which might also impede the production of lovastatin. While the culture treated with glycerol displayed favourable pellet morphology (short hair) and known to be slowly-metabolised, only moderate lovastatin concentrations were produced. Glycerol treatment produced the highest biomass which may reflect that this substrate favours the production of biomass rather than metabolite production, as discussed earlier. The group with significantly higher lovastatin (>40 mg/L) production than glycerol consists of D-xylose, Dfructose, D-sucrose and D-ribose. We observed that the carbohydrate consumptions varied between moderate to high (r=4.70 – 5.53), but not as fast as D-glucose (r=5.85), and not as slow as α-lactose (r=3.81) or glycerol (r=3.56). Our findings suggest that one of the prerequisite for good lovastatin production is the fungus’s ability to significantly improve its production during stationary phase (day 4, when biomass started to stabilise. Refer to Fig. 4) after the initial production at day 2. In particular, lovastatin production improved by 23%, 37%, 48% and 59% in D-ribose, D-sucrose, D-fructose and D-xylose, respectively after day 2. The uptake of D-xylose [56.46 g/L] potentially involves mstA transporter similar to D-glucose, but no repressive features that may block lovastatin production has been reported. Furthermore, D-xylose is one of the most abundant carbohydrates
found in nature. Therefore, A. terreus most likely have developed the ability to uptake this carbohydrate for growth and metabolite production. As for D-ribose, despite being an epimer (at C’2) of D-arabinose, it was metabolised quite efficiently by A. terreus with a considerable lovastatin yield (42.97 mg/L). It is likely that Dribose is more readily utilised than D-arabinose due to its structural difference and high availability in nature (found in all living cells). On the other hand, D-fructose and D-sucrose (with a yield of 51.08 mg/L and 47.40 mg/L, respectively) have been traditionally used for fermentation involving different fungus species including A. terreus with desirable lovastatin production (Casas López et al., 2003). Lopez et al. showed a satisfactory production of lovastatin using D-fructose which was thought to result from increased biomass growth; however this was not observed in our study. This may be explained by variations in experimental conditions (for example, the use of rich media culture (Casas López et al., 2003)) which may affect A. terreus growth and metabolism differently. D-sucrose may be a good carbohydrate for lovastatin production compared to other disaccharides as its hydroxylation would yield D-fructose (and D-glucose). The presence of D-fructose, hypothetically, may be the main factor that improves the production of lovastatin. 4.0 Conclusion Our findings show that the fermentation of lovastatin is highly dependent on the type of the carbohydrate sources used. Furthermore, the effects of certain carbons on pellet morphology may play a major role in defining the amount of lovastatin being produced. Conversely, the metabolism rate of the carbons was not as important as initially thought. Unlike the morphology, the biomass is not likely to be related to lovastatin production as no apparent trend between biomass and lovastatin production was observed in this experiment.
Conflict of interest statement The authors have declared no conflict of interest
Acknowledgement Authors would like to thank Bosch Mass Spectrometry Facility, University of Sydney, for the HPLC system and Malaysian Government for the scholarships. This research did not receive any specific grant from funding agencies in the public, commercial, or not-forprofit sectors
References
Abd Rahim, M.H., Hasan, H., Montoya, A., Abbas, A., 2015. Lovastatin and (+)-geodin production by Aspergillus terreus from crude glycerol. Eng. Life Sci. 15, 220–228. Barrios-González, J., Miranda, R., 2010. Biotechnological production and applications of statins. Appl. Microbiol. Biotechnol. 85, 869–883. Bettiga, M., Bengtsson, O., Hahn-Hägerdal, B., Gorwa-Grauslund, M., 2009. Arabinose and xylose fermentation by recombinant Saccharomyces cerevisiae expressing a fungal pentose utilization pathway. Microb. Cell Fact. 8, 1–12. Bizukojc, M., Ledakowicz, S., 2009. Physiological, morphological and kinetic aspects of lovastatin biosynthesis by Aspergillus terreus. Biotechnol. J. 4, 647–664. Bizukojc, M., Ledakowicz, S., 2008. Biosynthesis of lovastatin and (+)-geodin by Aspergillus terreus in batch and fed-batch culture in the stirred tank bioreactor. Biochem. Eng. J. 42, 198–207. Bizukojc, M., Ledakowicz, S., 2007. A macrokinetic modelling of the biosynthesis of lovastatin by Aspergillus terreus. J. Biotechnol. 130, 422–435. Bizukojc, M., Pawlowska, B., Ledakowicz, S., 2007. Supplementation of the cultivation media with B-group vitamins enhances lovastatin biosynthesis by Aspergillus terreus. J. Biotechnol. 127, 258–268. Bizukojc, M., Pecyna, M., 2011. Lovastatin and (+)-geodin formation by Aspergillus terreus ATCC 20542 in a batch culture with the simultaneous use of lactose and glycerol as carbon sources. Eng. Life Sci. 11, 272– 282. Bonnarme, P., Gillet, B., Sepulchre, A.M., Role, C., Beloeil, J.C., Ducrocq, C., 1995. Itaconate biosynthesis in Aspergillus terreus. J. Bacteriol. 177, 3573–8. Casas López, J.. L., Sánchez Pérez, J.. A., Fernández Sevilla, J.. M., Acién Fernández, F.. G., Molina Grima, E., Chisti, Y., Casas Lopez, J.L., Sanchez Perez, J.A., Fernandez Sevilla, J.M., Acien Fernandez, F.G., 2003. Production of lovastatin by Aspergillus terreus: effects of the C:N ratio and the principal nutrients on growth and metabolite production. Enzyme Microb. Technol. 33, 270–277. Casas López, J.L., Sánchez Pérez, J.A., Fernández Sevilla, J.M., Acién Fernández, F.G., Molina Grima, E., Chisti, Y., 2004. Fermentation optimization for the production of lovastatin by Aspergillus terreus: use of response surface methodology. J. Chem. Technol. Biotechnol. 79, 1119–1126. Casas López, J.L., Sánchez Pérez, J.A., Fernández Sevilla, J.M., Rodríguez Porcel, E.M., Chisti, Y., 2005. Pellet morphology, culture rheology and lovastatin production in cultures of Aspergillus terreus. J. Biotechnol. 116, 61–77. Christensen, U., Gruben, B.S., Madrid, S., Mulder, H., Nikolaev, I., de Vries, R.P., 2011. Unique regulatory mechanism for D-galactose utilization in Aspergillus nidulans. Appl. Environ. Microbiol. 77, 7084–7. Debakey, L., Endo, A., 2008. A gift from nature: the birth of the statins. Nat. Med. 14, 1050–2. Edgecombe, A.E., 1938. The Effect of Galactose on the Growth of Certain Fungi. Mycologia 30, 601–624 CR– Copyright 169; 1938 Mycological So. Fernandes, S., Murray, P., n.d. Metabolic engineering for improved microbial pentose fermentation. Bioeng. Bugs 1, 424–8. Gonyeau, M.J., 2005. Statins and osteoporosis: a clinical review. Pharmacotherapy 25, 228–43. Greene Lilly, V., Barnett, H.L., 1956. The Utilization of D- and L-Arabinose by Fungi. Am. J. Bot. 43, 709– 714.
Hajjaj, H., Niederberger, P., Duboc, P., 2001. Lovastatin Biosynthesis by Aspergillus terreus in a Chemically Defined Medium. Appl. Environ. Microbiol. 67, 2596–2602. 1 Hindler, K., Cleeland, C.S., Rivera, E., Collard, C.D., 2006. The role of statins in cancer therapy. Oncologist 11, 306–15. doi:10.1634/theoncologist.11-3-306 Jia, Z., Zhang, X., Cao, X., 2009a. Effects of carbon sources on fungal morphology and lovastatin biosynthesis by submerged cultivation of Aspergillus terreus. Asia-Pacific J. Chem. Eng. 4, 672–677. Jia, Z., Zhang, X., Zhao, Y., Cao, X., 2009b. Effects of divalent metal cations on lovastatin biosynthesis from Aspergillus terreus in chemically defined medium. World J. Microbiol. Biotechnol. 25, 1235–1241. Klaubauf, S., Ribot, C., Melayah, D., Lagorce, A., Lebrun, M.-H., de Vries, R.P., 2013. The pentose catabolic pathway of the rice-blast fungus Magnaporthe oryzae involves a novel pentose reductase restricted to few fungal species. FEBS Lett. 587, 1346–1352. Kumar, M.S., Kumar, P.M., Sarnaik, H.M., Sadhukhan, A.K., 2000. A rapid technique for screening of lovastatin-producing strains of Aspergillus terreus by agar plug and Neurospora crassa bioassay. J Microbiol Methods 40, 99–104. Lai, L.-S.T., Hung, C.-S., Lo, C.-C., 2007. Effects of lactose and glucose on production of itaconic acid and lovastatin by Aspergillus terreus ATCC 20542. J. Biosci. Bioeng. 104, 9–13. Lai, L.-S.T., Tsai, T.-H., Wang, T.C., Cheng, T.-Y., 2005. The influence of culturing environments on lovastatin production by Aspergillus terreus in submerged cultures. Enzyme Microb. Technol. 36, 737–748. Lehninger, A.L., Nelson, D.L., Cox, M.M., 2005. Lehninger Principles of Biochemistry. W. H. Freeman. Manzoni, M., Rollini, M., Bergomi, S., Cavazzoni, V., 1998. Production and purification of statins from Aspergillus terreus strains. Biotechnol. Tech. 12, 529–532. Marcin Bizukojc, Ledakowicz, S., 2005. Influence of environmental factors on mycelial growth and biosynthesis of lovastatin by Aspergillus terreus. Biotechnologia-monografie 2, 25–36. Masuko, T., Minami, A., Iwasaki, N., Majima, T., Nishimura, S.-I., Lee, Y.C., 2005. Carbohydrate analysis by a phenol-sulfuric acid method in microplate format. Anal. Biochem. 339, 69–72. Pecyna, M., Bizukojc, M., 2011. Lovastatin biosynthesis by Aspergillus terreus with the simultaneous use of lactose and glycerol in a discontinuous fed-batch culture. J. Biotechnol. 151, 77–86. Peterson, W.H., Fred, E.B., Anderson, J.A., 1922. THE Fermentation of hexoses and related compounds by certain pentose fermenting bacteria. J. Biol. Chem. 53, 111–123. Rodríguez Porcel, E.M., Casas López, J.L., Sánchez Pérez, J.A., Fernández Sevilla, J.M., Chisti, Y., 2005. Effects of pellet morphology on broth rheology in fermentations of Aspergillus terreus. Biochem. Eng. J. 26, 139–144. Rodríguez Porcel, E.M., Casas López, J.L., Sánchez Pérez, J.A., Fernández Sevilla, J.M., García Sánchez, J.L., Chisti, Y., 2006. Aspergillus terreus Broth Rheology, Oxygen Transfer, and Lovastatin Production in a Gas-Agitated Slurry Reactor. Ind. Eng. Chem. Res. 45, 4837–4843. doi:10.1021/ie0600801 Seiboth, B., Metz, B., 2011. Fungal arabinan and l-arabinose metabolism. Appl. Microbiol. Biotechnol. 89, 1665–1673. doi:10.1007/s00253-010-3071-8 Sparks, D.L., 2011. Alzheimer disease: Statins in the treatment of Alzheimer disease. Nat Rev Neurol 7, 662– 663. Szakács, G., Morovján, G., Tengerdy, R., 1998. Production of lovastatin by a wild strain of Aspergillus terreus. Biotechnol. Lett. 20, 411–415. Vankuyk, P.A., Diderich, J.A., MacCabe, A.P., Hererro, O., Ruijter, G.J.G., Visser, J., 2004. Aspergillus niger mstA encodes a high-affinity sugar/H+ symporter which is regulated in response to extracellular pH. Biochem. J. 379, 375–83.
Figure 1: The chemical structure of monosaccharides and disaccharides (pentoses and hexoses) used in this study. The chemical structures have been shown to influence the carbon consumption and metabolism in several microorganisms (Fernandes and Murray, n.d.; Peterson et al., 1922). From top left: A) D-glucose B) D-mannose C) D-galactose D) Dribose F) D-fructose G) D-arabinose H) α-lactose I) D-maltose J) D-sucrose
Figure 2: HPLC peaks obtained from the fermentation broth of A. terreus. Untreated samples taken directly from fermentation showed a good peak separation under isocratic elution mode using UV detector. Co-metabolites of lovastatin, sulochrin and (+)-geodin appear first around fourth and seventh minutes, followed by lovastatin at eleventh minutes. 238 nm is used as the main signal wavelength with 280 nm as the reference wavelength.
Figure 3: Three major pellet morphologies observed during fermentation of A. terreus at day 6. From left – no-hair (e.g. D-glucose-treated), short-hair (e.g. D-fructose-treated) and longhair (e.g. D-arabinose-treated) morphologies. Pellets with no-hair feature were more dense with minimal tendency to clump. Pellets with short-hair feature was less dense with higher tendency to clump, and exhibited patchy hair growth. Pellets with long-hair feature generally have thick filamentous hair and hollow, porous pellet.
70 60 Lovastatin (mg/L)
50 40 Day 2
30
Day 4
20
Day 6
10 0
Carbohydrates 12 11
Biomass (g/L)
10 9 8 7
Day 2
6
Day 4
5
Day 6
4
Carbohydrates
Figure 4: Lovastatin and biomass production under different carbohydrates treatment over the span of 6 days using 4 g/L yeast extract as nitrogen source. Error bar represents 95% confidence.
Figure 5: The pathway of metabolite production by A. terreus ATCC 20542. Note that all metabolite production requires Acetyl CoA as its main precursor. The pathways for lovastatin, sulochrin, (+)-geodin and fatty acids production require the combination of Acetyl CoA and Malonyl CoA. The channelling of more precursors towards one pathway may potentially cause deprivation of precursors for other pathways.
Table 1: The effects of different carbohydrates on yield and fermentation parameters by A. terreus after 6-days fermentation. a Carbohydrate
Lovastatin
YLOV/X YLOV/S
µmax
XFINAL
Final pH
(-saccharides)
Yield (mg/L)
(mg/g) (mg/g/L)
(1/d)
(g/L)
D-ribose
42.97 ± 1.8****
6.56
2.42
2.48
9.36 ± 0.5
7.06 ± 0.02
D-xylose
56.46 ± 4.8****
9.69
3.49
1.97
8.64 ± 1.2
7.19 ± 0.04
D-mannose
28.14 ± 5.7ns
5.20
1.55
2.41
8.02± 1.4
7.16 ± 0.03
D-arabinose
5.78 ± 0.9****
1.19
0.66
2.41
7.28 ± 0.2
6.25 ± 0.07
3.12
0.96
2.07
8.74 ± 0.7
7.46 ± 0.03
Mono-
*
D-glucose
17.28 ± 2.3
D-fructose
51.81 ± 1.2****
8.68
3.11
2.47
8.78 ± 0.3
7.01 ± 0.3
D-galactose
11.16 ± 1.2****
1.68
1.02
2.07
9.44 ± 0.3
6.19 ± 0.08
à-lactose
30.53 ± 0.6ns
4.47
2.01
2.47
9.64 ± 0.2
6.43 ± 0.05
D-maltose
31.94 ± 2.3ns
6.06
2.08
1.67
8.08 ± 0.6
6.91 ± 0.06
D-sucrose
47.40 ± 2.6****
7.15
2.94
2.17
9.44 ± 0.6
6.87 ± 0.08
25.68 ± 2.9
3.46
1.74
2.47
10.44 ± 0.7
6.88 ± 0.02
Di-
Control Glycerol
* The data are presented as means ± 95% confidence level. Analysis of variance (one-way ANOVA) using Tukey’s post-hoc test against the control (glycerol) was conducted. The number of (*) indicates significance when compared to glycerol. ns – not significant
The conditions for fermentation were minimal nutrient media
(KH2PO4, 0.4 g/L, MgSO4•7H2O, 0.2 g/L, NaCl2, 0.4 g/L, yeast extract, 4 g/L and the substrate of interest), temperature at 30 ˚C, initial pH at 6.5 and speed at 180 rpm in a shaking incubator. Seed culture was performed 24 to 30 hours prior to the actual experiment.
Table 2: The lovastatin production reported in previous studies
Carbon sourcea
Nitrogen sourcea
Product titreb
Strain and references
Glucose (20 g/L)
Glutamic acid (9.8 g/L)
(Hajjaj et al., 2001)
Glucose (45 g/L)
YE (12.5 g/L)
37 mg/L, but only produced after glucose is exhausted 12 mg/L
Fructose (20 g/L)
YE (1.33 g/L)
120 mg/L
Lactose (20 g/L)
YE (1.33 g/L)
90 mg/L
Lactose (20 g/L)
YE (8 g/L)
35 mg/L
Lactose (45 g/L)
Glutamic acid (9.8 g/L)
25 mg/L
Lactose (5-40 g/L)
YE (2-12 mg/L)
5 -110 mg/L
Sucrose (50 g/L)
CSL (10 g/L)
Glycerol (20 g/L)
YE (1.33 g)
40 % less than lactose 90 mg/L
Glycerol (20 g/L)
Glutamic acid (9.8 g/L)
6 mg/L
Glycerol (40 g/L)
YE (4 g/L)
40 mg/L
(Marcin Bizukojc and Ledakowicz, 2005) (Casas López et al., 2003) (Casas López et al., 2003) (Marcin Bizukojc and Ledakowicz, 2005) (Hajjaj et al., 2001) (Bizukojc and Ledakowicz, 2007) (Szakács et al., 1998) (Casas López et al., 2003) (Hajjaj et al., 2001) (Bizukojc and Pecyna, 2011)
a
Only studies that utilised single carbon or nitrogen sources are included in this table. To the authors’ knowledge, lovastatin production on galactose, arabinose, ribose, xylose and maltose have not been reported.
b
Product titres differ significantly due to the large variations in experimental conditions and the timepoint of lovastatin extraction.
Table 3: The carbohydrates consumption rate, r by A. terreus in the basic mediaa for the period of 4 daysb.
Carbohydrates
Consumption rate, r (day 2) Consumption rate, r (day 4)
Monosaccharides D-arabinose
2.97
1.42
D-galactose
3.10
2.37
D-glucose
5.85
3.14
D-mannose
6.48
2.58
D-ribose
5.54
3.28
D-fructose
5.14
3.19
D-xylose
4.70
3.23
D-sucrose
4.32
3.74
D-maltose
4.60
3.07
Α-lactose
3.81
3.80
3.56
3.82
Disaccharides
Control Glycerol
a
The fermentation was carried out under minimal nutrient media (KH2PO4, 0.4 g/L, MgSO4•7H2O, 0.2 g/L, NaCl2, 0.4 g/L, yeast extract, 4 g/L and the substrate of interest), temperature at 30 ˚C, initial pH at 6.5 and speed at 180 rpm in a shaking incubator. Seed culture was performed 24 to 30 hours prior to the actual experiment.
b
Day 6 consumption rate was not shown as most of the carbohydrates were already exhausted and might give potentially inaccurate results.
PII: DOI: Reference:
S1878-8181(17)30045-2 http://dx.doi.org/10.1016/j.bcab.2017.04.011 BCAB547
To appear in: Biocatalysis and Agricultural Biotechnology Received date: 23 January 2017 Revised date: 22 February 2017 Accepted date: 20 April 2017 Cite this article as: Muhamad Hafiz Abd Rahim, Hanis H. Harith, Alejandro Montoya and Ali Abbas, Growth And Lovastatin Production by Aspergillus Terreus Under Different Carbohyrates as Carbon Sources, Biocatalysis and Agricultural Biotechnology, http://dx.doi.org/10.1016/j.bcab.2017.04.011 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Growth And Lovastatin Production by Aspergillus Terreus Under Different Carbohyrates as Carbon Sources Muhamad Hafiz Abd Rahim1,2, Hanis H Harith3, Alejandro Montoya1, Ali Abbas1* 1
School of Chemical and Biomolecular Engineering, The University of Sydney, Sydney, Australia
2
The Department of Food Science, Faculty of Food Science and Technology, Universiti Putra Malaysia,
Serdang, Malaysia 3
Department of Biomedical Science, Faculty of Medicine and Health Sciences, Universiti Putra Malaysia,
Serdang, Malaysia *Corresponding author: Associate Professor Ali Abbas, Director of the Laboratory for Multiscale Systems, School of Chemical and Biomolecular Engineering, J01 - Chemical Engineering Building, The University of Sydney. Ph: +61 2 9351 3002; E-mail: [email protected]
ABSTRACT Carbon source is a key component of metabolites synthesis in microorganisms. This work examined the effects of selected carbon sources in the form of carbohydrates, on the growth of Aspergillus terreus ATCC 20542 and the production of lovastatin. Slowly metabolised carbohydrates, such as D-galactose (consumption rate, r=3.11), produced a high microbial biomass, XFINAL (9.44 g/L) compared to other carbohydrates, but with a low biomass yield coefficient (YLOV/X=1.68). In contrast, D-ribose (YLOV/X=) which showed moderate biomass growth (XFINAL=8.78 g/L) and consumption rate (r=5.44 g/day), produced the highest lovastatin amount (51.81 mg/L, day 6). These indicate little correlation between biomass growth and lovastatin production. Notably, culture consisting of pellets with short hairy surface feature is associated with enhanced lovastatin production. Our findings suggest that the production of lovastatin by Aspergillus terreus is highly influenced by the choice of carbohydrates that will shape the pellet morphology rather than the rate of carbohydrates metabolism. Keywords: Aspergillus terreus; carbon source; lovastatin ; fermentation, filamentous fungi
Abbreviation(s) r, consumption rate; YLOV/X, biomass yield coefficient; YLOV/S, substrate yield coefficient; XFINAL, biomass growth; umax, specific growth rate 1.0 INTRODUCTION Lovastatin (C24H36O5) is a commonly prescribed cholesterol-lowering drug belonging to the statins group that acts by competitively inhibiting the formation of 3-hydroxy-3-methyl glutaryl CoA (HGM-CoA) reductase, a rate limiting enzyme of cholesterol biosynthesis in the liver. This compound is produced naturally by a few fungus species including the commonly known Aspergillus terreus ATCC 20542 (Manzoni et al., 1998). Although unclear, lovastatin is thought to have a protective role in fungus on the basis that it is effective at reducing sterols important for the growth of other microbes (Debakey and Endo, 2008). Recent studies also suggest that lovastatin may have pharmacological potential beyond cholesterol-lowering ability, such as for the treatment of cancer (Hindler et al., 2006), Alzheimer’s disease (Sparks, 2011) and osteoporosis (Gonyeau, 2005). The factors that influence the production of lovastatin include the types of carbon and nitrogen source (Hajjaj et al., 2001), vitamin B content (Bizukojc et al., 2007) and physical factors such as aeration rate (Bizukojc and Ledakowicz, 2008), dissolved oxygen concentration (Casas López et al., 2004), pH (Lai et al., 2005), and types of fermentation (Barrios-González and Miranda, 2010). One of the most important aspects, carbon source, has been widely studied in terms of its impact on the growth and metabolism of A. terreus (Casas López et al., 2003; Hajjaj et al., 2001; Jia et al., 2009a, 2009b; Kumar et al., 2000). A higher ratio of carbon to nitrogen (also known as ‘nitrogen starvation’) is thought to be an important factor in lovastatin production (Casas López et al., 2003). Similarly, slowly-metabolised carbon sources, such as α-lactose and glycerol, are better carbon choices which favour higher lovastatin production (Casas López et al., 2003). The use of rapidly-metabolised carbon sources, such as as D-glucose, may lead to reduced lovastatin production via catabolite repression mechanism (Lai et al., 2007). It also favours the production of primary metabolite such as itaconic acid (Lai et al., 2007). Despite this, some studies have not reported repressed lovastatin yield when these carbons (D-glucose and fructose) were used (Casas López et al., 2003; Hajjaj et al., 2001). The present study investigated the effects of different sources of carbon in the form of carbohydrate on the growth and metabolite production of A. terreus under similar culture conditions. Here, we evaluated the effects of different carbohydrates comprised of monosaccharides (D-glucose, D-xylose, D-fructose, D-galactose, Dmannose, D-ribose, D-arabinose) and disaccharides (α-lactose, maltose, D-sucrose) on lovastatin production and the morphology of A. terreus.
2.0 MATERIALS AND METHODS 2.1 Culture condition The culture conditions used in this study have been described previously with slight modifications (Abd Rahim et al., 2015). The basal media contained minimal salt media (containing KH2PO4, 0.4 g/L, MgSO4•7H2O, 0.2 g/L, and NaCl2 0.4 g/L) to ensure that the effect on the yield and morphology is due to the carbon sources. 1 mL aliquot of spore suspension (107 spores/mL) was inoculated into 125 mL conical flask containing 50 mL of media at 30 ˚C, 180 rpm in a shaking incubator. Seed culture was performed 24 to 30 hours prior to the actual experiment, using 8 g/L of yeast extract and 10 g/L of α-lactose in the basal salt media to allow the fungus to achieve exponential growth before cultivation. Ten different carbohydrates comprised of monosaccharides (D-glucose, D-xylose, D-galactose, D-mannose, Darabinose, D-fructose, D-ribose) and disaccharides (D-maltose, α-lactose, D-sucrose) were used in this study at 20 g/L (Fig. 1). Glycerol was used in control experiment. Initial pH was kept at 6.5. No further pH control was applied during the duration of the experiment. 2.2 Dry cell weight determination The biomass yield was determined gravimetrically. The recovery of fungus biomass was carried out through filtration on No. 2 Whatman filter paper. The biomass was washed twice with distilled water, followed by drying at 80 ºC for 24 hours or until constant weight is achieved. 2.3 Analytical method Lovastatin (Sigma Aldrich, Sydney, Australia) standard was prepared according to manufacturer’s instruction. The main analysis was performed using High Performance Liquid Chromatography (HPLC), Agilent 1200, using a UV detector at a wavelength of 238 nm, with a reference wavelength of 360 nm. The column used was XDB Eclipse Zorbax C-18. Parameters used for HPLC were as follows: Column temperature at 30˚C, sample chamber temperature at 4˚C, flow rate at 1 ml/min, injection volume at 10 µL and 95% acetonitrile and 0.1% phosphoric acid solution as mobile phases. The resultant HPLC peaks are shown in Fig. 2. Glycerol concentration was determined using free glycerol calorimetric detection kit (Sigma-Aldrich), which is based on fluorescence reading at 575 nm. The sugar content in the samples was determined using phenolsulfuric acid method (Masuko et al., 2005) with slight modification. 2.4 Image capture procedure The images of the fungal pellet were taken using Digital Blue QX7 Computer Microscope (Digital Blue, Atlanta, USA), a light microscope equipped with camera. The fungal pellets were harvested from the fermentation system at day 4 (where the fully formed morphology was observed). The pellets were placed on a
microscope slide and the excess liquid around the pellet was dried using Kimtech Wipes. The images were taken immediately following the drying procedure. 2.5 Kinetic calculations The common growth kinetic stoichiometric parameters were used to describe the growth of A. terreus. These include: The substrate consumption rate, r is calculated in days r= Si is the initial concentration; while So is the final concentration of substrate (s).
The yields, Y are both measured in relation to the biomass production and total of substrate consumed in the reaction YLOV/X = YLOV/S = Pmax in this equation equals to maximum lovastatin concentration, and Pi is the initial lovastatin concentration.
μmax is specific growth rate obtained by plotting the specific biomass weight versus time μmax = Xt is the production of biomass at specific time while Xi is the initial biomass weight.
XFINAL is the final biomass weight at day 6, respectively. 2.6 Statistical analysis All experiments were conducted at least in quadruplicates. Data obtained for metabolite production involving lovastatin was analysed using one-way ANOVA with Tukey post hoc test. All statistical analysis was performed using Graphpad Prism, version 7.0. Error bars represent 95% confidence interval.
3.0 RESULTS AND DISCUSSION 3.1 The relationship between kinetic parameters, morphology and lovastatin production of A. terreus Morphology of fungus is one of the key factors that influence the production of metabolites. The pellet form is known to produce better metabolites yield in submerged fermentation compared to filamentous fungus, therefore the pellets were used for seed culture (Casas López et al., 2005; Rodríguez Porcel et al., 2005). In general, we observed spherical, round pellets with all treatments. However, there were three distinct morphologies of A. terreus, mainly grouped based on their hair characteristics. Fig. 3 showed the images of these fungal pellets under the light microscope, which are classified as short-haired, long-haired or no-hair characteristics. These hair characteristics will serve as our point of reference throughout the discussion. Generally, we observed that the production of lovastatin is mainly mixed-growth associated with all the substrates tested, as lovastatin was detected in the early and late phase of biomass production. This was also observed by Bizukojc et al., where similar pattern of production was observed in the presence of α-lactose (Bizukojc and Ledakowicz, 2007). The kinetic parameters for the production of lovastatin by A. terreus using different carbohydrates are shown in Table 1. Both lowest and highest yield coefficients (YLOV/X and YLOV/S) were detected within monosaccharides group. Lowest yield coefficients were detected in the presence of Dgalactose, D-arabinose and D-glucose; while the highest was observed in the presence of monosaccharides, specifically D-xylose and D-fructose. Disaccharides, in general, gave a satisfactory production of lovastatin relative to the biomass and substrate consumption, as their YLOV/X and YLOV/S values were high, with maltose giving the highest yield coefficient. This observation is also in contrasts with the theory that slowly metabolised carbon is better for lovastatin production (Bizukojc and Ledakowicz, 2009; Casas López et al., 2003). It may be due to the limited range of carbons tested in this study (only a maximum of three carbons were tested) or the different compositions of media which may affect the production of lovastatin. The carbohydrates with high lovastatin yield, namely D-xylose, D-fructose, D-ribose, D-maltose and D-sucrose exhibited similar features throughout the investigation. These include short-haired pellet morphology, moderate specific growth rate (umax), and moderate final biomass (Xfinal), with pH approaching 7. Previous studies have shown that short-haired pellets (Casas López et al., 2005) and moderate to high ØFINAL (Casas López et al., 2005; Rodríguez Porcel et al., 2006, 2005) are associated with good lovastatin production. This may be due to the balance between growth and oxygen transfer, since short hair and higher diameter allow better gas absorption. Conversely, we observed pellets with long hair feature, displaying gelatinous-like morphology in Darabinose (Fig. 3c), which corresponded with lowest X FINAL. This is consistent with previous reports suggesting that long hair morphology may reduce the nutrient transfer ability (Jia et al., 2009a). Others with low yield coefficient, namely D-glucose, D-galactose, and α-lactose exhibited no-hair and compact morphologies with pH away from neutral. The absence of hair may limit the uptake of nutrients into the fungal pellet, which could lead to lower lovastatin production. Nevertheless, there is still no plausible explanation on why or how hair characteristics contribute to lovastatin production.
3.2 Lovastatin production and carbohydrates consumption by A. terreus Slowly-metabolised carbons, particularly α-lactose and glycerol, have been described as the optimal carbon source for lovastatin production (Bizukojc and Ledakowicz, 2009; Casas López et al., 2003). Table 2 demonstrated several lovastatin yields from previous findings using sole carbon and nitrogen sources. From the table, it is clear that the production of lovastatin vary in several studies, most likely due to different experimental conditions. In addition, only certain carbon sources were used regularly in previous studies (glucose, glycerol, lactose, fructose) while other carbon sources (maltose, xylose, arabinose, galactose) were hardly used during the investigations. Our findings show that there are other carbohydrates substrates with higher efficiency for lovastatin production such as D-xylose and D-fructose. Table 3 shows the production of lovastatin over the span of 4 days, while Fig. 4 demontrates the carbohydrates consumption rate over the span of 6 days. Unlike some fungal system (Bettiga et al., 2009; Klaubauf et al., 2013), we observed that the group or structure of sugars (mono- and disaccharides, or pentoses and hexoses) have no effects on the production of biomass and lovastatin. In general, these carbohydrates can be characterised according to the lovastatin production, low (5 – 11 mg/L), medium (18 – 32 mg/L) and high (43 – 57 mg/L). Carbohydrates which produced significantly lower lovastatin yield compared to the control are D-arabinose (5.78 mg/L), D-galactose (11.16 mg/L) and D-glucose (17.28 mg/L). Of note, D-arabinose and D-galactose had low uptake rate and both were detectable in the media even after 6 days of fermentation. These correlated with a halt in lovastatin production at day 4, suggesting that both sources may not be suitable for the production of metabolites. Like other microorganisms, fungi are known for their ability to develop a mechanism to readily uptake available carbon sources in nature for survival. Given that L-arabinose is more common and readily available than D-arabinose, this may explain why A. terreus was unable to efficiently use the latter as the carbohydrate source. In fact, the most established pathway known for D-arabinose catabolism in fungi is the oxidoreductase pathway, which involves L-arabinose and not Darabinose (Seiboth and Metz, 2011). D-galactose, on the other hand, has been shown to be poorly metabolised by certain types of fungi, such as A. niger (Edgecombe, 1938). The inhibitory effect of D-galactose on lovastatin production is most likely due to the lack of both galR and galX regulators that control the utilisation of Dgalactose for growth in A. terreus (Christensen et al., 2011). Given that only metabolite production was inhibited but not its biomass growth, this indicates that additional factors or pathways may be involved, and thus reflects the complex nature of the fungi. Although D-glucose produced significantly lower lovastatin than glycerol, the amount (17.28 mg/L) is significantly higher than both D-arabinose and D-galactose. Apart from the unfavourable morphology as discussed previously, the lower production of lovastatin may also be attributed to the well-known catabolite repression mechanism of glucose on metabolite production (Hajjaj et al., 2001). Our data also showed that this carbohydrate was metabolised rapidly, perhaps due to its abundance in nature (Lehninger et al., 2005) using established pathways such as the classical Embden-Meyerhof or pentose phosphate pathway (Bonnarme et al., 1995; Greene Lilly and Barnett, 1956).
The medium group consist of glycerol (25.58 mg/L), D-mannose (28.14 mg/L), α-lactose (30.53 mg/L) and Dmaltose (31.94 mg/L) and were grouped together as they showed no significance when compared to glycerol (control). Except for D-mannose, all these carbohydrates were slowly consumed by A. terreus. Although the production of lovastatin in α-lactose and glycerol started off quite slow, the production continued up to day 6. In contrast, lovastatin production in D-mannose peaked at day 2 with little change observed until day 6. This may suggest that the production in D-mannose was not limited by carbohydrate availability, but rather, the production was halted after certain amount of lovastatin was produced. While this may indicate that D-mannose is unfavourable for the fermentation of A. terreus, the fungus appeared to tolerate D-mannose better than Dgalactose and D-arabinose (the low-production group) as it was consumed until exhausted, accompanied with higher production of lovastatin. The consumption of D-mannose may be aided by mstA, a high-affinity transporter present in A. terreus which efficiently transports D-glucose, D-mannose and D-xylose into cells (Vankuyk et al., 2004). Nevertheless, it is quite surprising to find that α-lactose and glycerol did not produce as much lovastatin, compared to previous reports (Casas López et al., 2003; Pecyna and Bizukojc, 2011; Szakács et al., 1998). Apart from the limited types of carbon comparison in those studies, these inconsistencies may be attributed to the variation in the media composition. For example, complete media consisting of trace metals and extra nutrients (for example, biotin, manganese chloride and sodium molybdate) was used in previous studies which is likely to affect the interactions and relationships between carbons and fungus utilisation (Casas López et al., 2003). More importantly, both α-lactose and glycerol also produced high biomass. This suggests that these carbohydrate types may favour fatty acid biosynthetic pathway (refer to Fig. 5 for metabolite production pathway) instead of metabolite production. As such, this may compete with the lovastatin production pathway. For α-lactose in particular, the limited lovastatin production may also be explained by the catabolism of α-lactose into its respective monomers which usually involve two pathways - extracellular or intracellular hydrolysis (Seiboth and Metz, 2011). Perhaps, given that α-lactose can be hydrolysed to D-galactose and D-glucose, this may explain the limited production of lovastatin as both D-galactose and D-glucose are not the optimal carbohydrates for lovastatin production. Similarly, D-maltose can be hydrolysed into two units of glucose, which might also impede the production of lovastatin. While the culture treated with glycerol displayed favourable pellet morphology (short hair) and known to be slowly-metabolised, only moderate lovastatin concentrations were produced. Glycerol treatment produced the highest biomass which may reflect that this substrate favours the production of biomass rather than metabolite production, as discussed earlier. The group with significantly higher lovastatin (>40 mg/L) production than glycerol consists of D-xylose, Dfructose, D-sucrose and D-ribose. We observed that the carbohydrate consumptions varied between moderate to high (r=4.70 – 5.53), but not as fast as D-glucose (r=5.85), and not as slow as α-lactose (r=3.81) or glycerol (r=3.56). Our findings suggest that one of the prerequisite for good lovastatin production is the fungus’s ability to significantly improve its production during stationary phase (day 4, when biomass started to stabilise. Refer to Fig. 4) after the initial production at day 2. In particular, lovastatin production improved by 23%, 37%, 48% and 59% in D-ribose, D-sucrose, D-fructose and D-xylose, respectively after day 2. The uptake of D-xylose [56.46 g/L] potentially involves mstA transporter similar to D-glucose, but no repressive features that may block lovastatin production has been reported. Furthermore, D-xylose is one of the most abundant carbohydrates
found in nature. Therefore, A. terreus most likely have developed the ability to uptake this carbohydrate for growth and metabolite production. As for D-ribose, despite being an epimer (at C’2) of D-arabinose, it was metabolised quite efficiently by A. terreus with a considerable lovastatin yield (42.97 mg/L). It is likely that Dribose is more readily utilised than D-arabinose due to its structural difference and high availability in nature (found in all living cells). On the other hand, D-fructose and D-sucrose (with a yield of 51.08 mg/L and 47.40 mg/L, respectively) have been traditionally used for fermentation involving different fungus species including A. terreus with desirable lovastatin production (Casas López et al., 2003). Lopez et al. showed a satisfactory production of lovastatin using D-fructose which was thought to result from increased biomass growth; however this was not observed in our study. This may be explained by variations in experimental conditions (for example, the use of rich media culture (Casas López et al., 2003)) which may affect A. terreus growth and metabolism differently. D-sucrose may be a good carbohydrate for lovastatin production compared to other disaccharides as its hydroxylation would yield D-fructose (and D-glucose). The presence of D-fructose, hypothetically, may be the main factor that improves the production of lovastatin. 4.0 Conclusion Our findings show that the fermentation of lovastatin is highly dependent on the type of the carbohydrate sources used. Furthermore, the effects of certain carbons on pellet morphology may play a major role in defining the amount of lovastatin being produced. Conversely, the metabolism rate of the carbons was not as important as initially thought. Unlike the morphology, the biomass is not likely to be related to lovastatin production as no apparent trend between biomass and lovastatin production was observed in this experiment.
Conflict of interest statement The authors have declared no conflict of interest
Acknowledgement Authors would like to thank Bosch Mass Spectrometry Facility, University of Sydney, for the HPLC system and Malaysian Government for the scholarships. This research did not receive any specific grant from funding agencies in the public, commercial, or not-forprofit sectors
References
Abd Rahim, M.H., Hasan, H., Montoya, A., Abbas, A., 2015. Lovastatin and (+)-geodin production by Aspergillus terreus from crude glycerol. Eng. Life Sci. 15, 220–228. Barrios-González, J., Miranda, R., 2010. Biotechnological production and applications of statins. Appl. Microbiol. Biotechnol. 85, 869–883. Bettiga, M., Bengtsson, O., Hahn-Hägerdal, B., Gorwa-Grauslund, M., 2009. Arabinose and xylose fermentation by recombinant Saccharomyces cerevisiae expressing a fungal pentose utilization pathway. Microb. Cell Fact. 8, 1–12. Bizukojc, M., Ledakowicz, S., 2009. Physiological, morphological and kinetic aspects of lovastatin biosynthesis by Aspergillus terreus. Biotechnol. J. 4, 647–664. Bizukojc, M., Ledakowicz, S., 2008. Biosynthesis of lovastatin and (+)-geodin by Aspergillus terreus in batch and fed-batch culture in the stirred tank bioreactor. Biochem. Eng. J. 42, 198–207. Bizukojc, M., Ledakowicz, S., 2007. A macrokinetic modelling of the biosynthesis of lovastatin by Aspergillus terreus. J. Biotechnol. 130, 422–435. Bizukojc, M., Pawlowska, B., Ledakowicz, S., 2007. Supplementation of the cultivation media with B-group vitamins enhances lovastatin biosynthesis by Aspergillus terreus. J. Biotechnol. 127, 258–268. Bizukojc, M., Pecyna, M., 2011. Lovastatin and (+)-geodin formation by Aspergillus terreus ATCC 20542 in a batch culture with the simultaneous use of lactose and glycerol as carbon sources. Eng. Life Sci. 11, 272– 282. Bonnarme, P., Gillet, B., Sepulchre, A.M., Role, C., Beloeil, J.C., Ducrocq, C., 1995. Itaconate biosynthesis in Aspergillus terreus. J. Bacteriol. 177, 3573–8. Casas López, J.. L., Sánchez Pérez, J.. A., Fernández Sevilla, J.. M., Acién Fernández, F.. G., Molina Grima, E., Chisti, Y., Casas Lopez, J.L., Sanchez Perez, J.A., Fernandez Sevilla, J.M., Acien Fernandez, F.G., 2003. Production of lovastatin by Aspergillus terreus: effects of the C:N ratio and the principal nutrients on growth and metabolite production. Enzyme Microb. Technol. 33, 270–277. Casas López, J.L., Sánchez Pérez, J.A., Fernández Sevilla, J.M., Acién Fernández, F.G., Molina Grima, E., Chisti, Y., 2004. Fermentation optimization for the production of lovastatin by Aspergillus terreus: use of response surface methodology. J. Chem. Technol. Biotechnol. 79, 1119–1126. Casas López, J.L., Sánchez Pérez, J.A., Fernández Sevilla, J.M., Rodríguez Porcel, E.M., Chisti, Y., 2005. Pellet morphology, culture rheology and lovastatin production in cultures of Aspergillus terreus. J. Biotechnol. 116, 61–77. Christensen, U., Gruben, B.S., Madrid, S., Mulder, H., Nikolaev, I., de Vries, R.P., 2011. Unique regulatory mechanism for D-galactose utilization in Aspergillus nidulans. Appl. Environ. Microbiol. 77, 7084–7. Debakey, L., Endo, A., 2008. A gift from nature: the birth of the statins. Nat. Med. 14, 1050–2. Edgecombe, A.E., 1938. The Effect of Galactose on the Growth of Certain Fungi. Mycologia 30, 601–624 CR– Copyright 169; 1938 Mycological So. Fernandes, S., Murray, P., n.d. Metabolic engineering for improved microbial pentose fermentation. Bioeng. Bugs 1, 424–8. Gonyeau, M.J., 2005. Statins and osteoporosis: a clinical review. Pharmacotherapy 25, 228–43. Greene Lilly, V., Barnett, H.L., 1956. The Utilization of D- and L-Arabinose by Fungi. Am. J. Bot. 43, 709– 714.
Hajjaj, H., Niederberger, P., Duboc, P., 2001. Lovastatin Biosynthesis by Aspergillus terreus in a Chemically Defined Medium. Appl. Environ. Microbiol. 67, 2596–2602. 1 Hindler, K., Cleeland, C.S., Rivera, E., Collard, C.D., 2006. The role of statins in cancer therapy. Oncologist 11, 306–15. doi:10.1634/theoncologist.11-3-306 Jia, Z., Zhang, X., Cao, X., 2009a. Effects of carbon sources on fungal morphology and lovastatin biosynthesis by submerged cultivation of Aspergillus terreus. Asia-Pacific J. Chem. Eng. 4, 672–677. Jia, Z., Zhang, X., Zhao, Y., Cao, X., 2009b. Effects of divalent metal cations on lovastatin biosynthesis from Aspergillus terreus in chemically defined medium. World J. Microbiol. Biotechnol. 25, 1235–1241. Klaubauf, S., Ribot, C., Melayah, D., Lagorce, A., Lebrun, M.-H., de Vries, R.P., 2013. The pentose catabolic pathway of the rice-blast fungus Magnaporthe oryzae involves a novel pentose reductase restricted to few fungal species. FEBS Lett. 587, 1346–1352. Kumar, M.S., Kumar, P.M., Sarnaik, H.M., Sadhukhan, A.K., 2000. A rapid technique for screening of lovastatin-producing strains of Aspergillus terreus by agar plug and Neurospora crassa bioassay. J Microbiol Methods 40, 99–104. Lai, L.-S.T., Hung, C.-S., Lo, C.-C., 2007. Effects of lactose and glucose on production of itaconic acid and lovastatin by Aspergillus terreus ATCC 20542. J. Biosci. Bioeng. 104, 9–13. Lai, L.-S.T., Tsai, T.-H., Wang, T.C., Cheng, T.-Y., 2005. The influence of culturing environments on lovastatin production by Aspergillus terreus in submerged cultures. Enzyme Microb. Technol. 36, 737–748. Lehninger, A.L., Nelson, D.L., Cox, M.M., 2005. Lehninger Principles of Biochemistry. W. H. Freeman. Manzoni, M., Rollini, M., Bergomi, S., Cavazzoni, V., 1998. Production and purification of statins from Aspergillus terreus strains. Biotechnol. Tech. 12, 529–532. Marcin Bizukojc, Ledakowicz, S., 2005. Influence of environmental factors on mycelial growth and biosynthesis of lovastatin by Aspergillus terreus. Biotechnologia-monografie 2, 25–36. Masuko, T., Minami, A., Iwasaki, N., Majima, T., Nishimura, S.-I., Lee, Y.C., 2005. Carbohydrate analysis by a phenol-sulfuric acid method in microplate format. Anal. Biochem. 339, 69–72. Pecyna, M., Bizukojc, M., 2011. Lovastatin biosynthesis by Aspergillus terreus with the simultaneous use of lactose and glycerol in a discontinuous fed-batch culture. J. Biotechnol. 151, 77–86. Peterson, W.H., Fred, E.B., Anderson, J.A., 1922. THE Fermentation of hexoses and related compounds by certain pentose fermenting bacteria. J. Biol. Chem. 53, 111–123. Rodríguez Porcel, E.M., Casas López, J.L., Sánchez Pérez, J.A., Fernández Sevilla, J.M., Chisti, Y., 2005. Effects of pellet morphology on broth rheology in fermentations of Aspergillus terreus. Biochem. Eng. J. 26, 139–144. Rodríguez Porcel, E.M., Casas López, J.L., Sánchez Pérez, J.A., Fernández Sevilla, J.M., García Sánchez, J.L., Chisti, Y., 2006. Aspergillus terreus Broth Rheology, Oxygen Transfer, and Lovastatin Production in a Gas-Agitated Slurry Reactor. Ind. Eng. Chem. Res. 45, 4837–4843. doi:10.1021/ie0600801 Seiboth, B., Metz, B., 2011. Fungal arabinan and l-arabinose metabolism. Appl. Microbiol. Biotechnol. 89, 1665–1673. doi:10.1007/s00253-010-3071-8 Sparks, D.L., 2011. Alzheimer disease: Statins in the treatment of Alzheimer disease. Nat Rev Neurol 7, 662– 663. Szakács, G., Morovján, G., Tengerdy, R., 1998. Production of lovastatin by a wild strain of Aspergillus terreus. Biotechnol. Lett. 20, 411–415. Vankuyk, P.A., Diderich, J.A., MacCabe, A.P., Hererro, O., Ruijter, G.J.G., Visser, J., 2004. Aspergillus niger mstA encodes a high-affinity sugar/H+ symporter which is regulated in response to extracellular pH. Biochem. J. 379, 375–83.
Figure 1: The chemical structure of monosaccharides and disaccharides (pentoses and hexoses) used in this study. The chemical structures have been shown to influence the carbon consumption and metabolism in several microorganisms (Fernandes and Murray, n.d.; Peterson et al., 1922). From top left: A) D-glucose B) D-mannose C) D-galactose D) Dribose F) D-fructose G) D-arabinose H) α-lactose I) D-maltose J) D-sucrose
Figure 2: HPLC peaks obtained from the fermentation broth of A. terreus. Untreated samples taken directly from fermentation showed a good peak separation under isocratic elution mode using UV detector. Co-metabolites of lovastatin, sulochrin and (+)-geodin appear first around fourth and seventh minutes, followed by lovastatin at eleventh minutes. 238 nm is used as the main signal wavelength with 280 nm as the reference wavelength.
Figure 3: Three major pellet morphologies observed during fermentation of A. terreus at day 6. From left – no-hair (e.g. D-glucose-treated), short-hair (e.g. D-fructose-treated) and longhair (e.g. D-arabinose-treated) morphologies. Pellets with no-hair feature were more dense with minimal tendency to clump. Pellets with short-hair feature was less dense with higher tendency to clump, and exhibited patchy hair growth. Pellets with long-hair feature generally have thick filamentous hair and hollow, porous pellet.
70 60 Lovastatin (mg/L)
50 40 Day 2
30
Day 4
20
Day 6
10 0
Carbohydrates 12 11
Biomass (g/L)
10 9 8 7
Day 2
6
Day 4
5
Day 6
4
Carbohydrates
Figure 4: Lovastatin and biomass production under different carbohydrates treatment over the span of 6 days using 4 g/L yeast extract as nitrogen source. Error bar represents 95% confidence.
Figure 5: The pathway of metabolite production by A. terreus ATCC 20542. Note that all metabolite production requires Acetyl CoA as its main precursor. The pathways for lovastatin, sulochrin, (+)-geodin and fatty acids production require the combination of Acetyl CoA and Malonyl CoA. The channelling of more precursors towards one pathway may potentially cause deprivation of precursors for other pathways.
Table 1: The effects of different carbohydrates on yield and fermentation parameters by A. terreus after 6-days fermentation. a Carbohydrate
Lovastatin
YLOV/X YLOV/S
µmax
XFINAL
Final pH
(-saccharides)
Yield (mg/L)
(mg/g) (mg/g/L)
(1/d)
(g/L)
D-ribose
42.97 ± 1.8****
6.56
2.42
2.48
9.36 ± 0.5
7.06 ± 0.02
D-xylose
56.46 ± 4.8****
9.69
3.49
1.97
8.64 ± 1.2
7.19 ± 0.04
D-mannose
28.14 ± 5.7ns
5.20
1.55
2.41
8.02± 1.4
7.16 ± 0.03
D-arabinose
5.78 ± 0.9****
1.19
0.66
2.41
7.28 ± 0.2
6.25 ± 0.07
3.12
0.96
2.07
8.74 ± 0.7
7.46 ± 0.03
Mono-
*
D-glucose
17.28 ± 2.3
D-fructose
51.81 ± 1.2****
8.68
3.11
2.47
8.78 ± 0.3
7.01 ± 0.3
D-galactose
11.16 ± 1.2****
1.68
1.02
2.07
9.44 ± 0.3
6.19 ± 0.08
à-lactose
30.53 ± 0.6ns
4.47
2.01
2.47
9.64 ± 0.2
6.43 ± 0.05
D-maltose
31.94 ± 2.3ns
6.06
2.08
1.67
8.08 ± 0.6
6.91 ± 0.06
D-sucrose
47.40 ± 2.6****
7.15
2.94
2.17
9.44 ± 0.6
6.87 ± 0.08
25.68 ± 2.9
3.46
1.74
2.47
10.44 ± 0.7
6.88 ± 0.02
Di-
Control Glycerol
* The data are presented as means ± 95% confidence level. Analysis of variance (one-way ANOVA) using Tukey’s post-hoc test against the control (glycerol) was conducted. The number of (*) indicates significance when compared to glycerol. ns – not significant
The conditions for fermentation were minimal nutrient media
(KH2PO4, 0.4 g/L, MgSO4•7H2O, 0.2 g/L, NaCl2, 0.4 g/L, yeast extract, 4 g/L and the substrate of interest), temperature at 30 ˚C, initial pH at 6.5 and speed at 180 rpm in a shaking incubator. Seed culture was performed 24 to 30 hours prior to the actual experiment.
Table 2: The lovastatin production reported in previous studies
Carbon sourcea
Nitrogen sourcea
Product titreb
Strain and references
Glucose (20 g/L)
Glutamic acid (9.8 g/L)
(Hajjaj et al., 2001)
Glucose (45 g/L)
YE (12.5 g/L)
37 mg/L, but only produced after glucose is exhausted 12 mg/L
Fructose (20 g/L)
YE (1.33 g/L)
120 mg/L
Lactose (20 g/L)
YE (1.33 g/L)
90 mg/L
Lactose (20 g/L)
YE (8 g/L)
35 mg/L
Lactose (45 g/L)
Glutamic acid (9.8 g/L)
25 mg/L
Lactose (5-40 g/L)
YE (2-12 mg/L)
5 -110 mg/L
Sucrose (50 g/L)
CSL (10 g/L)
Glycerol (20 g/L)
YE (1.33 g)
40 % less than lactose 90 mg/L
Glycerol (20 g/L)
Glutamic acid (9.8 g/L)
6 mg/L
Glycerol (40 g/L)
YE (4 g/L)
40 mg/L
(Marcin Bizukojc and Ledakowicz, 2005) (Casas López et al., 2003) (Casas López et al., 2003) (Marcin Bizukojc and Ledakowicz, 2005) (Hajjaj et al., 2001) (Bizukojc and Ledakowicz, 2007) (Szakács et al., 1998) (Casas López et al., 2003) (Hajjaj et al., 2001) (Bizukojc and Pecyna, 2011)
a
Only studies that utilised single carbon or nitrogen sources are included in this table. To the authors’ knowledge, lovastatin production on galactose, arabinose, ribose, xylose and maltose have not been reported.
b
Product titres differ significantly due to the large variations in experimental conditions and the timepoint of lovastatin extraction.
Table 3: The carbohydrates consumption rate, r by A. terreus in the basic mediaa for the period of 4 daysb.
Carbohydrates
Consumption rate, r (day 2) Consumption rate, r (day 4)
Monosaccharides D-arabinose
2.97
1.42
D-galactose
3.10
2.37
D-glucose
5.85
3.14
D-mannose
6.48
2.58
D-ribose
5.54
3.28
D-fructose
5.14
3.19
D-xylose
4.70
3.23
D-sucrose
4.32
3.74
D-maltose
4.60
3.07
Α-lactose
3.81
3.80
3.56
3.82
Disaccharides
Control Glycerol
a
The fermentation was carried out under minimal nutrient media (KH2PO4, 0.4 g/L, MgSO4•7H2O, 0.2 g/L, NaCl2, 0.4 g/L, yeast extract, 4 g/L and the substrate of interest), temperature at 30 ˚C, initial pH at 6.5 and speed at 180 rpm in a shaking incubator. Seed culture was performed 24 to 30 hours prior to the actual experiment.
b
Day 6 consumption rate was not shown as most of the carbohydrates were already exhausted and might give potentially inaccurate results.