Production of the sesquiterpenoid (+)-nootkatone by metabolic engineering of Pichia pastoris.

Metabolic Engineering ∎ (∎∎∎∎) ∎∎∎–∎∎∎

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Q1 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66

Contents lists available at ScienceDirect

Metabolic Engineering journal homepage: www.elsevier.com/locate/ymben

Production of the sesquiterpenoid ( þ)-nootkatone by metabolic engineering of Pichia pastoris Tamara Wriessnegger a, Peter Augustin a, Matthias Engleder e, Erich Leitner b, Monika Müller c, Iwona Kaluzna c, Martin Schürmann c, Daniel Mink c, Günther Zellnig d, Helmut Schwab a,e, Harald Pichler a,e,n a

ACIB—Austrian Centre of Industrial Biotechnology, Petersgasse 14, 8010 Graz, Austria Institute of Analytical Chemistry and Food Chemistry, Graz University of Technology, Stremayrgasse 9, 8010 Graz, Austria c DSM Innovative Synthesis B.V., Urmonderbaan 22, 6167 RD Geleen, The Netherlands d Institute of Plant Sciences, University of Graz, Schubertstrasse 51, 8010 Graz, Austria e Institute of Molecular Biotechnology, Graz University of Technology, Petersgasse 14/2, 8010 Graz, Austria b

ar t ic l e i nf o

a b s t r a c t

Article history: Received 5 December 2013 Received in revised form 2 April 2014 Accepted 8 April 2014

The sesquiterpenoid (þ )-nootkatone is a highly demanded and highly valued aroma compound naturally found in grapefruit, pummelo or Nootka cypress tree. Extraction of ( þ)-nootkatone from plant material or its production by chemical synthesis suffers from low yields and the use of environmentally harmful methods, respectively. Lately, major attention has been paid to biotechnological approaches, using cell extracts or whole-cell systems for the production of (þ )-nootkatone. In our study, the yeast Pichia pastoris initially was applied as whole-cell biocatalyst for the production of ( þ)-nootkatone from ( þ)-valencene, the abundant aroma compound of oranges. Therefore, we generated a strain co-expressing the premnaspirodiene oxygenase of Hyoscyamus muticus (HPO) and the Arabidopsis thaliana cytochrome P450 reductase (CPR) that hydroxylated extracellularly added ( þ)-valencene. Intracellular production of (þ )-valencene by co-expression of valencene synthase from Callitropsis nootkatensis resolved the phase-transfer issues of (þ )-valencene. Bi-phasic cultivations of P. pastoris resulted in the production of trans-nootkatol, which was oxidized to ( þ )-nootkatone by an intrinsic P. pastoris activity. Additional overexpression of a P. pastoris alcohol dehydrogenase and truncated hydroxy-methylglutaryl-CoA reductase (tHmg1p) significantly enhanced the ( þ)-nootkatone yield to 208 mg L 1 cell culture in bioreactor cultivations. Thus, metabolically engineered yeast P. pastoris represents a valuable, whole-cell system for high-level production of (þ )-nootkatone from simple carbon sources. & 2014 International Metabolic Engineering Society. Published by Elsevier Inc. All rights reserved.

Keywords: Pichia pastoris Metabolic engineering Membrane protein Cytochrome P450 Terpenoid Nootkatone

1. Introduction

Abbreviations: HPO, Hyoscyamus muticus premnaspirodiene oxygenase; CPR, cytochrome P450 reductase; ValS, valencene synthase; ADH, alcohol dehydrogenase; AOX1, alcohol oxidase; CDW, cell dry weight; YNB, yeast nitrogen base; HRP, horse radish peroxidase n Corresponding author at: Institute of Molecular Biotechnology, Graz University of Technology, Petersgasse 14/2, 8010 Graz, Austria. Fax: þ 43 316 873 4071. E-mail addresses: [email protected] (T. Wriessnegger), [email protected] (P. Augustin), [email protected] (M. Engleder), [email protected] (E. Leitner), [email protected] (M. Müller), [email protected] (I. Kaluzna), [email protected] (M. Schürmann), [email protected] (D. Mink), [email protected] (G. Zellnig), [email protected] (H. Schwab), [email protected] (H. Pichler).

(þ)-Nootkatone was first isolated from the Nootka cypress tree (Callitropsis nootkatensis, aka Alaska yellow cedar) (Erdtman and Hirose, 1962) and trace amounts were later also identified in grapefruit (Citrus paradisi) (MacLeod and Buigues, 1964) or pummelo (Citrus grandis) (Ortuno et al., 1995). Like many other sesquiterpenoids, (þ)-nootkatone exhibits unique odor characteristics, rendering it a highly sought-after flavor and fragrance compound for the food and cosmetics industries. Pharmaceutical industry has become interested in the application of (þ )-nootkatone as it has been shown to be an effective repellent of insects (Dietrich et al., 2006; Flor-Weiler et al., 2011; Jordan et al., 2012; Zhu et al., 2001). Recently, interesting therapeutic activities of (þ)-nootkatone and its metabolites have been reported, such as anti-platelet aggregation effects in rats (Seo et al., 2011), anti-proliferative activity towards cancer cell lines (Gliszczyńska et al., 2011) and enhancement of energy

http://dx.doi.org/10.1016/j.ymben.2014.04.001 1096-7176/& 2014 International Metabolic Engineering Society. Published by Elsevier Inc. All rights reserved.

Please cite this article as: Wriessnegger, T., et al., Production of the sesquiterpenoid (þ )-nootkatone by metabolic engineering of Pichia pastoris. Metab. Eng. (2014), http://dx.doi.org/10.1016/j.ymben.2014.04.001i

67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87

2

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66

T. Wriessnegger et al. / Metabolic Engineering ∎ (∎∎∎∎) ∎∎∎–∎∎∎

Fig. 1. Conversion of (þ )-valencene (1) to trans-nootkatol (2) and (þ)-nootkatone (3).

metabolism through AMP-activated protein kinase activation in skeletal muscle and liver (Murase et al., 2010). Extraction of (þ )-nootkatone from natural sources, e.g. citrus fruits, typically suffers from inadequate yields owing to slow biomass accumulation, low overall (þ)-nootkatone concentrations and annual harvest fluctuations. Thus, chemical methods for (þ )-nootkatone synthesis have been applied to satisfy the high industrial demand. As chemical synthesis often involved toxic heavy metals, highly flammable compounds or strong oxidants (Salvador and Clark, 2002; Wilson and Shaw, 1978), more attention is being paid to environment-friendly and safe methods for (þ )-nootkatone synthesis. Several approaches for de novo (þ )-nootkatone synthesis or the biotransformation of the abundantly available (þ )-valencene to the rare ( þ)-nootkatone (Fig. 1) have been reported (Fraatz et al., 2009a). Whole-cell systems employing bacteria (Dhavlikar and Albroscheit, 1973; Girhard et al., 2009; Okuda et al., 1994; Sowden et al., 2005), fungi (Cankar et al., 2011, 2014; Kaspera et al., 2005; Krügener et al., 2010) and plants (Drawert et al., 1984; Furusawa et al., 2005; Sakamaki et al., 2005) or applications of cell extracts and/or purified proteins as biocatalysts (Bouwmeester et al., 2007; De Kraker et al., 2003; Gavira et al., 2013; Muller et al., 1998; Takahashi et al., 2007a; Zelena et al., 2012) have been described. Many of these biotransformation reactions are catalyzed by enzymes of the cytochome P450 monooxygenase (CYP) superfamily, although the screening and identification of efficient and regio-selective P450 enzymes for commercial applications is still challenging (Urlacher and Schmid, 2006). A number of limitations have restricted the use of CYPs in industrial processes including narrow substrate specificity, association with membranous structures, the need for co-expression of cytochrome P450-reductases, cofactor regeneration issues, overall low activity and process stability (Urlacher and Girhard, 2012). Recently, a comparative study involving several different yeast species and Escherichia coli has indicated that the methylotrophic yeast Pichia pastoris is an excellent host for the functional expression of membrane-associated cytochrome P450 enzymes (Geier et al., 2012). P. pastoris features additional advantages over the wellstudied yeast Saccharomyces cerevisiae, like the ability to grow to very high cell densities in simple media and the availability of the strong and tightly regulated alcohol oxidase (AOX1) promoter (Cereghino and Cregg, 2000; Cregg et al., 2000; Macauley-Patrick et al., 2005; Ramón and Marín, 2011). Interestingly, P. pastoris has been described as host system for production of carotenoids like β-carotene, lycopene and astaxanthin (Araya-Garay et al., 2012a, 2012b). S. cerevisiae has been used in several metabolic engineering approaches for enhanced production of industrially relevant terpenoids (Asadollahi et al., 2010; Chang and Keasling, 2006; Dumas et al., 2006; Muntendam et al., 2009; Nevoigt, 2008; Siddiqui et al., 2012; Takahashi et al., 2007b). For example, a plant cytochrome P450 enzyme, Hyoscyamus muticus premnaspirodiene oxygenase (HPO), was characterized for the regio- and stereo-specific oxidation of various sesquiterpenes including (þ)-valencene (Takahashi et al., 2007b). Engineering of the substrate recognition site of HPO enhanced trans-nootkatol formation, but HPO still failed to synthesize (þ)-nootkatone in bioconversion reactions using S. cerevisiae. Relying on our experience with P. pastoris for recombinant expression

of cytochrome P450 enzymes (Geier et al., 2012), we aimed at the development of a P. pastoris whole-cell biocatalyst for high-level production of (þ)-nootkatone employing the engineered HPO variant V482I/A484I (Takahashi et al., 2007b). Co-expression of HPO and cytochrome P450 reductase (CPR) from Arabidopsis thaliana in P. pastoris cells led to the hydroxylation of externally added (þ)-valencene to trans-nootkatol. Moreover, we identified and overexpressed an endogenous P. pastoris ADH leading to highly efficient transnootkatol to (þ )-nootkatone conversion. In order to eliminate phasetransfer problems triggered by the external addition of (þ)-valencene, we introduced a valencene synthase from Nootka cypress to produce (þ )-valencene in vivo (Beekwilder et al., 2013). Thus, we created P. pastoris strains that can form industrially interesting quantities of (þ)-valencene, trans-nootkatol and (þ)-nootkatone, upon cultivation in two-phase systems using n-dodecane for trapping the produced terpenoid compounds.

2. Materials and methods 2.1. Chemicals and media Restriction enzymes were purchased from Thermo Scientific, St. Leon-Rot, Germany. Monoclonal anti-FLAGs M2 antibody produced in mouse and anti-c-myc antibody produced in rabbit were obtained from Sigma-Aldrichs, Vienna, Austria. Difco™ yeast nitrogen base w/o amino acids (YNB), Bacto™ peptone and Bacto™ yeast extract were obtained from Becton Dickinson and Company, Schwechat, Austria. Zeocin™ was purchased from InvivoGen (Eubio), Vienna, Austria. Kanamycin monosulfate, geneticin sulfate (G418S) and hygromycin B were purchased from Formedium™ (Norfolk, United Kingdom). Sterile water was acquired from Fresenius Kabi, Graz, Austria. Standard laboratory reagents were purchased from Carl Roth GmbH & Co. KG, Karlsruhe, Germany. Terpenoid standards were supplied by DSM Innovative Synthesis B.V., Geleen, The Netherlands. Pichia cultures were either grown in YPD (1% yeast extract, 2% peptone and 2% glucose) or buffered complex glycerol medium, BMGY (1% yeast extract, 2% peptone, 100 mM potassium phosphate, pH 6.0, 1.34% YNB, 4 10 5% biotin, 1% glycerol). BMMY (1% yeast extract, 2% peptone, 100 mM potassium phosphate, pH 6.0, 1.34% YNB, 4 10 5% biotin, 1% methanol) was used as induction medium. Minimal dextrose (MD) plates (1.34% YNB, 4 10 5% biotin, 2% dextrose) were used for selection of strains containing the pPpHIS4 expression vector. E. coli was cultivated in LB medium (Lennox) purchased from Carl Roth GmbH & Co. KG, Karlsruhe, Germany. Media for plates were solidified by addition of agar to 1.5%. 2.2. Vector and strain construction E. coli TOP 10F’ from life technologies (Vienna, Austria) (F'[lacIq Tn10(tetR)] mcrA Δ(mrr-hsdRMS-mcrBC) φ80lacZΔM15 ΔlacX74 deoR nupG recA1 araD139 Δ(ara-leu)7697 galU galK rpsL(StrR) endA1 λ ) was used for all cloning experiments and propagation of expression vectors. The P. pastoris strain CBS7435 his4 ku70 was used as host strain for all further strain constructions (Näätsaari et al., 2012). Plasmids employed for strain constructions in P. pastoris have been described in the same work. Phusions High Fidelity DNA polymerase (Thermo Fisher Scientific Inc., St. Leon-Rot, Germany) was used for gene amplification according to the recommended PCR protocol. Native and codon optimized gene variants of HPO (H. muticus premnaspirodiene oxygenase, GenBank number of native gene: EF569601.1) and A. thaliana CPR (Cytochrome P450 reductase, GenBank number of native gene: NM_118585.3) were designed manually by applying the Pichia


67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66

codon usage. Amino acid exchanges V482I/A484I were encoded in the HPO sequences to achieve higher catalytic efficiency as reported (Takahashi et al., 2007b). HPO and CPR genes with desired EcoRI/NotI restriction sites for cloning were purchased from GeneArts (Sup. Data 1 and 2). The CPR gene was subcloned into the EcoRI and NotI digested expression vector pPpHIS4 containing a synthetic variant of the AOX1 promoter and the HIS4 gene as auxotrophic marker. The HPO gene was cloned into the EcoRI/NotI site of the pPpT4 expression vector harboring a Zeocin™ resistance marker cassette for selection. C-terminal FLAG- and myc-tags were added to the HPO and CPR sequence, respectively, by PCR (Sup. Table 1). The generated pPpT4[HPO] vector was cut with BglII and BamHI to obtain the HPO gene flanked by AOX1 promoter and terminator regions. The purified fragment was ligated into the BamHI digested vector pPpHIS4[CPR]. P. pastoris ADH-C3, (GenBank: XM_002492172). P. pastoris ADH-C1 (GenBank: XM_002489969) and Sphingobium yanoikuyae alcohol dehydrogenase (GenBank: EU427523.1) genes were amplified from genomic DNA using primers containing EcoRI and NotI restriction sites as well as a FLAG-tag sequence (Sup. Table 1). ADH genes were subcloned into the pPpKan expression vector harboring the AOX1 promoter and the kanamycin/ geneticin selection cassette. A plasmid with a codon-optimised sequence of valencene synthase from C. nootkatensis (Beekwilder et al., 2013) was obtained from GeneArts and was used as template for PCR amplification attaching EcoRI and NotI restriction sites (Sup. Table 1 and Sup. Data 3). The valencene synthase gene was subcloned into the pPpB1 expression vector (AOX1 promoter, Zeocin™ resistance marker gene). Truncated HMG1 (tHMG1, GenBank number: NM_001182434.1) was amplified from genomic DNA of S. cerevisiae using primer pairs FwtHMG1 and RvtHMG1 for SpeI/AscI cloning into pPpHYG (AOX1 promoter, hygromycin resistance). All expression vectors were checked by sequencing the expression cassette and were linearized with BglII for integration into the genome of P. pastoris. Routinely, P. pastoris cells were transformed with 2 mg of linearized plasmids according to the protocol of Lin-Cereghino (LinCereghino et al., 2005). Competent P. pastoris cells and respective amounts of DNA were transferred to ice-cold electroporation cuvettes (0.2 cm; Biozyme Scientific GmbH, Hessisch Oldendorf, Germany) and pulsed at 200 Ω, 25 mF and 1.5 kV. Five hundred mL of icecold 1 M sorbitol was added immediately and after 1 h of incubation at 28 1C the same volume of YPD was added for further 2 h of regeneration at the same temperature. Aliquots were plated on histidin-free minimal media and on YPD plates containing 100 mg L 1 Zeocin™ or 400 mg L 1 geneticin, respectively. All strains described in this work are listed in Table 1. Expression cassette integrations into the genome of P. pastoris were confirmed by colony PCR. The reactions were accomplished according to the manual of the Maximas Hot Start Green PCR Master Mix Kit (Thermo Scientific, St. Leon-Rot, Germany) with control primers listed in Sup. Table 1.

3

2.3. Cultivation and screening of transformants in 96-deep well plates (DWPs) and shake flasks Screening of randomly chosen transformants was carried out in 96-DWPs as previously described, with some modifications (Weis et al., 2004). In brief, cells were pre-cultivated in 250 mL of BMGY medium for 60 h at 28 1C, 320 rpm and 80% humidity. Induction was started by addition of 250 mL of BMMY containing 2% methanol. Methanol was added every 12 h to a final concentration of 1% until 48 h of induction. Transformants from DWPs were screened for potential multi-copy gene integration events by pinning of cell cultures onto plates containing 2 mg mL 1 Zeocin™. Cell lysates were prepared from DWP cultures for western blot analysis (see Section 2.4). Cultivation of strains co-expressing HPO and CPR was scaled up to 50 mL total volume using widenecked, baffled 300 mL shake flasks covered with two layers of cotton cloth. Therefore, 25 mL of pre-cultures in BMGY medium were grown for 60 h at 28 1C, 140 rpm, followed by the addition of 25 mL of induction medium BMMY containing 2% methanol. Induction was performed as for DWP cultivations. 2.4. Protein analysis Cell lysates were prepared by glass bead lysis according to the “Pichia Expression Kit” manual (life technologies, Vienna, Austria) with minor modifications. In brief, cell pellets were disrupted with equal volumes of glass beads either in 1.5 mL reaction tubes or in DWPs with 200 and 100 mL of breaking buffer (50 mM NaH2PO4, pH 7.4, 1 mM PMSF, 1 mM EDTA, 5% glycerol), respectively. The cell suspensions were intermittently vortexed for 30 s and incubated on ice for 30 s, which was repeated for 10 cycles. After cell disruption, total cell lysates were centrifuged at 1500 g in a tabletop centrifuge to remove unbroken cells and cell debris. Protein amounts in cell lysates were quantified using the Bio-Rad (Bradford) protein assay with bovin serum albumin as a standard. Twenty mg of protein were precipitated with trichloroacetic acid (TCA) and solubilized in NuPAGEs sample buffer (life technologies, Vienna, Austria) for subsequent SDS-PAGE separation. Protein expression levels were checked by western blot analysis using primary antibodies against FLAG- or myc-tags. HRPconjugated secondary antibodies and enhanced chemiluminescent signal detection solution (SuperSignal ™, Pierce Biotechnology, Rockford, IL) were used to visualize immunoreactive bands. SDS-PAGE and western blot analyses were performed according to the manual of the NuPAGEs SDS-PAGE Gel System (life technologies, Vienna, Austria). 2.5. Resting cells assay After 48 h of induction, OD600 of the cell cultures was determined and culture volumes corresponding to 300 OD600 units

Table 1 P. pastoris strains used in this study. Name

Description

Wildtype (WT) Pp[HPO/CPR] PpWT/ADH-C1 PpWT/ADH-C3 Pp[HPO/CPR]ADH-C1 Pp[HPO/CPR]ADH-C3 PpValS Pp[HPO/CPR]ValS Pp[HPO/CPR]ValS/ADH-C1 Pp[HPO/CPR]ValS/ADH-C3 Pp[HPO/CPR]ValS/SyADH Pp[HPO/CPR]ValS/ADH-C3/tHMG1

CBS7435 CBS7435 CBS7435 CBS7435 CBS7435 CBS7435 CBS7435 CBS7435 CBS7435 CBS7435 CBS7435 CBS7435

his4 his4 his4 his4 his4 his4 his4 his4 his4 his4 his4 his4

Source ku70 ku70, ku70, ku70, ku70, ku70, ku70, ku70, ku70, ku70, ku70, ku70,

pPpHIS4[HPO/CPR] pPpKan[ADH-C1] pPpKan[ADH-C3] pPpHIS4[HPO/CPR], pPpKan[ADH-C1] pPpHIS4[HPO/CPR], pPpKan[ADH-C3] pPpB1-Zeocin[ValS] pPpHIS4[HPO/CPR], pPpB1-Zeocin[ValS] pPpHIS4[HPO/CPR], pPpB1-Zeocin[ValS], pPpHIS4[HPO/CPR], pPpB1-Zeocin[ValS], pPpHIS4[HPO/CPR], pPpB1-Zeocin[ValS], pPpHIS4[HPO/CPR], pPpB1-Zeocin[ValS],

pPpKan[ADH-C1] pPpKan[ADH-C3] pPpKan[SyADH] pPpKan[ADH-C3], pPpHYG[tHMG1]

Näätsaari et al. (2012) This study This study This study This study This study This study This study This study This study This study This study


67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132


4

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66

were transferred to sterile PYREXstubes. The cells were pelleted at 3220 g for 5 min in an Eppendorf 5810R centrifuge and the supernatants were discarded. The pellets were resuspended in 1 mL of 100 mM KPi, pH 7.4, and the assay was started by the addition of 50 mL of the (þ )-valencene or trans-nootkatol stock solutions (100 mM (þ )-valencene or trans-nootkatol and 1% Triton X-100 (v/v) dissolved in DMSO). PYREXs tubes with cell suspensions were incubated at 28 1C and 215 rpm. One hundred mL samples were withdrawn immediately after substrate addition and at indicated time points. All samples were extracted with 2 mL ethyl acetate by mixing for 30 min on an IKA Vibrax at full speed. Upon spinning for 5 min at 3220 g, the organic phase was transferred into GC vials for subsequent GC-FID measurement. 2.6. Bi-phasic assays for terpenoid production in DWPs and shake flasks Transformants producing ( þ)-valencene in vivo were screened in DWPs for production of terpenoid compounds. Concomitant with the induction medium, 0.2 volumes of n-dodecane were added to each well and induction took place as described in Section 2.3. After 48 h of induction, the cultures were spun at 3220 g for 5 min and the organic layers were transferred into GC vials for GC-FID analysis. The assay was scaled-up to 50 mL in 300 mL shake flasks for biotransformation experiments with strains selected from preceding DWP assays. 2.7. Product analysis by GC–MS and GC-FID Terpenoid extracts in organic solvents, i.e. ethyl acetate as well as n-dodecane, were initially analyzed by GC–MS for identification of compounds using reference standards and comparing the derived mass fragmentation spectra. A 30 m HP column (0.25 mm 0.25 mm) was used on a Hewlett-Packard 5890 Series II plus GC equipped with a 5972 series mass selective detector (MSD). Sample aliquots of 1 μL were injected in split mode (split ratio 20:1) at 220 1C injector and 280 1C detector temperatures with helium as carrier gas at constant flow rate of 32 cm s 1. The oven temperature program was as follows: 70 1C for 1 min, 10 1C min 1 ramp to 200 1C, and 30 1C min 1 ramp to 290 1C (2 min). MSD was operated in a mass range of 40–250 amu with 3.5 scans/s and at electron multiplier voltage of 1635 V. A GC-FID method was developed for routine analyses of terpenoid samples. Therefore, we used a HP-5 column (crosslinked 5% Ph-Me Siloxane; 10 m 0.10 mm 0.10 μm) on a HewlettPackard 6890 GC equipped with a flame ionization detector (FID). Sample aliquots of 1 mL were injected in split mode (split ratio 30:1) at 250 1C injector temperature and 320 1C detector temperature with hydrogen as carrier gas and a flow rate set to 0.4 mL min 1 in constant flow mode (49 cm s 1 linear velocity). The oven temperature program was as follows: 100 1C for 1 min, 20 1C min 1 ramp to 250 1C, and 45 1C min 1 ramp to 280 1C (0.5 min). The use of a high-speed/high-resolution column reduced the total run time to 9 min per sample, without any loss of chromatographic resolution. 2.8. CO-difference spectra Carbon-monoxide (CO)-difference spectra were recorded to estimate the amount of active cytochrome P450 in total cell lysates. Four hundred mL of cell lysates were diluted with 1.6 mL of 100 mM KPi, pH 7.4, 20% glycerol (v/v). After the addition of 100 mL of 200 mM KCN, pH 7.7, to mask the oxidases, a spatula tip of dithionite was added to reduce the cytochrome P450 enzymes. The samples were divided equally into two cuvettes and the baseline was recorded between 380 and 510 nm using a Beckman

Coulter DU 800 spectrophotometer (WinASpec software). After bubbling CO into one cuvette for 1 min the spectrum was recorded. The amount of functional HPO was calculated using an extinction coefficient of 91 mM 1 cm 1 at 450 nm (Omura and Sato, 1964; Schenkman and Jansson, 2006). 2.9. Bi-phasic fed-batch cultivation in bioreactors 2.9.1. Culture media composition Pre-cultures were grown in BMGY medium as described. Batch cultivation was performed in defined basal salt medium (BSM, amounts per liter: 0.17 g CaSO4 2H2O, 2.86 g K2SO4, 0.64 g KOH, 2.32 g MgSO4 7H2O, 0.6 g EDTA, 4.25 g H3PO4, 0.22 g NaCl, 33 g glucose monohydrate). Per liter of BSM, 4.35 mL of 0.02% biotin, 40 mL of 25% NH4Cl, 0.2 mL Antifoam Struktol J650 and 4.35 mL trace salt solution (PTM1, containing per liter 5.0 mL of H2SO4 (69%), 5.99 g CuSO4 5H2O, 1.18 g KJ, 3.0 g MnSO4 H2O, 0.2 g Na2MoO4 2H2O, 0.02 g H3BO3, 0.92 g CoCl2 6H2O, 42.18 g ZnSO4 7H2O and 65.0 g FeSO4 7H2O) were added. For fed-batch media, either glucose (60%) or methanol (99%) was supplemented with 12 mL of PTM1 solution and 12 mL of 0.02% of biotin.

2.9.2. Glucose batch phase For up-scaling of the P. pastoris whole-cell biotransformation reactions a DASGIP parallel bioreactor system was used (DS1500TPSS; DASGIP AG, Jülich, Germany). Pre-cultures of the respective strains were grown for 48 h at 28 1C and 140 rpm in 300 mL baffled shake flasks containing 50 mL of BMGY medium. The seed cultures in 1 L baffled shake flasks containing 100 mL BSM were inoculated with the pre-cultures to an OD600 of 1 and were grown at 28 1C and 140 rpm. At an OD600 of 15, the seed cultures were used to inoculate aseptically 600 mL of sterile BSM medium in the bioreactor vessel to a final OD600 of 1. Dissolved oxygen (dO2) was monitored with a dO2 electrode (OxyProbes, Broadley James Corp., USA). The inlet-gas flow rate was at least 1 vvm to keep dissolved oxygen levels430%. The pH was measured with an autoclavable pH-electrode (FermProbes, Broadley James Corp., Irvine, CA) and controlled at pH 6.0 by automatic addition of 12.5% NH4OH. Temperature was set to 28 1C and agitation was 500 rpm. The temperature, agitation, dO2 and pH were monitored and controlled using the DASGIP monitoring and control system.

2.9.3. Glucose/methanol fed-batch phases Batch culture was grown until glucose was completely consumed, which was indicated by dO2 increase. At this time point, glucose fed-batch was initiated by exponential addition of glucose solution (60%) containing PTM1 and biotin. Feeding was continued over 9 h according to the function f(t) ¼2.19e0.2 t (in g of carbon per h). After the complete consumption of glucose, which was again indicated by an increase of dissolved oxygen (dO2 spike), methanol feeding – supplemented with PTM1 and biotin – was started at a feeding rate of 5 mL h 1. Simultaneous with the start of methanol feeding, n-dodecane was added to the culture at a feeding rate of 20 mL h 1 for 6 h in total, leading to a final organic solvent concentration of 10% (v/v) in the cultivation. Biomass concentration in cultivation broth was determined gravimetrically as cell dry weight (CDW). One mL samples of the cell culture were centrifuged in pre-weighed 1.5 mL Eppendorf tubes for 5 min at 14 000 rpm. The supernatants were withdrawn and the pellets were dried at 80 1C in an oven for at least 2 days to constant weight. Samples from the organic phase were centrifuged and the clear n-dodecane layer was transferred to GC vials for GC-FID analysis.


67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66

5

2.10. Electron microscopy P. pastoris strains were cultivated either under standard conditions or as described for the bi-phasic whole-cell assays in shake flasks (see Section 2.6). After 48 h of methanol induction, i.e. 48 h of terpenoid production in the respective strains, the cells were harvested at 3220 g for 5 min in an Eppendorf 5810R centrifuge and the cell pellets were washed with distilled H2O. The cells were fixed for 5 min in 1% aqueous KMnO4 at room temperature, washed with distilled H2O, and fixed in 1% aqueous KMnO4 for 20 min. Fixed cells were washed four times in distilled water and incubated in 0.5% aqueous uranyl acetate over night at 4 1C. The samples were dehydrated for 20 min, each, in a graded series of ethanol (50%, 70%, 90%, and 100%). Pure ethanol was then exchanged by propylene oxide, and specimen were gradually infiltrated with increasing concentrations (30%, 50%, 70% and 100%) of Agar 100 epoxy resin mixed with propylene oxide for a minimum of 3 h per step. Samples were embedded in pure, fresh Agar 100 epoxy resin and polymerized at 60 1C for 48 h. Ultra-thin sections of 80 nm were stained for 3 min with lead citrate and viewed with a Philips CM 10 transmission electron microscope.

3. Results 3.1. Co-expression of HPO and CPR HPO and CPR were expressed in P. pastoris under the control of the AOX1 promoter either from separately transformed expression cassettes or from the co-expression construct described in the Materials and methods section. While the former strategy – in principle – allowed for fine-tuning of the relative expression levels of HPO and CPR, the latter strategy required only one selection marker, and was thus preferred in this work. Immunological detection of FLAG- and myc-tagged proteins showed that the expression levels in the strain with the HPO/CPR co-expression cassette and in the strain with separately integrated cassettes were similar (Fig. 2). The immunoblot indicated that CPR expression was slightly weaker if co-expressed with HPO. As there was no detectable difference in HPO activity between the different co-expression strains (data not shown) we further on used the Pp [HPO/CPR] co-expression strain in our study and as host for further genetic modifications (Table 1). The abundance of functional HPO present in total cell lysate was determined by recording CO-difference spectra and was calculated to be 8 pmol mg 1 of total protein in strain Pp[HPO/CPR]. 3.2. Whole-cell biotransformation of (þ)-valencene A suitable GC method for fast and efficient separation of the terpenoids (þ)-valencene, cis- as well as trans-nootkatol and

Fig. 2. Western blot analysis using antibodies against FLAG-tag (HPO-FLAG) and myc-tag (CPR-myc). After 48 h of methanol induction, 20 mg of TCA-precipitated total protein of P. pastoris wild type (1), PpHPO (2), PpCPR (3), PpHPO/CPR (4) and Pp[HPO/CPR] (5, co-expression cassette) were loaded.

Fig. 3. Fast GC-FID method for medium-throughput analysis of terpenoid products. Chromatogram of reference substance mix containing ( þ )-valencene (1), cisnootkatol (2), trans-nootkatol (3), ( þ)-nootkatone (4) is shown.

(þ)-nootkatone was established (Fig. 3). We chose to monitor (þ)-valencene conversion in whole-cell systems to circumvent the need for the addition of NADPH cofactor. Employing 300 OD600 units of Pp[HPO/CPR] resting cells for in vitro conversion of (þ)-valencene it became obvious that the addition of 0.02% Triton-X100 to the reaction mixture enhanced ( þ)-valencene conversion (Fig. 4A, data without Triton X-100 not shown). Most probably, the detergent increased the solubility of the hydrocarbon compound in the aqueous reaction mixture, thus enhancing HPOmediated biohydroxylation of (þ)-valencene. If (þ)-valencene was added to wild type P. pastoris cells, no hydroxylation was observed underscoring that trans-nootkatol formation required HPO/CPR expression. A major issue in these initial experiments, however, was the loss of ( þ)-valencene presumably due to evaporation. Re-extraction of ( þ)-valencene immediately upon addition to the assay mixture yielded in recoveries of roughly 90%, while after 12 h of contact with wild type resting cells only about 30% of valencene were recovered (Fig. 4A). Unspecific reaction products of (þ)-valencene were not observed by GC analysis (Sup. Fig. 1). Within 2 h, resting cells harboring HPO and CPR converted 36% of the added substrate to trans-nootkatol and 14% to (þ)-nootkatone (Fig. 4A). However, after 12 h of biotransformation about 50% of added (þ)-valencene was converted to (þ)-nootkatone without residual trans-nootkatol or other by-products, but with a moderate overall yield of 48% due to high substrate loss over time. Biotransformation extracts were subjected to GC–MS analysis for unequivocal identification of (þ )-nootkatone by comparison of the mass fragmentation pattern to an authentic (þ)-nootkatone reference standard (Fig. 5). HPO had been shown to catalyze the mono-hydroxylation of (þ)-valencene to trans-nootkatol in S. cerevisiae (Takahashi et al., 2007b). In the same work, HPO apparently had not converted trans-nootkatol to ( þ)-nootkatone or other hydroxylation products with appreciable catalytic efficiency. (þ)-Nootkatone formation had been observed only at trans-nootkatol concentrations of 4 30 mM. We tested trans-nootkatol conversion in wild type and Pp[HPO/CPR] resting cells assay by replacing (þ)-valencene with the identical amount of trans-nootkatol (2 mM). Interestingly, within 24 h both the wild type cells and the Pp[HPO/CPR] cells almost completely converted the secondary alcohol to (þ)-nootkatone suggesting that the ketone was rather formed by an endogenous activity of P. pastoris than by HPO (Fig. 4B). Notably, the Pp[HPO/CPR] strain was slightly more efficient in the oxidation reaction than the wild type strain. The loss of transnootkatol was minimal during 24 h of biotransformation, due to apparently lower volatility of trans-nootkatol and ( þ)-nootkatone in the aqueous environment as compared to (þ )-valencene.


67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132

6

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66


C. nootkatensis for integration into the genome of P. pastoris wild type and HPO/CPR co-expressing strains (Beekwilder et al., 2013). Functional expression of ValS was determined in two-phase shake flask biotransformation assays by analyzing the n-dodecane phases via GC. Strain PpValS produced 51 mg of (þ )-valencene per liter of cell culture of (Fig. 6A; Sup. Fig. 3A). Co-expression of ValS with HPO and CPR, i.e. strain Pp[HPO/CPR]ValS, yielded reduced overall terpenoid levels amounting to roughly 10 mg L 1 of cell culture. Beside about 2 mg L 1 of (þ)-valencene and roughly 8 mg L 1 of trans-nootkatol, only 0.3 mg L 1 of (þ)-nootkatone were observed (Fig. 6A). We reasoned that in situ produced (þ)-valencene was largely converted to trans-nootkatol by HPO/CPR before being exported from the P. pastoris cells. Like (þ)-valencene, trans-nootkatol was obviously trapped in the n-dodecane layer preventing further conversion to (þ)-nootkatone. Residual amounts of terpenoids in cells were r3% of total terpenoids in the resting cells assays as detected by ethyl acetate extraction and GC-FID measurement. Culture medium was devoid of terpenoids. To analyze the effect of ValS co-expression on HPO/CPR expression levels, we performed western blot analysis (Fig. 6B). The ValS protein levels could not be monitored due to the lack of a specific antibody and the absence of an immunological tag on the recombinant protein. However, the immunoblot showed that the expression levels of HPO and CPR dropped significantly when ValS was co-expressed. It is likely that also ValS was less strongly expressed in the Pp[HPO/CPR]ValS strain than in PpValS considering the lower overall terpenoid production rate in cells expressing all three proteins from the same promoter (Fig. 6A). 3.4. ADH co-overexpression improved ( þ)-nootkatone production

Fig. 4. Time courses of ( þ )-valencene (A) and trans-nootkatol (B) conversion by wild type (light grey bars) and Pp[HPO/CPR] (dark grey bars) are depicted. Mean values and standard deviations of biological triplicates are given.

We speculated that the endogenous activity forming (þ )-nootkatone from trans-nootkatol might be conferred by a hitherto unknown alcohol dehydrogenase of P. pastoris. 3.3. In-situ production and hydroxylation of ( þ)-valencene At this stage, two major limitations for (þ)-nootkatone production in P. pastoris cells co-expressing HPO and CPR had been identified. First, the presumed evaporation of (þ)-valencene added to resting cells and, secondly, a bottleneck in the endogenous P. pastoris activity converting trans-nootkatol to (þ)-nootkatone (Fig. 4B). Handling the evaporation problem first, we resorted to perform (þ)-valencene bioconversions in bi-phasic systems, which were formed by addition of 10–20% (v/v) n-dodecane to the resting cell assays (Girhard et al., 2009). While this measure averted the ( þ)-valencene evaporation problem, it created another serious issue as (þ)-valencene was quantitatively trapped in the n-dodecane phase and would neither evaporate nor be converted by Pp[HPO/CPR] cells to appreciable amounts (data not shown). The straightforward strategy to overcome these problems was the co-expression of a valencene synthase (ValS) leading to (þ )-valencene production from farnesyl pyrophosphate (FPP) in vivo. Several ValS genes from citrus fruits have been described in literature, but upon expression in yeasts most of these genes do not permit the production of commercially relevant amounts of (þ )-valencene (Asadollahi et al., 2008; Takahashi et al., 2007a). We selected a codon-optimized variant of a ValS gene from

Having found a solution for the evaporation problem, we next focused on improving the oxidation of trans-nootkatol to (þ)-nootkatone. Due to the efficient conversion of trans-nootkatol to (þ )-nootkatone found in resting wild type cells (Fig. 4B), we suspected that P. pastoris harbors at least one alcohol dehydrogenase that catalyzes this reaction. We aimed at overexpressing this activity for oxidizing as much of intracellular trans-nootkatol to (þ)-nootkatone as possible before the terpenoids would be trapped in the n-dodecane phase. Therefore, three different ADH genes were chosen for co-overexpression in the Pp[HPO/CPR]ValS strain. On the one hand, we searched the P. pastoris genome database (http://pichiagenome. org) for alcohol dehydrogenase genes with supposedly broad substrate specificity and found two of in total six ADH genes in P. pastoris, PpADH-C3 (XM_002492172) and PpADH-C1 (XM_002489969), which we chose for overexpression in the strains Pp[HPO/CPR], Pp[HPO/CPR] ValS and in wild type (De Schutter et al., 2009; Küberl et al., 2011; Mattanovich et al., 2009). On the other hand, we selected a S. yanoikuyae ADH (EU427523.1) gene from an in-house collection and tested its co-expression effect on trans-nootkatol oxidation (Lavandera et al., 2008). Both ADHs of P. pastoris were overexpressed in wild type and Pp[HPO/CPR] strains as demonstrated by western blot analyses, whereupon PpADH-C1 was expressed to much higher protein levels in both strains compared to PpADH-C3 (Sup. Fig. 2C and D). Overexpression of PpADH-C3 improved conversion of trans-nootkatol to (þ)-nootkatone as shown in both wild type and Pp[HPO/CPR] strains (Sup. Fig. 2A and B). Despite the abundant PpADH-C1 protein, strains overexpressing ADH-C1 did not show enhanced (þ)-nootkatone production as compared to wild type and Pp[HPO/CPR] reference strains. Thus, it was specifically the overexpression of PpADH-C3 that conferred (þ)-nootkatone production in P. pastoris. ( þ)-Nootkatone production levels of the newly created in situ production strains co-expressing HPO/CPR, valencene synthase and different ADHs were determined in bi-phasic, whole-cell biotransformation assays after 48 h of induction. In contrast to the other two ADHs (data not shown), PpADH-C3 co-expression


67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66

7

Fig. 5. GC chromatogram and mass fragmentation patterns of authentic ( þ)-nootkatone standard ((A) and (B), respectively) as well as the same data sets of (þ )-nootkatone produced by resting cell sample of strain Pp[HPO/CPR] after 12 h of ( þ)-valencene conversion ((C) and (D), respectively) are compared.

improved ( þ)-nootkatone formation about 20-fold in the Pp[HPO/ CPR]ValS strain background (Fig. 6A). Thus, we had probably identified the endogenous P. pastoris activity responsible for efficient conversion of trans-nootkatol to ( þ)-nootkatone. At least, overexpressing PpADH-C3 from the AOX1 promoter enhanced (þ)-nootkatone synthesis in bi-phasic shake flask cultivations to roughly 7 mg L 1 of cell culture making (þ )-nootkatone the most prominent terpenoid compound in the n-dodecane phase (Fig. 6A; Sup. Fig. 3B). Overall, about 14 mg terpenoids per liter of cell culture were formed in the Pp[HPO/CPR]ValS/ADH-C3 strain. The homologous overexpression of C-terminally FLAG-tagged ADH did not significantly influence HPO and CPR expression levels in the Pp [HPO/CPR]ValS strain background (Fig. 6B). 3.5. Improvement of terpenoid production by tHMG1 co-overexpression

Fig. 6. (A) De novo production of terpenoids in PpValS, Pp[HPO/CPR]ValS (PpHCV), Pp[HPO/CPR]ValS/ADH-C3 (PpHCVA) and Pp[HPO/CPR]ValS/ADH-C3/tHMG1 (PpHCVA/tHMG1), respectively, after 48 h of methanol induction. Mean values and standard deviations of biological triplicates are given. (B) Immunological detection of HPO-FLAG, PpADH-C3-FLAG and CPR-myc by western blot analysis using antibodies directed against the affinity tags. (1) Pp[HPO/CPR]; (2) Pp[HPO/ CPR]Vals (ValS was un-tagged); (3) Pp[HPO/CPR]ValS/ADH-C3; (4) Pp[HPO/CPR] ValS/ADH-C3/tHMG1; Samples were withdrawn after 48 h of methanol induction. Twenty mg of TCA-precipitated total protein were loaded on the SDS gel.

Multiple studies have shown that overexpression of the catalytic domain of Hmg1p (truncated Hmg1p, tHMG1) alone or in combination with other metabolic engineering strategies led to elevated production of terpenoids in yeast (Asadollahi et al., 2008, 2010; Engels et al., 2008; Jackson et al., 2003; Rico et al., 2010; Tokuhiro et al., 2009; Wriessnegger and Pichler, 2013). Thus, we expressed S. cerevisiae tHMG1 from the AOX1 promoter in the Pp [HPO/CPR]ValS/ADH-C3 strain background and tested the effect on (þ)-nootkatone formation in bi-phasic shake flask cultures (Fig. 6A). Overall, tHMG1 overexpression roughly tripled terpenoid amounts to approximately 42 mg L 1 of cell culture in the Pp


67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132

8

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66


[HPO/CPR]ValS/ADH-C3/tHMG1 strain after 48 h of methanol induction. ( þ)-Nootkatone levels were 2.5-fold higher than in the Pp[HPO/CPR]ValS/ADH-C3 strain reaching about 17 mg L 1 of cell culture, eventually. Nevertheless, after 48 h of terpenoid production about 18 mg of trans-nootkatol per liter of cell culture were accumulating. Prolonging the induction time to 72 h led to virtually complete oxidation of trans-nootkatol to (þ)-nootkatone and to accumulation of 35 mg L 1 of (þ)-nootkatone. tHMG1 overexpression obviously increased the available FPP pool providing ValS with enhanced substrate concentrations to form elevated levels of valencene. In the Pp[HPO/CPR]ValS/ADH-C3/tHMG1 strain, apparently, both the conversion of ( þ )-valencene by HPO/ CPR as well as ( þ)-nootkatone formation appeared to be limiting

steps in shake flasks assays (Fig. 6A), although HPO, CPR and ADH-C3 protein levels were only marginally decreased by cooverexpression of tHMG1 (Fig. 6B). We were curious whether the overexpression of several proteins, high-level terpenoid production or the unusual culturing conditions in bi-phasic systems had any recognizable effect on the subcellular structures of P. pastoris. Therefore, we carried out ultra-structural analyses by electron microscopy (Sup. Fig. 4). Proliferation of peroxisome structures was observed as was expected for methanol-induced P. pastoris cells (Wriessnegger et al., 2007). Electron micrographs showed that all strains co-expressing HPO and CPR formed stacked layers of ER membranes (see image enlargement of Sup. Fig. 4B). These stacked layers were reminiscent of the so-called ‘karmellae’ structures

Fig. 7. Bioreactor production of terpenoids in bi-phasic systems employing strains Pp[HPO/CPR]ValS ((A) and (D)), Pp[HPO/CPR]ValS/ADH-C3 (B) and (E) and Pp[HPO/CPR] ValS/ADH-C3/tHMG1 ((C) and (F)). One mL samples of n-dodecane phase of each time point were analyzed by GC-FID ((D)–(F)). Western Blot analyses of samples taken at indicated time points. Twenty mg of TCA-precipitated total protein were loaded on SDS gels. Mean values of terpenoid production and standard deviation of three independent bioreactor cultivations are given.


67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66

first observed upon HMG1 overexpression in S. cerevisiae (Wright et al., 1988). Comparing the (þ )-nootkatone producing strains with the Pp [HPO/CPR] strain no major differences in cell morphology were observed. Additionally, we assessed the influence of n-dodecane on the ultrastructure of the Pp[HPO/CPR]ValS/ADH-C3/tHMG1 strain (compare Sup. Fig. 4E and F) and found that n-dodecane did not have any obvious effect on the subcellular structures.

3.6. Fed-batch cultivation of (þ )-nootkatone producing strains One of the advantages of the P. pastoris host system over the more extensively studied S. cerevisiae host platform is the higher cell density obtained in controlled fed-batch cultivations. P. pastoris has been reported to accumulate more than 150 g CDW L 1 in bioreactor cultivations (Jahic et al., 2006). Fed-batch cultivations in lab scale are the ultimate test whether it is feasible to scale up a biotransformation to commercial scale. Thus, we cultivated the (þ )-nootkatone producing strains Pp[HPO/CPR]ValS, Pp[HPO/CPR] ValS/ADH-C3 and Pp[HPO/CPR]ValS/ADH-C3/tHMG1 in parallel in bioreactors (DASGIP, Jülich, Germany). Aliquots of the cell cultures and the n-dodecane phases were withdrawn at different timepoints of methanol induction (Fig. 7). Determination of cell dry weights (CDW) showed steady increases of the total biomass during methanol fed-batch cultivations. After 108 h, the CDW of all tested strains reached values of 100–140 g L 1 of culture. On average, the CDW almost tripled in the course of 108 h of methanol induction. Overall, steady increases of total terpenoid levels were observed (Fig. 7A–C). As already observed in shake flask experiments, Pp[HPO/CPR]ValS produced predominantly trans-nootkatol and lower amounts of (þ)-nootkatone in the bioreactor, yet formed higher total terpenoid levels than in shake flask. After 108 h of methanol induction, we obtained 136 mg L 1 transnootkatol and 64 mg L 1 (þ )-nootkatone (Fig. 7A). ( þ)-nootkatone yield was clearly enhanced by the co-expression of PpADH-C3 as the Pp[HPO/CPR]ValS/ADH-C3 strain made 97 mg of (þ)-nootkatone per liter of cell culture within 108 h of methanol induction (Fig. 7B). Highest ( þ)-nootkatone productivity was recorded for the Pp[HPO/CPR]ValS/ADH-C3/tHMG1 strain yielding ( þ)-nootkatone levels of 208 mg L 1 cell culture (Fig. 7C). Remarkably, this strain had produced additional 166 mg L 1 of ( þ)-valencene and 44 mg L 1 of trans-nootkatol after 108 h of induction in the bioreactor culture. Comparing the specific ( þ)-nootkatone production levels obtained after 48 h of induction in shake flask (0.34 mg L 1/OD600) and bioreactor cultivations (0.35 mg L 1/ OD600) suggested that the higher (þ )-nootkatone yield in bioreactors was solely due to the higher cell densities, and was not triggered by improved cultivation parameters in the bioreactors. Western blot analyses revealed that the levels of HPO and CPR proteins reached a maximum after 24–48 h of methanol induction and declined until the end of the induction time, suggesting that protein biosynthesis was reduced or protein degradation was enhanced after a certain time of bioreactor cultivation (Fig. 7D– F). Ongoing work in our laboratory hints at instability of HPO and/ or CPR causing overall destabilization of the HPO/CPR-interaction after prolonged induction time (data not shown). We currently do not see any indication that cellular stress responses or genetic instabilities are to blame for the reduced level of HPO and CPR proteins after prolonged bioreactor cultivation times. However, the western blot results correlated with terpenoid formation in the bioreactor, as the amount of trans-nootkatol did not further increase after 48 h of induction – in contrast to the amount of (þ )-nootkatone – indicating a diminished activity of the enzymes HPO and CPR, respectively.

9

4. Discussion The production of ( þ)-nootkatone by biotechnological means is of high industrial interest. Numerous reports have been published using whole-cell systems for the specific conversion of (þ)-valencene to (þ )-nootkatone (Table 2). Comparing the published volumetric activities to the results described in this work suggests that the P. pastoris system appears to be one of the most efficient production hosts for (þ )-nootkatone so far. Many cell systems, including whole-plants or algae (Drawert et al., 1984; Furusawa et al., 2005; Sakamaki et al., 2005), bacteria (Girhard et al., 2009; Okuda et al., 1994) or fungi (Cankar et al., 2011; Cankar et al. 2014; Fraatz et al., 2009b; Kaspera et al., 2005; Krügener et al., 2010) suffer either from long cultivation times or from low conversion rates, or – in the worst case – from the inapplicability for large scale processes. Remarkably, whole-cells of P. pastoris can either be applied to convert externally added ( þ)-valencene to (þ)-nootkatone with moderate efficiency (Fig. 4A), or can be metabolically engineered to form commercially interesting levels of ( þ)-nootkatone from simple carbon sources (Figs. 6 and 7). Up-scaling (þ)-valencene conversion with P. pastoris in aqueous phase will require sophisticated process development to deal with vast amounts of mostly ( þ)-valencene vapors that need to be recycled in order to increase the overall yields, but also to reduce the hydrocarbon burden on the environment. in vivo production of (þ)-nootkatone in P. pastoris, on the other hand, may be further optimized in terms of terpenoid yield and overall purity, which should facilitate downstream processing, e.g. isolation of (þ)-nootkatone from the organic phase of the cultivation. To the best of our knowledge, there are only two reports from one group on metabolically engineered yeast cells that can produce (þ)-nootkatone in vivo from simple carbon source (Table 2). Production of ( þ)-nootkatone by S. cerevisiae cells overexpressing ValS and a chicory cytochrome P450 monooxygenase CYP71AV8 yielded in volumetric productivities of 1.36 mg L 1 for ( þ)-valencene and 0.04 mg L 1 for ( þ)-nootkatone (Cankar et al., 2011). Recently, a valencene oxidase from C. nootkatensis was described to produce (þ)-nootkatone upon co-expression with a valencene synthase (CnVS) in S. cerevisiae (Cankar et al., 2014). Production of 0.14 mg of ( þ)-nootkatone per liter yeast culture was reported after three days of single-phase shake flask cultivation. In contrast, the P. pastoris strain overexpressing ValS described here produced 51 mg ( þ)-valencene per liter cell culture under shake flask conditions. Upon co-expression of HPO/CPR and (þ)-valencene synthase volumetric productivities of roughly 2 mg L 1 of (þ )-valencene, 8 mg L 1 of trans-nootkatol and about 0.3 mg L 1 of (þ )-nootkatone were observed before any optimization steps (Fig. 6A). In all these studies, the coexpression of ValS with appropriate cytochrome P450 systems was key to successful production of ( þ)-nootkatone. In several studies, using S. cerevisiae or E. coli systems, the addition of ndodecane as second phase has been described to be beneficial in whole-cell biotransformations, serving as a reservoir for volatile, hydrophobic or toxic products (Asadollahi et al., 2008; Girhard et al., 2009; Newman et al., 2006; Ro et al., 2006; Scalcinati et al., 2012). The organic phase prevents toxicity effects of high concentrations of terpenoids on yeasts and attenuates their potential inhibition of cytochrome P450-mediated hydroxylation reaction. Such effects have recently been described as hampering (þ)-valencene biohydroxylation in S. cerevisiae (Gavira et al., 2013). Interestingly, while the S. cerevisiae strain co-expressing CnVO/CnVs did not hydroxylate much of the formed ( þ)-valencene in the biphasic system, almost all trans-nootkatol formed in the mono-phasic system was further processed to ( þ)-nootkatone (Cankar et al., 2014). In the P. pastoris system described here, (þ)-valencene produced in situ is efficiently hydroxylated to trans-


67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132


10

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66

Table 2 List of biotransformation approaches using whole-cells of different organisms for the production of ( þ )-nootkatone. Organism

Reaction set up/time

( þ )-Valencene ( þ )-Nootkatone (transuseda [mg L 1] nootkatol) produceda [mg L 1]

mol % References

Bacterial whole-cell conversions Enterobacter

nd

nd

nd

11 (nd)

Rhodococcus

nd

500 mg

2.5 mg (nd)

B. subtilis CYP109B1/expression in E. coli (two-phase system:isooctane/water) Plant/algae whole-cell conversions C. paradisii

Culture preparation: 24 h Conversion: 8h

408

16 (318)

0.5 (nd) 4 (72)

Culture preparation: 9 months plant cell nd culture growth; Conversion: 6 h Culture preparation: callus growth of the 900 plant; Conversion: 20 days Culture preparation: 7 days algae culture 400 growth; Conversion: 18 days

1.1 (nd)

G. pentaphyllum Chlorella fusca var. vacuolata Yeast/fungi whole-cell conversions Chaetomium globosum

597 (91.3)

19 (nd) 62 (9)

252 (0)

59 (0)

2 ( o1) 38 (8) 11 (3) 6 (nd)

5 Days preculture, 3 days inoculation

460

8 (1)

P. sapidus/fungi lyophilisate

Preparation of culture, 9 days Conversion: 24 h

552, 2 760

225 (50), 325 (90)

P. sapidus/homogenized fresh mycelium, fedbatch

nd

10 000

600 (nd)

3 days

80

4 (11)

2 days,

0.04 (0.92)

5 days,

In-situ production In-situ production 473

Preparation of culture: 2 days, Conversion: 48 h (shake flask)

In-situ production

17 (18)

Fed-batch cultivation in bioreactor Preparation of cultures: 2 days, Conversion: 108 h

In-situ production

208 (43)

CYP71D51v2 from tobacco/expression in S. cerevisiae Preparation of culture: Conversion: 24 h Chicory CYP71AV8/expression in S. cerevisiae (in situ Preparation of culture: valencene production) Conversion: 3 days C. nootkatensis CnVO/expression in S. cerevisiae Preparation of culture: (in situ valencene production (CnVS)) Conversion: 3 days H. muticus HPO/expression in P. pastoris (coPreparation of culture: expression of PpADH-C3) Conversion: 24 h H. muticus HPO/expression in P. pastoris (in situ ( þ)-valencene production, PpADH-C3 and tHmg1p co-expression H. muticus HPO/expression in P. pastoris (in situ ( þ)-valencene production, PpADH-C3 and tHmg1p co-expression a

2 days,

0.14 (0) 317 (0)

63 (0)

Dhavlikar and Albroscheit (1973) Okuda et al. (1994) Girhard et al. (2009) Drawert et al. (1984) Sakamaki et al. (2005) Furusawa et al. (2005) Kaspera et al. (2005) Fraatz et al. (2009a) Zorn et al. (2009) Gavira et al. (2013) Cankar et al. (2011) Cankar et al. (2014) This study, 2013 (data not shown) This study, 2013 This study, 2013

Amounts calculated per liter assay volume; nd: not determined/not described.

nootkatol, and further converted to ( þ)-nootkatone, before terpenoids are excreted and trapped in the organic phase. This is the first report that P. pastoris is suitable for application in a bi-phasic biotransformation reaction using n-dodecane. Growth tests with n-dodecane at concentrations of 10–20% revealed excellent solvent compatibility of P. pastoris as growth rates were indistinguishable from those without n-dodecane (data not shown). Very recently, it was suggested that it is most likely an endogenous yeast enzyme that converts trans-nootkatol into (þ )-nootkatone and that this reaction cannot be credited to cytochrome P450 enzymes applied to make trans-nootkatol (Gavira et al., 2013; Takahashi et al., 2007b). By overexpressing an ADH of P. pastoris, we have employed an enzyme that boosts (þ )-nootkatone formation 4 20-fold in shake flask cultivations (Fig. 6A). In bioreactor cultivations, PpADH-C3 co-overexpression enhanced ( þ)-nootkatone production 1.5-fold as judged from comparing the productivities of Pp[HPO/CPR]/ValS/ADH-C3 and Pp[HPO/CPR]ValS strains (Fig. 7). Supposedly, the endogenous P. pastoris activity responsible for the trans-nootkatol to (þ )-nootkatone conversion is more actively expressed or more active in bioreactor cultivations than in shake flasks. In the latter, ( þ)-nootkatone makes up only about 3% of total terpenoids produced in Pp [HPO/CPR]ValS (Fig. 6A), while it accounts for about 22% of total terpenoids in the same strain in the bioreactors (Fig. 7). Like in many reports on terpenoid production in baker's yeast (Asadollahi et al., 2008, 2010; Engels et al., 2008; Jackson et al., 2003;

Rico et al., 2010; Tokuhiro et al., 2009; Wriessnegger and Pichler, 2013), co-overexpressing a truncated version of hydroxymethylglutaryl CoA reductase (tHMG1) increased (þ)-nootkatone productivity clearly in this work. Sesquiterpenes, like (þ)-valencene are derived from farnesyl pyrophosphate (FPP), which is a product of the mevalonate pathway. Regulation of the mevalonate pathway in eukaryotes is complex, but HMG1 activity and the FPP branch-point are two key regulation sites in the pathway (Basson et al., 1987; Donald et al., 1997; Maury et al., 2005; Polakowski et al., 1998). The overexpression of the catalytic domain of HMG1 in Pp[HPO/CPR]ValS/ ADH-C3/tHMG1 resulted in an overall increase in terpenoid levels and a final (þ )-nootkatone titer in bioreactor cultivations of 208 mg L 1 cell culture (Fig. 7C). This indicates (i) that mevalonate pathway regulation is at least partially similar in P. pastoris and S. cerevisiae, and that (ii) the concept of metabolic engineering of the FPP pool can be transferred from S. cerevisiae to P. pastoris successfully. Starting from a volumetric yield of 208 mg L 1 of (þ)-nootkatone in the bioreactor cultivations, it is not too big a leap to a commercially viable process in the g L 1 range. We foresee a couple of immediate improvements that can further increase (þ)-nootkatone yield in the P. pastoris system described in this work. Based on the high amount of (þ)-valencene in the ndodecane phase of the Pp[HPO/CPR]ValS/ADH-C3/tHMG1 strain in bioreactor cultivation (Fig. 7C) and the decrease of HPO and CPR protein signals detected by western blot analysis in the course of bioreactor cultivation (Fig. 7F), it should be the cytochrome P450


67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 Q2 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66

activity that is limiting in this case. Novel or engineered cytochrome P450 enzymes with superior catalytic properties will be beneficial to enhance the potential of P. pastoris cells in converting (þ )-valencene. Alternatively, fine-tuned expression of ValS, HPO, CPR, ADH-C3 and tHmg1p might aid in quantitatively transforming ( þ)-valencene into ( þ)-nootkatone, before the latter is exported from the yeast cells and trapped in the n-dodecane phase. Moreover, we assume that the ( þ)-nootkatone yield can be further improved by bioprocess optimization (Gasser et al., 2010; Bora et al., 2012) and further metabolic engineering approaches, e. g. FPP-pathway engineering, as reported to be beneficial for the production of other sesquiterpenoids in yeast (Asadollahi et al., 2010; Paradise et al., 2008; Ro et al., 2006; Takahashi et al., 2007a). P. pastoris is known as very efficient expression system for correctly folded soluble proteins producing up to several grams of protein per liter of culture (Macauley-Patrick et al., 2005). Lately, major efforts have been made in applying P. pastoris also as expression host for membrane proteins and as production platform for hydrophobic compounds (Freigassner et al., 2009; Hirz et al., 2013; Ramón and Marín, 2011; Bö rgel et al., 2012; Araya-Garay et al., 2012b). However, there are still only few reports on cytochrome P450 expression studies in this yeast (Andersen and Møller, 2002; Dietrich et al., 2005; Geier et al., 2012; Kolar et al., 2007; Trant, 1996; Wang et al., 2010). Data reported in this work suggest that P. pastoris has excellent potential for (i) functional expression of (plant-derived) membrane-attached cytochrome P450 enzymes and for (ii) the in vivo production and conversion of hydrophobic compounds, such as (þ)-valencene and (þ )-nootkatone.

Acknowledgments We thank Theo Sonke for fruitful discussions. This work has been supported by the Federal Ministry of Science, Research and Economy (BMWFW), the Federal Ministry of Traffic, Innovation and Technology (bmvit), the Styrian Business Promotion Agency SFG, the Standortagentur Tirol and ZIT—Technology Agency of the City of Vienna through the COMET-Funding Program managed by the Austrian Research Promotion Agency FFG.

Appendix A. Supporting information Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.ymben.2014.04.001. References Andersen, M.D., Møller, B.L., 2002. Use of methylotropic yeast Pichia pastoris for expression of cytochromes P450. Methods Enzymol. 357, 333–342. Araya-Garay, J.M., Feijoo-Siota, L., Rosa-Dos-Santos, F., Veiga-Crespo, P., Villa, T.G., 2012a. Construction of new Pichia pastoris X-33 strains for production of lycopene and β-carotene. Appl. Microbiol. Biotechnol. 93, 2483–2492. Araya-Garay, J.M., Ageitos, J.M., Vallejo, J.A., Veiga-Crespo, P., Sánchez-Pérez, A., Villa, T.G., 2012b. Construction of a novel Pichia pastoris strain for production of xanthophylls. AMB Express 2, 24. Asadollahi, M.A., Maury, J., Schalk, M., Clark, A., Nielsen, J., 2010. Enhancement of farnesyl diphosphate pool as direct precursor of sesquiterpenes through metabolic engineering of the mevalonate pathway in Saccharomyces cerevisiae. Biotechnol. Bioeng. 106, 86–96. Asadollahi, M.A., Maury, J., Møller, K., Nielsen, K.F., Schalk, M., Clark, A., Nielsen, J., 2008. Production of plant sesquiterpenes in Saccharomyces cerevisiae: effect of ERG9 repression on sesquiterpene biosynthesis. Biotechnol. Bioeng. 99, 666–677. Basson, M.E., Moore, R.L., O'Rear, J., Rine, J., 1987. Identifying mutations in duplicated functions in Saccharomyces cerevisiae: recessive mutations in HMG-CoA reductase genes. Genetics 117, 645–655. Beekwilder, J., van Houwelingen, A., Cankar, K., van Dijk, A.D.J., de Jong, R.M., Stoopen, G., Bouwmeester, H., Achkar, J., Sonke, T., Bosch, D., 2013. Valencene synthase from the heartwood of Nootka cypress (Callitropsis nootkatensis) for biotechnological production of valencene. Plant Biotechnol. J., 25 ([Epub ahead of print] Sep).

11

Bora, N., Bawa, Z., Bill, R.M., Wilks, M.D.B., 2012. The implementation of a design of experiments strategy to increase recombinant protein yields in yeast (review). Methods Mol. Biol. 866, 115–127. Bö rgel, D., van den Berg, M., Hüller, T., Andrea, H., Liebisch, G., Boles, E., Schorsch, C., van der Pol, R., Arink, A., Boogers, I., van der Hoeven, R., Korevaar, K., Farwick, M., Kö hler, T., Schaffer, S., 2012. Metabolic engineering of the non-conventional yeast Pichia ciferrii for production of rare sphingoid bases. Metab. Eng. 14, 412–426. Bouwmeester, H., de Kraker, J., Schurink, M., Bino, R., de Groot, A., Franssen, M., 2007. Plant Enzymes for Bioconversion. US Patent 7214507. Cankar, K., van Houwelingen, A., Bosch, D., Sonke, T., Bouwmeester, H., Beekwilder, J., 2011. A chicory cytochrome P450 mono-oxygenase CYP71AV8 for the oxidation of ( þ )-valencene. FEBS Lett. 585, 178–182. Cankar, K., van Houwelingen, A., Goedbloed, M., Renirie, R., de Jong, R.M., Bouwmeester, H., Bosch, D., Sonke, T., Beekwilder, J., 2014. Valencene oxidase CYP706M1 from Alaska cedar (Callitropsis nootkatensis). FEBS Lett. 25 ([Epub ahead of print] Feb). Cereghino, J.L., Cregg, J.M., 2000. Heterologous protein expression in the methylotrophic yeast Pichia pastoris. FEMS Microbiol. Rev. 24, 45–66. Chang, M.C.Y., Keasling, J.D., 2006. Production of isoprenoid pharmaceuticals by engineered microbes. Nat. Chem. Biol. 2, 674–681. Cregg, J.M.L. Cereghino, Shi, J., J.R.Higgins, D., 2000. Recombinant protein expression in Pichia pastoris. Mol. Biotechnol. 16, 23–52. Dhavlikar, R., Albroscheit, G., 1973. Mikrobiologische Umsetzung von Terpenen: Valencene. Dragoco Rep 12, 251–258. De Kraker, J.W., Schurink, M., Franssen, M.C., Kö nig, W.A., de Groot, A., Bouwmeester, H.J., 2003. Hydroxylation of sesquiterpenes by enzymes from chicory (Cichorium intybus L.) roots. Tetrahedron 59, 409–418. De Schutter, K., Lin, Y.C., Tiels, P., Van Hecke, A., Glinka, S., Weber-Lehmann, J., Rouzé, P., Van de Peer, Y., Callewaert, N., 2009. Genome sequence of the recombinant protein production host Pichia pastoris. Nat. Biotechnol. 27, 561–566. Dietrich, G., Dolan, M.C., Peralta-Cruz, J., Schmidt, J., Piesman, J., Eisen, R.J., Karchesy, J.J., 2006. Repellent activity of fractioned compounds from Chamaecyparis nootkatensis essential oil against nymphal Ixodes scapularis (Acari: Ixodidae). J. Med. Entomol. 43, 957–961. Dietrich, M., Grundmann, L., Kurr, K., Valinotto, L., Saussele, T., Schmid, R.D., Lange, S., 2005. Recombinant production of human microsomal cytochrome P450 2D6 in the methylotrophic yeast Pichia pastoris. Chembiochem 6, 2014–2022. Donald, K.A., Hampton, R.Y., Fritz, I.B., 1997. Effects of overproduction of the catalytic domain of 3-hydroxy-3-methylglutaryl coenzyme A reductase on squalene synthesis in Saccharomyces cerevisiae. Appl. Environ. Microbiol. 63, 3341–3344. Drawert, F., Berger, R.G., Godelmann, R., 1984. Regioselective biotransformation of valencene in cell suspension cultures of Citrus sp. Plant Cell Rep. 3, 37–40. Dumas, B., Brocard-Masson, C., Assemat-Lebrun, K., Achstetter, T., 2006. Hydrocortisone made in yeast: metabolic engineering turns a unicellular microorganism into a drug-synthesizing factory. Biotechnol. J 1, 299–307. Engels, B., Dahm, P., Jennewein, S., 2008. Metabolic engineering of taxadiene biosynthesis in yeast as a first step towards Taxol (Paclitaxel) production. Metab. Eng. 10, 201–206. Erdtman, H., Hirose, Y., 1962. The chemistry of the natural order Cupressales. 46: the structure of nootkatone. Acta Chem. Scand. 16, 1311–1314. Flor-Weiler, L.B., Behle, R.W., Stafford, K.C., 2011. Susceptibility of four tick species, Amblyomma americanum, Dermacentor variabilis, Ixodes scapularis, and Rhipicephalus sanguineus (Acari: Ixodidae), to nootkatone from essential oil of grapefruit. J. Med. Entomol. 48, 322–326. Fraatz, M., Berger, R.G., Zorn, H., 2009a. Nootkatone—a biotechnological challenge. Appl. Microbiol. Biotechnol. 83, 35–41. Fraatz, M., Riemer, S.J.L., Stö ber, R., Kaspera, R., Nimtz, M., Berger, R.G., Zorn, H., 2009b. A novel oxygenase from Pleurotus sapidus transforms valencene to nootkatone. J. Mol. Catal. B: Enzym. 61, 202–207. Freigassner, M., Pichler, H., Glieder, A., 2009. Tuning microbial hosts for membrane protein production. Microb. Cell Fact. 8, 69. Furusawa, M., Hashimoto, T., Noma, Y., Asakawa, Y., 2005. Highly efficient production of nootkatone, the grapefruit aroma, from valencene by biotransformation. Chem. Pharm. Bull. (Tokyo). 53, 1513–1514. Gasser, B., Dragosits, M., Mattanovich, D., 2010. Engineering of biotin-prototrophy in Pichia pastoris for robust production processes. Metab. Eng. 12, 573–580. Gavira, C., Hö fer, R., Lesot, A., Lambert, F., Zucca, J., Werck-Reichhart, D., 2013. Challenges and pitfalls of P450-dependent ( þ)-valencene bioconversion by Saccharomyces cerevisiae. Metab. Eng. 18, 25–35. Geier, M., Braun, A., Emmerstorfer, A., Pichler, H., Glieder, A., 2012. Production of human cytochrome P450 2D6 drug metabolites with recombinant microbes—a comparative study. Biotechnol. J 7, 1346–1358. Girhard, M., Machida, K., Itoh, M., Schmid, R.D., Arisawa, A., Urlacher, V.B., 2009. Regioselective biooxidation of (þ)-valencene by recombinant E. coli expressing CYP109B1 from Bacillus subtilis in a two-liquid-phase system. Microb. Cell Fact. 8, 36. Gliszczyńska, A., Łysek, A., Janeczko, T., Świtalska, M., Wietrzyk, J., Wawrzeńczyk, C., 2011. Microbial transformation of ( þ )-nootkatone and the antiproliferative activity of its metabolites. Bioorg. Med. Chem. 19, 2464–2469. Hirz, M., Richter, G., Leitner, E., Wriessnegger, T., Pichler, H., 2013. A novel cholesterol-producing Pichia pastoris strain is an ideal host for functional expression of human Na,K-ATPase α3β1 isoform. Appl. Microbiol. Biotechnol. 97, 9465–9478. Jackson, B.E., Hart-Wells, E.A., Matsuda, S.P.T., 2003. Metabolic engineering to produce sesquiterpenes in yeast. Org. Lett. 5, 1629–1632.


67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132

12

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56


Jahic, M., Veide, A., Charoenrat, T., Teeri, T., Enfors, S.O., 2006. Process technology for production and recovery of heterologous proteins with Pichia pastoris. Biotechnol. Prog. 22, 1465–1473. Jordan, R.A., Schulze, T.L., Dolan, M.C., 2012. Efficacy of plant-derived and synthetic compounds on clothing as repellents against Ixodes scapularis and Amblyomma americanum (Acari: Ixodidae). J. Med. Entomol. 49, 101–106. Kaspera, R., Krings, U., Nanzad, T., Berger, R.G., 2005. Bioconversion of ( þ )valencene in submerged cultures of the ascomycete Chaetomium globosum. Appl. Microbiol. Biotechnol. 67, 477–483. Kolar, N.W., Swart, A.C., Mason, J.I., Swart, P., 2007. Functional expression and characterisation of human cytochrome P45017alpha in Pichia pastoris. J. Biotechnol. 129, 635–644. Krügener, S., Krings, U., Zorn, H., Berger, R.G., 2010. A dioxygenase of Pleurotus sapidus transforms ( þ)-valencene regio-specifically to ( þ )-nootkatone via a stereo-specific allylic hydroperoxidation. Bioresour. Technol 101, 457–462. Küberl, A., Schneider, J., Thallinger, G.G., Anderl, I., Wibberg, D., Hajek, T., Jaenicke, S., Brinkrolf, K., Goesmann, A., Szczepanowski, R., Pühler, A., Schwab, H., Glieder, A., Pichler, H., 2011. High-quality genome sequence of Pichia pastoris CBS7435. J. Biotechnol. 154, 312–320. Lavandera, I., Kern, A., Resch, V., Ferreira-Silva, B., Glieder, A., Fabian, W.M.F., de Wildeman, S., Kroutil, W., 2008. One-way biohydrogen transfer for oxidation of sec-alcohols. Org. Lett. 10, 2155–2158. Lin-Cereghino, J., Wong, W.W., Xiong, S., Giang, W., Luong, L.T., Vu, J., Johnson, S.D., Lin-Cereghino, G.P., 2005. Condensed protocol for competent cell preparation and transformation of the methylotrophic yeast Pichia pastoris. Biotechniques 38 (44 46), 48. Macauley-Patrick, S., Fazenda, M.L., McNeil, B., Harvey, L.M., 2005. Heterologous protein production using the Pichia pastoris expression system. Yeast 22, 249–270. MacLeod, W.D., Buigues, N.M., 1964. Sesquiterpenes. I. Nootkatone, A New Grapefruit Flavor Constituent. J. Food Sci. 29, 565–568. Mattanovich, D., Graf, A., Stadlmann, J., Dragosits, M., Redl, A., Maurer, M., Kleinheinz, M., Sauer, M., Altmann, F., Gasser, B., 2009. Genome, secretome and glucose transport highlight unique features of the protein production host Pichia pastoris. Microb. Cell Fact. 8, 29. Maury, J., Asadollahi, M.A., Møller, K., Clark, A., Nielsen, J., 2005. Microbial isoprenoid production: an example of green chemistry through metabolic engineering. Adv. Biochem. Eng. Biotechnol. 100, 19–51. Muller, B., Dean, C., Schmidt, C., Kuhn, J.-C., 1998. Process for the Preparation of Nootkatone. US Patent 5847226. Muntendam, R., Melillo, E., Ryden, A., Kayser, O., 2009. Perspectives and limits of engineering the isoprenoid metabolism in heterologous hosts. Appl. Microbiol. Biotechnol. 84, 1003–1019. Murase, T., Misawa, K., Haramizu, S., Minegishi, Y., Hase, T., 2010. Nootkatone, a characteristic constituent of grapefruit, stimulates energy metabolism and prevents diet-induced obesity by activating AMPK. Am. J. Physiol. Endocrinol. Metab. 299, E266–E275. Näätsaari, L., Mistlberger, B., Ruth, C., Hajek, T., Hartner, F.S., Glieder, A., 2012. Deletion of the Pichia pastoris KU70 homologue facilitates platform strain generation for gene expression and synthetic biology. PLoS One 7, e39720. Nevoigt, E., 2008. Progress in metabolic engineering of Saccharomyces cerevisiae. Microbiol. Mol. Biol. Rev. 72, 379–412. Newman, J.D., Marshall, J., Chang, M., Nowroozi, F., Paradise, E., Pitera, D., Newman, K.L., Keasling, J.D., 2006. High-level production of amorpha-4,11-diene in a twophase partitioning bioreactor of metabolically engineered Escherichia coli. Biotechnol. Bioeng. 95, 684–691. Okuda, M., Sonohara, H., Takigawa, H., Tajima, K., Ito, S., 1994. Nootkatone Manufacture with Rhodococcus from Valencene. JP Patent 06303967. Omura, T., Sato, R., 1964. The carbon monoxide-binding pigment of liver microsomes. I. Evidence for its hemoprotein nature. J. Biol. Chem. 239, 2370–2378. Ortuno, A., Garcia-Puig, D., Fuster, M.D., Perez, M.L., Sabater, F., Porras, I., GarciaLidon, A., Del Rio, J.A., 1995. Flavanone and nootkatone levels in different varieties of grapefruit and pummelo. J. Agric. Food Chem. 43, 1–5. Paradise, E.M., Kirby, J., Chan, R., Keasling, J.D., 2008. Redirection of flux through the FPP branch-point in Saccharomyces cerevisiae by down-regulating squalene synthase. Biotechnol. Bioeng. 100, 371–378. Polakowski, T., Stahl, U., Lang, C., 1998. Overexpression of a cytosolic hydroxymethylglutaryl-CoA reductase leads to squalene accumulation in yeast. Appl. Microbiol. Biotechnol. 49, 66–71. Ramón, A., Marín, M., 2011. Advances in the production of membrane proteins in Pichia pastoris. Biotechnol. J 6, 700–706.

Rico, J., Pardo, E., Orejas, M., 2010. Enhanced production of a plant monoterpene by overexpression of the 3-hydroxy-3-methylglutaryl coenzyme A reductase catalytic domain in Saccharomyces cerevisiae. Appl. Environ. Microbiol. 76, 6449–6454. Ro, D.K., Paradise, E.M., Ouellet, M., Fisher, K.J., Newman, K.L., Ndungu, J.M., Ho, K.A., Eachus, R.A., Ham, T.S., Kirby, J., Chang, M.C., Withers, S.T., Shiba, Y., Sarpong, R., Keasling, J.D., 2006. Production of the antimalarial drug precursor artemisinic acid in engineered yeast. Nature 440, 940–943. Sakamaki, H., Itoh, K., Taniai, T., Kitanaka, S., Takagi, Y., Chai, W., Horiuchi, C.A., 2005. Biotransformation of valencene by cultured cells of Gynostemma pentaphyllum. J. Mol. Catal. B: Enzym. 32, 103–106. Salvador, J.A.R., Clark, J.H., 2002. The allylic oxidation of unsaturated steroids by tert-butyl hydroperoxide using surface functionalised silica supported metal catalysts. Green Chem. 4, 352–356. Scalcinati, G., Knuf, C., Partow, S., Chen, Y., Maury, J., Schalk, M., Daviet, L., Nielsen, J., Siewers, V., 2012. Dynamic control of gene expression in Saccharomyces cerevisiae engineered for the production of plant sesquitepene α-santalene in a fed-batch mode. Metab. Eng. 14, 91–103. Schenkman, J.B., Jansson, I., 2006. Spectral analyses of cytochromes P450. Methods Mol. Biol. 320, 11–18. Seo, E.J., Lee, D.U., Kwak, J.H., Lee, S.M., Kim, Y.S., Jung, Y.S., 2011. Antiplatelet effects of Cyperus rotundus and its component ( þ)-nootkatone. J. Ethnopharmacol. 135, 48–54. Siddiqui, M.S., Thodey, K., Trenchard, I., Smolke, C.D., 2012. Advancing secondary metabolite biosynthesis in yeast with synthetic biology tools. FEMS Yeast Res. 12, 144–170. Sowden, R.J., Yasmin, S., Rees, N.H., Bell, S.G., Wong, L.L., 2005. Biotransformation of the sesquiterpene (þ )-valencene by cytochrome P450cam and P450BM-3. Org. Biomol. Chem. 3, 57–64. Takahashi, S., Yeo, Y., Greenhagen, B.T., McMullin, T., Song, L., Maurina-Brunker, J., Rosson, R., Noel, J.P., Chappell, J., 2007a. Metabolic engineering of sesquiterpene metabolism in yeast. Biotechnol. Bioeng. 97, 170–181. Takahashi, S., Yeo, Y.S., Zhao, Y., O'Maille, P.E., Greenhagen, B.T., Noel, J.P., Coates, R. M., Chappell, J., 2007b. Functional characterization of premnaspirodiene oxygenase, a cytochrome P450 catalyzing regio- and stereo-specific hydroxylations of diverse sesquiterpene substrates. J. Biol. Chem. 282, 31744–31754. Tokuhiro, K., Muramatsu, M., Ohto, C., Kawaguchi, T., Obata, S., Muramoto, N., Hirai, M., Takahashi, H., Kondo, A., Sakuradani, E., Shimizu, S., 2009. Overproduction of geranylgeraniol by metabolically engineered Saccharomyces cerevisiae. Appl. Environ. Microbiol. 75, 5536–5543. Trant, J.M., 1996. Functional expression of recombinant spiny dogfish shark (Squalus acanthias) cytochrome P450c17 (17 alpha-hydroxylase/C17,20-lyase) in yeast (Pichia pastoris). Arch. Biochem. Biophys. 326, 8–14. Urlacher, V.B., Girhard, M., 2012. Cytochrome P450 monooxygenases: an update on perspectives for synthetic application. Trends Biotechnol. 30, 26–36. Urlacher, V.B., Schmid, R.D., 2006. Recent advances in oxygenase-catalyzed biotransformations. Curr. Opin. Chem. Biol. 10, 156–161. Wang, J., Liu, Y., Cai, Y., Zhang, F., Xia, G., Xiang, F., 2010. Cloning and functional analysis of geraniol 10-hydroxylase, a cytochrome P450 from Swertia mussotii Franch. Biosci. Biotechnol. Biochem. 74, 1583–1590. Weis, R., Luiten, R., Skranc, W., Schwab, H., Wubbolts, M., Glieder, A., 2004. Reliable high-throughput screening with Pichia pastoris by limiting yeast cell death phenomena. FEMS Yeast Res. 5, 179–189. Wilson, C.W., Shaw, P.E., 1978. Synthesis of nootkatone from valencene. J. Agric. Food Chem. 26, 1430–1432. Wriessnegger, T., Gübitz, G., Leitner, E., Ingolic, E., Cregg, J., de la Cruz, B.J., Daum, G., 2007. Lipid composition of peroxisomes from the yeast Pichia pastoris grown on different carbon sources. Biochim. Biophys. Acta 1771, 455–461. Wriessnegger, T., Pichler, H., 2013. Yeast metabolic engineering—targeting sterol metabolism and terpenoid formation. Prog. Lipid Res. 52, 277–293. Wright, R., Basson, M., D'Ari, L., Rine, J., 1988. Increased amounts of HMG-CoA reductase induce “karmellae”: a proliferation of stacked membrane pairs surrounding the yeast nucleus. J. Cell Biol. 107, 101–114. Zelena, K., Krings, U., Berger, R.G., 2012. Functional expression of a valencene dioxygenase from Pleurotus sapidus in E. coli. Bioresour. Technol. 108, 231–239. Zhu, B.C., Henderson, G., Chen, F., Maistrello, L., Laine, R.A., 2001. Nootkatone is a repellent for Formosan subterranean termite (Coptotermes formosanus). J. Chem. Ecol. 27, 523–531. Zorn, H., Fraatz, M.A., Berger, R.G., Riemer, S.J.L., Krings, U., Marx, S., 2009. Enzymatische synthese von nootkaton. Evol. Psychol., 08171148.


57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111

Metabolic engineering of Pichia pastoris to produce ricinoleic acid, a hydroxy fatty acid of industrial importance.

Model based engineering of Pichia pastoris central metabolism enhances recombinant protein production.

Autolysis of Pichia pastoris induced by cold.

Production and analysis of perdeuterated lipids from Pichia pastoris cells.

Combining Protein and Strain Engineering for the Production of Glyco-Engineered Horseradish Peroxidase C1A in Pichia pastoris.

Production of Chimeric Acidic α-Amylase by the Recombinant Pichia pastoris and Its Applications.

Cloning and upscale production of monoamine oxidase N (MAO-N D5) by Pichia pastoris.

Enhanced production of leech hyaluronidase by optimizing secretion and cultivation in Pichia pastoris.

Upscale of recombinant α-L-rhamnosidase production by Pichia pastoris Mut(S) strain.

Overexpression of alcohol oxidase in Pichia pastoris.

GH fusion protein.

Engineering the Pichia pastoris N-Glycosylation Pathway Using the GlycoSwitch Technology.

Tannase sequence from a xerophilic Aspergillus niger Strain and production of the enzyme in Pichia pastoris.

Recent advances in the production of recombinant subunit vaccines in Pichia pastoris.

Expression of HpaI in Pichia pastoris and optimization of conditions for the heparinase I production.

Enhancing ergosterol production in Pichia pastoris GS115 by overexpressing squalene synthase gene from Glycyrrhiza uralensis.

Effect of codon optimisation on the production of recombinant fish growth hormone in Pichia pastoris.

Comparative genomics and transcriptomics of Pichia pastoris.

The trade-off of availability and growth inhibition through copper for the production of copper-dependent enzymes by Pichia pastoris.

Production and immunogenicity of VP2 protein of porcine parvovirus expressed in Pichia pastoris.

Comprehensive reconstruction and evaluation of Pichia pastoris genome-scale metabolic model that accounts for 1243 ORFs.

Engineering Pichia pastoris for improved NADH regeneration: A novel chassis strain for whole-cell catalysis.

Recombinant Bordetella pertussis pertactin (P69) from the yeast Pichia pastoris: high-level production and immunological properties.

Gene expression in yeast: Pichia pastoris.