Heterologous Carotenoid-Biosynthetic Enzymes: Functional Complementation and Effects on Carotenoid Profiles in Escherichia coli

Abstract

A limited number of carotenoid pathway genes from microbial sources have been studied for analyzing the pathway complementation in the heterologous host Escherichia coli. In order to systematically investigate the functionality of carotenoid pathway enzymes in E. coli, the pathway genes of carotenogenic microorganisms (Brevibacterium linens, Corynebacterium glutamicum, Rhodobacter sphaeroides, Rhodobacter capsulatus, Rhodopirellula baltica, and Pantoea ananatis) were modified to form synthetic expression modules and then were complemented with Pantoea agglomerans pathway enzymes (CrtE, CrtB, CrtI, CrtY, and CrtZ). The carotenogenic pathway enzymes in the synthetic modules showed unusual activities when complemented with E. coli. For example, the expression of heterologous CrtEs of B. linens, C. glutamicum, and R. baltica influenced P. agglomerans CrtI to convert its substrate phytoene into a rare product—3,4,3′,4′-tetradehydrolycopene—along with lycopene, which was an expected product, indicating that CrtE, the first enzyme in the carotenoid biosynthesis pathway, can influence carotenoid profiles. In addition, CrtIs of R. sphaeroides and R. capsulatus converted phytoene into an unusual lycopene as well as into neurosporene. Thus, this study shows that the functional complementation of pathway enzymes from different sources is a useful methodology for diversifying biosynthesis as nature does.

INTRODUCTION

Carotenoids, which are isoprenoids produced by many microorganisms and plants, belong to a class of pigment chemicals found in nature. These structurally diverse pigments have different biological functions such as coloration, photo protection, light harvesting, and serving as precursors for many hormones (1). Carotenoids are commercially used as food colorants and animal feed supplements, and more recently, they are being used as nutraceuticals and cosmetic and pharmaceutical compounds (2). The increasing industrial importance of carotenoids has led to renewed efforts in developing bioprocesses for large-scale production of a range of carotenoids, including lycopene, β-carotene, and other more structurally diverse carotenoids (3–6).

In addition to in vitro evolution (6, 27), combinatorial biosynthesis with carotenoid-modifying enzymes in a heterologous host has often been used to generate structurally diverse carotenoids (7). This combinatorial approach basically relies on the functional coordination/complementation of pathway enzymes from different sources in a heterologous host (2, 8, 9). Among the carotenogenic microorganisms (2, 8), Pantoea (formerly Enterobacter) is a well-known carotenogenic bacterium that produces zeaxanthin and its derivatives (8). The carotenogenic pathway consists of 6 enzymes: geranylgeranyl diphosphate (GGPP) synthase CrtE, phytoene synthase CrtB, phytoene desaturase CrtI, lycopene cyclase CrtY, β-carotene hydroxylase CrtZ, and the zeaxanthin glucosyltransferase CrtX (Fig. 1). The carotenoid biosynthetic pathways of Pantoea have been successfully used as basic pathways for many carotenoid studies (11–13). The pathway enzymes from different sources have occasionally shown unusual activities or substrate affinities in a heterologous host (14–16). Carotenogenic enzymes, especially carotenoid-modifying enzymes, tend to be promiscuous in their substrate specificity (14) and show unexpected/hidden activities (3, 16) when expressed in heterologous host microorganisms. Therefore, the activities of carotenoid-modifying enzymes (e.g., spheroidene monooxygenase CrtA and CrtZ) or important pathway branch point enzymes (e.g., CrtY) in heterologous hosts have been the focus of study. In contrast, little attention has been paid to carotenoid backbone-generating enzymes such as CrtE, CrtB, and CrtI.

Fig 1.

Carotenoid pathways of Pantoea agglomerans wild-type and reconstructed pathways in E. coli. Carotenogenic enzymes of P. agglomerans are indicated by bold letters. Shown are IDI (IPP isomerase), IspA (FPP synthase), CrtE (GGPP synthase), CrtB (phytoene synthase), CrtI (phytoene desaturase), CrtI (lycopene cyclase), CrtZ (β-carotene hydrolase), and CrtX (zeaxanthin glucosyltransferase). Solid arrows refer to reconstructed carotenoid pathways in E. coli, and dotted arrows refer to wild-type carotenoid pathways in P. agglomerans.

In this study, in order to systematically investigate the activity and functionality of carotenoid-backbone-generating pathway enzymes, as well as important pathway branch point enzymes, the pathway enzymes of carotenogenic microorganisms (Brevibacterium linens, Corynebacterium glutamicum, Rhodobacter sphaeroides, Rhodobacter capsulatus, Rhodopirellula baltica, and Pantoea ananatis) were complemented with the Pantoea agglomerans pathway enzymes in synthetic expression modules in Escherichia coli. In particular, we showed that CrtE, the first upstream enzyme in the carotenoid biosynthesis pathway, can influence carotenoid profiles in E. coli.

MATERIALS AND METHODS

Bacterial strains, plasmids, and growth conditions.

The bacterial strains and plasmids used in this study are listed in Table 1. E. coli was grown in Luria-Bertani (LB) medium (28) at 37°C for gene cloning and in Terrific broth (TB) medium (28) at 30°C for carotenoid production on a rotary shaker at 250 rpm. Chloramphenicol (50 μg/ml), ampicillin (100 μg/ml), and/or kanamycin (30 μg/ml) (Sigma) were added as required. Pantoea agglomerans KCTC 2479 was grown in LB medium at 37°C.

Table 1.

Strains and plasmids used in this study

Strains and plasmids	Relevant properties	Source or reference
Strains
Pantoea agglomerans	C40 carotenoid pathway	KCTC 2479
Pantoea ananatis	C40 carotenoid pathway	DSM 30080
Brevibacterium linens	C40 carotenoid pathway	DSM 20426
Rhodobacter sphaeroides	C40 carotenoid pathway	KCTC 12085
Rhodobacter capsulatus	C40 carotenoid pathway	KCTC 2583
Rhodopirellula baltica	C40 carotenoid pathway	DSM 10527
Corynebacterium glutamicum	C50 carotenoid pathway	KCTC 1445
E. coli SURE	endA1 glnV44 thi-1 gyrA96 relA1 lac recB recJ sbcC umuC::Tn5 uvrC e14- Δ(mcrCB-hsdSMR-mrr)171 F′[proAB+ lacIq lacZΔM15 Tn10]	Stratagene
E. coli XL1-Blue	endA1 gyrA96(Nalr) thi-1 recA1 relA1 lac glnV44 F′[::Tn10 proAB+ lacIq Δ(lacZ)M15] hsdR17(rK− mK+)	Stratagene
Plasmids
pUC19	Cloning vector, lac promoter, Ap	NEB
pUCM	Cloning vector modified from pUC19; constitutive lac promoter, Ap	15
pACYC184	Expression vector that is compatible with pMB1 or ColE1-related plasmids	NEB
pBBR1MCS-2	Cloning vector, Km	17
pUCM-EPAG	Constitutively expressed crtE gene from P. agglomerans	This study
pUCM-BPAG	Constitutively expressed crtB gene from P. agglomerans	This study
pUCM-IPAG	Constitutively expressed crtI gene from P. agglomerans	This study
pUCM-YPAG	Constitutively expressed crtY gene from P. agglomerans	This study
pUCM-ZPAG	Constitutively expressed crtZ gene from P. agglomerans	This study
pACM-EPAG	Constitutively expressed crtE gene from P. agglomerans	This study
pACM-BPAG	Constitutively expressed crtB gene from P. agglomerans	This study
pACM-IPAG	Constitutively expressed crtI genes from P. agglomerans	This study
pACM-EPAG-BPAG	Constitutively expressed crtE and crtB genes from P. agglomerans	This study
pACM-EPAG-IPAG	Constitutively expressed crtE and crtI genes from P. agglomerans	This study
pACM-BPAG-IPAG	Constitutively expressed crtB and crtI genes from P. agglomerans	This study
pACM-EPAG-BPAG-IPAG	Constitutively expressed crtE, crtB, and crtI genes from P. agglomerans to produce lycopene	This study
pACM-EPAG-BPAG-IPAG-YPAG	Constitutively expressed crtE, crtB, crtI, and crtY genes from P. agglomerans to produce β-carotene	This study
pACM-EPAG-BPAG-IPAG-YPAG-ZPAG	Constitutively expressed crtE, crtB, crtI, crtY, and crtZ genes from P. agglomerans to produce zeaxanthin	This study
pUCM-EBL	Constitutively expressed crtE gene from B. linens	15
pUCM-ERB	Constitutively expressed crtE gene from R. baltica	Unpublished
pUCM-EGG	Constitutively expressed crtE gene from C. glutamicum	Unpublished
pUCM-ERS	Constitutively expressed crtE gene from R. sphaeroides	Unpublished
pUCM-ERC	Constitutively expressed crtE gene from R. capsulatus	Unpublished
pUCM-BBL	Constitutively expressed crtB gene from B. linens	15
pUCM-BRB	Constitutively expressed crtB gene from R. baltica	Unpublished
pUCM-B1CG	Constitutively expressed crtB1 gene from C. glutamicum	Unpublished
pUCM-BRS	Constitutively expressed crtB gene from R. sphaeroides	Unpublished
pUCM-BRC	Constitutively expressed crtB gene from R. capsulatus	Unpublished
pUCM-IBL	Constitutively expressed crtI gene from B. linens	15
pUCM-I2CG	Constitutively expressed crtI2 gene from C. glutamicum	Unpublished
pUCM-IRS	Constitutively expressed crtI gene from R. sphaeroides	Unpublished
pUCM-IRC	Constitutively expressed crtI gene from R. capsulatus	Unpublished
pUCM-IRB	Constitutively expressed crtI gene from R. baltica	Unpublished
pUCM-YPAN	Constitutively expressed crtY gene from P. ananatis	18
pUCM-YcYdBL	Constitutively expressed crtYcYd gene from B. linens	15
pUCM-YeYfCG	Constitutively expressed crtYeYf gene from C. glutamicum	Unpublished
pUCM-ZPAN	Constitutively expressed crtZ gene from P. ananatis	4

Recombinant DNA techniques.

Restriction enzymes, polymerase, and DNA ligase were purchased from New England BioLabs, TaKaRa, or Stratagene. Genomic DNA of P. agglomerans was isolated using a genomic DNA (gDNA) extraction kit (GeneAll, South Korea). Plasmid DNA was prepared with the DNA-spin Plasmid DNA Purification kit (Intron, South Korea). PCR was carried out using a DNA Engine Thermal Cycler (Bio-Rad) with Vent DNA polymerase or PfuUltra II Fusion HS DNA polymerase, according to the manufacturer's instructions.

Gene cloning.

P. agglomerans carotenoid pathway genes crtE, crtB, crtI, crtY, and crtZ were amplified from gDNA of P. agglomerans KCTC 2479 by PCR using primers that were designed according to the corresponding gene sequences from P. agglomerans pv. millettiae (AB076662) and Pantoea sp. C1B1Y (AY876938) (see Table S1 in the supplemental material). The PCR products were then cloned into pUCM (15), which has been modified to facilitate the constitutive expression of a cloned gene, resulting in pUCM-X_PAG (where X is a pathway gene) (Table 1).

Functional assembly of carotenoid biosynthetic genes.

To assemble the carotenogenic pathway genes of P. agglomerans, genes together with a modified constitutive lac-promoter and a ribosome binding site were sequentially subcloned from pUCM-X (where X is a pathway gene) into pACYC184 vector as described previously (15). Briefly, a CrtE expression module was subcloned from pUCM-E_PAG into the HindIII site of pACYC184, resulting in pACM-E_PAG; a CrtB expression module was subcloned from pUCM-B_PAG into the SalI site of pACYC184, resulting in pACM-B_PAG; and a CrtI expression module was subcloned from pUCM-I_PAG into the BamHI site of pACYC184, resulting in pACM-I_PAG (Table 1). Next, the CrtI expression module was subcloned from pUCM-I_PAG into pACM-E_PAG to generate the plasmid pACM-E_PAG-I_PAG, which expresses CrtE and CrtI together. Similarly, the CrtE expression module from pUCM-E_PAG and the CrtB expression module from pUCM-B_PAG were subcloned into pACM-B_PAG and pACM-I_PAG, respectively, to generate 2 expression plasmids: pACM-E_PAG-B_PAG, expressing CrtE and CrtB together, and pACM-B_PAG-I_PAG, expressing CrtB and CrtI together. The functionality of the resulting synthetic modules expressing 2 pathway enzymes was confirmed by complementation with a third gene: for example, complementing CrtB on pUCM-B_PAG with a synthetic module of CrtE and CrtI on pACM-E_PAG-I_PAG. Finally, the lycopene expression module was constructed by assembling the CrtB expression module into pACM-E_PAG-I_PAG, resulting in the plasmid pACM-E_PAG-B_PAG-I_PAG. Similarly, to reconstruct the β-carotene biosynthetic pathway, the CrtY expression module was subcloned from pUCM-Y_PAG into the XmnI site of pACM-E_PAG-B_PAG-I_PAG, resulting in pACM-E_PAG-B_PAG-I_PAG-Y_PAG. A zeaxanthin expression module was constructed by assembling a CrtZ expression module into pACM-E_PAG-B_PAG-I_PAG-Y_PAG, resulting in pACM-E_PAG-B_PAG-I_PAG-Y_PAG-Z_PAG (Table 1).

Phylogenetic tree.

The amino acid sequences of carotenoid genes were aligned using the BioEdit program (20). Phylogenetic and molecular evolutionary analyses were conducted using MEGA version 4 (21).

Extraction of carotenoids from recombinant E. coli.

For carotenoid production, recombinant E. coli cells harboring carotenogenic plasmids were cultivated in 100 ml of TB medium supplemented with the appropriate selective antibiotics for 48 h at 30°C and 250 rpm. The cells were pelleted by centrifugation (4°C, 4,000 rpm) and extracted repeatedly with a total volume of 15 ml acetone until all visible pigments were extracted. After centrifugation (4°C, 4,000 rpm), the colored supernatants were pooled and reextracted with an equal volume of hexane after the addition of an equal volume of double-distilled water. The collected color fractions were dried completely using an EZ-2 Plus centrifugal evaporator (Genevac Inc., NY), and resuspended in 0.5 ml of acetone. After centrifugation (13,000 rpm, 20 min), the extracts were filtered (0.45-μm GHP membrane; Pall) to remove the fine particles.

Analysis of carotenoids.

Thin-layer chromatography (TLC) analysis was performed using a 100% hexane solvent system for the initial analysis (4). High-pressure liquid chromatography (HPLC) analysis was carried out with a Zorbax Eclipse XDB-C18 column (4.6 by 150 mm, 5 μm; Agilent Technologies); the elution was performed under isocratic conditions with a solvent system containing 80% acetonitrile, 15% methanol, and 5% isopropanol at a flow rate of 1 ml/min by using an Agilent 1200 system equipped with a photodiode array detector. For structural elucidation, carotenoids were identified using a combination of HPLC retention times, UV-visible (UV-Vis) absorption spectra, and mass fragmentation spectra. Mass fragmentation spectra were monitored in the mass range of m/z 300 to 800 on a liquid chromatography-mass spectrometer (LC/MS) (Agilent 6150) equipped with an atmospheric pressure chemical ionization (APCI) or an electrospray ionization (ESI) ion source (Agilent Technologies, CA). For the reduction of carotenoids with NaBH₄, carotenoids were dried and resuspended in 100% ethanol and then incubated for 1 h in the dark at 30°C after a few crystals of NaBH₄ (22) were added.

In vitro CrtZ activity assay.

In vitro CrtZ activity was assayed by the method of Hundle et al. (23) with some modifications. E. coli cells containing the plasmid pUCM (as the negative control) and pUCM-Z_PAG were grown in LB medium (300 ml) at 30°C for 12 h. The cells were pelleted, washed, resuspended in 300 μl sodium phosphate buffer (100 mM, pH 7.0), and incubated for 15 min at 25°C after the addition of 3 μl of lysozyme (100 mg/ml). The cells were then sonicated on ice (65% power with a cycle of 5 s pulse-on and 10 s pulse-off for 75 s), and the cell debris was removed by centrifugation at 13,000 rpm for 30 min at 4°C. The crude protein extracts were collected, quantified using a Bio-Rad protein assay kit (Bio-Rad), and used directly for in vitro assays. The reactions were started at 30°C for 14 h by adding 300 μl of protein extract (∼5 μg) into a 200-μl mixture of sodium phosphate buffer (100 mM, pH 7.0) containing 3 μl of FeCl₂ (20 mM), 0.3 μl of Tween 80, and 1.2 nmol of β-carotene.

Nucleotide sequence accession numbers.

The GenBank accession numbers of carotenogenic enzymes of P. agglomerans KCTC 2479 are as follows: CrtE, JX871355; CrtB, JX871356; CrtI, JX871357; CrtY, JX871358; CrtZ, JX871359.

RESULTS AND DISCUSSION

Construction of platform modules expressing P. agglomerans carotenoid pathway enzymes.

For the construction of basic carotenoid expression modules for functional complementation, P. agglomerans carotenoid biosynthetic genes were chosen because these pathway genes have been extensively used as a model system in bacteria and yeasts (8). Unidentified carotenoid pathway genes of yellow P. agglomerans KCTC 2479 were obtained by performing PCR on the basis of the corresponding gene sequences of P. agglomerans pv. millettiae and Pantoea sp. C1B1Y (Table 2). In sequence analysis, 5 cloned genes from P. agglomerans showed a high homology to those from P. agglomerans pv. millettiae (97 to 99%), Pantoea sp. C1B1Y (98 to 99%), and P. ananatis LMG20103 (73 to 87%) (see Table S1 in the supplemental material). Each gene encoding a carotenoid pathway enzyme (CrtE_PAG, CrtB_PAG, CrtI_PAG, CrtY_PAG, or CrtZ_PAG) was individually modulated and then sequentially assembled into the pACYC184 vector to generate pACM-E_PAG-B_PAG, pACM-B_PAG-I_PAG, pACM-E_PAG-I_PAG, pACM-E_PAG-B_PAG-I_PAG, pACM-E_PAG-B_PAG-I_PAG-Y_PAG, and pACM-E_PAG-B_PAG-I_PAG-Y_PAG-Z_PAG. To investigate the functionality of the heterologous carotenoid biosynthetic enzymes, we selected 15 heterogeneous lycopene biosynthetic enzymes (CrtE, CrtB, and CrtI) from 5 carotenogenic microorganisms (Brevibacterium linens, Corynebacterium glutamicum, Rhodobacter sphaeroides, Rhodobacter capsulatus, and Rhodopirellula baltica) and 3 carotene cyclases (CrtYs) from 3 carotenogenic microorganisms (Brevibacterium linens, Corynebacterium glutamicum, and P. ananatis) (Table 1).

Table 2.

Primers used for cloning Pantoea agglomerans in this study

Gene	Sequencea	Enzyme site
crtE	F: 5′-GCTCTAGAAGGAGGATTACAAAATGATGGTCTGCGCA-3′	XbaI
	R: 5′-CCGGAATTCTTAACTGACGGCAGC-3′	EcoRI
crtB	F: 5′-GCTCTAGAAGGAGGATTACAAAATGGMGGTKGGMTCG-3′	XbaI
	R: 5′-CCGGAATTCTTAGATGGGGCGCTG-3′	EcoRI
crtI	F: 5′-GC TCTAGAAGGAGGATTACAAAATGAAWMSAACTACR-3′	XbaI
	R: 5′-CCGGAATTCTCAWRYCAGRTCCYC-3′	EcoRI
crtX	F: 5′-GCTCTAGAAGGAGGATTACAAAATGAGCCACTTTGCGGTC-3′	XbaI
	R: 5′-CCGGAATTCTCATACCGCGGCATAGT-3′	EcoRI
crtY	F: 5′-GCTCTAGAAGGAGGATTACAAAATGCCGCGGTATGATCTG-3′	XbaI
	R: 5′-CCGGAATTCTCATTGCATCGCCTGTTG-3′	EcoRI
crtZ	F: 5′-GCTCTAGAAGGAGGATTACAAAATGTTGTGGATTTGGAATGC-3′	XbaI
	R: 5′-CCGGAATTCTTACTTCCCGGGTGGCG-3′	EcoRI

^aUnderlining indicates restriction enzyme sites, and bold indicates mixed nucleotides (M = A and C; R = A and G; W = A and T; S = C and G; Y = C and T). F, forward; R, reverse.

Functional complementation of heterogeneous phytoene desaturases (CrtIs) from 5 carotenogenic microorganisms.

First, we investigated the heterologous complementations of 5 CrtIs from 5 carotenogenic microorganisms (BLI, CrtI of B. linens; CGI, CrtI of C. glutamicum; RSI, CrtI of R. sphaeroides; RCI, CrtI of R. capsulatus; and RBI, CrtI of R. baltica) and the homologous complementation of CrtI from P. agglomerans (PAI) with the P. agglomerans carotenogenic module expressing CrtE_PAG + CrtB_PAG. Figure 2A shows the phylogenetic distances between the 6 CrtIs (PAI, BLI, CGI, RSI, RCI, and RBI). Coexpression of each CrtI with the module expressing CrtE_PAG + CrtB_PAG changed the color densities of the recombinant E. coli cells (Fig. 2B). HPLC and LC/MS analysis detected carotenoid peaks in the extracts of 5 recombinant E. coli cells but not in a complementation of RBI + CrtE_PAG + CrtB_PAG or in a negative control expressing empty vectors (Fig. 2C). Interestingly, RSI and RCI produced neurosporene and lycopene at a similar ratio in this study (Fig. 2D), a result that conflicts with the reported activity of CrtI (RCI) from R. capsulatus, which is known to produce only neurosporene as an end product (29). Unexpectedly, the carotenoids 3,4,3′,4′-tetradehydrolycopene and 3,4-didehydrolycopene were detected in the homologous complementation PAI + CrtE_PAG + CrtB_PAG and the heterologous complementation CGI + CrtE_PAG + CrtB_PAG, respectively. In other studies, the rare, fully saturated 3,4,3′,4′-tetradehydrolycopene was produced when a CrtI generated by directed evolution was involved (18) or wild-type CrtI of P. ananatis was expressed at high levels (11). BLI, CGI, and RBI, which belong to an outgroup of the phylogenetic tree, produced a far smaller amount of carotenoids, suggesting that these heterologous complementations were less efficient in converting phytoene into lycopene. This hypothesis is supported by the fact that the 3 enzymes were able to produce more lycopene with their own pathway enzymes (unpublished data for CGI and RBI; see reference 15 for BLI). There were no distinguishable differences in the expression observed by performing SDS-PAGE (see the supplemental material).

Fig 2.

Functional complementations of 6 CrtIs from carotenogenic microorganisms in E. coli expressing CrtE and CrtB from Pantoea agglomerans. (A) Phylogenetic tree of 6 CrtIs from 6 different carotenogenic microorganisms. The neighbor-joining tree is based on the CrtI protein sequence. Bootstrap values expressed as a percentage of 1,000 replications are shown next to each node. (B) Pellets (upper row) and extracts (lower row) of recombinant E. coli expressing lycopene biosynthetic enzymes (CrtE, CrtB, and CrtI). (C) HPLC analysis of the carotenoid profiles of recombinant E. coli expressing lycopene biosynthetic enzymes. (D) Relative production of carotenoids in recombinant E. coli expressing lycopene biosynthetic enzymes. PAI, CrtI of P. agglomerans; BLI, CrtI of B. linens; CGI, CrtI of C. glutamicum; RSI, CrtI of R. sphaeroides; RCI, CrtI of R. capsulatus; RBI, CrtI of R. baltica. The following carotenoids were identified: 3,4,3′,4′-tetradehydrolycopene (λ_max, 493, 516, and 540); 3,4-didehydrolycopene (λ_max, 460, 491, and 523); lycopene (λ_max, 447, 473, and 503); and neurosporene (λ_max, 419, 442, and 470). ND, not detected.

Functional complementation of heterogeneous phytoene synthases (CrtBs) from 5 carotenogenic microorganisms.

Next, the heterologous complementations of 5 CrtBs (BLB, CrtB of B. linens; CGB, CrtB of C. glutamicum; RSB, CrtB of R. sphaeroides; RCB, CrtB of R. capsulatus; and RBB, CrtB of R. baltica) and the homologous complementation of CrtB (PAB) were investigated in E. coli cells expressing CrtE_PAG + CrtI_PAG. Phylogenetic distances between the 6 CrtBs (PAB, BLB, CGB, RSB, RCB, and RBB) were similar to those between the CrtIs (Fig. 3A). However, when each CrtB was complemented with the module expressing CrtE_PAG + CrtI_PAG, recombinant E. coli cells showed a deeper color (Fig. 3B) than that observed in the case of E. coli expressing CrtI from different sources. Unlike the carotenoid profiles observed in the CrtI complementations, only lycopene was detected in 5 of the CrtB complementations. No carotenoids were detected for the RBB complementation (Fig. 3C). Notably, RSB and RCB produced only lycopene without neurosporene (8, 19), unlike RSI and RCI (compare Fig. 2D and Fig. 3D). Furthermore, complementations of BLB and CGB, which belong to an outgroup of the phylogenetic tree, produced more lycopene than the complementations of BLI and CGI did (compare Fig. 2D and 3D). This suggests that BLB and CGB are better fitted than BLI and CGI to assemble the heterologous carotenogenic machinery in E. coli (4). The highest production of lycopene was observed with RCB, followed by CGB, PAB, RSB, and BLB. Interestingly, even though RCB and RSB come from the same genus (Rhodobacter) and are separated by a short phylogenetic distance (Fig. 3A), they showed a different heterologous complementation pattern in E. coli, as seen in a different pattern of carotenogenic enzymes from P. agglomerans and P. ananatis (24).

Fig 3.

Functional complementations of 6 CrtBs from carotenogenic microorganisms in E. coli expressing CrtE and CrtI from Pantoea agglomerans. (A) Phylogenetic tree of 6 CrtBs from 6 different carotenogenic microorganisms. The neighbor-joining tree is based on the CrtB protein sequence. Bootstrap values expressed as a percentage of 1,000 replications are shown next to each node. (B) Pellets (upper row) and extracts (lower row) of recombinant E. coli expressing lycopene biosynthetic enzymes (CrtE, CrtB, and CrtI). (C) HPLC analysis of the carotenoid profiles of recombinant E. coli expressing lycopene biosynthetic enzymes. (D) Relative production of carotenoids in recombinant E. coli expressing lycopene biosynthetic enzymes. PAB, CrtB of P. agglomerans; BLB, CrtB of B. linens; CGB, CrtB of C. glutamicum; RSB, CrtB of R. sphaeroides; RCB, CrtB of R. capsulatus; RBB, CrtB of R. baltica. The following carotenoid was identified: lycopene (λ_max, 447, 473, and 503). ND, not detected.

Functional complementation of heterogeneous GGPP synthases (CrtEs) from 5 carotenogenic microorganisms.

The heterologous complementations of 5 CrtEs (BLE, CrtE of B. linens; CGE, CrtE of C. glutamicum; RSE, CrtE of R. sphaeroides; RCE, CrtE of R. capsulatus; and RBE, CrtE of R. baltica) and the homologous complementation of CrtE of P. agglomerans (PAE) were investigated with a module expressing CrtB_PAG + CrtI_PAG (Fig. 4A). Unlike the results obtained with CrtI and CrtB complementations, all CrtE complementations, including RBE, turned recombinant E. coli cells a deep reddish color (Fig. 4B). HPLC and LC/MS analysis showed that PAE, BLE, CGE, and RBE produced 3,4,3′,4′-tetradehydrolycopene as well as lycopene (Fig. 4C). Detection of 3,4,3′,4′-tetradehydrolycopene in CrtE complementations is very unusual because so far only CrtI has been reported to be capable of synthesizing 3,4,3′,4′-tetradehydrolycopene or 3,4-didehydrolycopene (2). Therefore, most research has focused on CrtI to understand the desaturation of phytoene (19). It is obvious that, as shown in Fig. 1, CrtE is not directly responsible for the biosynthesis of 3,4,3′,4′-tetradehydrolycopene and lycopene. However, for the first time, to our knowledge, our results clearly show that CrtE induces, through an unknown mechanism, the carotenogenic machinery to synthesize 3,4,3′,4′-tetradehydrolycopene in the heterologous host E. coli. Similar to RSB and RCB, RSE and RCE produced only lycopene without neurosporene accumulation (compare Fig. 3D and 4D). Interestingly, with the exception of RBE, heterologous complementation of CrtE (PAE, BLE, CGE, and RBE) produced more total carotenoid (3,4,3′,4′-tetradehydrolycopene + lycopene) than did homologous complementation of PAE (Fig. 4D). CrtE has been considered one of the rate-limiting enzymes in carotenoid biosynthesis (24). Therefore, the enhancement of carotenoid production can be explained by CrtEs with different affinities for farnesyl pyrophosphate (FPP) having flexibility in the heterologous carotenogenic machinery assembly. How CrtE influenced the activity of CrtI to produce the unusual 3,4,3′,4′-tetradehydrolycopene product as well as lycopene, however, remains to be answered.

Fig 4.

Functional complementations of 6 CrtEs from carotenogenic microorganisms in E. coli expressing CrtB and CrtI from Pantoea agglomerans. (A) Phylogenetic tree of 6 CrtEs from 6 different carotenogenic microorganisms. The neighbor-joining tree is based on the CrtE protein sequence. Bootstrap values expressed as a percentage of 1,000 replications are shown next to each node. (B) Pellets (upper row) and extracts (lower row) of recombinant E. coli expressing lycopene biosynthetic enzymes (CrtE, CrtB, and CrtI). (C) HPLC analysis of the carotenoid profiles of recombinant E. coli expressing lycopene biosynthetic enzymes. (D) Relative production of carotenoids in recombinant E. coli expressing lycopene biosynthetic enzymes. PAE, CrtE of P. agglomerans; BLE, CrtE of B. linens; CGE, CrtE of C. glutamicum; RSE, CrtE of R. sphaeroides; RCE, CrtE of R. capsulatus; RBE, CrtE of R. baltica. The following carotenoids were identified: 3,4,3′,4′-tetradehydrolycopene (λ_max, 493, 516, and 540) and lycopene (λ_max, 447, 473, and 503).

Functional complementation of heterogeneous lycopene cyclases (CrtYs) from 3 carotenogenic microorganisms.

Next, we further extended our complementation study to the β-carotene pathway in E. coli. Because the microbial sources for lycopene cyclase CrtY were relatively limited, 3 lycopene cyclases (PNY, CrtY of P. ananatis; BLY, CrtY_cY_d of B. linens; and PAY, CrtY of P. agglomerans) and 1 carotene cyclase (CGY, CrtY_eY_f of C. glutamicum) were chosen and then complemented with a lycopene module expressing PAE + PAB + PAI in E. coli. Both heterologous BLY and homologous PAY completely converted lycopene into β-carotene, while heterologous PNY did not completely convert lycopene into β-carotene (Fig. 5). As expected, CGY was not functionally complemented with a lycopene module because CGY has a substrate specificity for carotenoid backbones with C₄₅ and C₅₀ rather than C₄₀ lycopene (25). CGY was functionally complemented with homologous upstream pathway enzymes in E. coli (data not shown).

Fig 5.

Functional complementations of 4 CrtYs from carotenogenic microorganisms in E. coli expressing CrtE, CrtB, and CrtI from Pantoea agglomerans. HPLC analysis of the carotenoid profiles of recombinant E. coli expressing a lycopene cyclase with the lycopene biosynthetic enzymes. PAY, CrtY of P. agglomerans; BLY, CrtY of B. linens; CGY, CrtY of C. glutamicum; PNY, CrtY of P. ananatis. The following carotenoids were identified: β-carotene (λ_max, 427, 453, and 481) and lycopene (λ_max, 447, 473, and 503).

Functional complementation of β-carotene hydroxylase CrtZ.

Two CrtZ enzymes (PNZ, CrtZ of P. ananatis; and PAZ, CrtZ of P. agglomerans) were further complemented with a β-carotene module expressing PAE + PAB + PAI + PAY in E. coli. Both heterologous PNZ and homologous PAZ produced zeaxanthin without accumulation of an intermediate β-cryptoxanthin derivative (Fig. 6A). However, unlike PNZ, PAZ complementation produced additional polar compounds (Fig. 6A). The polar compound could be a carotenoid with a keto group(s) in its structure, which has been deduced on the basis of the reduction of the new polar compound with NaBH₄ and the UV-Vis spectra of the reduced compounds (Fig. 6B). Interestingly, the conversion of β-carotene to zeaxanthin without formation of polar compounds was observed only in vitro (Fig. 6C). Therefore, we assume that a specific in vivo environment may account for the unusual activity of PAZ in E. coli. It is worth noting that CrtZ has a relatively broad substrate preference (26) and is bifunctional as a hydroxylase/ketolase (9). Detailed structural analysis of these compounds is under way.

Fig 6.

Functional complementations of 2 CrtZs from carotenogenic microorganisms in E. coli expressing CrtE, CrtB, CrtI, and CrtY from Pantoea agglomerans. (A) HPLC analysis of the carotenoid profiles of recombinant E. coli expressing a β-carotene hydroxylase with the β-carotene biosynthetic enzymes. PAZ, CrtZ of P. agglomerans; PNZ, CrtZ of P. ananatis. The following carotenoids were identified: peak 1, β-carotene (λ_max, 427, 453, and 481); peak 2, zeaxanthin (λ_max, 427, 453, and 481); peak 3, unidentified keto-hydroxy compounds. (B) HPLC analysis of unidentified keto-hydroxy compounds before (a) and after (b) reduction with NaBH₄. UV-Vis spectra of peaks 3 and 4 (c) and peaks 5 and 6 (d). Peaks 3 and 4, oxidized unidentified keto-hydroxy compounds; peaks 5 and 6, reduced unidentified keto-hydroxy compounds. (C) HPLC analysis of the in vitro activity of CrtZ with β-carotene.

In conclusion, microbial carotenogenic pathway enzymes in synthetic expression modules showed unusual activities that differed from those in previous reports when complemented with P. agglomerans pathway enzymes in the heterologous host E. coli. In particular, the expression of heterologous CrtEs from B. linens, C. glutamicum, and R. baltica and the homologous CrtE from P. agglomerans induced, by an unknown mechanism, the P. agglomerans CrtI to convert its substrate phytoene into a rare 3,4,3′,4′-tetradehydrolycopene as well as the expected lycopene. This indicates that single expression of CrtE, the first upstream enzyme in the carotenoid biosynthesis pathway (Fig. 1), can influence carotenoid profiles in E. coli. Additionally, heterologous CrtIs (from R. sphaeroides, R. capsulatus, and C. glutamicum) and homologous CrtI (from P. agglomerans) converted phytoene into unusual carotenoids (for example, lycopene by CrtIs from R. sphaeroides and R. capsulatus) as well as into the expected end product carotenoids. Therefore, the functional complementation of carotenoid pathway enzymes from different sources is a useful methodology for diversifying biosynthesis as nature does (8).