A Phenylpyruvic Acid Reductase Is Required for Biosynthesis of Tropane Alkaloids

Article abstract of DOI:10.1021/acs.orglett.8b03236

Solanaceous medicinal plants produce tropane alkaloids (TAs). We discovered a novel gene from Atropa belladonna, AbPPAR, which encodes a phenylpyruvic acid reductase required for TA biosynthesis. AbPPAR was specifically expressed in root pericycles and endodermis. AbPPAR was shown to catalyze reduction of phenylpyruvic acid to phenyllactic acid, a precursor of TAs. Suppression of AbPPAR disrupted TA biosynthesis through reduction of phenyllactic acid levels. In summary, we identified a novel enzyme involved in TA biosynthesis.

Full text of DOI:10.1021/acs.orglett.8b03236

Letter
pubs.acs.org/OrgLett
Cite This: Org. Lett. XXXX, XXX, XXX−XXX
A Phenylpyruvic Acid Reductase Is Required for Biosynthesis of
Tropane Alkaloids
Fei Qiu,† Chunxian Yang,† Lina Yuan,† Dan Xiang,† Xiaozhong Lan,‡ Min Chen,§ and Zhihua Liao*,†
†Key Laboratory of Eco-environments in Three Gorges Reservoir Region (Ministry of Education), School of Life Sciences,
Southwest University, Chongqing 400715, China
‡TAAHC-SWU Medicinal Plant Joint R&D Centre, Xizang Agricultural and Husbandry College, Nyingchi of Tibet 860000, China
§College of Pharmaceutical Sciences, Key Laboratory of Luminescent and Real-Time Analytical Chemistry (Ministry of Education),
Southwest University, Chongqing 400715, China
S
* Supporting Information
ABSTRACT: Solanaceous medicinal plants produce tropane
alkaloids (TAs). We discovered a novel gene from Atropa
belladonna, AbPPAR, which encodes a phenylpyruvic acid
reductase required for TA biosynthesis. AbPPAR was
specifically expressed in root pericycles and endodermis.
AbPPAR was shown to catalyze reduction of phenylpyruvic
acid to phenyllactic acid, a precursor of TAs. Suppression of
AbPPAR disrupted TA biosynthesis through reduction of
phenyllactic acid levels. In summary, we identified a novel
enzyme involved in TA biosynthesis.
Scheme 1. Proposed Biosynthetic Pathway of Tropane
Alkaloids in Solanaceae
he tropane alkaloids (TAs) of Solanaceae, such as
hyoscyamine (1), anisodamine (2), and scopolamine
a
T
(3), are widely used as anticholinergic agents to treat motion
sickness, postoperative nausea, arrhythmia, and Parkinson’s
diseases.1,2 These TAs are specifically produced in genera of
Solanaceae including Atropa, Datura, Hyoscyamus, and etc.3−5
The medicinal plants of Solanaceae are the only commercial
source for producing these valuable compounds, but the
alkaloid contents in planta are very low, thus prompting plant
scientists to find new approaches to produce TAs. Metabolic
engineering and synthetic biology offer promising approaches
to enhance TA production. On a more fundamental level,
however, elucidation of the entire TA biosynthetic pathway is
essential.
Although TA biogenic routes are clear (Scheme 1), TA
biosynthesis is not completely understood. Fortunately, several
TA biosynthesis enzymes have been identified from different
TA-producing plant species of Solanaceae (Scheme 1). Such
enzymes include putrescine N-methyltransferase (PMT), N-
methylputrescine oxidase (MPO), tropinone reductase I
(TRI), littorine mutase (LM, CYP80F1), aromatic amino
acid aminotransferase (AT4), and hyoscyamine 6β-hydroxylase
(H6H). PMT catalyzes N-methylation of putrescine to
produce N-methylputrescine, which is subsequently converted
to 4-methylaminobutanal by MPO. After unknown enzymatic
reactions, tropinone is synthesized. TRI converts tropinone to
tropine, providing the tropane moiety of TAs. Further, tropine
is condensed with a phenyllactic acid moiety to produce
littorine (4). AT4 catalyzes the transamination of phenyl-
alanine to produce phenylpyruvic acid (5) that is then reduced
a
AR, arginase. ODC, ornithine decarboxylase. PMT, putrescine N-
methyltransferase. MPO, N-methylputrescine oxidase. TRI, tropine-
forming reductase. CYP80F1, littorine mutase. H6H, hyoscyamine
6β-hydroxylase. AT4, aromatic amino acid aminotransferase. PPAR,
phenylpyruvic acid reductase.
to phenyllactic acid (6). Littorine mutase CYP80F1 converts 4
into hyoscyamine aldehyde, which is the direct precursor of 1.
Received: October 11, 2018
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.8b03236
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Finally, H6H catalyzes 6β-hydroxylation of 1 to 2, as well as
the subsequent epoxidation of 2 to 3.4,6−11
thesis genes were specifically or highly expressed in secondary
roots, and AbPPAR was exclusively expressed in roots, with
much higher levels in the secondary roots (Figure 1A).
To date, the full TA biosynthetic pathway has not been
completely resolved. Uncharacterized enzymes involved in TA
biosynthesis may catalyze the formation of tropinone, the
reduction of 5 to 6, the production of 4, and the conversion of
hyoscyamine aldehyde into 1. Isotope-labeling experiments
clearly demonstrated 6 was incorporated at high levels into
TAs including 4, 1, and 3, confirming that 6 is a precursor of
TAs.12 However, the enzyme catalyzing the reduction of 5 has
not been found in the plant kingdom. In this study, we
identified a novel root-expressed gene encoding a phenyl-
pyruvic acid reductase from Atropa belladonna (AbPPAR),
which is required for TA biosynthesis.
It is well known that hydroxyphenylpyruvate reductase
(HPPR) catalyzes the reduction of hydroxyphenylpyruvate to
its respective lactate.13 However, HPPRs have not been shown
to reduce 5 to form 6 in plants. To date, a phenylpyruvic acid
reductase has not been found in plants. Theoretically, the
reduction of 5 to 6 is similar to the reaction catalyzed by
HPPR. In other words, phenylpyruvic acid reductase might be
a homologue of HPPRs. Biochemically, HPPR belongs to the
formate/glycerate dehydrogenase like family of proteins that
contain a hydroxyl acid dehydrogenase (HAD) domain
(PF00389). We used PF00389 in a Hidden Markov Model
(HMM) program to search sequenced transcriptome data of A.
belladonna and identified 38 sequences belonging to 16
unigenes. TAs are specifically synthesized in secondary roots of
A. belladonna, due to high or specific expression of TA
biosynthesis genes in roots. Therefore, we predicted that
unidentified TA biosynthesis genes might also be specifically or
highly expressed in secondary roots. To this end, the
expression relationship between previously identified TA
biosynthesis genes and the 16 unigenes were analyzed by a
Multiple Experiment Viewer. Solely, one of the 16 unigenes,
aba_locus_14073_iso_2_len_1166_ver_2, was highly ex-
pressed in secondary roots and not expressed in stems and
leaves (Supporting Information (SI) Figure S1), suggesting
that it might be the candidate phenylpyruvic acid reductase
gene (AbPPAR). To explore this possibility, full-length
AbPPAR cDNA (MK121981) was isolated and used for
functional identification (Figure S2).
A phylogenetic analysis indicated that plant and microbial
HADs were divided into different clades (Figure S3). Plant
HPPR sequences were highly conserved (81.5% identity). In
the clade of plant HADs, AbPPAR was clearly separated from
the group of HPPRs, suggesting functional divergence between
AbPPAR and plant HPPRs. Sequence comparison showed that
AbPPAR had low similarity (<40.5% identity) to plant HPPRs
(Figure S4). To date, only two phenylpyruvate reductases have
been characterized from microbes, including WfPPR and
LaPPR.14,15 Although AbPPAR showed very low similarity
(<24% identity) to these enzymes, it contained the conserved
substrate-binding site G97 and catalytic sites composed of
R253-E282-H300 (Figure S4), as found in plant HPPRs and
microbial PPRs.
Figure 1. Tissue profiles of TA biosynthesis genes in Atropa
belladonna. (A) Digital gene expression pattern of TA biosynthesis
genes. (B) Expression levels of AbPPAR in secondary root (SR),
primary root (PR), stem (S), and leaf (L). Different letters on
columns indicated significant difference at the level of P < 0.05 as
determined by Duncan’s test. (C) AbPPAR was localized in root
pericycles and endodermis. PC, pericycle. EN, endodermis. CX,
cortex. EP, epidermis.
Quantitative PCR (qPCR) was used to confirm the root-
specific expression of AbPPAR (Figure 1B). The expression
level of AbPPAR was 18.17-fold higher in the secondary roots
than that in primary roots. Coordinated expression of TA
biosynthesis genes facilitates efficient biosynthesis of related
metabolites in roots.
TA biosynthesis is also organized in a cell-specific manner in
roots. PMT and H6H transcripts are specifically localized in
root pericycles.18,19 In contrast, TRI is highly expressed in the
endodermis and outer cortex.20 Such differences in cellular
compartmentation suggest transport of tropine or other
alkaloid intermediates across different cell layers.21 In this
study, the AbPPAR promoter (MK121982) was cloned (Figure
S5) and used to drive the expression of the reporter gene (β-
glucuronidase, GUS) in root cultures of A. belladonna. The
GUS-stained tissues were observed in all cells of root pericycles
and some cells of endodermis, but were not found in the cortex
and epidermis (Figure 1C). These results suggest that 6 is
synthesized in pericycles and endodermis. It is well-known that
amino acids such as phenylalanine (which is converted into 5
by AT4) are transported through vascular tissues and then
unloaded in pericycles.21 Localization of AbPPAR in pericycles
may offer access to readily available precursors. Moreover, the
different tissue localization of AbPPAR (pericycles and
endodermis) and TRI (endodermis and cortex) also supports
the hypothesis that tropine and alkaloid intermediates are
transported between cell types. Importantly, since AbPPAR
and TRI are also expressed in the endodermis together, it is
reasonably deduced that the endodermis might be one of the
tissues to synthesize the TA intermediates downstream of
tropine and 6.
Genes involved in the same biosynthetic pathway often
exhibit highly similar tissue profiles. Artemisinin biosynthesis
genes are exclusively expressed in glandular secretory
trichomes of Artemisia annua, in which artemisinin is
synthesized.16 Anthocyanin biosynthesis genes in purple-
fleshed sweet potato are highly expressed in storage roots.17
The digital gene expression pattern showed that TA biosyn-
We next sought to determine the enzymatic activity of
AbPPAR. Recombinantly expressed, N-terminal His-tagged
AbPPAR was purified using Ni2+-chelating resin (Figure 2A).
Substrate-feeding assays indicated that AbPPAR was able to
reduce 5 to 6 (Figure 2B) and also catalyze the reduction of p-
hydroxyphenylpyruvic acid (7) to p-hydroxyphenyllactic acid
B
DOI: 10.1021/acs.orglett.8b03236
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Figure 2. Purification of AbPPAR and enzymatic assays. (A)
Purification of His-tagged AbPPAR using Ni2+-chelating resin. 1,
protein maker. 2, AbPPAR. (B) HPLC traces of phenylpyruvic acid
(PPA) and phenyllactic acid (PLA). (C) HPLC traces of p-
hydroxyphenylpyruvic acid (p-HPPA) and p-hydroxyphenyllactic
acid (p-HPLA). The purple traces indicated authentic samples. The
red traces showed the detected metabolites in the AbPPAR-catalyzed
reactions. The blue lines represented the detected metabolites in a
negative control using boiled AbPPAR. (D) MS of PLA. (E) MS of p-
HPLA.
Figure 3. Establishment of Atropa belladonna root cultures and
molecular detection. (A) Induction of transgenic roots from leaf discs
(1), a monoclony transgenic root cultured on solid MS medium (2),
and a single line of transgenic root cultured in liquid MS medium (3).
(B) Detection of the target genes, including rolB, rolC, NPTII, and
35S::AbPPAR RNAi in root cultures. (C) The expression levels of
AbPPAR in control root cultures and AbPPAR-RNAi root cultures. C,
control root lines. Ri-1 to -5 represent independently transformed
root lines, in which AbPPAR was suppressed by RNAi. The bars on
columns represented means SD (n ≥ 3). *** indicated significant
difference at the level of P < 0.001 as determined by t-test.
(8) (Figure 2C). These products were identified by LC-MS
(Figure 2D, 2E). These products were also identified by MS/
1
MS and H NMR (Figures S6−S8). The maximum activity of
cultures (Figure 3C). The expression levels of the five other
homologues of AbPPAR were also detected, and their
expression were not reduced in AbPPAR-RNAi root cultures
(Figure S10). The PCR and qPCR results confirm that
AbPPAR was specifically and substantially suppressed in
transgenic root cultures.
AbPPAR for 5 was found at pH 8.0 and 30 °C (Figure S9).
Enzymatic kinetic parameters were detected under optimum
conditions (Table 1). The Km values of AbPPAR for 5 and 7
were 2.77
0.39 mM and 2.25
0.34 mM (Figure S9),
respectively, suggesting that the affinity of AbPPAR for the two
substrates was essentially similar. The Kcat/Km value of
AbPPAR for 5 (108.30 4.33 s−1 M−1) was 8.41-fold higher
than that for 7 (12.88 0.44 s−1 M−1), demonstrating that the
catalytic efficiency of AbPPAR was much higher for 5 than for
7.
Finally, the production of 6, 1, 2, and 3 was analyzed in root
cultures of A. belladonna (Figure 4). The accumulation of each
compound was dramatically reduced in AbPPAR-RNAi root
cultures, compared to the control. Control root cultures
produced 6 at a level of 18.56 μg/g dry weight (DW), while
AbPPAR-RNAi root cultures produced 6 at levels between 0.09
and 1.19 μg/g DW (Figure 4A). We attribute this decreased
production of 6 to suppression of AbPPAR. Furthermore, due
to a reduced supply of 6, the contents of 1, 2, and 3 were also
markedly decreased in AbPPAR-RNAi root cultures. 1 content
was 4.05 mg/g DW in control root cultures and varied from
0.02 to 0.37 mg/g DW in AbPPAR-RNAi root cultures (Figure
4B). The production of 2 and 3 was respectively 2.18 mg/g
DW and 1.33 mg/g DW in control root cultures. In AbPPAR-
RNAi root cultures, the levels of 2 and 3 were 0.05 to 0.47 mg/
g DW (Figure 4C) and 0.06 to 0.54 mg/g DW (Figure 4D),
respectively. These results indicate that the biosynthesis of 6
was perturbed upon AbPPAR suppression, which consequently
decreased the production of tropane alkaloids. The content of
5 was also detected. Its production was unaltered in AbPPAR-
RNAi root cultures (Figure S11). Besides AT4, the five other
ATs were also expressed in secondary roots of A. belladonna,
resulting in the 5 production not only in root pericycles and
endodermis but also in other cells.10 Suppression of AbPPAR,
Suppression of target genes using RNAi technology is an
efficient method to study their roles in the biosynthesis of
target metabolites. To investigate the function of AbPPAR in
TA biosynthesis, AbPPAR-RNAi root cultures of A. belladonna
were established. The root cultures exhibited fast growth on
phytohormone-free medium, many lateral branches, and
plagiotropical growth (Figure 3A), consistent with typical
phenotypes of hairy roots.22 Gene fragments, corresponding to
two rooting genes (rolB and rolC) and a marker gene (NPTII),
were amplified from engineered Agrobacterium tumefaciens
C58C1 (positive control), control root cultures (without
AbPPAR RNAi), and AbPPAR-RNAi root cultures. The gene
fragments of 35S::AbPPAR-RNAi were amplified from
AbPPAR-RNAi root cultures and the positive control
(pBin19 harboring the RNAi fragment of AbPPAR), but
were not detected in control root cultures (Figure 3B).
Further, the expression levels of AbPPAR were detected in the
control and AbPPAR-RNAi root cultures. The expression levels
of AbPPAR were markedly decreased in AbPPAR-RNAi root
Table 1. Enzymatic Kinetics Parameters of AbPPAR
substrates
product
Km (mM)
Vmax (nKat mg−1
)
Kcat. (s−1
)
K
cat./Km (s‑1 M−1
)
PPA
p-HPPA
PLA
p-HPLA
2.77 0.39
2.25 0.34
7.16 0.29
0.71 0.04
0.300 0.012
0.029 0.001
108.30 4.33
12.88 0.44
C
DOI: 10.1021/acs.orglett.8b03236
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
(5) Zhao, K.; Zeng, J.; Zhao, T.; Zhang, H.; Qiu, F.; Yang, C.; Zeng,
L.; Liu, X.; Chen, M.; Lan, X.; Liao, Z. Front. Plant Sci. 2017, 8,
DOI: 10.3389/fpls.2017.01745.
(6) Geng, C.; Zhao, T.; Yang, C.; Zhang, Q.; Bai, F.; Zeng, J.; Zhang,
F.; Liu, X.; Lan, X.; Chen, M.; Liao, Z. Plant Physiol. Biochem. 2018,
127, 47−54.
(7) Naconsie, M.; Kato, K.; Shoji, T.; Hashimoto, T. Plant Cell
Physiol. 2014, 55, 436−444.
̈
(8) Drager, B.; Schaal, A. Phytochemistry 1994, 35, 1441−1447.
(9) Li, R.; Reed, D. W.; Liu, E.; Nowak, J.; Pelcher, L. E.; Page, J. E.;
Covello, P. S. Chem. Biol. 2006, 13, 513−520.
(10) Bedewitz, M. A.; Gongora-Castillo, E.; Uebler, J. B.; Gonzales-
Vigil, E.; Wiegert-Rininger, K. E.; Childs, K. L.; Hamilton, J. P.;
Vaillancourt, B.; Yeo, Y.-S.; Chappell, J.; DellaPenna, D.; Jones, A. D.;
Buell, C. R.; Barry, C. S. Plant Cell 2014, 26, 3745−3762.
(11) Li, J.; Van Belkum, M. J.; Vederas, J. C. Bioorg. Med. Chem.
2012, 20, 4356−4363.
(12) Robins, R. J.; Woolley, J. G.; Ansarin, M.; Eagles, J.;
Goodfellow, B. J. Planta 1994, 194, 86−94.
(13) Janiak, V.; Petersen, M.; Zentgraf, M.; Klebe, G.; Heine, A. Acta
Crystallogr., Sect. D: Biol. Crystallogr. 2010, 66, 593−603.
(14) Fujii, T.; Shimizu, M.; Doi, Y.; Fujita, T.; Ito, T.; Miura, D.;
Wariishi, H.; Takaya, N. Biochim. Biophys. Acta, Proteins Proteomics
2011, 1814, 1669−1676.
Figure 4. Metabolite detection in root cultures of Atropa belladonna.
(A) Phenyllactic acid. (B) Hyoscyamine. (C) Anisodamine. (D)
Scopolamine. DW, dry weight. The bars on columns represented
means SD (n ≥ 3). *, **, and *** indicated significant difference at
the levels of P < 0.05, P < 0.01, and P < 0.001, respectively, as
determined by t-test.
(15) Xu, G. C.; Zhang, L. L.; Ni, Y. J. Biotechnol. 2016, 222, 29−37.
(16) Polichuk, D. R.; Zhang, Y.; Reed, D. W.; Schmidt, J. F.;
Covello, P. S. Phytochemistry 2010, 71, 1264−1269.
(17) Mano, H.; Ogasawara, F.; Sato, K.; Higo, H.; Minobe, Y. Plant
Physiol. 2007, 143, 1252−1268.
specifically expressed in root pericycles and endodermis, was
insufficient to increase the accumulation of 5 in roots.
In summary, we identified a novel root-expressed phenyl-
pyruvic acid reductase localized in root pericycles and
endodermis that was involved in TA biosynthesis. This work
not only mapped a missing enzyme in the TA biosynthetic
pathway but also provided new insight into TA biosynthesis.
(18) Suzuki, K. I.; Yamada, Y.; Hashimoto, T. Plant Cell Physiol.
1999, 40, 289−297.
(19) Suzuki, K.; Yun, D. J.; Chen, X. Y.; Yamada, Y.; Hashimoto, T.
Plant Mol. Biol. 1999, 40, 141−152.
(20) Nakajima, K.; Hashimoto, T. Plant Cell Physiol. 1999, 40,
1099−1107.
ASSOCIATED CONTENT
* Supporting Information
■
(21) De Luca, V.; St Pierre, B. Trends Plant Sci. 2000, 5, 168−173.
(22) Qin, B.; Ma, L.; Wang, Y.; Chen, M.; Lan, X.; Wu, N.; Liao, Z.
Plant Cell, Tissue Organ Cult. 2014, 117, 483−490.
S
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.or-
glett.8b03236.
Experimental procedures and additional results (PDF)
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: zhliao@swu.edu.cn.
ORCID
Zhihua Liao: 0000-0002-1283-4649
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was financially supported by the NSFC projects
(31770335 and 31370333) and Chongqing Research Program
of Basic Research and Frontier Technology (cstc2017-
jcyjAX0208).
REFERENCES
■
́
́
̈
̃
(1) Hakkinen, S. T.; Moyano, E.; Cusido, R. M.; Palazon, J.; Pinol,
M. T.; Oksman-Caldentey, K. M. J. Exp. Bot. 2005, 56, 2611−2618.
(2) Grynkiewicz, G.; Gadzikowska, M. Pharmacol. Reports 2008, 60,
439−463.
(3) Jirschitzka, J.; Dolke, F.; D’Auria, J. C. Adv. Bot. Res. 2013, 68,
39−72.
(4) Qiang, W.; Xia, K.; Zhang, Q.; Zeng, J.; Huang, Y.; Yang, C.;
Chen, M.; Liu, X.; Lan, X.; Liao, Z. Phytochemistry 2016, 127, 12−22.
D
DOI: 10.1021/acs.orglett.8b03236
Org. Lett. XXXX, XXX, XXX−XXX