Oxidation of monolignols by members of the berberine bridge enzyme family suggests a role in plant cell wall metabolism

Article abstract of DOI:10.1074/jbc.M115.659631

Plant genomes contain a large number of genes encoding for berberine bridge enzyme (BBE)-like enzymes. Despite the wide-spread occurrence and abundance of this protein family in the plant kingdom, the biochemical function remains largely unexplored. In this study, we have expressed two members of the BBE-like enzyme family from Arabidopsis thaliana in the host organism Komagataella pastoris. The two proteins, termed AtBBE-like 13 and AtBBE-like 15, were purified, and their catalytic properties were determined. In addition, AtBBE-like 15 was crystallized and structurally characterized by x-ray crystallography. Here, we show that the enzymes catalyze the oxidation of aromatic allylic alcohols, such as coumaryl, sinapyl, and coniferyl alcohol, to the corresponding aldehydes and that AtBBE-like 15 adopts the same fold as vanillyl alcohol oxidase as reported previously for berberine bridge enzyme and other FAD-dependent oxidoreductases. Further analysis of the substrate range identified coniferin, the glycosylated storage form of coniferyl alcohol, as a substrate of the enzymes, whereas other glycosylated monolignols were rather poor substrates. A detailed analysis of the motifs present in the active sites of the BBE-like enzymes in A. thaliana suggested that 14 out of 28 members of the family might catalyze similar reactions. Based on these findings, we propose a novel role of BBE-like enzymes in monolignol metabolism that was previously not recognized for this enzyme family.

Full text of DOI:10.1074/jbc.M115.659631

THE JOURNAL OF BIOLOGICAL CHEMISTRY VOL. 290, NO. 30, pp. 18770–18781, July 24, 2015
© 2015 by The American Society for Biochemistry and Molecular Biology, Inc. Published in the U.S.A.
Author’s Choice
Oxidation of Monolignols by Members of the Berberine
Bridge Enzyme Family Suggests a Role in Plant Cell Wall
□
S
Metabolism*
Received for publication, April 27, 2015, and in revised form, May 28, 2015 Published, JBC Papers in Press, June 2, 2015, DOI 10.1074/jbc.M115.659631
Bastian Daniel‡, Tea Pavkov-Keller§¶, Barbara Steiner‡, Andela Dordic§¶, Alexander Gutmannʈ, Bernd Nidetzkyʈ,
Christoph W. Sensen**, Eric van der Graaff‡‡, Silvia Wallner‡, Karl Gruber§, and Peter Macheroux‡1
From the Institutes of ‡Biochemistry, ʈBiotechnology and Biochemical Engineering, and **Molecular Biotechnology, Graz University
of Technology, 8010 Graz, Austria, the §Institute of Molecular Biosciences, University of Graz, 8010 Graz, Austria, the ¶ACIB GmbH,
8010 Graz, Austria, and the ‡‡Section for Crop Sciences, Copenhagen University, 2630 Copenhagen, Denmark
Background: Berberine bridge enzyme-like proteins are a multigene family in plants.
Results: Members of the berberine bridge enzyme-like family were identified as monolignol oxidoreductases.
Conclusion: Berberine bridge enzyme-like enzymes play a role in monolignol metabolism and lignin formation.
Significance: Our results indicate a novel and unexpected role of berberine bridge enzyme-like enzymes in plant biochemistry
and physiology.
bridge enzyme (BBE)2-like proteins (pfam 08031) can be set
apart due to their unusual bicovalent attachment of the FAD
cofactor. The namesake of this protein family is BBE from
Eschscholzia californica (California poppy) that catalyzes the
formation of the so-called “berberine bridge” by oxidation of
the N-methyl group of (S)-reticuline yielding (S)-scoulerine (1).
This step constitutes a branch point in the biosynthesis of
numerous isoquinoline alkaloids (2, 3). In recent years, a large
number of genes encoding BBE-like enzymes have been identi-
fied in plants and bacteria in the course of genome sequencing
efforts. The number of BBE-like genes in individual plant spe-
cies varies considerably from a single gene in the moss
Physcomitrella patens to 64 in the western poplar (Populus
trichocarpa) (4). In the model plant Arabidopsis thaliana, 28
BBE-like genes were identified (termed AtBBE-like 1–28 (4)).
However, the role of these BBE homologs in A. thaliana and
most of the other plants is unknown, as most do not synthesize
alkaloids of the benzylisoquinoline family. Examination of
microarray data (5) revealed the expression of AtBBE-like genes
during certain developmental stages, such as root elongation
and maturation, and proliferation as well as embryonal devel-
opment (Table 1). Similarly, osmotic stress and pathogen attack
were shown to cause up-regulation of AtBBE-likes by up to
400-fold. This up-regulation was also seen in other plants, such
as citrus fruit and poplar (6, 7). The role in pathogen defense
appears to be related to a carbohydrate oxidase activity that was
reported previously for tobacco (nectarin V), sunflower, and
lettuce (8, 9). Similar oxidation reactions occur in the biosyn-
thesis of antibiotics generated in various bacterial species, such
as Streptomyces (10–12). However, the role of BBE homologs in
plant development remains enigmatic.
Plant genomes contain a large number of genes encoding for
berberine bridge enzyme (BBE)-like enzymes. Despite the wide-
spread occurrence and abundance of this protein family in the
plant kingdom, the biochemical function remains largely unex-
plored. In this study, we have expressed two members of the
BBE-like enzyme family from Arabidopsis thaliana in the host
organism Komagataella pastoris. The two proteins, termed
AtBBE-like 13 and AtBBE-like 15, were purified, and their cata-
lytic properties were determined. In addition, AtBBE-like 15
was crystallized and structurally characterized by x-ray crystal-
lography. Here, we show that the enzymes catalyze the oxidation
of aromatic allylic alcohols, such as coumaryl, sinapyl, and
coniferyl alcohol, to the corresponding aldehydes and that
AtBBE-like 15 adopts the same fold as vanillyl alcohol oxidase as
reported previously for berberine bridge enzyme and other
FAD-dependent oxidoreductases. Further analysis of the sub-
strate range identified coniferin, the glycosylated storage form
of coniferyl alcohol, as a substrate of the enzymes, whereas other
glycosylated monolignols were rather poor substrates. A
detailed analysis of the motifs present in the active sites of the
BBE-like enzymes in A. thaliana suggested that 14 out of 28
members of the family might catalyze similar reactions. Based
on these findings, we propose a novel role of BBE-like enzymes
in monolignol metabolism that was previously not recognized
for this enzyme family.
Flavoproteins are a large and diverse protein family employ-
ing the FAD cofactor for catalysis. Among them the berberine
* This work was supported by the “Fonds zur Fo¨rderung der Wissenschaftli-
chen Forschung” through the Ph.D. Program “Molecular Enzymology” (to
K. G., B. N., and P. M.). The authors declare that they have no conflicts of
interest with the contents of this article.
For this study, we have chosen AtBBE-like 15 (At2g34790)
for further studies into the role of the BBE-like enzyme family in
plants because gene disruption resulted in defects in female
□S
This article contains supplemental Fig. S1 and Tables S1 and S2.
The atomic coordinates and structure factors (code 4UD8) have been deposited
in the Protein Data Bank (http://wwpdb.org/).
Author’s Choice—Final version free via Creative Commons CC-BY license.
1 To whom correspondence should be addressed. Tel.: 43-316-873-6450; Fax:
43-316-873-6952; E-mail: peter.macheroux@tugraz.at.
2 The abbreviations used are: BBE, berberine bridge enzyme; THCA, ⌬1-tetra-
hydrocannabinolic acid synthase; VAO, vanillyl alcohol oxidase; GOOX,
glucooligosaccharide oxidase.
18770 JOURNAL OF BIOLOGICAL CHEMISTRY
VOLUME 290•NUMBER 30•JULY 24, 2015

Oxidation of Monolignols by BBE-like Enzymes
TABLE 1
Expression of AtBBE-homologs in planta
Data were retrieved from the Arabidopsis eFP browser.
Most significant expression value/ Second significant expression value/
Subgroup
Name
up-regulation
up-regulation
Localization, MS/MS
Locus
1
1
1
2
2
2
2
2
2
3
3
4
4
4
5
5
5
5
5
5
6
6
6
6
6
7
7
AtBBE-like 18 822, seed linear cotyledon
Unknown
AT4G20820.1
AT5G44410.1
AT5G44440.1
AT1G26380.1
AT1G26390.1
AT1G26400.1
AT1G26410.1
AT1G26420.1
AT4G20800.1
AT1G30720.1
AT1G30730.1
AT1G30710.1
AT1G34575.1
AT2G34810.1
AT1G01980.1
AT1G11770.1
AT1G30700.1
AT1G30740.1
AtBBE-like 27 1117, root
Unknown
AtBBE-like 28 6000, shift dark to light
927, quiescent center root
Extracellular
Endoplasmic reticulum
Unknown
AtBBE-like 3
AtBBE-like 4
AtBBE-like 5
AtBBE-like 6
AtBBE-like 7
2457/438-fold osmotic stress seedling 2318/457-fold, infection P. infestans
2070/78-fold infection, P. syringae
665, cotyledons
391, root regeneration
475, cotyledon greens
Unknown
604/42-fold, osmotic stress
1562/412-fold, osmotic stress
278/36-fold infection, P. syringae
800/163-fold, mock infection
Unknown
Unknown
AtBBE-like 17 7772, micropylar endosperm
AtBBE-like 10 4899, lateral root cap
AtBBE-like 11 4988, root epidermis
Unknown
912/57-fold, mock infection
Unknown
762/10-fold, infection P. syringae
Extracellular
Unknown
AtBBE-like 9
7582, mature pollen
AtBBE-like 14 50, seed
Unknown
AtBBE-like 16 3682, root maturation zone
1537, stigma and ovaries
5509, mature pollen
Unknown
AtBBE-like 1
AtBBE-like 2
AtBBE-like 8
11,626, pollen
Plasma membrane, Golgi
Unknown
5088, pollen
2000, pollen tubes
2557, root maturation zone
1157/246-fold infection, P. syringae
Unknown
AtBBE-like 12 642, endosperm
Unknown
AtBBE-like 20 4000, root elongation zone
AtBBE-like 21 973, root elongation zone
AtBBE-like 13 2144, lateral root initiation
AtBBE-like 15 4927, xylem root
1394/10-fold infection P. syringae
Extracellular, plasma membrane AT4G20830.2
Plasma membrane
Unknown
AT4G20840.1
AT1G30760.1
1594, xylem root maturation zone
439, seed stage 8 curled
Extracellular, plasma membrane AT2G34790.1
AtBBE-like 24 32,075, lateral root cap
AtBBE-like 25 491, procambium root
AtBBE-like 26 604/5-fold osmotic stress root
AtBBE-like 22 1641, lateral root initiation
AtBBE-like 23 824, root
1412, cotyledon
Extracellular
AT5G44380.1
AT5G44390.1
AT5G44400.1
AT4G20860.1
AT5G44360.1
164/5-fold infection P. syringae
250/3-fold infection P. syringae
1427/13-fold infection P. syringae
649, seed cotyledon heart stage
Extracellular
Extracellular
Plasma membrane, cytosol
Unknown
gametophyte development (i.e. unfused polar nuclei and endo- erences or have distinct spatial (e.g. different plant tissues) or
sperm development arrest) (13). Furthermore, mass spectro- temporal (e.g. in response to pathogens or herbivores) functions
metric analysis has shown that AtBBE-like 15 is located in the in planta.
cell wall and thus might participate in as yet undefined reac-
Experimental Procedures
tions relevant for the formation of the cell wall (14). In addition,
we have selected AtBBE-like 13 (At1g30760) for further analy-
Chemicals—All chemicals were from Sigma and were of the
sis due to its close sequence relationship of 81% identity to highest grade commercially available. Restriction enzymes
AtBBE-like 15.
were obtained from Fermentas (Waltham, MA). Nickel-Sep-
Structural analysis of recombinant AtBBE-like 15 by x-ray harose 6 Fast Flow column material was from GE Healthcare
crystallography confirmed a topology similar to that of berber- (Little Chalfont, UK). Synthetic genes coding for AtBBE-like 15
ine bridge enzyme from E. californica (EcBBE) and ⌬1-tetrahy- and AtBBE-like 13 were obtained from Life Technologies, Inc.,
drocannabinolic acid synthase (THCA synthase) from Canna- with codon usage optimized for Komagataella pastoris.
bis sativa (15, 16). However, composition and architecture of
Molecular Cloning—The proteins were expressed using
the active site of AtBBE-like 15 indicated distinct catalytic Komagataella pastoris as expression host, according to the
properties. To identify potential substrates, we screened a small EasySelectTM Pichia expression kit provided by Invitrogen. The
compound library with regard to its effect on the protein’s ther- genes were adapted to the K. pastoris codon usage, and a C-ter-
mal stability. This approach produced several hits and minal His tag was added. SignalP was used to identify the native
allowed the identification of cinnamyl alcohol as a lead signal sequence of 30 and 27 amino acids for AtBBE-like 13 and
structure. Further analysis showed that the p-hydroxylated AtBBE-like 15, respectively (17). The genes lacking the signal
derivatives of cinnamyl alcohol, i.e. the monolignols p-cou- sequence were cloned into pPICZ␣ vectorꢀ (Invitrogen) using
maryl-, coniferyl-, and sinapyl alcohol are rapidly oxidized to standard techniques. K. pastoris strain KM71H was trans-
their corresponding aldehydes. Furthermore, the ␤-O-glycosy- formed with the pPICK-PDI vector harboring the gene for the
lated form of coniferyl alcohol (coniferin) is also accepted as a protein-disulfide isomerase from Saccharomyces cerevisiae.
substrate. Because monolignols and their ␤-glycosylated deriv- The modified KM71H strain was transformed with the linear-
atives (coniferin, syringin, and p-coumaryl-␤-glycoside) are ized [pPICZ␣-AtBBE-like 13] or [pPICZ␣-AtBBE-like 15] con-
important building blocks and monolignol storage forms, struct using electroporation. Optimal expression strains were
respectively, the catalytic reactions performed by AtBBE-like identified using the method proposed by Weis et al. (18).
13 and AtBBE-like 15 constitute a novel link between the phe-
Protein Expression and Purification—Expression was carried
nylpropanoid pathway and the formation of plant polymeriza- out using a BBI CT5-2 fermenter (Sartorius, Go¨ttingen, Ger-
tion products, such as lignin and suberin. The characteristic many) using a basal salt minimal medium as described by
active site found in AtBBE-like 13 and AtBBE-like 15 is con- Schrittwieser et al. (19). After 96 h of methanol induction, the
served in the majority of BBE-like enzymes in A. thaliana (14 of pH was set to 8.0 with sodium hydroxide, and imidazole was
28) suggesting that they either exhibit different substrate pref- added to a final concentration of 10 mM. The cells were
JULY 24, 2015•VOLUME 290•NUMBER 30
JOURNAL OF BIOLOGICAL CHEMISTRY 18771

Oxidation of Monolignols by BBE-like Enzymes
removed by centrifugation at 4000 rpm at 4 °C for 30 min. The consisting of 0.5 ␮l of protein stock solution (37 mg/ml), 0.2 ␮l
supernatant was incubated with 50 ml of nickel-Sepharose 6 of the seeding stock, and 0.5 ␮l of the JCSGϩ screen solutions
Fast Flow material at 4 °C for 45 min. Then the affinity material from the reservoir. The best diffracting crystals appeared after 4
was packed into a column and washed with 5 column volumes and a half weeks in a serial dilution setup done manually in
of 50 mM phosphate buffer, pH 8.0, containing 150 mM NaCl Crystal Clear duo plates for sitting drop (Douglas Instruments).
and 20 mM imidazole. The protein was eluted using 50 mM The 1.2-␮l drops consisted of 0.5 ␮l of protein stock solution
phosphate buffer, pH 8.0, containing 150 mM NaCl and 150 mM (36 mg/ml), 0.2 ␮l of the seeding stock, and 0.5 ␮l of the JCSGϩ
imidazole. Fractions containing AtBBE-like 15 were concen- screen condition 2-33 (0.1 M potassium thiocyanate and 30%
trated using Amicon Ultracentrifugal filter units and subse- w/v PEG 2000 MME) from the reservoir.
quently loaded on a Superdex 200 gel filtration column using an
Data Collection and Processing—X-ray diffraction data were
¨
Akta system and separated using 50 mM Tris buffer, pH 8.0, collected at 100 K at Elettra (Trieste) without additional cryo-
containing 150 mM sodium chloride. The purity of purified pro- protectant. Data processing was performed with the XDS pro-
tein was monitored by SDS-PAGE. The final yield from 3.5 gram package (21). Unit cell parameters and assigned space
liters of fermentation supernatant was 520 mg of purified pro- groups as well as data statistics are shown in Table 3. The sol-
tein. The large scale expression of AtBBE-like 13 was conducted vent content was estimated based on the calculated Matthews
as described for AtBBE-like15. In contrast to AtBBE-like 15, coefficient (22). Molecular replacement was performed with
AtBBE-like 13 was not secreted, and hence the protein was iso- Phaser (23) using the structure of EcBBE as the search template.
lated from cells and not the cultivation medium. Briefly, the cell Structure refinement was done with Phenix (24) followed by
pellet was harvested and redissolved (1:3 w/v) in cell lysis buffer manual inspection and model rebuilding in Coot (25). The pro-
(50 mM NaH2PO4, 300 mM NaCl, 20 mM imidazole, pH 8.0), gram Readyset! (included in Phenix) was used to generate
disrupted by zirconia/silica beads using a Merckenschlager restraints for metal coordination and the PEG fragment
homogenizer. The lysate was separated from the cell debris by included in the refinement. In total, three sodium and one
centrifugation (18000 rpm, 30 min, 4 °C), filtered, and incu- potassium ion were found to bind to the protein surface with-
bated with nickel-Sepharose 6 Fast Flow material (GE Health- out structural impact. The TLSMD server was used to generate
care). The column was packed with the Sepharose material, and partitioning groups (three for chain A and four for chain B) to
after extensive washing (50 mM NaH2PO4, 300 mM NaCl, 50 mM improve refinement (26). Several refinement cycles were done
imidazole, pH 8.0), the protein was eluted with elution buffer until all visible residues were assigned, and no significant
containing 500 mM imidazole. Subsequent purification was changes in R and Rfree were observed. Structure validation was
performed as described for AtBBE-like 15.
performed using MolProbity (27).
Site-directed Mutagenesis, Generation of the L182V Vari-
Chain A of the model consists of residues Ser-27 to Gly-532.
ant—Site-directed mutagenesis was performed according to Chain B starts with residue Gln-30 and extends to Gly-532.
the instructions of the QuikChange XL kit (Stratagene, La Jolla, Some residues in the N-terminal region could not be modeled
CA) for site-directed mutagenesis. The pPICZ␣-AtBBE-like 15 into the electron density most likely due to their high flexibility
vector was used as template for the polymerase chain reaction. (Val-43 and Ser-44 in chain A as well Gln-40, Ser-41, and
The introduction of the desired codon causing a change of leu- Asp-42 in chain B). In addition, no clear electron density was
cine in position 182 to valine was verified by sequencing. Trans- visible for a loop region (residues 301–305) in both chains.
formation of the gene was performed as described before.
Thermofluor Experiments—Thermofluor experiments were
Expression of the mutated gene and purification of the AtBBE- performed using a CFX Connect real time PCR system (Bio-
like 15 L182V variant was performed as described for wild-type Rad). The experiments were performed using Syproꢀ Orange as
AtBBE-like 15.
fluorescent dye in 50 mM MES buffer, pH 7.0. Stock solutions of
Crystallization—Initial crystal screening of AtBBE-like all substrates were prepared in water with a concentration of 10
15(27–532) was performed with an Oryx8 robot (Douglas mg/ml. For substrates with a lower solubility, a saturated solu-
Instruments, Berkshire, UK) using commercially available tion was used. The total volume in each well was 25 ␮l with a
Index (Hampton Research, Aliso Viejo, CA) and JCSGϩ protein concentration of 0.4 and 2 mg/ml substrate. The start-
(Molecular Dimensions, Suffolk, UK) screens. Screening was ing temperature of 20 °C was kept for 5 min and then the tem-
performed in Swissci triple well plates (Molecular Dimensions) perature was increased at a rate of 0.5 °C/min to 95 °C. Melting
using the sitting drop method with a reservoir volume of 33 ␮l. temperatures were determined using the program Bio-Rad
Drops of 1 ␮l were pipetted in a 1:1 ratio of protein and reser- CFX Manager 3.0.
voir solution with two different protein stock solutions, 22 and
Rapid Reaction Kinetics Using Stopped-flow Spectropho-
50 mg/ml in 50 mM Tris buffer, pH 8.0, containing 150 mM tometry—Reductive and oxidative half-reactions were deter-
sodium chloride, respectively. Plates were sealed with the ther- mined with a stopped-flow device (SF-61DX2, TgK Scientific,
mal seal A sealing films (Sigma) and were incubated at 289 K. UK) at 25 °C under anoxic conditions in a glove box (Belle
First crystals appeared after 1 week in JCSGϩ screen condition Technology, Weymouth, UK). Oxygen was removed from the
1–14 (0.2 M sodium thiocyanate and 20% w/v PEG 3350) with sample by flushing with nitrogen and subsequent incubation of
the higher protein concentration. Crystals from this drop were the samples in the glove box for 1 h. Spectral changes of the
pipetted out and crushed in 100 ␮l of the original 1–14 solution flavin cofactor were followed at a wavelength of 450 nm using a
and used as a seeding stock. Cross-seeding setups (20) were KinetaScanT diode array detector (MG-6560, Hi-Tech)
performed in the same manner as initial screening with drops employing the Kinetic Studio software (TgK Scientific, UK).
18772 JOURNAL OF BIOLOGICAL CHEMISTRY
VOLUME 290•NUMBER 30•JULY 24, 2015

Oxidation of Monolignols by BBE-like Enzymes
Reductive half-reactions were assayed by mixing 60 ␮M AtBBE- sis. We chose EcBBE as the outgroup sequence. The tree shown
like 15 or 13 in 50 mM potassium phosphate buffer, pH 7.0, with in supplemental Fig. S1 was visualized using FigTree (Tree Fig-
various substrate concentrations. The observed rate constants ure Drawing Tool, version 1.4.0 by Andrew Rambaut). The
at different substrate concentrations (kobs) were determined accession codes for the sequences used for the analysis are given
using the Kinetic Studio software by fitting the data to an expo- in the legend of supplemental Fig. S1.
nential function. If suitable, the dissociation constant was
Docking—Docking of coniferyl alcohol was performed using
determined performing a sigmoidal fit of the observed rate con- AutoDock Vina using the default docking parameters (33, 34).
stants employing Origin 7 (OriginLab Corp., Northampton). Point charges were assigned according to the AMBER03 force
The oxidative half-reaction was determined by mixing oxygen field (35). The setup of the receptor was done with the YASARA
free photo-reduced 60 ␮M AtBBE-like 15, AtBBE-like 15 L182V molecular modeling program (36). Twenty five Vina docking
or AtBBE-like 13 with air-saturated buffer. Photoreduction was runs were performed, and after clustering all runs, four distinct
carried out according to Massey and Hemmerich (28).
complex conformations with a binding energy of Ϫ6.4, Ϫ6.1,
Product Identification—Products resulting from the conver- Ϫ5.0, and Ϫ4.6 kJ molϪ1 were found. The complex with a bind-
sion of monolignols and their glycosides were separated and ing energy of Ϫ6.4 kJ molϪ1 and a conformation that is in agree-
identified using HPLC. Reactions were performed in 1.5-ml ment with the catalytic activity found for AtBBE-like 15 is
reaction vials in 50 mM potassium phosphate buffer, pH 7.0, at shown in Fig. 3. 8% of the docks cluster in this top pose.
30 °C and 700 rpm. Substrates were used at a concentration of
Results
2.5 mM and 30 ␮g/ml AtBBE-like 15. After 24 h, the reaction
was quenched by mixing a 500-␮l sample with 500 ␮l of meth-
Enzymatic Properties and Identification of Substrates of
anol. Samples were spun for 10 min at 9400 ϫ g in a bench-top AtBBE-like 13 and AtBBE-like 15—AtBBE-like 13 and AtBBE-
centrifuge before the clear supernatant was applied to the like 15 were expressed in K. pastoris and purified from the cul-
HPLC. The products were identified by retention time and by ture medium by nickel-Sepharose affinity chromatography and
comparing their UV absorption spectra with authentic stan- subsequent gel filtration yielding ϳ10 and 150 mg of protein
dards. HPLC analyses were done using a Dionex UltiMate 3000 from 1 liter of fermentation culture, respectively. To identify
HPLC (Thermo Fisher Scientific, Waltham, MA) equipped potential substrates, we inspected the chemical nature of sub-
with an Atlantisꢀ dC18 5 ␮m (4.6 ϫ 250 mm) column. Separa- strates recently identified for vanillyl alcohol oxidase (VAO)
tion of all compounds was achieved using a linear gradient with superfamily members. This revealed that oxidation of primary
water with 0.1% TFA as solvent A and acetonitrile with 0.1 TFA and secondary alcohol groups is a prevailing feature in all of
as solvent B and a flow rate of 0.5 ml/min. Separations were these reactions (4). In view of this recurring motif, we have
started with a mobile phase of 80% solvent A and 20% solvent B. assembled a library comprising predominantly compounds
The concentration of solvent B was increased to 50% within 20 with one or more hydroxyl groups (see supplemental Table S1).
min followed by a steep ramp to 100% solvent B in 10 min. At Because strong protein-ligand interactions are known to
the end of the protocol, the concentration of solvent B was increase the thermal stability of proteins, we first screened our
again decreased to 20% over 5 min. Retention times of authentic library with regard to an increase in the melting temperature of
standard compounds were determined using the described pro- AtBBE-like 15 using the ThermoFluor method (37). This
tocol. Under these experimental conditions, the following approach produced a number of promising hits, yielding a sub-
retention times were observed: coniferyl alcohol, 16.06 min; stantial increase of the melting temperature (up to 17 °C, for a
coniferyl aldehyde, 21.39 min; ferulic acid, 18.27 min; coumaryl complete list see supplemental Table S1). The compounds with
alcohol, 15.52 min; coumaryl aldehyde, 21.07 min; p-coumaric the strongest effect on the thermal stability were tested as sub-
acid, 17.67 min; sinapyl alcohol, 15.40 min; sinapyl aldehyde, strates revealing that cinnamyl alcohol was oxidized to cin-
21.01 min, and sinapic acid, 17.76 min.
namyl aldehyde by AtBBE-like 15. Using cinnamyl alcohol as a
Synthesis of Monolignol Glycosides—A glucosyltransferase lead structure, we tested several other naturally occurring aro-
from apple (UGT71A15, Malus x domestica) was used for the matic alcohols as putative substrates and found that AtBBE-like
synthesis of 4-O-␤-D-glucopyranosides of the monolignols (29). 15 oxidizes the monolignols p-coumaryl-, coniferyl-, and sina-
Heterologous expression in Escherichia coli and purification of pyl alcohol to their corresponding aldehydes. To gain more
the glycosyltransferase will be published elsewhere. Briefly, glu- detailed information on substrate specificity and kinetic pa-
cosylation of the aglyca (5 mM) from 7.5 mM uridine UDP-glu- rameters, reductive half-reactions were studied using stopped-
cose was performed in 50 mM Tris/Cl buffer, pH 7.5, containing flow spectrophotometry (a summary of kinetic parameters is
50 mM MgCl2, 0.13% BSA, and 10% DMSO in the presence of 6 given in Table 2). Reduction of AtBBE-like 15 and 13, using
␮M UGT71A15.
either coniferyl or sinapyl alcohol, was very fast and essentially
Phylogenetic Tree Construction—M-Coffee was used to cre- complete within the dead time of the instrument (ϳ5 ms) indi-
ate a multiple sequence alignment, including all 28 AtBBE-like cating that the rate of reduction exceeds 500 sϪ1. In the case of
family members (30). The alignment was edited by hand using p-coumaryl alcohol, the rate of reduction could be analyzed as a
Jalview (31). The PHYLIP package (PHYLIP 3.69) was used to function of substrate concentration, yielding a dissociation
create a bootstrapped phylogenetic tree using the programs constant of 700 ␮M and a limiting rate of 332 sϪ1. Because
SEQBOOT, PROTDIST, FITCH, and CONSENSE (32). We cre- monosaccharides, such as D-glucose, had such a profound effect
ated 1000 Jackknife sub-alignments with SEQBOOT, which were on the protein’s thermal stability, we also analyzed the ␤-O-
subsequently subjected to a bootstrapped protein-distance analy- glycosylated derivatives of the monolignols and found that
JULY 24, 2015•VOLUME 290•NUMBER 30
JOURNAL OF BIOLOGICAL CHEMISTRY 18773

Oxidation of Monolignols by BBE-like Enzymes
TABLE 2
Chemical structures of substrates and their kinetic parameters for AtBBE-like 13 and AtBBE-like 15
Dissociation constants and observed rates of reduction of AtBBE-like 13 and AtBBE-like 15 with different monolignols and their glycosylated derivatives are shown.
* Data were not determined due to solubility limitation.
** Reaction was too fast to be measured.
coniferin was accepted as a substrate exhibiting a dissociation FAD with dioxygen is very slow, we assume that an alternative
Ϫ1
constant of 660 ␮M and a limiting rate of reduction of 171 s
.
electron acceptor is required for reoxidation of the reduced
FAD cofactor of AtBBE-like 13 and AtBBE-like 15 in planta.
Crystal Structure of AtBBE-like 15—To better understand
the enzymatic reaction mechanism and substrate binding to the
active site, we have elucidated the three-dimensional structure
of AtBBE-like 15 by means of x-ray crystallography (Protein
Data Bank code 4UD8). Protein crystals were obtained using
sitting drop conditions yielding diffraction data to 2.1 Å. The
structure was solved by molecular replacement using EcBBE
(Protein Data Bank code 3D2H) with two molecules in the
asymmetric unit (data collection and statistics are given in
Table 3). AtBBE-like 15 adopts the same fold as other BBE-like
enzymes, i.e. the VAO-fold that is characterized by a FAD-
binding and a substrate-binding domain (Fig. 2C, shown in
green and orange, respectively). The substrate-binding domain
consists of a seven-stranded antiparallel ␤-sheet, which is cov-
ered by ␣-helices. The substrate-binding pocket and the active
site are formed by two short loops (termed loop ␤2-␤3 and loop
␣L-␣M), the “oxygen reactivity motif” consisting of loop
␣D-␣E, helix ␣E, and loop ␣E-␤6. Additionally, the three
strands ␤14, ␤15, and ␤16 contribute to the active site (Fig. 2B,
nomenclature of secondary structure elements according to
Ref. 15). The FAD cofactor is bicovalently linked to the peptide
However, the ␤-O-glycosylated derivatives of p-coumaryl and
sinapyl alcohol (syringin) were relatively poor substrates of the
enzyme. An equivalent set of experiments was conducted with
AtBBE-like 13 demonstrating similar catalytic properties (sum-
marized in Table 2). The observed rate constants for selected
substrates determined for AtBBE-like 15 and 13 are shown in
Fig. 1, A and B. Thus, we propose that AtBBE-like 13 and
AtBBE-like 15 are in fact monolignol oxidoreductases.
Reoxidation of photo-reduced AtBBE-like 13 and AtBBE-like
15 by molecular dioxygen was very slow yielding an oxidative
rate constant of 3.3 Ϯ 0.6 and 27.5 Ϯ 1.4 MϪ1 sϪ1, respectively,
thus indicating that the enzymes suppress the reduction of oxy-
gen by the reduced FAD cofactor. Recently, we have reported
that oxygen reactivity in the BBE family is controlled by the side
chain of a gatekeeper residue in the “oxygen reactivity loop” on
the re-side of the isoalloxazine ring. In the case of AtBBE-like
15, the gatekeeper residue is a leucine (Leu-182), and thus oxy-
gen reactivity is suppressed because the side chain blocks access
to the oxygen pocket (38). Replacement of Leu-182 to valine in
AtBBE-like 15 increased the rate of reoxidation ϳ400-fold to
1ꢁ104
M
s
Ϫ1, i.e. very similar to the rate found with EcBBE
Ϫ1
(kox ϭ 5ꢁ104 Ϫ1 sϪ1) (39), which also possesses a valine residue
M
in the gatekeeper position. Because the reaction of the reduced
18774 JOURNAL OF BIOLOGICAL CHEMISTRY
VOLUME 290•NUMBER 30•JULY 24, 2015

Oxidation of Monolignols by BBE-like Enzymes
FIGURE 1. Rate constants observed for the reductive half-reaction of AtBBE-like 15 (A) and AtBBE-like 13 (B). A stopped-flow device was used to
determine the rate constants of the reductive half-reaction of AtBBE-like 15 and 13 with p-coumaryl alcohol (solid square), cinnamyl alcohol (green triangle),
coniferylalcohol␤-glycoside(orangecircle), andsinapylalcohol␤-glycoside(invertedbluetriangle). Thecourseofthereactionwasdeterminedbyfollowingthe
absorption of the oxidized flavin at 450 nm. As the reactions follow pseudo-first order kinetics, rates were calculated from a monoexponential fit.
TABLE 3
fore, these residues were defined as substrate coordination
Data acquisition and refinement parameters for x-ray crystallography
motif. The hydroxyl groups of Tyr-117, Tyr-479, and Tyr-193
AtBBE-like 15
point toward the active site with distances to the N5 locus of 4.8,
Beamline
Elettra XRD1
4.6, and 5.0 Å, respectively. The two phenolic hydroxyl groups
of Tyr-479 and Tyr-193 are within hydrogen bond distance (2.4
Å, Fig. 2A). The ⑀-amino group of Lys-436 is located 3.1 Å above
the plane of the aromatic ring of Tyr-193 and engages in a
cation-␲ interaction. This arrangement of Tyr-479, Tyr-193,
and Lys-436 presumably favors deprotonation of the phenolic
hydroxyl group of Tyr-193 and activates this residue as a puta-
tive active site base, and thus these residues were defined as a
catalytic base motif.
Wavelength (Å)
Unit cell parameters (Å,°)
Space group
0.971670
63.6, 94.7, 188.3, 90, 90, 90
P212121
Resolution limits (Å)
Rmeas
50-2.09 (2.21-2.09)
0.209 (0.865)
0.191 (0.794)
415,743 (58,987)
67,102 (9501)
7.73 (2.23)
97.8 (87.0)
6.2
Rmerge
Total no. of observations
Total no. unique
͗I/␴(I)͘
Completeness (%)
Redundancy
Wilson B factor (Å2)
Matthews coefficient (Å3 Da
Molecules per ASU
Solvent content (%)
32.02
Ϫ1
)
2.36
Docking of Substrates to the Active Site—Because co-crystal-
lization with the identified substrates was unsuccessful, we con-
ducted docking experiments using coniferyl alcohol as a model
substrate to gain further insights into the interaction of sub-
strates with amino acid residues in the active site. At the si-side
of the isoalloxazine ring, a hydrophobic pocket is formed by
residues Phe-373, Phe-377, Leu-407, and Leu-440 and the di-
methylbenzene moiety of the isoalloxazine ring. As shown in Fig.
3, the substrate fits into this hydrophobic pocket (binding
energy of Ϫ6.4 kJ molϪ1). The docking result suggests that the
orientation of the allylic alcohol toward the proposed catalytic
base Tyr-193 is possible, although the aromatic ring is located
in the hydrophobic pocket, and thus this orientation is consis-
tent with the productive catalysis found for AtBBE-like 15 with
this ligand.
2
48
Refinement
Resolution (Å)
R (%)
47.37-2.09
18.35
R
free (%)
22.09
Root mean square deviation
stereochemistry
Bond lengths (Å)
0.008
0.825
7946
247
Bond angles (°)
No. of protein atoms
No. of non-protein atoms
No. of water molecules
Average -B-factor (Å2)
615
25.54
Ramachandran analysis
Favored (%)
95.96
3.94
0.1
Allowed (%)
Disallowed (%)
PDB code
4ud8
Functional and Structural Characterization of the BBE-like
Family in A. thaliana—The identification of critical active site
residues with a role in substrate binding and catalysis in AtBBE-
like 15 raised the question whether these are conserved in all
AtBBE-like family members. To investigate this issue, we have
constructed a protein distance tree (Fig. 4, left). Seven distinct
groups can be identified in the phylogenetic tree within the
family of BBE-like enzymes in A. thaliana, in good agreement
with previously published work (9). Based on these distinct
groups, we have created sequence logos for each of the elements
chain via a covalent bond of His-115 to the 8␣-methyl group
and of Cys-179 to the C6-position of the isoalloxazine ring.
As hydride transfer from the substrate to the N5 locus of the
flavin cofactor is a generally accepted mode of reduction, a
sphere with a radius of 10 Å around this locus was examined for
putative catalytic residues. As shown in Fig. 2A, Tyr-117 and
Gln-438 are engaged in a hydrogen bond on the si-side of the
isoalloxazine ring (2.6 Å). These residues are positioned
between the active site and the binding pocket and thus must
interact with the substrate when it enters the active site. There- forming the active site, i.e. the three strands at the ceiling of the
JULY 24, 2015•VOLUME 290•NUMBER 30 JOURNAL OF BIOLOGICAL CHEMISTRY 18775

Oxidation of Monolignols by BBE-like Enzymes
FIGURE 2. A, close-up view of the amino acids with a proposed role in substrate binding and catalysis. Tyr-117 and Gln-438 form a hydrogen bond. Lys-436 and
Tyr-193 areinvolved ina cation-␲ interaction, andTyr-193formsahydrogenbond toTyr-479. Twowelldefined water moleculesarefound intheactive site (red
spheres). The electron density was cut off at a ␴ value of 1.5 Å. B, close-up view of the active site. Loop ␤2-␤3 and the oxygen reactivity motif originate from the
FAD-bindingdomainandharborHis-115andCys-179, whichformthetwocovalentlinkagestotheisoalloxazineringoftheFAD. Strands ␤14, ␤15, and␤16and
loop ␣L-␣M originate from the substrate-binding domain. They harbor amino acids defining the si-face near the N5 locus (Ͻ10 Å) that are putatively involved
in the catalytic mechanism. C, overall topology of AtBBE-like 15. AtBBE-like 15 adopts the VAO topology consisting of a FAD-binding domain (green) and a
substrate-binding domain (orange). Furthermore, it features a bicovalently attached FAD typical for BBE-like proteins. The FAD cofactor is shown in stick
representation in yellow.
FIGURE 3. Substrate binding to AtBBE-like 15. The substrate-binding cavity and active site of AtBBE-like 15 are shown in stereo view. The isoalloxazine ring
is shown in yellow and coniferyl alcohol in light blue as a stick model. Coniferyl alcohol was docked into the cavity using YASARA. The aromatic moiety is located
in the hydrophobic binding pocket formed by Phe-377, Phe-373, Leu-407, Leu-440, and Tyr-117, and the isoalloxazine ring, whereas the allyl alcohol is facing
the active site. All residues shown in this figure are conserved in AtBBE-like 13.
active site cavity and the three loops in the vicinity of the oriented toward the ␣-helices (Fig. 4, upward pointing arrows)
isoalloxazine ring (Fig. 4, right). The histidine residue (His-115) are conserved because they are important for the structural
in loop ␤2-␤3 that forms the covalent bond to the 8␣-methyl integrity of the protein. For example, Lys-380 and Asp-382 in
group of the flavin isoalloxazine ring is conserved in all AtBBE- strand ␤14 form salt bridges to Asp-481 (2.7 Å) and Arg-473
like enzymes, and thus it can be assumed that the covalent link- (3.3 Å), respectively. Thus, these amino acids are invariant in
age is present in all family members. In contrast, the cysteine the whole family, because all of them adopt the same overall
residue (Cys-179) located in the oxygen reactivity motif is only topology. In contrast, amino acid residues pointing toward the
conserved in phylogenetic groups 2, 4, 5 and 7, whereas the active site (Fig. 4, downward pointing arrows) are only con-
remaining groups feature other amino acids, such as histi- served within certain phylogenetic groups, indicating that the
dine (group 1), serine (group 3), or tyrosine (group 1 and 6). We composition of the active site differs markedly. In the case of
assume that these four AtBBE-like enzymes possess a single AtBBE-like 15 (group 6 in Fig. 4), the active site residues Tyr-
covalent linkage, whereas the remaining 24 feature a bicovalent 117, Tyr-193, Lys-436, Gln-438, and Tyr-479 are important for
linkage of the FAD cofactor as already shown for EcBBE (40), substrate binding and catalysis (boxed residues in Fig. 4 and
Dbv29 (41), glucooligosaccharide oxidase (GOOX) (42), and shown as stick models in Fig. 5A). While residues Tyr-193, Tyr-
THCA synthase (15).
479, and Lys-436 concertedly act as an active site base (Fig. 5A,
Further analysis of the sequence logos generated for the three “catalytic motif,” shown in green), Tyr-117 and Gln-438 appear
strands of the ␤-sheet provided important insights into the to be involved in determining the substrate preference (“sub-
functions of amino acid side chains. Generally, amino acids strate coordination motif,” shown in cyan). Both motifs are
18776 JOURNAL OF BIOLOGICAL CHEMISTRY
VOLUME 290•NUMBER 30•JULY 24, 2015

Oxidation of Monolignols by BBE-like Enzymes
FIGURE 4. Chemical nature of the active sites present in different phylogenetic groups is represented by sequence logos of the active site forming
secondary structure elements. Left, protein distance tree of all BBE-like enzymes from A. thaliana with EcBBE as outgroup. Right, sequence logos representing
the structural elements forming the active site of the seven phylogenetic groups reveal highly conserved motifs. Arrows indicate the orientation of the residues
inthe␤-strands. Anupwardarrowindicatesthattheresiduepointsawayfromtheactivesite. Theseresidueswerefoundtobestructurallyrelevant. A downward
arrow indicates the residue points toward the active site and thereby contributes to the decoration. Residues highlighted in a green and cyan boxes define the
substratecoordinationandcatalyticmotifs, respectively. Bothmotifsareputativelyinvolvedinthecatalyticmechanism(seealsoFig. 5and6). Positionslabeled
with an asterisk highlight the two potential sites of covalent linkage.
alcohols to their corresponding aldehyde products. This analy-
sis was extended to other plants revealing that the active site
signature found in AtBBE-like 13 and AtBBE-like 15 is con-
served in soybean (Glycine max; 21 homologs of 43 BBE-like
enzymes), eucalyptus (Eucalyptus grandis; 18 of 27), poplar (P.
trichocarpa; 17 of 64), and potato (Solanum tuberosum; 6 of 18).
Interestingly, monocotyledonous species such as maize (Zea
mays; 16 BBE-like enzymes) and rice (Oriza sativa, 11 BBE-like
enzymes) appear to lack monolignol oxidoreductase activity,
because the catalytic and substrate coordination motifs are not
found in any of the BBE-like enzymes.
In contrast, major variations in the catalytic and/or substrate
coordination motif are seen in groups one, three, and seven. In
group one, Lys-436 in the catalytic motif is replaced by glutamic
FIGURE 5. Active site composition predominantly present in different phy-
acid (Fig. 4, strand ␤16, and Fig. 5B) and Tyr-479 by phenylala-
logenetic groups. The active site compositions shown in B, C, and D were visu-
nine or leucine, respectively (Fig. 4, ␣L-␣M), although the sub-
alized using homology models created with YASARA. A, active site composition
of AtBBE-like 15 that is representative for groups 2 and 4–6. B, this active site
compositionisfoundingroup1(numberingaccordingtoAtBBE28). Thecatalytic
base as found in AtBBE-like 15 has been disrupted. Lys-436 has been replaced by
strate coordination motif is conserved. In group three, the sub-
strate coordination motif (Tyr-117 and Gln-438 are replaced by
a glutamic acid that putatively can recover the catalytic base function. C, this
phenylalanine and glutamic acid, respectively) as well as the
active site composition is found in group 3 (numbering according to AtBBE11).
catalytic motif residues (Lys-438 replaced by asparagine) are
The catalytic base motif is inconsistent; Gln-438 has been replaced by a glutamic
modified (Fig. 5C). Group seven features a conserved substrate
acid that putatively recovers the catalytic base function. D, active site composi-
tion as found in group 7 (numbering according to AtBBE22). In the catalytic base,
motif Lys-436 has been replaced by an asparagine and Tyr-193 by a phenylala-
nine. In ␤15, a glutamic acid can be found that putatively recovers the catalytic
coordination motif, but in the catalytic motif residues Lys-436
and Tyr-193 are replaced by asparagine and phenylalanine,
base function in this active site.
respectively (Fig. 5D). It is worthwhile noting that in all cases
where the catalytic motif appears to be disrupted, an alternative
catalytic base is present, for example glutamic acid instead of
strictly conserved in 14 members of the BBE-like enzymes in A.
thaliana (see Table 4), indicating that their catalytic mecha- Lys-436 or Asn-411 in groups one and seven, respectively. Sim-
nism and substrate preference will be similar, i.e. oxidation of ilarly, Gln-438 in group three is replaced by glutamic acid.
JULY 24, 2015•VOLUME 290•NUMBER 30
JOURNAL OF BIOLOGICAL CHEMISTRY 18777

Oxidation of Monolignols by BBE-like Enzymes
Overall, we conclude that the BBE-like enzymes of these groups lytic mechanism. Tyr-193 is positioned such that the phenolic
have distinct catalytic properties and act on different yet hydroxyl group points toward the substrate’s alcohol group.
unidentified substrates. Interestingly, the presence of a catalytic
active base appears to be an important feature in BBE-like
enzymes, as it was also observed in EcBBE (16), Dbv29 (41), GilR
(43), GOOX (42), and THCA synthase (15).
Lys-436 forms a cation-␲ interaction with the aromatic ring of
Tyr-193, thereby lowering the pKa value and thus stabilizing
the deprotonated state. We also assume that Tyr-479 stabilizes
the deprotonated state of Tyr-193 by hydrogen bonding (Figs. 2
and 3). Thus, these three amino acids make up the core catalytic
machinery for deprotonation of the substrate’s hydroxyl group
that results in the concomitant transfer of a hydride from the
C␥ position to the N5 locus of the isoalloxazine ring of the
flavin. A schematic reaction mechanism for the oxidation of
monolignols by AtBBE-like 15 is shown in Fig. 6. To test this
hypothesis, we have initiated a site-directed mutagenesis study
of the pertinent active site residues.
Discussion
Catalytic Mechanism of Monolignol Oxidation—Based on
the crystallographic structure of AtBBE-like 15 and the results
obtained by substrate docking, we propose the following cata-
TABLE 4
Composition of the substrate coordination (Tyr-117 and Gln-438) and
catalytic motif (Tyr-193, Lys-436, and Tyr-479)
Structural Comparison with Other Members of the VAO
Family—In recent years, the structures of several other mem-
bers of the VAO family were determined that share a high
degree of structural similarity despite low similarity on the
sequence level. Structural superposition of AtBBE-like 15 with
Dbv29, GOOX, AknOx, and EcBBE gives a root mean square
deviation of 1.45 Å (for 1652 backbone atoms), 1.36 Å (for 1556
backbone atoms), 1.47 Å (for 1526 backbone atoms), and 0.95 Å
(for 2068 backbone atoms) respectively, whereas sequence
identities of 22, 22, 23, and 41%, respectively, are found. The
bacterial oxidases Dbv29 and AknOx catalyze similar oxida-
tions of alcohol groups and thus possess a very similar catalytic
motif consisting of both tyrosine residues and the lysine residue
also present in AtBBE-like 15, whereas the substrate recogni-
tion motif is clearly different as depicted in Fig. 7 (C and D,
respectively). In any case, these enzymes act on alcohol sub-
strates that show no similarity to the substrates identified for
AtBBE-like 15 and 13 and, moreover, are not related to lignin
metabolism. Importantly, for Dbv29 and AknOx it was found
that both tyrosine residues are crucial for catalysis thus sup-
porting the proposed role of Tyr-193 and Tyr-479 in the mech-
anism shown in Fig. 6 for AtBBE-like 15 (and AtBBE-like 13)
(10, 41). Although Dbv29 and AknOx resemble AtBBE-like 15
and 13 concerning the composition of the active site on the
si-side of the isoalloxazine ring, the residue that determines the
reactivity toward dioxygen, i.e. Leu-182 in the case of AtBBE-
like 15, is a valine in both enzymes, defining them as oxidases
(38). The active site of GOOX has the same substrate recogni-
tion motif but has a distinct catalytic motif featuring a gluta-
mate residue (shown in blue in Fig. 7E) instead of the tyrosine
FIGURE 6. Proposed reaction mechanism for substrate oxidation. Tyr-193, Tyr-479, and Lys-436 concertedly act as an active site base abstracting a proton
from the alcohol with subsequent hydride transfer to the N5 of the isoalloxazine ring.
18778 JOURNAL OF BIOLOGICAL CHEMISTRY
VOLUME 290•NUMBER 30•JULY 24, 2015

Oxidation of Monolignols by BBE-like Enzymes
ucts of monolignols but the aldehydes are also incorporated
into the cell wall (45). It has been shown by mass spectrometry
that AtBBE-like 15 is located in the apoplastic fluid of A. thali-
ana, and thus our findings that AtBBE-like 15 (and AtBBE-like
13) oxidizes monolignols to their corresponding aldehydes
indicates that they are likely physiological substrates (46). In
addition, the glycosylated form of coniferyl alcohol (coniferin)
is also accepted as substrate. Additionally, oxidation of conife-
rin to its corresponding aldehyde has been postulated by Tsuji
et al. (47) for the plant Gingko biloba. According to their
hypothesis, oxidation of the ␤-glycosylated monolignols pre-
cedes hydrolysis to the free monolignols and thus plays an
important role for the mobilization of monolignols from their
glycosidic storage forms. Hence, the demonstrated activity of
AtBBE-like 13 and AtBBE-like 15 in vitro suggests that the
enzymes may participate in the mobilization and oxidation of
monolignols required for polymerization processes in the plant
cell wall (e.g. lignification). Interestingly, more than 20 different
glycoside hydrolases were also identified during this proteome
analysis, suggesting that AtBBE-like 15 and potentially also
other members of the family work in concert with these hydro-
lases to mobilize building blocks from their storage forms. This
concept receives further support by co-expression data
retrieved from the ATTET II server (CoExSearch Version 4.1).
Among the enzymes co-expressed with AtBBE-like 15 are
FIGURE 7. Comparison of overall topology and active site composition of
various BBE-like enzymes and VAO superfamily members. A, overall
topology of AtBBE-like 15 and superposition with EcBBE. B–F show the sub-
strate coordination and catalytic motif in cyan and green, respectively, for the
followingBBE-likeproteins. B, AtBBE-like 15, the substrate coordination motif, phenylalanine ammonia-lyase 4 (PAL4), an important rate-de-
consists of Tyr-117 and Gln-438, and the catalytic motif is formed by Tyr-479,
termining entry point of the phenylpropanoid pathway, and
Tyr-193, and Lys-436. C, Dbv29, the catalytic motif is invariant, whereas the
substrate coordination motif is different. D, AknOx, the catalytic motif is
invariant, whereas the substrate coordination motif differs. E, GOOX, the cat-
alytic motif is disrupted, and it appears to be restored by the presence of
another catalytic base, i.e. Asp-355. F, EcBBE, both motifs are disrupted with
Glu-417 serving as a catalytic base.
␤-glucosidase 41 clustering with the monolignol glucoside
hydrolases At1g61810 (BGLU45) and At1g61820 (BGLU46)
(see supplemental Table S2) (48, 49). Thus, we suggest that
AtBBE-like 13 and AtBBE-like 15 participate in the adaptation
of the extracellular monolignol pool prior to polymerization
and are thus involved in cell wall formation required during
and lysine seen in AtBBE-like 15 (compare B and E in Fig. 7). In
certain growth phases and in response to stressors. Because the
fact, this active site composition is similar to group one of the
active site motifs found in AtBBE-like 13 and AtBBE-like 15 are
AtBBE-like family, and thus it is conceivable that these BBE
conserved in 12 other members of the BBE-like enzyme family
homologs are also involved in the oxidation of carbohydrates as
in A. thaliana, we predict that they play similar roles in the
already reported for some plant BBE-like enzymes (8, 9). Not
plant potentially with diverging substrate specificities or, alter-
surprisingly, the biggest difference in active site composition is
natively, with different spatial (i.e. different tissues) or temporal
seen with EcBBE as this enzyme catalyzes an unusual oxidative
(i.e. induced by different biotic or abiotic stressors) features.
The presence of homologs of AtBBE-like 13 and AtBBE-like 15
in many other plants (see above) indicates a general role of this
enzyme family in the plant kingdom. In contrast, enzymes such
as EcBBE and THCA synthase serve in “secondary” pathways
restricted to certain plant families or genera, and hence these
have evolved, e.g. by gene duplication and subsequent diversifi-
cation by mutation, from the primordial monolignol oxi-
doreductases, a phenomenon well documented for the evolu-
tion of new plant defense compounds (50).
ring closure reaction in alkaloid biosynthesis; although the
overall topology shows high similarity (Fig. 7A), except for the
tyrosine in the substrate recognition motif (Tyr-106), all other
residues vary (compare B and F in Fig. 7). Moreover, Glu-417
replacing the glutamine of the substrate recognition motif
assumes a crucial role in the reaction mechanism of EcBBE and
thus becomes part of the catalytic machinery of the enzyme
(16). This comparison illustrates that this enzyme family can be
tuned for new reactivities by strategic single or multiple
exchanges of amino acids in the active site. In fact, even within
a given active site different (oxidation), reactions are supported
depending on the available substrates further demonstrating
the plasticity of the active site (44).
From the analysis of the active site signatures, it is also evi-
dent that AtBBE-like enzymes in phylogenetic groups one,
three, and seven possess distinct catalytic machineries and
probably act on as yet unidentified substrates. These BBE ho-
mologs are currently the target of further studies to reveal their
biochemical properties and physiological function. In addition,
we have initiated the generation, identification, and character-
Proposed Role of AtBBE-like 15/Monolignol Oxidoreductases
in Plants—Monolignols are secreted into the apoplast and
polymerized to various cell wall components. Although mono-
lignols are thought to be secreted in their alcohol forms, not all
monolignol-derived cell wall components are coupling prod- ization of gene knock-out plants with the aim to investigate the
JULY 24, 2015•VOLUME 290•NUMBER 30
JOURNAL OF BIOLOGICAL CHEMISTRY 18779

Oxidation of Monolignols by BBE-like Enzymes
zyme controlling the psychoactivity of Cannabis sativa. J. Mol. Biol. 423,
96–105
role of AtBBE-likes in cell wall metabolism during plant devel-
opment and stress response.
16. Winkler, A., Lyskowski, A., Riedl, S., Puhl, M., Kutchan, T. M., Macher-
oux, P., and Gruber, K. (2008) A concerted mechanism for berberine
bridge enzyme. Nat. Chem. Biol. 4, 739–741
Author Contributions—B. D., K. G., and P. M. initiated and designed
the research; B. D., B. S., and S. W. carried out biochemical research;
A. G. and B. N. synthesized glycosylated monolignol derivatives;
A. D., T. P.-K., B. D., and K. G. crystallized proteins and solved the
x-ray crystal structure; B. D., C. S., and E. v. d. G. performed phylo-
genetic analysis and evaluated the data; B. D., E. v. d. G., K. G., and
P. M. wrote the manuscript.
17. Petersen, T. N., Brunak, S., von Heijne, G., and Nielsen, H. (2011) SignalP
4.0: discriminating signal peptides from transmembrane regions. Nat.
Methods 8, 785–786
18. Weis, R., Luiten, R., Skranc, W., Schwab, H., Wubbolts, M., and Glieder, A.
(2004) Reliable high-throughput screening with Pichia pastoris by limiting
yeast cell death phenomena. FEMS Yeast Res. 5, 179–189
19. Schrittwieser, J. H., Resch, V., Wallner, S., Lienhart, W. D., Sattler, J. H.,
Resch, J., Macheroux, P., and Kroutil, W. (2011) Biocatalytic organic syn-
thesis of optically pure (S)-scoulerine and berbine and benzylisoquinoline
alkaloids. J. Org. Chem. 76, 6703–6714
References
1. Winkler, A., Hartner, F., Kutchan, T. M., Glieder, A., and Macheroux, P.
(2006) Biochemical evidence that berberine bridge enzyme belongs to a
novel family of flavoproteins containing a bi-covalently attached FAD
cofactor. J. Biol. Chem. 281, 21276–21285
20. D’Arcy, A., Villard, F., and Marsh, M. (2007) An automated microseed
matrix-screening method for protein crystallization. Acta Crystallogr. D
Biol. Crystallogr. 63, 550–554
21. Kabsch, W. (2010) Integration, scaling, space-group assignment and post-
refinement. Acta Crystallogr. D Biol. Crystallogr. 66, 125–132
22. Kantardjieff, K. A., and Rupp, B. (2003) Matthews coefficient probabilities:
improved estimates for unit cell contents of proteins, DNA, and protein-
nucleic acid complex crystals. Protein Sci. 12, 1865–1871
2. Facchini, P. J., Penzes, C., Johnson, A. G., and Bull, D. (1996) Molecular
characterisation of berberine bridge enzyme genes from opium poppy.
Plant Physiol. 112, 1669–1677
3. Kutchan, T. M., and Dittrich, H. (1995) Characterization and mechanism
of the berberine bridge enzyme, a covalently flavinylated oxidase of ben-
zophenanthridine alkaloid biosynthesis in plants. J. Biol. Chem. 270,
24475–24481
23. McCoy, A. J., Grosse-Kunstleve, R. W., Adams, P. D., Winn, M. D., Sto-
roni, L. C., and Read, R. J. (2007) Phaser crystallographic software. J. Appl.
Crystallogr. 40, 658–674
4. Wallner, S., Dully, C., Daniel, B., and Macheroux, P. (2013) in Handbook of
Flavoproteins, Oxidases, Dehydrogenases and Related Systems (Hille, R.,
Miller, S. M., and Palfrey, B., eds) Vol. 1, De Gruyter, Berlin
24. Adams, P. D., Afonine, P. V., Bunko´czi, G., Chen, V. B., Davis, I. W., Echols,
N., Headd, J. J., Hung, L.-W., Kapral, G. J., Grosse-Kunstleve, R. W., Mc-
Coy, A. J., Moriarty, N. W., Oeffner, R., Read, R. J., Richardson, D. C., et al.
(2010) PHENIX: a comprehensive Python-based system for macromolec-
ular structure solution. Acta Crystallogr. D Biol. Crystallogr. 66, 213–221
25. Emsley, P., and Cowtan, K. (2004) Coot: Model-building tools for molec-
ular graphics. Acta Crystallogr. D Biol. Crystallogr. 60, 2126–2132
26. Painter, J., and Merritt, E. A. (2006) Optimal description of a protein
structure in terms of multiple groups undergoing TLS motion. Acta Crys-
tallogr. D Biol. Crystallogr. 62, 439–450
5. Winter, D., Vinegar, B., Nahal, H., Ammar, R., Wilson, G. V., and Provart,
N. J. (2007) An “Electronic Fluorescent Pictograph” browser for exploring
and analyzing large-scale biological data sets. PLoS One 2, e718
6. Gonza´lez-Candelas, L., Alamar, S., Sa´nchez-Torres, P., Zacarías, L., and
Marcos, J. F. (2010) A transcriptomic approach highlights induction of
secondary metabolism in citrus fruit in response to Penicillium digitatum
infection. BMC Plant Biol. 10, 194
7. Attila, C., Ueda, A., Cirillo, S. L., Cirillo, J. D., Chen, W., and Wood, T. K.
(2008) Pseudomonas aeruginosa PAO1 virulence factors and poplar tree
response in the rhizosphere Microb. Biotechnol. 1, 17–29
27. Chen, V. B., Arendall, W. B., 3rd, Headd, J. J., Keedy, D. A., Immormino,
R. M., Kapral, G. J., Murray, L. W., Richardson, J. S., and Richardson, D. C.
(2010) MolProbity: all-atom structure validation for macromolecular
crystallography. Acta Crystallogr. D Biol. Crystallogr. 66, 12–21
28. Massey, V., and Hemmerich, P. (1978) Photoreduction of flavoproteins
and other biological compounds catalyzed by deazaflavins. Biochemistry
17, 9–16
8. Carter, C. J., and Thornburg, R. W. (2004) Tobacco nectarin V is a flavin-
containing berberine bridge enzyme-like protein with glucose oxidase ac-
tivity. Plant Physiol. 134, 460–469
9. Custers, J. H., Harrison, S. J., Sela-Buurlage, M. B., van Deventer, E.,
Lageweg, W., Howe, P. W., van der Meijs, P. J., Ponstein, A. S., Simons,
B. H., Melchers, L. S., and Stuiver, M. H. (2004) Isolation and characteri-
sation of a class of carbohydrate oxidases from higher plants, with a role in
active defence. Plant J. 39, 147–160
29. Gosch, C., Halbwirth, H., Schneider, B., Ho¨lscher, D., and Stich, K. (2010)
Cloning and heterologous expression of glycosyltransferases from Malus
x domestica and Pyrus communis, which convert phloretin to phloretin
2Ј-O-glucoside (phloridzin). Plant Sci. 178, 299–306
10. Alexeev, I., Sultana, A., Ma¨ntsa¨la¨, P., Niemi, J., and Schneider, G. (2007)
Aclacinomycin oxidoreductase (AknOx) from the biosynthetic pathway
of the antibiotic aclacinomycin is an unusual flavoenzyme with a dual
active site. Proc. Natl. Acad. Sci. U.S.A. 104, 6170–6175
30. Wallace, I. M., O’Sullivan, O., Higgins, D. G., and Notredame, C. (2006)
M-Coffee: combining multiple sequence alignment methods with T-Cof-
fee. Nucleic Acids Res. 34, 1692–1699
31. Waterhouse, A. M., Procter, J. B., Martin, D. M., Clamp, M., and Barton,
G. J. (2009) Jalview Version 2–a multiple sequence alignment editor and
analysis workbench. Bioinformatics 25, 1189–1191
11. Carlson, J. C., Li, S., Gunatilleke, S. S., Anzai, Y., Burr, D. A., Podust, L. M.,
and Sherman, D. H. (2011) Tirandamycin biosynthesis is mediated by
co-dependent oxidative enzymes. Nat. Chem. 3, 628–633
32. Felsenstein, J. (2005) PHYLIP–Phylogeny Inference Package, Version 3.6.
Department of Genome Sciences, University of Washington, Seattle
12. Mo, X., Huang, H., Ma, J., Wang, Z., Wang, B., Zhang, S., Zhang, C., and Ju,
J. (2011) Characterization of TrdL as a 10-hydroxy dehydrogenase and
generation of new analogues from a tirandamycin biosynthetic pathway. 33. Morris, G. M., Goodsell, D. S., Huey, R., and Olson, A. J. (1996) Distributed
Org. Lett. 13, 2212–2215
automated docking of flexible ligands to proteins: parallel applications of
13. Pagnussat, G. C., Yu, H. J., Ngo, Q. A., Rajani, S., Mayalagu, S., Johnson,
AutoDock 2.4. J. Comput. Aided Mol. Des. 10, 293–304
C. S., Capron, A., Xie, L. F., Ye, D., and Sundaresan, V. (2005) Genetic and 34. Goodsell, D. S., Morris, G. M., and Olson, A. J. (1996) Automated docking
molecular identification of genes required for female gametophyte devel-
opment and function in Arabidopsis. Development 132, 603–614
14. Hooper, C. M., Tanz, S. K., Castleden, I. R., Vacher, M. A., Small, I. D., and
Millar, A. H. (2014) SUBAcon: a consensus algorithm for unifying the
subcellular localization data of the Arabidopsis proteome. Bioinformatics
30, 3356–3364
of flexible ligands: applications of AutoDock. J. Mol. Recognit. 9, 1–5
35. Duan, Y., Wu, C., Chowdhury, S., Lee, M. C., Xiong, G., Zhang, W., Yang,
R., Cieplak, P., Luo, R., Lee, T., Caldwell, J., Wang, J., and Kollman, P.
(2003) A point-charge force field for molecular mechanics simulations of
proteins based on condensed-phase quantum mechanical calculations.
J. Comput. Chem. 24, 1999–2012
15. Shoyama, Y., Tamada, T., Kurihara, K., Takeuchi, A., Taura, F., Arai, S., 36. Krieger, E., Koraimann, G., and Vriend, G. (2002) Increasing the precision
Blaber, M., Shoyama, Y., Morimoto, S., and Kuroki, R. (2012) Structure
and function of 1-tetrahydrocannabinolic acid (THCA) synthase, the en-
of comparative models with YASARA NOVA–a self-parameterizing force
field. Proteins 47, 393–402
18780 JOURNAL OF BIOLOGICAL CHEMISTRY
VOLUME 290•NUMBER 30•JULY 24, 2015

Oxidation of Monolignols by BBE-like Enzymes
37. Cummings, M. D., Farnum, M. A., and Nelen, M. I. (2006) Universal
screening methods and applications of ThermoFluor. J. Biomol. Screen.
11, 854–863
mechanism of an unusual oxidoreductase, GilR, involved in gilvocarcin V
biosynthesis. J. Biol. Chem. 286, 23533–23543
44. Winkler, A., Puhl, M., Weber, H., Kutchan, T. M., Gruber, K., and Mach-
eroux, P. (2009) Berberine bridge enzyme catalyzes the six electron oxida-
tion of (S)-reticuline to dehydroscoulerine. Phytochemistry 70, 1092–1097
38. Zafred, D., Steiner, B., Teufelberger, A. R., Hromic, A., Karplus, P. A.,
Schofield, C. J., Wallner, S., and Macheroux, P. (2015) Rationally engi-
neered flavin-dependent oxidase reveals steric control of dioxygen reac- 45. Morreel, K., Ralph, J., Kim, H., Lu, F., Goeminne, G., Ralph, S., Messens, E.,
tivity. FEBS J. 10.1111/febs.13212
and Boerjan, W. (2004) Profiling of oligolignols reveals monolignol cou-
pling conditions in lignifying poplar xylem. Plant Physiol. 136, 3537–3549
46. Jamet, E., Canut, H., Boudart, G., and Pont-Lezica, R. F. (2006) Cell wall
proteins: a new insight through proteomics. Trends Plant Sci. 11, 33–39
39. Winkler, A., Kutchan, T. M., and Macheroux, P. (2007) 6-S-Cysteinylation
of bi-covalently attached FAD in berberine bridge enzyme tunes the redox
potential for optimal activity. J. Biol. Chem. 282, 24437–24443
40. Winkler, A., Motz, K., Riedl, S., Puhl, M., Macheroux, P., and Gruber, K. 47. Tsuji, Y., Chen, F., Yasuda, S., and Fukushima, K. (2005) Unexpected be-
(2009) Structural and mechanistic studies reveal the functional role of
bicovalent flavinylation in berberine bridge enzyme. J. Biol. Chem. 284,
19993–20001
havior of coniferin in lignin biosynthesis of Ginkgo biloba L. Planta 222,
58–69
48. Escamilla-Trevin˜o, L. L., Chen, W., Card, M. L., Shih, M. C., Cheng,
C. L., and Poulton, J. E. (2006) Arabidopsis thaliana ␤-glucosidases
BGLU45 and BGLU46 hydrolyse monolignol glucosides. Phytochemis-
try 67, 1651–1660
41. Liu, Y. C., Li, Y. S., Lyu, S. Y., Hsu, L. J., Chen, Y. H., Huang, Y. T., Chan,
H. C., Huang, C. J., Chen, G. H., Chou, C. C., Tsai, M. D., and Li, T. L.
(2011) Interception of teicoplanin oxidation intermediates yields new an-
timicrobial scaffolds. Nat. Chem. Biol. 7, 304–309
49. Xu, Z., Escamilla-Trevin˜o, L., Zeng, L., Lalgondar, M., Bevan, D., Winkel,
B., Mohamed, A., Cheng, C. L., Shih, M. C., Poulton, J., and Esen, A. (2004)
Functional genomic analysis of Arabidopsis thaliana glycoside hydrolase
family 1. Plant Mol. Biol. 55, 343–367
42. Huang, C.-H., Lai, W.-L., Lee, M.-H., Chen, C.-J., Vasella, A., Tsai, Y.-C.,
and Liaw, S.-H. (2005) Crystal structure of glucooligosaccharide oxidase
from Acremonium strictum. A novel flavinylation of 6-S-cysteinyl, 8␣-N1-
histidyl FAD. J. Biol. Chem. 280, 38831–38838
50. Ober, D., and Kaltenegger, E. (2009) Pyrrolizidine alkaloid biosynthesis,
evolution of a pathway in plant secondary metabolism. Phytochemistry 70,
1687–1695
43. Noinaj, N., Bosserman, M. A., Schickli, M. A., Piszczek, G., Kharel, M. K.,
Pahari, P., Buchanan, S. K., and Rohr, J. (2011) The crystal structure and
JULY 24, 2015•VOLUME 290•NUMBER 30
JOURNAL OF BIOLOGICAL CHEMISTRY 18781

Enzymology:
Oxidation of Monolignols by Members of
the Berberine Bridge Enzyme Family
Suggests a Role in Plant Cell Wall
Metabolism
Bastian Daniel, Tea Pavkov-Keller, Barbara
Steiner, Andela Dordic, Alexander Gutmann,
Bernd Nidetzky, Christoph W. Sensen, Eric
van der Graaff, Silvia Wallner, Karl Gruber
and Peter Macheroux
J. Biol. Chem. 2015, 290:18770-18781.
doi: 10.1074/jbc.M115.659631 originally published online June 2, 2015
Access the most updated version of this article at doi: 10.1074/jbc.M115.659631
Find articles, minireviews, Reflections and Classics on similar topics on the JBC Affinity Sites.
Alerts:
• When this article is cited
• When a correction for this article is posted
Click here to choose from all of JBC's e-mail alerts
Supplemental material:
http://www.jbc.org/content/suppl/2015/06/02/M115.659631.DC1.html
This article cites 48 references, 16 of which can be accessed free at
http://www.jbc.org/content/290/30/18770.full.html#ref-list-1