The Arabidopsis phenylalanine ammonia lyase gene family: Kinetic characterization of the four PAL isoforms

Article abstract of DOI:10.1016/j.phytochem.2004.05.006

In Arabidopsis thaliana, four genes have been annotated as provisionally encoding PAL. In this study, recombinant native AtPAL1, 2, and 4 were demonstrated to be catalytically competent for L-phenylalanine deamination, whereas AtPAL3, obtained as a N-terminal His-tagged protein, was of very low activity and only detectable at high substrate concentrations. All four PALs displayed similar pH optima, but not temperature optima; AtPAL3 had a lower temperature optimum than the other three isoforms. AtPAL1, 2 and 4 had similar K_m values (64-71 μM) for L-Phe, with AtPAL2 apparently being slightly more catalytically efficacious due to decreased K_m and higher k_cat values, relative to the others. As anticipated, PAL activities with L-tyrosine were either low (AtPAL1, 2, and 4) or undetectable (AtPAL3), thereby establishing that L-Phe is the true physiological substrate. This detailed knowledge of the kinetic and functional properties of the various PAL isoforms now provides the necessary biochemical foundation required for the systematic investigation and dissection of the organization of the PAL metabolic network/gene circuitry involved in numerous aspects of phenylpropanoid metabolism in A. thaliana spanning various cell types, tissues and organs.

Full text of DOI:10.1016/j.phytochem.2004.05.006

PHYTOCHEMISTRY
Phytochemistry 65 (2004) 1557–1564
www.elsevier.com/locate/phytochem
The Arabidopsis phenylalanine ammonia lyase gene family:
kinetic characterization of the four PAL isoforms q
*
Fiona C. Cochrane, Laurence B. Davin, Norman G. Lewis
Institute of Biological Chemistry, Washington State University, Clark Hall 467, Pullman, WA 99164-6340, USA
Received 31 January 2004; received in revised form 27 April 2004
Available online 20 June 2004
Abstract
In Arabidopsis thaliana, four genes have been annotated as provisionally encoding PAL. In this study, recombinant native At-
PAL1, 2, and 4 were demonstrated to be catalytically competent for -phenylalanine deamination, whereas AtPAL3, obtained as a
L
N-terminal His-tagged protein, was of very low activity and only detectable at high substrate concentrations. All four PALs dis-
played similar pH optima, but not temperature optima; AtPAL3 had a lower temperature optimum than the other three isoforms.
AtPAL1, 2 and 4 had similar Km values (64–71 lM) for
due to decreased Km and higher kcat values, relative to the others. As anticipated, PAL activities with
(AtPAL1, 2, and 4) or undetectable (AtPAL3), thereby establishing that -Phe is the true physiological substrate. This detailed
L
-Phe, with AtPAL2 apparently being slightly more catalytically efficacious
L
-tyrosine were either low
L
knowledge of the kinetic and functional properties of the various PAL isoforms now provides the necessary biochemical foundation
required for the systematic investigation and dissection of the organization of the PAL metabolic network/gene circuitry involved in
numerous aspects of phenylpropanoid metabolism in A. thaliana spanning various cell types, tissues and organs.
Ó 2004 Elsevier Ltd. All rights reserved.
Keywords: Arabidopsis thaliana; Cruciferae; Kinetics; Phenylalanine ammonia lyase; Molecular genomics and proteomics; Recombinant protein;
Multigene families
1. Introduction
nium ion (Fig. 1); the latter compound is recycled via
action of GS/GOGAT to ultimately regenerate arogen-
ate (Razal et al., 1996; van Heerden et al., 1996; Singh
et al., 1998; Towers et al., 1998), thereby enabling phe-
nylpropanoid metabolism to continue as needed, while
cinnamic acid (3) is further metabolized to afford vari-
ous phenylpropanoids and their derivatives (Dixon,
1999; Lewis et al., 1999; Anterola et al., 2002). In
Phenylalanine ammonia lyase (PAL; E.C. 4.3.1.5) has
been extensively studied since it was first described in
1961 (Koukol and Conn, 1961). It catalyses the first step
of the phenylpropanoid pathway where L-phenylalanine
-Phe) (1) is non-oxidatively deaminated to form the
(
7,8-unsaturated trans-cinnamic acid (3) and an ammo-
L
monocots, PAL can also utilize
substrate, yielding p-coumaric acid (4) and an ammo-
L-tyrosine (L-Tyr) (2) as
Abbreviations: ABRC, Arabidopsis Biological Resource Center;
CAD, cinnamyl alcohol dehydrogenase; IMAC, immobilized metal
€
nium ion (Neish, 1961; Rosler et al., 1997). Typically,
affinity chromatography; IPTG, isopropyl b-D-thiogalactoside; pkat/
PAL is encoded by a small multigene family (Cramer
et al., 1989), which is in accordance with the first dem-
onstration of multiple PAL isoforms in Phaseolus vul-
garis (Bolwell et al., 1985). Potato, however, appears to
be an exception with more than 40 copies reported (Joos
and Hahlbrock, 1992); by contrast, Arabidopsis thaliana
(L.) Heynh has four putative PAL isoenzymes.
lg, pmoles of substrate converted to product per second per lg
protein; ORF, open reading frame; PAL, phenylalanine ammonia
lyase; RT-PCR, reverse transcription-polymerase chain reaction;
UTR, untranslated region.
q Data Deposition: All sequences reported in this paper have been
deposited in the GenBank database [Accession Nos. AY303128
(AtPAL1), AY303129 (AtPAL2), AY528562 (AtPAL3) and
AY303130 (AtPAL4)].
* Corresponding author. Tel.: +1-509-335-8382; fax: +1-509-335-8206.
E-mail address: lewisn@wsu.edu (N.G. Lewis).
The extensive in vitro biochemical characterization of
PAL has been carried out for several plant species, such
0031-9422/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.phytochem.2004.05.006

1558
F.C. Cochrane et al. / Phytochemistry 65 (2004) 1557–1564
9
chloride) and pathogen attack (Ohl et al., 1990; Mauch-
–
–
COO
COO
Mani and Slusarenko, 1996; Rookes and Cahill, 2003).
In this study, all four putative Arabidopsis pal genes
were cloned, with the corresponding recombinant pro-
teins expressed, purified to apparent homogeneity, and
characterized with regard to kinetic and physical pa-
rameters. This baseline biochemical information is es-
sential prior to establishing if and how the proposed
PAL metabolic networks are functionally organized at
the cellular/cell type levels during all stages of growth
and development, including identification of the down-
stream metabolic processes involved and the nature of
the defense responses, in A. thaliana.
8
+
7
NH3
1
4
6
2
3
5
+
NH4
R
L
R = OH,
R
R = H,
-Phenylalanine (1)
-Tyrosine (2)
R = H, trans-Cinnamic acid (3)
R = OH, p-Coumaric acid (4)
L
Fig. 1. PAL and TAL conversions, showing substrates and products.
as parsley (Petroselinum crispum) (Appert et al., 1994),
but not for the crucifer Arabidopsis. Accordingly, this
particular study was directed to define unambiguously
both the biochemical functions and kinetic parameters
of the proteins encoded by the putative pal genes, as part
of a National Science Foundation (USA) initiative ul-
timately aimed at establishing the true physiological
function, or functions, of each Arabidopsis gene by 2010.
From a physiological perspective, PAL is expressed
constitutively in Arabidopsis and can also be induced
upon exposure to various stresses. Data from public
websites such as MPSS (www.mpss.udel.edu) and TAIR
(www.arabidopsis.org) indicate that Atpal1 and 2 ex-
pression patterns are very similar for various tissues, e.g.
callus, inflorescence and roots, but give no definitive
indication of physiological function and the metabolic
processes involved. As well, expression levels of Atpal1
and 2 are higher than that of Atpal4 while Atpal3 is
barely detectable. In A. thaliana, Atpal1 has been pro-
posed as involved in lignification (Ohl et al., 1990;
Mauch-Mani and Slusarenko, 1996; Rookes and Cahill,
2003) although available EST data does not fully sup-
port this hypothesis (Costa et al., 2003); however, this
latter interpretation is cautionary, because of limited
sampling of the aerial (vascular) plant tissue used to
create the EST data as well as the limited number of
ESTs obtained. Based on analysis of the same EST data,
it would appear that either AtPAL2 is as likely a can-
didate for lignification or that (more likely) overlapping
PAL metabolic networks are operative in vascular tis-
sues (e.g. using both AtPAL1 and 2) (Costa et al., 2003).
The latter would assure functional capability even if one
pal gene were impaired in its expression/catalytic prop-
erties. Such overlapping metabolic networks leading to
functional redundancy have been noted, for example,
with the cinnamyl alcohol dehydrogenase (CAD) family
during Arabidopsis stem vasculature lignification (Kim
et al., 2004). By contrast, Atpal3 appears to be expressed
mainly in roots and leaves, albeit at low levels (Wanner
et al., 1995; Raes et al., 2003), while Atpal4 is more
abundant in developing seed tissue (Costa et al., 2003;
Raes et al., 2003). Of the four isoenzymes, Atpal1 and 2
are preferentially induced upon treatment with factors
such as light, wounding, heavy metals (e.g. mercuric
2. Results and discussion
The four putative pal genes in Arabidopsis encode
proteins that share P 74% identity and P 78% amino
acid similarity. ClustalW analysis of the four PAL ge-
nomic sequences (i.e. 50, 30 UTR regions, ORF including
introns) showed that Atpal1 and 2 and Atpal3 and 4
form two distinct groups (data not shown), in agreement
with the phylogenetic analysis carried out by Raes et al.
(2003). This 2 + 2 pattern most likely arose from dupli-
cation as AtPAL1 and 2 share 90.4% amino acid identity
while AtPAL3 and 4 have 83.5% identity. Intron num-
ber also differs between the two groups (Wanner et al.,
1995) with Atpal1 and 2 possessing a single intron
whereas Atpal3 and 4 have two, which may indicate
differential post-transcriptional processing for the two
groups. While the significance of chromosome location
is as yet unknown, the four PAL genes are distributed
over three of the five Arabidopsis chromosomes, chro-
mosome II (Atpal1), chromosome III (Atpal2 and 4) and
chromosome V (Atpal3). In the latter case, Atpal3 is
located near the telomere region of chromosome V.
Interestingly, Atpal3 possesses a non-consensus splice
site at the 30 end of the second exon (Wanner et al.,
1995) and, as a result, two distinct amino acid sequences
have been reported, with one possessing four additional
amino acids i.e. 337DRYVLGYALRTS versus
337DRYALRTS. In this context, we are proposing that
this alternative gene splicing may arise as a result of
different stresses experienced (Bournay et al., 1996;
Marrs and Walbot, 1997). ClustalW alignment of all
known PAL isoforms from various other plant species,
however, revealed no other examples of additional
amino acids in any of the other sequences reported
(http://www.ncbi.nih.gov). Indeed, in our hands, the
AtPAL3 clone isolated encoded a sequence lacking all
four ‘extra’ amino acids.
Gene specific primers (Table 1) were designed for
each Atpal cDNA sequence for PCR amplification of the
individual target genes. To obtain each PAL clone (ex-
cept Atpal1 which was provided by the Arabidopsis Bi-

F.C. Cochrane et al. / Phytochemistry 65 (2004) 1557–1564
1559
Table 1
Gene specific primers for amplification of Arabidopsis PAL genes Atpal1– 4
Gene locus
At2g37040
Gene size (bp)
2178
Gene specific primers
Atpal1
Atpal2
Atpal3
Atpal4
FOR 50 ATGGAGATTAACGGGGCACA
REV 50 TTAACATATTGGAATGGGAGCTCC
FOR 50 ATGGATCAAATCGAAGCAATG
REV 50 TTAGCAAATCGGAATCGGAGC
FOR 50 ATGGAGTTTCGTCAACCAAAC
REV 50 TTAGCAGATAGAAATCGGAGCA
FOR 50 ATGGAGCTATGCAATCAAAAC
REV 50 TCAACAGATTGAAACCGGAGC
At3g53260
At5g04230
At3g10340
2154
2085
2124
ological Resource Center, ABRC), total RNA was iso-
lated from 8-day-old Arabidopsis seedlings with RNA
used to synthesize first strand cDNA, this in turn being
required to amplify Atpal2, 3, and 4 individually. Atpal1
cDNA, on the other hand, was amplified using purified
plasmid #U10120 as template. Initially, genes encoding
AtPAL2 and 4 were cloned into pTrcHis TOPOÒ ex-
pression vectors; however, only low protein yields of
AtPAL2 and 4 were obtained (as estimated by SDS–
PAGE, data not shown), in spite of a wide variety of
IPTG concentrations tested (0.1–1 mM) at different in-
cubation temperatures (20–37 °C) and times (6–24 h).
Accordingly, we next employed the pET Blue-1 expres-
sion vector system to establish if successful overexpres-
sion of the putative AtPAL isoforms in recombinant
native form could be obtained instead. Optimal condi-
tions for good overexpression levels of AtPAL1, 2, and 4
(e.g. 2 mg purified AtPAL2 from 2 l culture) were
identified by varying temperature (20–30 °C), time (4–20
h) and IPTG concentration (0.1–1 mM). Best conditions
found were 22 °C, 14–16 h and 0.35–0.5 mM IPTG; by
contrast, none of these conditions resulted in AtPAL3
overexpression.
Fig. 2. SDS–PAGE showing steps involved in purification of AtPAL2
from induced (0.5 mM IPTG) E. coli cell-free extracts (lysate) with
detection by silver staining: crude cell lysate (lane 1) and purity level
following acetone precipitation (lane 2), first ultrafiltration (lane 3),
anion-exchange column (POROS 20HQ) (lane 4), second ultrafiltra-
tion (lane 5) and gel filtration (lane 6) showing both purified PAL
(ꢀ77.5 kDa) and traces of PAL aggregate (ꢀ160 kDa); protein stan-
dards (lane 7).
the known ability of PAL to aggregate (Parkhurst and
Hodgins, 1971; Schopfer, 1971). SDS–PAGE estima-
tions of apparent molecular masses of each PAL
monomer were in close agreement with values calculated
for each isoform (78.7, 77.6 and 76.9 kDa for AtPAL1, 2
and 4, respectively).
Initial assays using crude cell lysate to estimate PAL
L
activities utilized
-[U-14C]Phe (1; 1 mM, 4 lCi/mmol)
with product analysis (i.e. for [U-14C]cinnamic acid (3))
involving detection and quantification by HPLC anal-
yses followed by radioactivity measurements. For At-
PAL1, 2, and 4, PAL activities were readily detectable
after 20 min incubation, whereas with AtPAL3, even
overnight assays did not yield detectable radioactive [U-
14C]cinnamic acid (3) (data not shown).
Since the expression of AtPAL3 as a native enzyme
was not detectable above basal levels (as revealed by
SDS–PAGE analyses and radioactive assays with radi-
olabeled
L
-[U-14C]Phe (1), data not shown), we next
investigated as to whether a functional AtPAL3 coupled
to a His6-tag could be obtained. Using the pTrc-His
TOPOÒ vector to introduce an in-frame His6-tag at the
N-terminus, the clone was transformed into Tuner cells
and AtPAL3 formation was most efficiently induced at 1
mM IPTG (22 °C, 20 h). Soluble protein was purified
using immobilized metal affinity chromatography
(IMAC) to apparent homogeneity (as estimated by
SDS–PAGE, data not shown), this giving circa 0.6 mg
purified AtPAL3 from 2 l culture. Its apparent molec-
ular mass (with the attached His6-tag of ꢀ3–4 kDa) was
ꢀ 79.1 kDa, as determined by SDS–PAGE, this being
the expected size since the calculated mass of the native
protein is ꢀ76.2 kDa.
AtPAL1, 2, and 4 were next individually purified
using a combination of acetone precipitation, ultrafil-
tration, anion exchange (POROS 20HQ) and gel filtra-
tion (Sephacryl S-300) chromatographic steps. A small
protein (ꢀ28 kDa) contaminant was initially present
after the first gel filtration step (as revealed by SDS–
PAGE analysis), this being removed by repetition of the
preceding gel filtration step. In all cases, this yielded
essentially homogenous PAL as visualized by silver
staining (Fig. 2, AtPAL2 as example; lane 6), but with
trace amounts of a larger protein (ꢀ160 kDa) being
detected in the final purification step; this results from

1560
F.C. Cochrane et al. / Phytochemistry 65 (2004) 1557–1564
Table 2
Kinetic and physical properties of recombinant Arabidopsis PAL1–4
L
-Phe (1)
L-Tyr (2)
Km (lM)
Vmax (pkat/lg
protein)
kcat (sꢁ1
)
k
cat=Km
pH opti-
mum
Temperature
optimum (°C)
kcat=Km
(Mꢁ1 sꢁ1
)
(Mꢁ1 sꢁ1
)
AtPAL1
AtPAL2
AtPAL3a
AtPAL4
68 ꢂ 5
64 ꢂ 3.5
2560 ꢂ 340
71 ꢂ 3
5.5 ꢂ 0.03
10.5 ꢂ 0.05
0.4 ꢂ 0.03
9.9 ꢂ 0.04
1.8
3.2
0.1
3.0
25,500
51,200
50
8.4–9.2
8.4–8.9
8.7–8.9
8.4–8.8
46–48
48
31
75 ꢂ 6
40 ꢂ 12
n.d.
43,100
46–48
44 ꢂ 1
n.d. ¼ not determined.
a Note: AtPAL3 overexpression is as the N-terminal His6-tagged recombinant derivative, whereas AtPAL1, 2, and 4 are overexpressed as re-
combinant native proteins.
All products of enzyme assays, with either L-Phe
(1) or -Tyr (2) as potential substrates, were verified
L
by HPLC analyses, using UV, mass spectroscopy
(atmospheric pressure chemical ionization, APCI) and
radiochemical detection systems as needed. AtPAL1,
2, and 4 each displayed very similar properties for
L-
Phe (1) deamination, as regards to the temperature
and pH optima (Table 2). The temperature optima
(46–48 °C) obtained for each isoform were quite
comparable to native PALs isolated from other
Brassicaceae members, such as Brassica campestris and
B. juncea (both 45 °C; www.brenda.uni-koeln.de and
references therein). The pH optima, however, ranged
from 8.4 to 8.8 (AtPAL4) to 8.9 (AtPAL2) and to 9.2
(AtPAL1), these values being within the ranges known
for PAL from Hordeum vulgare (Koukol and Conn,
1961), Sinapis alba (Gupta and Acton, 1979) and
Petroselinum crispum (Appert et al., 1994), respec-
tively. Additionally, although AtPAL3 had a similar
pH optimum (8.7–8.9), its temperature optimum (31
°C) was lower than that of the other isoforms, an
observation which might possibly result from inter-
ference by the attached N-terminal His6-tag with
packing of the native tetramer form.
Kinetic parameters of AtPAL1, 2 and 4 were deter-
mined (Table 2) with Hanes plots being linear (Fig. 3)
indicating that the three homologues followed standard
Michaelis–Menten kinetics; calculated kinetic parame-
ters gave values of apparent Km (64–71 lM), Vmax (5.5–
10.5 pkat/lg protein) and kcat (1.8–3.2 sꢁ1), respectively,
for -Phe (1). The Km values were thus similar to those
described for Glycine max (Hanson and Havir, 1981)
and Medicago sativa (Jorrin and Dixon, 1990) PAL
L
Fig. 3. Hanes plots for L-Phe (1) with AtPAL1 (a), AtPAL2 (b) and
AtPAL4 (c). [Phe]/velocity is mM lg protein pkatꢁ1
.
(both 70 lM); whereas apparent kcat values (1.8–3.2 sꢁ1
were roughly comparable (only 3–5-fold lower) to maize
)
decrease in Km (AtPAL2 only) and an increase in kcat
(AtPAL2 and 4), respectively. Similar kinetic properties
amongst isoenzymes of high homology were also ob-
PAL (10.6 sꢁ1with
L-Phe (1) at pH 8.7; (Rosler et al.,
€
1997)). As well, the AtPAL1, 2, and 4 kcat/Km values
(25,000–51,000 Mꢁ1 sꢁ1) were within range of the kcat
/
;
served for the four parsley PAL proteins (Km’s [L-Phe]
Km value determined for maize PAL (39,000 Mꢁ1 sꢁ1
from 15–25 lM) (Appert et al., 1994), these sharing very
high identity (94.7–99.4%) and possessing nearly indis-
tinguishable catalytic properties. While this correlation
of high sequence similarity and comparable kinetic
properties has been suggested to arise from relatively
€
(Rosler et al., 1997)). Comparison of relative enzyme
efficacies also showed that AtPAL2 and AtPAL4, when
compared with AtPAL1, were catalytically very slightly
more efficacious, this being a consequence of a small

F.C. Cochrane et al. / Phytochemistry 65 (2004) 1557–1564
1561
recent divergence of the gene within Petroselinum (Ku-
mar and Ellis, 2001), it cannot be extended to Arabid-
opsis where AtPAL4 displays similar kinetic properties
to AtPAL1 and 2 yet does not cluster with these two
proteins (see above). In contrast to AtPAL1, 2 and 4, the
N-terminal His6-tagged AtPAL3 was of very low cata-
poor level of understanding. Indeed, many studies are
carried out at the crude whole tissue level and thus
neither address nor give good insight into such aspects:
for example, Raes et al. (2003) reported, using RT-PCR,
that in Arabidopsis Atpal1, 2, and 4 were present in stem
tissues, with Atpal1 and 2 transcripts apparently in-
creasing during development; however, neither the
metabolic significance of this is known, nor the cellular
specificities of the tissues involved. Furthermore, the
analysis of Atpal3 transcripts has given mixed results;
Wanner et al. (1995) only detected Atpal3 in 10-day-old
Arabidopsis seedlings and in roots of mature (4–6-weeks
old) plants, whereas Mizutani et al. (1997) reported that
AtPAL3 mRNA could not be detected in roots of 3-
week-old plants; Raes et al. (2003) have since proposed
(using RT-PCR), that AtPAL3 is expressed in all tissues
but mainly in the leaves.
Accordingly, having now systematically established
the catalytic functions of each of the recombinant PAL
isoforms in vitro as bona fide PALs, current work is
underway using Arabidopsis transformants harboring
the individual Atpal promoter::GUS constructs. This, in
turn, should permit the unequivocal identification of
temporal and spatial patterns of expression of the genes
encoding each PAL isoform in the specific cell types
during Arabidopsis growth and development. Of partic-
ular interest is the extent to which the PAL genes are
co-expressed in various cell types thereby ensuring
functional redundancy, as well as delineation of the
precise involvement of each gene in various downstream
metabolic processes. These studies, together with analy-
sis of specific PAL isoform ‘‘knockouts’’ and application
of RNAi technology (to silence one and/or all of the four
PAL isoforms) should provide definitive insight into not
only the physiological function of each AtPAL isoform
in vivo, but also how the proposed PAL metabolic net-
works/gene circuits are actually organized in vivo.
lytic activity with
Michaelis–Menten plot reaching saturation by ꢀ3.5 mM
-Phe (1). In this case, assays required both increased
L-Phe (1) as substrate, with its
L
protein amounts ( P 50 lg versus ꢀ0.5–1 lg for the
other AtPAL isoforms) and an extended incubation
period (reaction was still linear at 20 h). Indeed, At-
PAL3 was estimated to have a catalytic efficacy 500–
1000-fold lower than that of the other PAL isoforms,
this resulting from its higher Km (2.6 mM) and its very
low kcat (0.1 sꢁ1) values, respectively. This overall low
efficacy in activity may result from either the enzyme
being catalytically defective or, as indicated earlier, from
introduction of the N-terminal His6-tag, and thus, may
not be a reflection of the true catalytic properties of
native AtPAL3.
Additionally, it was deemed instructive to examine
whether each of the AtPALs (1–4) were capable of
converting
PAL can utilize both
al., 1997), albeit with
L
-Tyr (2) to p-coumaric acid (4), given that
€
L-Tyr (2) (Rosler et
-Phe (1) being preferred. How-
L
-Phe (1) and
L
ever, assays carried out to determine kinetic parameters
did not reach enzyme saturation due to solubility limi-
tations with
ꢀ10) was required for
a limited number of substrate concentrations (up to 13
mM -Tyr (2)) were possible. For the AtPAL1, 2, and 4
homologues, enzyme efficacy with -Tyr (2) was greatly
reduced as compared to -Phe (1) deamination [340-
L
-Tyr (2). Indeed, a high alkaline pH (pH
L
-Tyr (2) solubilization and only
L
L
L
(AtPAL1), 1200- (AtPAL2) and 970- (AtPAL4) fold
decreases, respectively], whereas AtPAL3 was essentially
inactive. This finding was not unexpected since dicoty-
ledons such as Arabidopsis are not known to effectively
employ L-Tyr (2) as an alternative starting point for
phenylpropanoid biosynthesis; this result is thus similar
to that of Appert et al. (1994) who determined that
3. Experimental
enzyme efficiency of parsley PAL with
decreased by four orders of magnitude as compared to
L
-Tyr (2) was
3.1. Materials
L
-Phe (1).
The establishment of multiple bona fide PAL genes in
All solvents and chemicals were of reagent or high
performance liquid chromatography (HPLC) grade un-
less otherwise specified. Cinnamic (3) and p-coumaric (4)
Arabidopsis now raises several important questions, in
particular concerning the true physiological function of
each PAL isoform. This would include establishing to
what extent functionally redundant metabolic networks/
gene circuits exist for the PAL family, particularly since
most research reported has focused only on Atpal1
leaving Atpal2, 3, and 4 largely unexplored. In this
context, our current understanding of the nature of the
metabolic networks leading ultimately to not only dif-
ferent metabolic products, but also involving overlap-
ping networks in specific cell types, is still at a relatively
acids as well as
from Sigma, whereas
L
-Phe (1) and
L
L-Tyr (2) were purchased
-[U-14C]Phe (1) (496 mCi/mmol)
was from Perkin–Elmer Life Sciences, Inc. RNeasyÒ
Plant Mini and QIAquick Gel Extraction Kits were
from Qiagen, while SuperscriptÔ First-Strand Synthesis
System for RT-PCR and pTrcHis TOPOÒ TA Expres-
sion Kits were purchased from Invitrogen. A WizardÒ
Plus SV Miniprep kit was obtained from Promega,
whereas pETBlueÔ-1 vector, Escherichia coli Tuner
(DE3)pLacI cells, BugbusterÒ Protein Extraction Re-

1562
F.C. Cochrane et al. / Phytochemistry 65 (2004) 1557–1564
agent and BenzonaseÒ were from Novagen and Expand
HiFi polymerase was from Roche. Custom oligonu-
cleotide primers for polymerase chain reaction (PCR)
and sequencing were synthesized by Invitrogen. DNA
concentrations were approximately determined by
comparison to a low DNA mass ladder (Invitrogen) in
0.8% agarose gels. CentriconÒ Plus-80 and Plus-20 mi-
crofugal filters, as well as 100,000 MWCO (YM100)
filtration membranes were purchased from Amicon,
Millipore. POROS 20HQ and POROS 20MC resins
were from PerSeptive Biosystems whereas Sephacryl
S300 was from Amersham Biosciences.
tension for 2.2 min at 72 °C. The final extension time
was 10 min at 72 °C. PCR products (ꢀ2 kb) were gel-
purified using a QIAquick Gel Extraction Kit (accord-
ing to manufacturer’s instructions) and cloned into
pETBlueÔ-1 vectors. Additionally, Atpal2, 3, and 4
ORFs were individually cloned into pTrcHis TOPOÒ
vectors. Both the orientation and the sequence of the
inserts were checked before transformation of E. coli
Tuner (DE3)pLacI for expression.
3.4. Overexpression and purification of recombinant PAL
A single colony from each transformed E. coli cell line
was incubated in 1 ml Luria-Bertani (LB) broth sup-
plemented with carbenicillin (0.1 mg/ml) for 12 h at 37
°C. Inocula were then individually added to 500 ml LB
containing carbenicillin (0.1 mg/ml) and grown at 37 °C
until an OD600 between 0.5 and 0.9 was reached at which
3.2. Instrumentation
PCR amplifications used an AmplitronÒII Thermal
cycler (Barnstead/Thermolyne Coop.), whereas RNA
and protein measurements utilized a Lambda 6 UV–vis-
ible spectrophotometer (Perkin–Elmer). For protein pu-
rification, both anion exchange and IMAC steps
employed a BioCAD (Perseptive Biosystems), while a
Pharmacia LKB FPLC system equipped with a Monitor
UV-M unit was used for gel filtration. An Amicon stirred
cell (Cell Model No. 8050) with reservoir (RC800) was
used for ultrafiltration. Radioactive samples were ana-
lyzed in Biodegradable Counting Scintillant (BSC,
Amersham Biosciences) and counted using a Liquid
Scintillation Analyzer (Tri-Carb 2100TR; Packard). Re-
versed-phase HPLC analyses were carried out using an
HPLC-PDA system that consisted of a WatersÔ 600
pump Controller, WatersÔ 717 plus autosampler and
WatersÔ 996 photodiode array detector, all controlled by
Millennium 2.1 software. As well, an APCI MSn, a LCQ
mass spectrometer (Thermo-Finnigan), equipped with
atmospheric pressure chemical ionization and an ion
trap, were used to verify enzyme assay products.
time isopropyl thio-b-D-galactoside (IPTG) was added,
with PAL expression for each cell line being induced
with 0.35 (AtPAL1), 0.5 (AtPAL2 and AtPAL4) or 1
mM (AtPAL3) IPTG, respectively. Cells containing
AtPALs 1, 2 and 4 were grown at 22 °C for 14–16 h,
while those harboring AtPAL3 were induced at 22 °C
for 20 h. Cells were individually harvested by centrifu-
gation at 3000g for 20 min and stored frozen at )80 °C
until required. Pellets were thawed at room temperature
for 5 min, suspended in a 1:1 mixture of 20 mM Tris–
HCl (pH 7.5; Buffer A) and BugBusterÔ Protein Ex-
traction reagent, then sonicated 10 ꢃ 15 s with 1 min
intervals while on ice. Cellular debris from each cell line
was removed by centrifugation at 15,000g for 30 min.
Supernatants containing AtPAL1, 2 and 4 were, re-
spectively, subjected to acetone precipitation, with the
25–55% acetone saturation fraction (also in the 25–55%
ammonium sulphate saturation range; data not shown)
centrifuged at 15,000g for 30 min. Each resulting pellet
was next resuspended in Buffer A and filtered through a
100,000 MWCO (YM100) membrane using an Amicon
stirred cell. Each retentate was then applied to a POROS
20HQ anion exchange column (100 mm ꢃ 4.6 mm; 10
ml/min flow rate) pre-equilibrated in Buffer A. After
washing the column with 15 ml Buffer A, each PAL was
individually eluted with a linear gradient from 0 to 1 M
NaCl in Buffer A with 1 ml fractions collected; PAL
isoforms were eluted with 0.37 M NaCl. Fractions
containing each PAL were then individually pooled,
desalted by ultrafiltration (YM100 membrane), further
concentrated using a CentriconÒ Plus-80 microfugal
filter and loaded onto a Sephacryl S-300 gel filtration
column (60 cm ꢃ 1.6 cm) pre-equilibrated in Buffer B
(Buffer A containing 200 mM NaCl) at a flow rate of 0.5
ml/min. Fractions (1 ml) containing each PAL isoform,
eluted with 48–58 ml of mobile phase, were pooled,
concentrated and reapplied to the column under the
same conditions. The resulting PAL fractions were in-
3.3. Cloning of PAL cDNAs
The Atpal 1 clone (#U10120) was obtained as a stab
culture from the ABRC (Ohio State University, Co-
lumbus, Ohio), with plasmid purified from a 5 ml
overnight culture, whereas total RNA (to obtain Atpal2,
3, and 4) was extracted from 8-day-old A. thaliana (L.)
Heynh. (ecotype Col.) seedlings using the RNeasyÒ
Plant Mini Kit. For RT-PCR, the RNA was reversely
transcribed by SuperscriptÔ First-Strand Synthesis
System using 5 lg for each transcription and random
hexamers for priming. To amplify the pal ORFs 5 lg
cDNA (or 80 ng plasmid) was used as template with 10
pmol of each gene-specific primer (Table 1, based on
gene sequences available through NCBI; www.ncbi.nlm.
nih.gov), 0.2 mM dNTPs, 2.5 mM MgCl2 and 1.25 units
of Expand HiFi polymerase in 25 ll reactions. PCR
conditions were: hot start (94 °C, 3 min), denaturation
94 °C for 1 min, annealing for 2 min at 45 °C and ex-

F.C. Cochrane et al. / Phytochemistry 65 (2004) 1557–1564
1563
dividually combined and concentrated using Centricon
Plus-20 concentrators. Purification levels of AtPALs 1, 2
and 4 were analyzed by SDS–PAGE using Laemmli’s
buffer system under denaturing and reducing conditions
(Laemmli, 1970). Each PAL was separated on a 4–20%
gradient gel and visualized by silver staining (Amersham
Bioscience Application Note).
above. Again, for AtPAL3, an increased protein amount
(50 lg) was used, with the assay time also extended to 15
h; assays were carried out at 31 °C. For all data points,
including both pH and temperature optima determina-
tions, controls either without enzyme or denatured
protein (10 min, 100 °C) were performed. All assays
were carried out in triplicate.
Supernatant from sonicated and pelleted E. coli
containing His-tagged AtPAL3 was filtered (0.45 lm
syringe filter AcrodiscÒ, Gelman Laboratory) prior to
purification by IMAC. Isolation of AtPAL3 was carried
out at room temperature using a POROS 20MC column
(100 mm ꢃ 4.6 mm) pre-equilibrated with Buffer C (20
mM Tris–HCl, pH 7.9; 500 mM NaCl; 5 mM imidazole)
with a flow rate of 10 ml/min. Aliquots (1.5 ml) of fil-
trate containing AtPAL3 were applied to the column
and, after washing with Buffer D (20 mM Tris–HCl, pH
7.9; 500 mM NaCl; 50 mM imidazole) to remove non-
specific proteins, AtPAL3 was eluted with a linear gra-
dient from Buffer D to Buffer E (20 mM Tris–HCl, pH
7.9; 500 mM NaCl; 500 mM imidazole) with 1 ml
fractions being collected; recombinant AtPAL3 was
eluted with 280 mM imidazole. Aliquots (12 ll) of every
other fraction were analyzed by SDS–PAGE as de-
scribed above. Fractions containing His-tagged AtPAL3
were pooled and concentrated using CentriconÒ Plus-80
concentrators, diluted with 20 mM Tris–HCl (pH 7.5)
and reconcentrated.
3.6. PAL assays – kinetic parameters
Enzyme assays for AtPAL1, 2 and 4 were performed at
37 °C with a 2 min pre-incubation time (whereas AtPAL3
assays were carried out at 31 °C) in a temperature-con-
trolled waterbath. For kinetic analyses, 16 concentrations
of -Phe (1) (43 lM to 2 mM) were used for AtPAL1, 2
L
and 4, while for AtPAL3 they were between 0.25 and 4
mM. Each assay (250 ll) contained 100 mM Bis–Tris
Propane (pH 8.4, AtPAL1, 2, and 4; pH 8.8 AtPAL3),
purified enzyme (0.39 lg AtPAL1, 0.92 lg AtPAL2, 50 lg
AtPAL3 and 0.47 lg AtPAL4) and various amounts of L-
Phe (1), respectively. Assays were initiated by addition of
substrate, incubated for 20 min (AtPAL3, 15 h) and
stopped by acidification (50 ll of glacial AcOH). An ali-
quot (80 ll) of each assay mixture was then subjected to
HPLC analysis as described below. Km and Vmax values for
AtPAL1, 2 and 4 for L-Phe (1) were obtained from Hanes
plots, while Michaelis–Menten plots were used to deter-
mine AtPAL3 kinetic parameters. For Km determinations
using L-Tyr (2), concentrations of 0.55–13 mM were used,
All protein concentrations were determined by the
Bradford (1976) method using BSA as standard.
with each assay (250 ll) containing 100 mM Bis–Tris
Propane (pH 8.4), purified enzyme (4.87 lg AtPAL1; 5.40
lg AtPAL2 and 3.92 lg AtPAL4) and various amounts of
3.5. Radiochemical PAL assays – temperature and pH
L
-Tyr (2). Incubation times were either 30 min (AtPAL4)
or 1 h (AtPAL1 and 2). kcat/Km values were determined
from Michaelis–Menten plots.
Temperature and pH optima were individually de-
termined for each PAL protein. For temperature op-
tima, incubations were carried out at a constant pH of
8.2, while varying the temperature (14–70 °C). The assay
mixture consisted of Bis–Tris Propane (100 mM, 100 ll,
pH 8.2), 10 lg of PAL (post-POROS 20HQ column;
3.7. Reversed-phase HPLC separation of PAL assays
Cinnamic acid (3) detection was accomplished using a
reversed-phase column (Nova-PakÒ C18, 3.9 ꢃ 150 mm,
Waters) eluted with a 3% AcOH in H2O (v/v; Solvent A)
and CH3CN (Solvent B) gradient at a flow rate of 1 ml/
min with detection at 278 nm. The column was pre-
equilibrated in A:B (23:2), which was held for 2 min
after introduction of the sample with a linear gradient to
A:B (1:19) applied over 20 min, this being held for 10
min. The gradient was returned to A:B (23:2) over 2 min
and held at this ratio for another 18 min for column
re-equilibration. The same system was used to separate
p-coumaric acid (4) with detection at 310 nm.
AtPAL1, 2, and 4), 20 mM Tris–HCl (pH 7.5) and L-[U-
14C]Phe (1 mM, 4 lCi/mmol) in a total volume of 250 ll.
Protein and buffers were pre-incubated for 2 min at each
temperature prior to assay initiation by addition of
substrate. After 20 min, assays were stopped by addition
of glacial AcOH (10 ll), with cinnamic acid (3; 100
nmol) added as radiochemical carrier before assay
mixtures were extracted with EtOAc (2 ꢃ 200 ll). The
combined EtOAc solubles were washed with distilled
water (1 ꢃ 200 ll), with an aliquot (200 ll) of the EtOAc
layer removed for liquid scintillation counting. Identical
assays were carried out for AtPAL3 but with larger
amounts of purified protein (50 lg) used and with the
incubation times extended to 15 h. For pH optima, as-
says were performed at 37 °C with Bis–Tris Propane
(100 mM; pH 6.5–9.2), TAPS (100 mM; pH 8–9.05) or
CAPSO (100 mM; pH 9.2–10.1) buffers as described
Acknowledgements
The authors thank the US National Science Foun-
dation Arabidopsis 2010 Project Grant MCB-0117260,

1564
F.C. Cochrane et al. / Phytochemistry 65 (2004) 1557–1564
Koukol, J., Conn, E.E., 1961. The metabolism of aromatic compounds
in higher plants. IV. Purification and properties of the phenylal-
anine deaminase of Hordeum vulgare. Journal of Biological
Chemistry 236, 2692–2698.
the National Aeronautics and Space Administration
(NAG 2-1513) and the G. Thomas and Anita Hargrove
Center for Plant Genomic Research for generous
support.
Kumar, A., Ellis, B.E., 2001. The phenylalanine ammonia-lyase gene
family in raspberry. Structure, expression, and evolution. Plant
Physiology 127, 230–239.
References
Laemmli, U.K., 1970. Cleavage of structural proteins during the
assembly of the head of bacteriophage T4. Nature 277,
680–685.
Anterola, A.M., Jeon, J.-H., Davin, L.B., Lewis, N.G., 2002. Tran-
scriptional control of monolignol biosynthesis in Pinus taeda:
factors affecting monolignol ratios and carbon allocation in
phenylpropanoid metabolism. Journal of Biological Chemistry
277, 18272–18280.
Lewis, N.G., Davin, L.B., Sarkanen, S., 1999. The nature and function
of lignins. In: Barton, Sir D.H.R., Nakanishi, K., Meth-Cohn, O.
(Eds.), Comprehensive Natural Products Chemistry, vol. 3. Else-
vier, Oxford, pp. 617–745.
Marrs, K.A., Walbot, V., 1997. Expression and RNA splicing of the
maize glutathione S-transferase Bronze2 gene is regulated by
cadmium and other stresses. Plant Physiology 113, 93–102.
Mauch-Mani, B., Slusarenko, A.J., 1996. Production of salicylic acid
precursors is a major function of phenylalanine ammonia-lyase in
the resistance of Arabidopsis to Peronospora parasitica. Plant Cell 8,
203–212.
Appert, C., Logemann, E., Hahlbrock, K., Schmid, J., Amrhein, N.,
1994. Structural and catalytic properties of the four phenylalanine
ammonia-lyase isoenzymes from parsley (Petroselinum crispum
Nym.). European Journal of Biochemistry 225, 491–499.
Bolwell, G.P., Bell, J.N., Cramer, C.L., Schuch, W., Lamb, C.J.,
Dixon, R.A., 1985. Phenylalanine ammonia-lyase from Phaseolus
vulgaris. Characterisation and differential induction of multiple
forms from elicitor-treated cell suspension cultures. European
Journal of Biochemistry 149, 411–419.
Mizutani, M., Ohta, D., Sato, R., 1997. Isolation of a cDNA and a
genomic clone encoding cinnamate 4-hydroxylase from Arabidopsis
and its expression manner in planta. Plant Physiology 113, 755–
763.
Bournay, A.-S., Hedley, P.E., Maddison, A., Waugh, R., Machray,
G.C., 1996. Exon skipping induced by cold stress in a potato
invertase gene transcript. Nucleic Acids Research 24, 2347–2351.
Neish, A.C., 1961. Formation of m- and p-coumaric acids by
enzymatic deamination of the corresponding isomers of tyrosine.
Phytochemistry 1, 1–24.
Bradford, M.M., 1976.
A rapid and sensitive method for the
quantitation of microgram quantities of protein utilizing the
principle of protein-dye binding. Analytical Biochemistry 72,
248–254.
Ohl, S., Hedrick, S.A., Chory, J., Lamb, C.J., 1990. Functional
properties of a phenylalanine ammonia-lyase promoter from
Arabidopsis. Plant Cell 2, 837–848.
Costa, M.A., Collins, R.E., Anterola, A.M., Cochrane, F.C., Davin,
L.B., Lewis, N.G., 2003. An in silico assessment of gene function
and organization of the phenylpropanoid pathway metabolic
networks in Arabidopsis thaliana and limitations thereof. Phyto-
chemistry 64, 1097–1112.
Parkhurst, J.R., Hodgins, D.S., 1971. Phenylalanine and tyrosine
ammonia-lyase activity in Sporobolomyces pararoseus. Phytochem-
istry 10, 2997–3000.
Raes, J., Rohde, A., Holst Christensen, J., Van de Peer, Y., Boerjan,
W., 2003. Genome-wide characterization of the lignification
toolbox in Arabidopsis. Plant Physiology 133, 1051–1071.
Razal, R.A., Ellis, S., Singh, S., Lewis, N.G., Towers, G.H.N., 1996.
Nitrogen recycling in phenylpropanoid metabolism. Phytochemis-
try 41, 31–35.
Cramer, C.L., Edwards, K., Dron, M., Liang, X., Dildine, S.L.,
Bolwell, G.P., Dixon, R.A., Lamb, C.J., Schuch, W., 1989.
Phenylalanine ammonia-lyase gene organization and structure.
Plant Molecular Biology 12, 367–383.
Dixon, R.A., 1999. Isoflavonoids: biochemistry, molecular biology,
and biological functions. In: Barton, Sir D.H.R., Nakanishi, K.,
Meth-Cohn, O. (Eds.), Comprehensive Natural Products Chemis-
try, vol. 1. Elsevier, Oxford, pp. 773–823.
Rookes, J.E., Cahill, D.M., 2003. A PAL1 gene promoter-green
fluorescent protein reporter system to analyse defence responses in
live cells of Arabidopsis thaliana. European Journal of Plant
Pathology 109, 83–94.
Gupta, S., Acton, G.J., 1979. Purification to homogeneity and some
properties of L-phenylalanine ammonia-lyase of irradiated mustard
€
Rosler, J., Krekel, F., Amrhein, N., Schmid, J., 1997. Maize
phenylalanine ammonia-lyase has tyrosine ammonia-lyase activity.
Plant Physiology 113, 175–179.
(Sinapis alba L.) cotyledons. Biochimica et Biophysica Acta 570,
187–197.
Schopfer, P., 1971. Die Phenylalaninammoniumlyase (PAL, EC
4.3.1.5) des Senfkeimlings (Sinapis alba L.), ein elektrophoretisch
einheitliches Enzym. Planta 99, 339–346.
Hanson, K.R., Havir, E.A., 1981. Phenylalanine ammonia lyase. In:
Conn, E.E. (Ed.), Biochemistry of Plants. vol. 7. Secondary Plant
Products. Academic Press, New York, pp. 577–625.
Singh, S., Lewis, N.G., Towers, G.H.N., 1998. Nitrogen recycling
during phenylpropanoid metabolism in sweet potato tubers.
Journal of Plant Physiology 153, 316–323.
Joos, H.-J., Hahlbrock, K., 1992. Phenylalanine ammonia-lyase in
potato (Solanum tuberosum L.). Genomic complexity, structural
comparison of two selected genes and modes of expression.
European Journal of Biochemistry 204, 621–629.
Towers, G.H.N., Singh, S., van Heerden, P.S., Zuiches, J., Lewis,
N.G., 1998. Integrating nitrogen and phenylpropanoid metabolic
pathways in plants and fungi. In: Lewis, N.G., Sarkanen, S. (Eds.),
Lignin and Lignan Biosynthesis. ACS Symposium Series, vol. 697,
Washington, DC, pp. 42–54.
Jorrin, J., Dixon, R.A., 1990. Stress responses in alfalfa (Medicago
sativa L.) II. Purification, characterization, and induction of
phenylalanine ammonia-lyase isoforms from elicitor-treated cell
suspension cultures. Plant Physiology 92, 447–455.
van Heerden, P.S., Towers, G.H.N., Lewis, N.G., 1996. Nitrogen
metabolism in lignifying Pinus taeda cell cultures. Journal of
Biological Chemistry 271, 12350–12355.
Kim, S.-J., Kim, M.-R., Bedgar, D.L., Moinuddin, S.G.A., Cardenas,
C.L., Davin, L.B., Kang, C.-H., Lewis, N.G., 2004. Functional
reclassification of the putative cinnamyl alcohol dehydrogenase
multigene family in Arabidopsis. Proceedings of the National
Academy of Sciences of the United States of America 101, 1455–
1460.
Wanner, L.A., Li, G., Ware, D., Somssich, I.E., Davis, K.R., 1995.
The phenylalanine ammonia-lyase gene family in Arabidopsis
thaliana. Plant Molecular Biology 27, 327–338.