Prolyl 4-hydroxylase: Substrate isosteres in which an (E)- or (Z)-alkene replaces the prolyl peptide bond

Article abstract of DOI:10.1021/acs.biochem.6b00976

Collagen prolyl 4-hydroxylases (CP4Hs) catalyze a prevalent posttranslational modification, the hydroxylation of (2S)-proline residues in protocollagen strands. The ensuing (2S,4R)-4-hydroxyproline residues are necessary for the conformational stability of the collagen triple helix. Prolyl peptide bonds isomerize between cis and trans isomers, and the preference of the enzyme is unknown. We synthesized alkene isosteres of the cis and trans isomers to probe the conformational preferences of human CP4H1. We discovered that the presence of a prolyl peptide bond is necessary for catalysis. The cis isostere is, however, an inhibitor with a potency greater than that of the trans isostere, suggesting that the cis conformation of a prolyl peptide bond is recognized preferentially. Comparative studies with a Chlamydomonas reinhardtii P4H, which has a similar catalytic domain but lacks an N-terminal substrate-binding domain, showed a similar preference for the cis isostere. These findings support the hypothesis that the catalytic domain of CP4Hs recognizes the cis conformation of the prolyl peptide bond and inform the use of alkenes as isosteres for peptide bonds.

Full text of DOI:10.1021/acs.biochem.6b00976

Article
pubs.acs.org/biochemistry
Prolyl 4‑Hydroxylase: Substrate Isosteres in Which an (E)- or
(Z)‑Alkene Replaces the Prolyl Peptide Bond
James D. Vasta,†,∥ Amit Choudhary,‡,⊥ Katrina H. Jensen,§,@ Nicholas A. McGrath,§,▽
and Ronald T. Raines*,†,§
†Department of Biochemistry, ‡Graduate Program in Biophysics, and §Department of Chemistry, University of WisconsinMadison,
Madison, Wisconsin 53706, United States
S
* Supporting Information
ABSTRACT: Collagen prolyl 4-hydroxylases (CP4Hs) catalyze a prevalent posttranslational modification, the hydroxylation of
(2S)-proline residues in protocollagen strands. The ensuing (2S,4R)-4-hydroxyproline residues are necessary for the
conformational stability of the collagen triple helix. Prolyl peptide bonds isomerize between cis and trans isomers, and the
preference of the enzyme is unknown. We synthesized alkene isosteres of the cis and trans isomers to probe the conformational
preferences of human CP4H1. We discovered that the presence of a prolyl peptide bond is necessary for catalysis. The cis isostere
is, however, an inhibitor with a potency greater than that of the trans isostere, suggesting that the cis conformation of a prolyl
peptide bond is recognized preferentially. Comparative studies with a Chlamydomonas reinhardtii P4H, which has a similar
catalytic domain but lacks an N-terminal substrate-binding domain, showed a similar preference for the cis isostere. These
findings support the hypothesis that the catalytic domain of CP4Hs recognizes the cis conformation of the prolyl peptide bond
and inform the use of alkenes as isosteres for peptide bonds.
carboxylate group forms hydrogen bonds and engages in
Coulombic interactions with a basic residue [often arginine or
lysine (Figure 1B)]. All FAKGDs are believed to effect catalysis
through a two-stage mechanism in which α-ketoglutarate is
decarboxylated oxidatively to generate a highly reactive
Fe(IV)O species that elicits substrate oxidation via a radical
rebound process.15−18
In vertebrates, CP4Hs exist as α2β2 tetramers. In these
tetramers, the α-subunit contains the C-terminal catalytic
domain and N-terminal peptide substrate-binding domain (PSB
domain). The β-subunit is protein disulfide isomerase, which is
a multifunctional protein that maintains the α-subunit in a
soluble form and active conformation.9 Humans have three
isoforms of the α-subunit: α(I), α(II), and α(III).9 All three of
these isoforms interact with a single β-subunit to form
tetramers, which are known as CP4H1−CP4H3, respectively.
ollagen is the principal component of the extracellular
1
Cmatrix, connective tissues, and bone of animals. The
overproduction of collagen is associated with a variety of
diseases, including fibrotic diseases2 and cancers.3−7 The
stability of collagen relies on posttranslational modifications.8
By far, the most common of these modifications is mediated by
collagen prolyl 4-hydroxylases (CP4Hs), which are Fe(II)- and
α-ketoglutarate-dependent dioxygenases (FAKGDs) that reside
in the lumen of the endoplasmic reticulum.9 Catalysis by
CP4Hs converts (2S)-proline (Pro) residues in protocollagen
strands into (2S,4R)-4-hydroxyproline (Hyp) residues (Figure
1A), which endow mature collagen triple helices with
conformational stability.10 CP4Hs, while essential for animal
life,11−13 are validated targets for treating fibrotic diseases14 and
metastatic breast cancer.6
Catalysis by CP4Hs and other FAKGDs requires Fe(II) and
the cosubstrates α-ketoglutarate and molecular oxygen.15−17
The Fe(II) is bound by a conserved His-X-Asp/Glu···Xn···His
motif, and α-ketoglutarate chelates to the enzyme-bound Fe(II)
by using its C-1 carboxylate and C-2 keto groups while its C-5
Received: September 24, 2016
Revised: December 1, 2016
Published: December 7, 2016
© 2016 American Chemical Society
219
DOI: 10.1021/acs.biochem.6b00976
Biochemistry 2017, 56, 219−227

Biochemistry
Article
Figure 1. (A) Reaction catalyzed by collagen prolyl 4-hydroxylase
(CP4H). (B) Putative enzymic ligands for CP4H.15−17 (C) Proline
analogues that are substrates for CP4H,30−33 despite substitutions at
Cβ, Cγ (i.e., C-4), or Cδ.
CP4H1 is the most prevalent of these holoenzymes and has
been characterized most extensively. Although the structure of
the tetrameric complex is unknown, those of the individual
domains of the α(I) subunit have provided insight into the
interaction of CP4Hs with the protocollagen substrate, as well
as the interaction of two α(I) subunits in the tetramer.19−22
Many features of the reaction catalyzed by CP4Hs are well-
known.9 Moreover, the development of CP4H inhibitors is of
keen interest.2,14,23,24 Like others,25 we have directed our
attention toward inhibitors that mimic the α-ketoglutarate
cosubstrate.12,26−28 Because many enzymes use α-ketoglutarate
as a substrate or cosubstrate, CP4H inhibitors based upon the
protocollagen substrate could offer improved selectivity. CP4H
is recalcitrant to substitutions in its Fe(II)-binding motif.29 The
enzyme can, however, tolerate alterations in its substrate.30−33
For example, CP4H catalyzes the hydroxylation of proline
analogues that bear a substitution at Cβ, Cγ, or Cδ (Figure 1C).
Little is known about the conformation of the peptide
substrate during catalysis. Proline is a secondary amine, and
prolyl peptide bonds are tertiary amides. Accordingly, the cis
isomer of prolyl peptide bonds (Figure 2A) occurs much more
frequently than with non-prolyl residues.34 Biochemical and
biophysical investigations have suggested that β-turns are sites
of prolyl hydroxylation.35−37 β-Turns can have cis or trans
prolyl peptide bonds,38 and the isomerization state of the
peptide bond in CP4H substrates is unknown, as is the
necessity of even having a peptide bond. Here, we probe this
fundamental issue by using alkene isosteres, which delete the
peptide bond while preserving its geometric constraints. We
describe the synthesis of collagen mimetic peptides in which a
prolyl peptide bond is replaced with an (E)- or (Z)-alkene,
which are the smallest isosteres that fix the bond as a trans or cis
isomer, respectively.39 We find that the presence of a peptide
bond is necessary for catalysis by both human CP4H1 and a
related P4H from the algae Chlamydomonas reinhardtii
(CrP4H140). These enzymes prefer to bind to the cis isostere
relative to the trans, and recognition of the cis isostere likely
Figure 2. Peptide substrates for CP4H and their alkene isosteres.
Tetrapeptide RGlyProProGlyOEt (1 and 2) can adopt a trans or cis
conformation in solution. The corresponding alkene isosteres are
locked as a trans (3 and 4) or cis (5 and 6) isomer.
occurs in the catalytic domain. These results provide new
insight into catalysis by prolyl 4-hydroxylases.
EXPERIMENTAL PROCEDURES
■
Peptide Synthesis. Detailed procedures for the synthesis
of peptides 1 and 2, trans isosteres 3 and 4 (Scheme 1), and cis
isosteres 5 and 6 (Scheme 2) are reported in the Supporting
Information.
Instrumentation. The progress of reactions catalyzed by
prolyl 4-hydroxylases was assessed by analytical UPLC using an
Acquity UPLC H-Class system from Waters (equipped with a
photodiode array detector, a quaternary solvent manager, a
sample manager with a flow-through needle, and Empower 3
software). The concentration of dansyl-labeled peptides was
determined with a Cary 60 UV−vis spectrometer from Agilent
Technologies (Santa Clara, CA) using an extinction coefficient
of 3900 M−1 cm−1. Solution concentrations of proteins were
calculated with the Beer−Lambert law from A280 values as
measured with a NanoVue Plus spectrophotometer from GE
Healthcare (Piscataway, NJ) and extinction coefficients of
290000 M−1 cm−1 for human CP4H141 and 44000 M−1 cm−1
for CrP4H-1 (which was estimated with EXPASY software by
assuming that all cysteine residues form disulfide bonds).
Values of IC50 were calculated from experimental data with
Prism version 6.0 from GraphPad Software (La Jolla, CA).
220
DOI: 10.1021/acs.biochem.6b00976
Biochemistry 2017, 56, 219−227

Biochemistry
Article
Scheme 1. Synthetic Route to GPtransPGOEt trans Isosteres 3 and 4
Scheme 2. Synthetic Route to GPcisPGOEt cis Isosteres 5 and 6
Production of Recombinant Human CP4H1. Human
CP4H containing the α(I) isoform was produced hetero-
logously in Origami B(DE3) Escherichia coli cells and purified
as described previously.41
Stock Solutions for Assays. Assay mixtures were made
from the following stock solutions. A stock solution of Tris-
HCl buffer (10×) was prepared at 500 mM and pH 7.8 in H2O.
A stock solution of α-ketoglutarate (10×) was prepared in
H2O. A stock solution of FeSO4 (10×) was prepared at 500 μM
in H2O. A stock solution of sodium ascorbate (10×) was
prepared at 20 mM in H2O. A stock solution of DTT (10×)
was prepared freshly on the day of its use at 1.0 mM in H2O. A
stock solution of catalase (10×) was prepared freshly on the
day of its use at 1.0 mg/mL in 50 mM Tris-HCl buffer (pH
7.8). A stock solution of BSA (10 mg/mL) was prepared in 50
mM Tris-HCl buffer (pH 7.8). Concentrated stock solutions of
the dansylGPPGOEt substrate and alkene isosteres were
prepared in EtOH and diluted to working stock solutions
(10×) in H2O.
Assay of Human CP4H1 Activity in the Presence of
Alkene Isosteres. The catalytic activity of human CP4H1 was
assayed as described previously.41 Briefly, activity assays were
221
DOI: 10.1021/acs.biochem.6b00976
Biochemistry 2017, 56, 219−227

Biochemistry
Article
performed at 30 °C in 100 μL of 50 mM Tris-HCl buffer (pH
7.8) containing human CP4H1 (100 nM), substrate (0−1.86
mM), alkene isostere (0−3.00 mM), α-ketoglutarate (13−1000
μM), FeSO4 (50 μM), sodium ascorbate (2 mM), DTT (100
μM), catalase (0.1 mg/mL to decompose any hydrogen
peroxide), and BSA (1 mg/mL). Unless noted otherwise,
reaction mixtures were prepared by adding concentrated stock
solutions of each component to concentrated assay buffer in
this order: FeSO4, DTT, sodium ascorbate, BSA, catalase,
CP4H1, peptide substrate, and alkene isostere (or vehicle).
Solutions prepared thusly were preincubated at 30 °C for 2
min, after which a reaction was initiated by the addition of α-
ketoglutarate. After 15 min, reactions were quenched by boiling
for 45 s, and reaction mixtures were subjected to centrifugation
at 10000g. The supernatant (5−10 μL) was injected into an
Acquity UPLC BEH C18 column (2.1 mm × 50 mm, 1.7 μm
particle size) from Waters. The column was eluted for 2.9 min
at a rate of 0.6 mL/min with a linear gradient of acetonitrile [20
to 68% (v/v)] in water containing TFA [0.1% (v/v)] while
monitoring the value of A289. All assays were performed in
triplicate. Data are reported as activity relative to control
reactions lacking an alkene isostere, where activity is
determined from the percent conversion of substrate to
product. Dose−response curves were generated for each
isostere by plotting the relative activity versus the log of
isostere concentration. Values of IC50 for each isostere were
interpolated from the dose−response curves by nonlinear
regression using the sigmoidal dose−response function
available in Prism. For substrate inhibition by dansylGlyPro-
ProGlyOEt, data were fitted to the substrate inhibition function
in Prism.
system based upon a small collagen mimetic tetrapeptide,
GlyProProGlyOEt (GPPGOEt), with corresponding alkene
isosteres GPtransPGOEt and GPcisPGOEt (Figure 2). Small
tetrapeptides of this nature that contain one collagen-like triplet
repeat have been found previously to be suitable substrates for
CP4Hs.27,32,33,37,41 We chose to use model peptides capped
with an N-terminal dansyl sulfonamide, which are well-known
to be substrates for human CP4H,27,41 as well as peptides
capped with an N-terminal Cbz group30,31 for supplemental
analyses (Figure 2). The model substrates dansylGPPGOEt (1)
and CbzGPPGOEt (2) were synthesized by standard solution-
phase peptide coupling procedures as described previously for
similar peptides.27
trans isosteres dansylGPtransPGOEt (3) and CbzGPtransP-
GOEt (4) were synthesized from Boc-Pro-OH in 11 steps by a
route (Scheme 1) adapted from previous reports by Etzkorn,
Koert, and co-workers.43−47 The facial selectivity in the
conversion of aldehyde 9 to alcohol 10 is particularly striking.
The Boc group in aldehyde 9 blocks the si face of the aldehyde,
leading to (S)-configured alcohol 10. (The stereochemistry of
alcohol 10 was confirmed by X-ray crystallography.) The key
step in the route relied on the stereoselective Ireland−Claisen
rearrangement of allyl ester 12 to γ,δ-alkenyl acid 13 in nearly
quantitative yield.48 Standard transformations led to acid 17,
which was then coupled with H-Gly-OEt by using a hindered
base, 2,4,6-collidine, to prevent isomerization of the olefin.
Lastly, deprotection of the N-terminal Boc group of alkene 18
followed by coupling with dansyl-Gly-OH or Cbz-Gly-OH
yielded trans isostere 3 or 4, respectively. These isosteres had
1
characteristic alkene H chemical shifts at 5−6 ppm, indicating
that the alkene did not isomerize to be in conjugation with the
amidic carbonyl group.
Production of Recombinant C. reinhardtii P4H-1. C.
reinhardtii P4H130−245 having an N-terminal hexahistidine
(His6) tag (NHis6-CrP4H-1) was produced heterologously in
Origami B(DE3) E. coli cells and purified as described
previously.42
The corresponding cis isosteres dansylGPcisPGOEt (5) and
CbzGPcisPGOEt (6) were synthesized from Bn-Pro-OH in
nine steps by a route (Scheme 2) inspired by the work of
Etzkorn and co-workers.43,49 A variety of reducing agents were
evaluated for the stereoselective reduction of ketone 21,
including L-Selectride, K-Selectride, NaBH4, DIBALH, and
LiAlH4. The highest yields of alcohol 22 were afforded by
LiAlH4, but as a nearly equimolar mixture of diastereomers.
Although the desired diastereomer of alcohol 22 was isolable,
the separation of diastereomers following alkylation to ether 23
was more facile. (The stereochemistry of the two diastereomers
of alcohol 22 was assigned by forming the two oxazolidinones
Assay of C. reinhardtii P4H-1 Activity in the Presence
of Alkene Isosteres. The catalytic activity of NHis6-CrP4H-1
was assessed in assays similar to those for human CP4H1 with
minor modification. Briefly, activity assays were carried out at
30 °C in 100 μL of 50 mM HEPES-HCl buffer (pH 6.8)
containing NHis6-CrP4H-1 (1.2 μM), substrate dansylGlyPro-
ProGlyOEt (500 μM), alkene isostere (0−100 μM), α-
ketoglutarate (1 mM), FeSO4 (200 μM), sodium ascorbate
(2 mM), DTT (100 μM), catalase (0.1 mg/mL), and BSA (1
mg/mL). Reaction mixtures were prepared by adding
concentrated stock solutions of each component to concen-
trated assay buffer in this order: FeSO4, DTT, sodium
ascorbate, BSA, catalase, NHis6-CrP4H-1, peptide substrate,
and alkene isostere (or vehicle). Solutions prepared thusly were
preincubated at 30 °C for 2 min, after which a reaction was
initiated by the addition of α-ketoglutarate. After 1 h, reactions
were quenched by adding EDTA (1 μL of a 0.5 M solution)
and boiling for 45 s. Quenched reaction mixtures were
subjected to centrifugation at 10000g for 5 min, after which
the supernatant (5−10 μL) was analyzed by UPLC as described
above.
1
and using H NMR spectroscopy.) The key step in the route
employed a stereoselective Still−Wittig rearrangement50 of
ether 23 to afford alkene 24 in 86% yield. As with the synthesis
of the trans isosteres, the coupling of H-Gly-OEt to acid 26
worked well when mediated by 2,4,6-collidine. Lastly,
deprotection of the N-terminal Boc group of alkene 27
followed by subsequent coupling of dansyl-Gly-OH or Cbz-
Gly-OH yielded cis isostere 5 or 6, respectively. Again, these
1
isosteres had characteristic alkene H chemical shifts at 5−6
ppm, indicating that the alkene did not isomerize to be in
conjugation with the amidic carbonyl group. Analytical UPLC
revealed that compounds 1−6 were of high purity.
Interaction of Model Peptides and Alkene Isosteres
with Human CP4H. With the model peptides and
corresponding alkene isosteres in hand, we set out to probe
their interaction with human CP4H. First, we tested the model
peptides (1 and 2) as substrates for human CP4H by
employing a UPLC assay that was adapted from an HPLC
assay described previously.27,32,41 Initial tests confirmed that
RESULTS AND DISCUSSION
■
Synthesis of Alkene Isosteres. The primary substrates for
CP4Hs in vivo are protocollagens and other large polymers (i.e.,
elastin).9 These large protein substrates are not amenable to
most chemical analyses. Hence, we chose to employ a model
222
DOI: 10.1021/acs.biochem.6b00976
Biochemistry 2017, 56, 219−227

Biochemistry
Article
both dansylGPPGOEt and CbzGPPGOEt were substrates for
human CP4H1, as expected from previous reports.27,41 In
contrast, neither the cis nor the trans isosteres (3−6) served as
substrates for the enzyme (data not shown), regardless of
concentration (0.1−3.0 mM for dansyl isosteres; 0.1−10 mM
for Cbz isosteres).
Although isosteres 3−6 did not serve as substrates for human
CP4H, we reasoned that they might serve as inhibitors. Using
the dansyl model peptides, we found that both dansylGP-
transPGOEt and dansylGPcisPGOEt were indeed inhibitors of
catalysis by human CP4H (Figure 3A), with the cis isostere
being significantly more potent than the trans isostere. In
subsequent dose−response experiments, the inhibition curves
for both dansyl isosteres were found to be sigmoidal (Figure
3B), with IC50 values of (75 6) and (410 30) μM for the cis
and trans isosteres, respectively. Moreover, this same trend in
which the cis isostere inhibits with greater potency than does
the trans isostere was also observed with the Cbz-capped
isosteres 4 and 6 (Figure 3C), indicating that the trend is
independent of the N-terminal group. These data are consistent
with a previous correlation indicating that substrates for human
CP4H1 tend to have a stronger preference for a cis prolyl
peptide bond.32 In contrast to isosteres 3−6, however, none of
those previous substrates were fixed as a cis (or trans) isomer.
Because the quantitation and sensitivity were more reliable than
those of the Cbz peptides, we proceeded with dansyl isosteres 3
and 5 in subsequent analyses.
With an interest in characterizing further the inhibition
observed for the dansyl isosteres, we next considered their
kinetic mechanism. CP4Hs have been shown previously to
display an ordered ter−ter mechanism in which α-ketoglutarate
first binds to the CP4H·Fe(II) complex, after which O2 and the
peptide substrate bind in an ordered fashion.51 We found that
turnover of dansylGPPGOEt did not display Michaelis−
Menten kinetics, but rather a pattern that could be recognized
as substrate inhibition (Figure 4A), which was evident in the
Figure 4. Substrate inhibition of human CP4H1 by dansylGPPGOEt
(1). (A) Velocity plot. Assays were performed at pH 7.8 with
dansylGPPGOEt (58−1860 μM) as the substrate and α-ketoglutarate
(1.00 mM) as the cosubstrate. (B) Lineweaver−Burke plot of the data
in panel A.
corresponding Lineweaver−Burke plot (Figure 4B). The data
in Figure 4A were fitted to the equation for substrate inhibition:
Figure 3. Inhibition of human CP4H1 by alkene isosteres 3−6. (A and
B) Assays were performed at pH 7.8 with dansylGPPGOEt (0.50 mM)
as the substrate, α-ketoglutarate (1.00 mM), and a dansyl isostere (100
μM in panel A). (C) Assays were performed at pH 7.8 in the presence
of a Cbz isostere (1.00 mM in panel C) with CbzGPPGOEt (1.00
mM) as a substrate and α-ketoglutarate (1.00 mM) as a cosubstrate. In
all panels, values are the means ( standard deviation) of three
independent experiments and represent the ratio of enzymatic activity
in the presence and absence of the isostere.
V
max[S]
v =
[S]2
Ki
KM + [S] +
where Ki is the dissociation constant for substrate binding in an
inhibitory manner. This fit (R2 = 0.98) yielded values for Vmax
KM, and Ki of (8 3) μM min−1, (2.0 0.8) mM, and (0.3
0.1) μM, respectively.
,
223
DOI: 10.1021/acs.biochem.6b00976
Biochemistry 2017, 56, 219−227

Biochemistry
Article
Substrate inhibition had been observed previously for a
CP4H during turnover of a longer collagen mimetic peptide
substrate, (ProProGly)10.51 Moreover, that substrate inhibition
was competitive with the cosubstrate α-ketoglutarate. Substrate
inhibition by dansylGPPGOEt in our assays was, however,
observed even at saturating concentrations of α-ketoglutarate,
suggesting that the inhibitory binding mode of dansylGPP-
GOEt could not be eliminated in this manner.
Given the complexities in the kinetics with respect to the
peptide substrate, we reasoned that, if the kinetic mechanism
entailed ordered binding, then the dansyl isosteres would
display uncompetitive inhibition versus the α-ketoglutarate
cosubstrate. We performed a Lineweaver−Burke analysis using
cis isostere 5 as a model inhibitor but found an inhibition
pattern consistent with noncompetitive inhibition versus α-
ketoglutarate (Figure 5A). The slope (Figure 5B) and intercept
(Figure 5C) replots for the noncompetitive inhibition were
linear, and Kis and Kii values of (56 13) and (18 8) μM,
respectively, were extracted from the slope (Kis) and intercept
(Kii) replots by linear regression analysis. This value for Kis is
indistinguishable from the IC50 value of (76 13) μM for cis
isostere 5. The observation of noncompetitive inhibition for cis
isostere 5 versus α-ketoglutarate suggests either that this
isostere binds in a CP4H subsite that is independent of the site
for hydroxylation or that the binding of the small collagenous
tetrapeptides is not strictly ordered relative to the cosubstrates
α-ketoglutarate and molecular oxygen. Kinetic experiments
cannot distinguish between these two possibilities.
Preliminarily, the inhibition pattern reported above is
consistent with the hypothesis that human CP4H preferentially
binds to the cis conformation compared to the trans
conformation in collagen-like peptides. The CP4H α-subunit
is composed of multiple domains, including a C-terminal
catalytic domain and an N-terminal PSB domain,9,14 both of
which likely interact with a typical polymeric substrate. The
structure of the PSB domain has been characterized and is
known to interact with peptides in a PPII helical conformation,
which has trans peptide bonds.20−22 Accordingly, we were not
surprised to learn that human CP4H has some affinity for trans
isostere 3, which is likely to adopt an extended conformation
similar to that of a PPII helix. Given its locked conformation,
however, cis isostere 5 is less likely to interact favorably with the
PSB domain, which raises the issue of whether the C-terminal
catalytic domain is responsible for preferential recognition of
the cis conformation.
To address this issue, we sought a more simple P4H model
system. We chose to investigate CrP4H-1, which is an algal
P4H.40 CrP4H-1 is a soluble, 29 kDa monomer with an amino
acid sequence 26% identical to that of the catalytic domain of
human CP4H140 and a known three-dimensional structure.52,53
FAKGD enzymes are generally conserved at the level of fold,16
and all essential active-site residues are conserved in the two
P4Hs.40 Moreover, although substrates for algal P4Hs such as
CrP4H-1 are typically proline- and serine-rich polymers of the
algal cell wall in vivo, CrP4H-1 has been observed to turn over
collagen-like peptides, albeit with less positional specificity (e.g.,
CrP4H-1 catalyzes the hydroxylation of proline residues in both
position X and position Y).40
Figure 5. Lineweaver−Burke analysis of inhibition of human CP4H1
by cis isostere 5. (A) A Lineweaver−Burke plot for 5 is indicative of
noncompetitive inhibition with respect to α-ketoglutarate, as the
intersecting pattern does not converge on the ordinate and the slope
(B) and intercept (C) replots are linear. Assays were performed at pH
7.8 with dansylGPPGOEt (500 μM) as the substrate and α-
ketoglutarate (13−100 μM) as the cosubstrate in the presence of cis
isostere 5 (0, 50, 100, or 150 μM). Values are the means ( standard
error) of three independent experiments.
as a substrate for CrP4H-1, albeit with a substrate inhibition
phenomenon similar to that observed for human CP4H1
(Figure 6). Again, we did not observe hydroxylation of dansyl
isostere 3 or 5 by CrP4H-1 (data not shown). We did find,
however, that both dansyl isosteres 3 and 5 served as inhibitors
of CrP4H-1. We observed an inhibition pattern virtually
identical to that observed for human CP4H1, again with cis
isostere 5 displaying potency significantly greater than that of
trans isostere 3 (Figure 7). Similar to human CP4H1, these data
suggest that CrP4H-1 preferentially recognizes the cis
conformation in peptide substrate analogues. Given the
similarities between human CP4H1 and algal CrP4H-1, as
well as CrP4H-1 being (in essence) a monomeric catalytic
domain, our data suggest that the catalytic domains of both
As CrP4H-1 can be viewed as a reasonable model system for
the catalytic domain of the human CP4H1 α-subunit, we
sought to investigate the interaction of dansyl peptides 1, 3, and
5 with CrP4H-1. Under an assay condition similar to that used
to assay human CP4H1, we found that model peptide 1 served
224
DOI: 10.1021/acs.biochem.6b00976
Biochemistry 2017, 56, 219−227

Biochemistry
Article
Neither the cis nor the trans prolyl peptide bond isostere
serves as a substrate for CP4H. This result suggests that CP4H
interrogates substitutions at the prolyl peptide bond of its
substrate more closely than those near Cγ (Figure 1C).
Numerous explanations are possible for this intolerance. The
substitution could eliminate a key hydrogen bond between the
enzyme and the amidic oxygen (Oi−1) of the prolyl peptide
bond, perturbing the orientation of the substrate. The
substitution could affect the substrate conformation by
preventing two potential n → π* interactions,46,59−61 one
with the C′i−1Oi−1 group as a donor and one with the
C′i−1Oi−1 group as an acceptor. Finally, the conformational
mobility of the prolyl peptide bond (even its cis−trans
isomerization) could play an integral role in catalysis by
CP4H. These scenarios are the basis for ongoing work in our
laboratory.
ASSOCIATED CONTENT
■
S
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.bio-
chem.6b00976.
Procedures for the synthesis of compounds 1−6 (PDF)
Figure 6. Substrate inhibition of CrP4H-1 by dansylGPPGOEt (1).
(A) Velocity plot. Assays were performed at pH 7.8 with
dansylGPPGOEt (212−2120 μM) as the substrate and α-ketoglutarate
(1.00 mM) as the cosubstrate. (B) Lineweaver−Burke plot of the data
in panel A.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: rtraines@wisc.edu.
ORCID
Ronald T. Raines: 0000-0001-7164-1719
Present Addresses
∥J.D.V.: Promega Corp., 2800 Woods Hollow Rd., Fitchburg,
WI 53711.
⊥A.C.: Department of Medicine, Harvard Medical School,
Boston, MA 02115.
@K.H.J.: Chemistry Program, Black Hills State University,
Spearfish, SD 57799.
▽N.A.M.: Department of Chemistry and Biochemistry,
University of WisconsinLa Crosse, La Crosse, WI 54601.
Funding
J.D.V. was supported by Molecular Biosciences Training Grant
T32 GM007215 [National Institutes of Health (NIH)] and a
fellowship from the Department of Biochemistry at the
University of WisconsinMadison. K.H.J. was supported by
Postdoctoral Fellowship F32 GM098032 (NIH). N.A.M. was
supported by Postdoctoral Fellowship F32 GM096712 (NIH).
This work was supported by Grant R01 AR044276 (NIH) and
made use of the National Magnetic Resonance Facility at
Madison, which is supported by Grant P41 GM103399 (NIH),
and a Micromass LCT mass spectrometer, which was obtained
with support from Grant CHE-9974839 (National Science
Foundation).
Figure 7. Inhibition of C. reinhardtii CrP4H-1 by alkene isosteres 3
and 5. Assays were performed at pH 7.8 with dansylGPPGOEt (0.50
mM) as the substrate, α-ketoglutarate (1.00 mM), and an alkene
isostere (100 μM). Values are the means ( standard deviation) of
three independent experiments and represent the ratio of enzymatic
activity in the presence and absence of isostere.
P4Hs preferentially recognize the cis conformation of prolyl
peptide bonds.
Product inhibition is often detrimental to enzyme
function.54−56 This problem could be especially acute with
CP4H because of the abundance of collagen strands in the
endoplasmic reticulum. The installation of a 4(R)-hydroxyl
group is known to decrease the fraction of prolyl peptide bonds
that reside in the cis conformation.57,58 Taken with our data,
this increase in the trans:cis ratio could slow product inhibition.
We note as well that the hydrophilicity conferred by
hydroxylation could diminish the affinity for hydrophobic
pockets of the PSB domain.22
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We are grateful to Drs. B. Gold, M. D. Shoulders, K. L. Gorres,
I. A. Guzei, and C. L. Jenkins (University of Wisconsin
Madison) for contributive discussions.
225
DOI: 10.1021/acs.biochem.6b00976
Biochemistry 2017, 56, 219−227

Biochemistry
REFERENCES
Article
(20) Hieta, R., Kukkola, L., Permi, P., Pirila, P., Kivirikko, K. I.,
Kilpelainen, I., and Myllyharju, J. (2003) The peptide-substrate
binding domain of human collagen prolyl 4-hydroxylases. Backbone
assignments, secondary structure, and binding of proline-rich peptides.
J. Biol. Chem. 278, 34966−34974.
̈
■
̈
(1) Shoulders, M. D., and Raines, R. T. (2009) Collagen structure
and stability. Annu. Rev. Biochem. 78, 929−958.
(2) Rosenbloom, J., Castro, S. V., and Jimenez, S. A. (2010) Narrative
review: Fibrotic diseases: Cellular and molecular mechanisms and
novel therapies. Ann. Intern. Med. 152, 159−166.
(21) Pekkala, M., Hieta, R., Bergmann, U., Kivirikko, K. I., Wierenga,
R. K., and Myllyharju, J. (2004) The peptide-substrate-binding domain
of collagen prolyl 4-hydroxylase is a tetratricopeptide repeat domain
with functional aromatic residues. J. Biol. Chem. 279, 52255−52261.
(22) Anantharajan, J., Koski, M. K., Kursula, P., Hieta, R., Bergmann,
U., Myllyharju, J., and Wierenga, R. K. (2013) The structural motifs for
substrate binding and dimerization of the α subunit of collagen prolyl
4-hydroxylase. Structure 21, 2107−2118.
(23) Hanauske-Abel, H. M. (1991) Prolyl 4-hydroxylase, a target
enzyme for drug development. Design of suppressive agents and the in
vitro effect of inhibitors and proinhibitors. J. Hepatol. 13 (Suppl. 3),
S8−S16.
(24) Friedman, S. L., Sheppard, D., Duffield, J. S., and Violette, S.
(2013) Therapy for fibrotic diseases: Nearing the starting line. Sci.
Transl. Med. 5, 167sr1.
(3) Provenzano, P. P., Inman, D. R., Eliceiri, K. W., Knittel, J. G., Yan,
L., Rueden, C. T., White, J. G., and Keely, P. J. (2008) Collagen
density promotes mammary tumor initiation and progression. BMC
Med. 6, 11.
(4) Conklin, M. W., Eickhoff, J. C., Riching, K. M., Pehlke, C. A.,
Eliceiri, K. W., Provenzano, P. P., Friedl, A., and Keely, P. J. (2011)
Aligned collagen is a prognostic signature for survival in human breast
carcinoma. Am. J. Pathol. 178, 1221−1232.
(5) Eisinger-Mathason, T. S., Zhang, M., Qiu, Q., Skuli, N.,
Nakazawa, M. S., Karakasheva, T., Mucaj, V., Shay, J. E. S.,
́
Stangenberg, L., Sadri, N., Pure, E., Yoon, S. S., Kirsch, D. G., and
Simon, M. C. (2013) Hypoxia-dependent modification of collagen
networks promotes sarcoma metastasis. Cancer Discovery 3, 1190−
1205.
(25) Rose, N. R., Mcdonough, M. A., King, O. N. F., Kawamura, A.,
and Schofield, C. J. (2011) Inhibition of 2-oxoglutarate dependent
oxygenases. Chem. Soc. Rev. 40, 4364−4397.
(26) Vasta, J. D., Higgin, J. J., Kersteen, E. A., and Raines, R. T.
(2013) Bioavailable affinity label for collagen prolyl 4-hydroxylase.
Bioorg. Med. Chem. 21, 3597−3601.
(27) Vasta, J. D., and Raines, R. T. (2015) Selective inhibition of
prolyl 4-hydroxylases by bipyridinedicarboxylates. Bioorg. Med. Chem.
23, 3081−3090.
(28) Vasta, J. D., Andersen, K. A., Deck, K. M., Nizzi, C. P.,
Eisenstein, R. S., and Raines, R. T. (2016) Selective inhibition of
collagen prolyl 4-hydroxylase in human cells. ACS Chem. Biol. 11,
193−199.
(29) Gorres, K. L., Pua, K. H., and Raines, R. T. (2009) Stringency of
the 2-His-1-Asp active-site motif in prolyl 4-hydroxylase. PLoS One 4,
e7635.
(30) Tandon, M., Wu, M., Begley, T. P., Myllyharju, J., Pirskanen, A.,
and Kivirikko, K. (1998) Substrate specificity of human prolyl-4-
hydroxylase. Bioorg. Med. Chem. Lett. 8, 1139−1144.
(31) Wu, M., Moon, H.-S., Begley, T. P., Myllyharju, J., and Kivirikko,
K. I. (1999) Mechanism-based inactivation of the human prolyl-4-
hydroxylase by 5-oxaproline-containing peptides: Evidence for a prolyl
radical intermediate. J. Am. Chem. Soc. 121, 587−588.
(32) Gorres, K. L., Edupuganti, R., Krow, G. R., and Raines, R. T.
(2008) Conformational preferences of substrates for human prolyl 4-
hydroxylase. Biochemistry 47, 9447−9455.
(33) Gorres, K. L., and Raines, R. T. (2009) Direct and continuous
assay for prolyl 4-hydroxylase. Anal. Biochem. 386, 181−185.
(34) Fischer, G. (2000) Chemical aspects of peptide bond
isomerisation. Chem. Soc. Rev. 29, 119−127.
(35) Rapaka, R. S., Renugopalakrishnan, V., Urry, D. W., and
Bhatnagar, R. S. (1978) Hydroxylation of proline in polytripeptide
models of collagen: Stereochemistry of polytripeptide−prolyl hydrox-
ylase interaction. Biochemistry 17, 2892−2989.
(36) Brahmachari, S. K., and Ananthanarayanan, V. S. (1979) β-
Turns in nascent procollagen are sites of posttranslational enzymatic
hydroxylation of proline. Proc. Natl. Acad. Sci. U. S. A. 76, 5119−5123.
(37) Atreya, P. L., and Ananthanarayanan, V. S. (1991) Interaction of
prolyl 4-hydroxylase with synthetic peptide substrates. J. Biol. Chem.
266, 2852−2858.
(38) Chou, P. Y., and Fasman, G. D. (1977) β-Turns in proteins. J.
Mol. Biol. 115, 135−175.
(39) Choudhary, A., and Raines, R. T. (2011) An evaluation of
peptide-bond isosteres. ChemBioChem 12, 1801−1807.
(40) Keskiaho, K., Hieta, R., Sormunen, R., and Myllyharju, J. (2007)
Chlamydomonas reinhardtii has multiple prolyl 4-hydroxylases, one of
which is essential for proper cell wall assembly. Plant Cell 19, 256−269.
(6) Gilkes, D. M., Chaturvedi, P., Bajpai, S., Wong, C. C., Wei, H.,
Pitcairn, S., Hubbi, M. E., Wirtz, D., and Semenza, G. L. (2013)
Collagen prolyl hydroxylases are essential for breast cancer metastasis.
Cancer Res. 73, 3285−3296.
(7) Cox, T. R., Bird, D., Baker, A.-M., Barker, H. E., Ho, M. W.-Y.,
Lang, G., and Erler, J. T. (2013) Lox-mediated collagen crosslinking is
responsible for fibrosis-enhanced metastasis. Cancer Res. 73, 1721−
1732.
(8) Myllyharju, J., and Kivirikko, K. I. (2004) Collagens, modifying
enzymes and their mutations in humans, flies and worms. Trends
Genet. 20, 33−43.
(9) Gorres, K. L., and Raines, R. T. (2010) Prolyl-4-hydroxylase. Crit.
Rev. Biochem. Mol. Biol. 45, 106−124.
(10) Berg, R. A., and Prockop, D. J. (1973) The thermal transition of
a non-hydroxylated form of collagen. Evidence for a role for
hydroxyproline in stabilizing the triple helix of collagen. Biochem.
Biophys. Res. Commun. 52, 115−120.
(11) Winter, A. D., and Page, A. P. (2000) Prolyl 4-hydroxylase is an
essential procollagen-modifying enzyme required for exoskeleton
formation and the maintenance of body shape in the nematode
Caenorhabditis elegans. Mol. Cell. Biol. 20, 4084−4093.
(12) Friedman, L., Higgin, J. J., Moulder, G., Barstead, R., Raines, R.
T., and Kimble, J. (2000) Prolyl 4-hydroxylase is required for viability
and morphogenesis in Caenorhabditis elegans. Proc. Natl. Acad. Sci. U. S.
A. 97, 4736−4741.
(13) Holster, T., Pakkanen, O., Soininen, R., Sormunen, R.,
Nokelainen, M., Kivirikko, K. I., and Myllyharju, J. (2007) Loss of
assembly of the main basement membrane collagen, Type IV, but not
fibril-forming collagens and embryonic death in collagen prolyl 4-
hydroxylase I null mice. J. Biol. Chem. 282, 2512−2519.
(14) Myllyharju, J. (2008) Prolyl 4-hydroxylases, key enzymes in the
synthesis of collagens and regulation of the response to hypoxia, and
their roles as treatment targets. Ann. Med. 40, 402−417.
(15) Hanauske-Abel, H. M., and Gunzler, V. (1982) A stereochemical
̈
concept for the catalytic mechanism of prolylhydroxylase: Applicability
to classification and design of inhibitors. J. Theor. Biol. 94, 421−455.
(16) Hausinger, R. P. (2004) Fe(II)/α-ketoglutarate-dependent
hydroxylases and related enzymes. Crit. Rev. Biochem. Mol. Biol. 39,
21−68.
(17) Costas, M., Mehn, M. P., Jensen, M. P., and Que, L. J. (2004)
Dioxygen activation at mononuclear nonheme iron active sites:
Enzymes, models, and intermediates. Chem. Rev. 104, 939−986.
(18) Groves, J. T., and McClusky, G. A. (1976) Aliphatic
hydroxylation via oxygen rebound. Oxygen transfer catalyzed by
iron. J. Am. Chem. Soc. 98, 859−861.
(19) Abraham, R. J., and McLauchlan, K. A. (1962) The proton
resonance spectra and conformations of the prolines. Part I. Mol. Phys.
5, 195−203.
226
DOI: 10.1021/acs.biochem.6b00976
Biochemistry 2017, 56, 219−227

Biochemistry
Article
(41) Kersteen, E. A., Higgin, J. J., and Raines, R. T. (2004)
Production of human prolyl 4-hydroxylase in Escherichia coli. Protein
Expression Purif. 38, 279−291.
(42) Vasta, J. D., and Raines, R. T. (2016) Human collagen prolyl 4-
hydroxylase is activated by ligands for its iron center. Biochemistry 55,
3224−3233.
(43) Wang, X. J., Hart, S. A., Xu, B., Mason, M. D., Goodell, J. R., and
Etzkorn, F. A. (2003) Serine−cis-proline and serine−trans-proline
isosteres: Stereoselective synthesis of (Z)- and (E)-alkene mimics by
Still−Wittig and Ireland−Claisen rearrangements. J. Org. Chem. 68,
2343−2349.
(44) Bandur, N. G., Harms, K., and Koert, U. (2005) First
stereoselective synthesis of a Pro−Pro E-alkene dipeptide isostere.
Synlett, 0773−0776.
(45) Bandur, N. G., Harms, K., and Koert, U. (2007) Synthesis of
two dipeptide isosteres containing di- and trisubstituted E-configured
double bonds. Synthesis 2007, 2720−2730.
(46) Dai, N., Wang, X. J., and Etzkorn, F. A. (2008) The effect of a
trans-locked Gly−Pro alkene isostere on collagen triple helix stability.
J. Am. Chem. Soc. 130, 5396−5397.
(47) Dai, N., and Etzkorn, F. A. (2009) Cis−trans proline
isomerization effects on collagen triple-helix stability are limited. J.
Am. Chem. Soc. 131, 13728−13732.
(48) Ireland, R. E., and Mueller, R. H. (1972) The Claisen
rearrangement of allyl esters. J. Am. Chem. Soc. 94, 5897−5898.
(49) Hart, S. A., Sabat, M., and Etzkorn, F. A. (1998) Enantio- and
regioselective synthesis of a (Z)-alkene cis-proline mimic. J. Org. Chem.
63, 7580−7581.
(50) Still, W. C., and Mitra, A. (1978) A highly stereoselective
synthesis of Z-trisubstituted olefins via [2,3]-sigmatropic rearrange-
ment. Preference for a psuedoaxially substituted transition state. J. Am.
Chem. Soc. 100, 1927−1928.
(51) Myllyla, R., Tuderman, L., and Kivirikko, K. I. (1977)
̈
Mechanism of the prolyl hydroxylase reaction. 2. Kinetic analysis of
the reaction sequence. Eur. J. Biochem. 80, 349−357.
(52) Koski, M. K., Hieta, R., Bollner, C., Kivirikko, K. I., Myllyharju,
̈
J., and Wierenga, R. K. (2007) The active site of an algal prolyl 4-
hydroxylase has large structural plasticity. J. Biol. Chem. 282, 37112−
37123.
(53) Koski, M. K., Hieta, R., Hirsila, M., Ronka, A., Myllyharju, J., and
̈
̈
̈
Wierenga, R. K. (2009) The crystal structure of an algal prolyl 4-
hydroxylase complexed with a proline-rich peptide reveals a novel
buried tripeptide binding motif. J. Biol. Chem. 284, 25290−25301.
(54) Walter, C., and Frieden, E. (2006) The prevalence and
significance of the product inhibition of enzymes. Adv. Enzymol. Relat.
Areas Mol. Biol. 25, 167−274.
(55) Frieden, E., and Walter, C. (1963) Prevalence and significance
of the product inhibition of enzymes. Nature 198, 834−837.
(56) Cook, P. F., and Cleland, W. W. (2007) Enzyme Kinetics and
Mechanism, pp 128−150, Garland Science, New York.
(57) Eberhardt, E. S., Panasik, N., Jr., and Raines, R. T. (1996)
Inductive effects on the energetics of prolyl peptide bond isomer-
ization: Implications for collagen folding and stability. J. Am. Chem.
Soc. 118, 12261−12266.
(58) Bretscher, L. E., Jenkins, C. L., Taylor, K. M., DeRider, M. L.,
and Raines, R. T. (2001) Conformational stability of collagen relies on
a stereoelectronic effect. J. Am. Chem. Soc. 123, 777−778.
(59) Jenkins, C. L., Vasbinder, M. M., Miller, S. J., and Raines, R. T.
(2005) Peptide bond isosteres: Ester or (E)-alkene in the backbone of
the collagen triple helix. Org. Lett. 7, 2619−2622.
(60) Horng, J.-C., and Raines, R. T. (2006) Stereoelectronic effects
on polyproline conformation. Protein Sci. 15, 74−83.
(61) Jakobsche, C. E., Choudhary, A., Miller, S. J., and Raines, R. T.
(2010) n→π* Interaction and n)(π Pauli repulsion are antagonistic for
protein stability. J. Am. Chem. Soc. 132, 6651−6653.
227
DOI: 10.1021/acs.biochem.6b00976
Biochemistry 2017, 56, 219−227