Mammalian peptide isomerase: Platypus-type activity is present in mouse heart

Article abstract of DOI:10.1002/cbdv.200900300

Male platypus (Ornithorhynchus anatinus) venom has a peptidyl aminoacyl L/D-isomerase (hereafter called peptide isomerase) that converts the second amino acid residue in from the N-terminus from the L- to the D-form, and vice versa. A reversed-phase high-performance liquid chromatography (RP-HPLC) assay has been developed to monitor the interconversion using synthetic hexapeptides derived from defensin-like peptide-2 (DLP-2) and DLP-4 as substrates. It was hypothesised that animals other than the platypus would have peptide isomerase with the same substrate specificity. Accordingly, eight mouse tissues were tested and heart was shown to have the activity. This is notable for being the first evidence of a peptide isomerase being present in a higher mammal and heralds finding the activity in man.

Full text of DOI:10.1002/cbdv.200900300

CHEMISTRY & BIODIVERSITY – Vol. 7 (2010)
1603
Mammalian Peptide Isomerase: Platypus-Type Activity Is Present in Mouse
Heart
by Jennifer M. S. Koh, Stephanie J. P. Chow, Ben Crossett, and Philip W. Kuchel*
School of Molecular Biosciences, Building G08, University of Sydney, NSW 2006, Australia
(phone: þ61293513709; fax: þ6129351 4726; e-mail: philip.kuchel@sydney.edu.au)
Male platypus (Ornithorhynchus anatinus) venom has a peptidyl aminoacyl l/d-isomerase (hereafter
called peptide isomerase) that converts the second amino acid residue in from the N-terminus from the l-
to the d-form, and vice versa. A reversed-phase high-performance liquid chromatography (RP-HPLC)
assay has been developed to monitor the interconversion using synthetic hexapeptides derived from
defensin-like peptide-2 (DLP-2) and DLP-4 as substrates. It was hypothesised that animals other than
the platypus would have peptide isomerase with the same substrate specificity. Accordingly, eight mouse
tissues were tested and heart was shown to have the activity. This is notable for being the first evidence of
a peptide isomerase being present in a higher mammal and heralds finding the activity in man.
Introduction1). – A peptidyl aminoacyl l/d-isomerase (hereafter called peptide
isomerase) was discovered in the venom of the male platypus (Ornithorhynchus
anatinus) [1][2]. This enzyme interconverts l- and d-amino acid residues at the second
position in from the N-terminus of a small series of different peptides [3]. The
discovery of the peptide isomerase occurred after finding two pairs of platypus venom
peptides that showed 1H-NMR-spectral differences primarily at the second amino acid
residue; these peptides were defensin-like peptides (DLP-2/DLP-4) and Ornithorhyn-
chus venom C-type natriuretic peptides (OvCNPa/OvCNPb). DLP-4 and OvCNPa
have all the amino acid residues in the normally occurring l-form, while DLP-2 and
OvCNPb have the second amino acid residue in the d-form [4]. As the protein
synthesis ꢀmachineryꢁ on ribosomes only generates peptides from l-amino acids, it is
generally accepted that DLP-2 and OvCNPb are post-translationally modified from
DLP-4 and OvCNPa, respectively. And it is the peptide isomerase that catalyses this l-
to d-conversion [1].
Isomerases that catalyse the conversion of l- to d-amino acid residues have already
been reported in spider venom [5], lobster nervous system [6], and frog skin secretion
[7]. All the isomerases that have been characterised to date, including platypus venom
peptide isomerase, isomerise amino acids at different positions in peptides either near
the N- or C-terminus of the respective peptide. However, platypus venom peptide
1
)
Abbreviations. – MeCN: acetonitrile; DAO: d-amino acid oxidase; DLP-2: defensin-like peptide-2;
DLP-4: defensin-like peptide-4; DEPC: diethylpyrocarbonate; EDTA: ethylenediaminetetraacetic
acid; MALDI-TOF-MS: matrix-assisted laser desorption/ionisation-time of flight mass spectro-
metry; Nle: norleucine; NMDA: N-methyl-d-aspartate; NMR: nuclear magnetic resonance;
OvCNPa: Ornithorhynchus C-type natriuretic peptide A; OvCNPb: Ornithorhynchus C-type
natriuretic peptide B; PBS: phosphate buffered saline; RP-HPLC: reversed-phase high-perfor-
mance liquid chromatography; TFA: CF3COOH.
ꢂ 2010 Verlag Helvetica Chimica Acta AG, Zꢃrich

1604
CHEMISTRY & BIODIVERSITY – Vol. 7 (2010)
isomerase is the only one which is of mammalian origin; thus, it is the first such enzyme
reported in mammals [3].
One proposed outcome of peptide isomerases is that the modified peptides are less
prone to peptidase-mediated degradation in the circulation or in tissues, so they persist
longer in vivo [1][2]. It is also possible that the d-amino acid-containing peptides are
more potent than their l-counterparts as is the case with dermophin [8].
Platypus venom peptide isomerase is ca. 55–65 kDa in size; it isomerises the second
amino acid residue in from the N-terminus in both l- to d- and d- to l-directions. The
enzyme remains active up to 558 [2], and it is inhibited by diethylpyrocarbonate
(DEPC) [9]. In previous work, we reported that the peptide isomerase can operate on
short hexapeptides with the N-terminal tripeptide sequences derived from DLP-2 and
DLP-4, with a short water-solubilising chain appended at the C-terminus [9]. Some
modifications have successfully been made to these synthetic hexapeptides; thus,
norleucine can replace methionine as the second amino acid residue and it displays
faster isomerisation kinetics. Also, the solubilising chain can be made with all d-amino
acids which has the benefit of protecting the hexapeptides from l-specific carboxy-
peptidases [9].
Although the DLP-2/DLP-4-derived hexapeptides have identical amino acid
sequences, their elution profiles on RP-HPLC are different [9]. This is because the
l-/d-amino acid residues at the second position cause differences in the overall fold of
the hexapeptides and thus change their binding affinity to C18 chromatographic
medium (Fig. 1). Therefore, a convenient isomerase assay was able to be established
using RP-HPLC in which we employed the differences in the elution times of each
isomeric species.
Fig. 1. Molecular structures of the N-terminal tripeptide sequences, IMF: a) DLP-4, with l-methionine;
and b) DLP-2, with d-methionine. The third amino acid residue, phenylalanine, is aligned in almost the
same direction in each structure. The different folding of the peptides is caused by the side chain of
methionine being positioned on different sides of the a-C-atom depending on whether it is the l- or d-
form. Such folding differences are the basis for distinctly different interactions with the RP-HPLC C18
chromatography medium even though the medium itself is not chirally selective. Structures were drawn
using Jmol [10][11].
The presence of peptide isomerase in higher mammals was postulated to be possible
once it had been detected in the platypus. However, due to the lack of easily accessible
techniques and convenient methods to monitor the l-/d-interconversion, the activity
has not been widely studied. In this work, the HPLC-based assay developed for

CHEMISTRY & BIODIVERSITY – Vol. 7 (2010)
1605
platypus venom peptide isomerase was employed to test for isomerase activity in
various mouse tissues. This is the beginning of a broad search for a wide range of
different peptide specificities in mammals, including man, because there appears to be
no a priori reason why only platypus-type specificity should exist and no others, such as
those from the phylogenetically older frogs [7], spiders [5], and crustaceans [6].
Results and Discussion. – Screening. Eight mouse tissues (brain, heart, kidney, liver,
lung, small intestine, spleen, and stomach) were tested for isomerase activity using the
synthetic hexapeptides derived from DLP-4, viz., IMFsrs and its mutant, I(Nle)Fsrs
(uppercase letters denote l-amino acids; lowercase letters denote d-amino acids).
When platypus-specific isomerisation occurs, the second amino acid residue of these
hexapeptides is converted into the d-amino acid that is differentiated from the all-l-
amino acid hexapeptides on the basis of their different elution times on RP-HPLC.
Platypus Control. A control reaction was performed by incubating I(Nle)Fsrs with a
partially purified platypus venom-gland extract at 378. As seen in Fig. 2, I(Nle)Fsrs was
converted to I(nle)Fsrs after incubation, based on the elution time being the same as
that of the known, pure synthetic I(nle)Fsrs. To ascertain if the converted product was
indeed I(nle)Fsrs, the eluted fraction was collected and MALDI-TOF-MS was
performed to determine the main molecular mass. The product peak corresponded
to a molecular species with a mass that was the same (within 0.2 mass units) of that of
I(Nle)Fsrs; thus the isomerisation was confirmed.
Mouse Tissue Extracts. Similar results were obtained from assays using mouse heart
extract. Two peptide substrates were used, viz., IMFsrs and I(Nle)Fsrs. They both
showed isomerisation; a peak appeared in the chromatogram at the time where the d-
amino acid-containing hexapeptide was eluted (see Figs. 3 and 4). Again, the identities
of the main constituents of the corresponding fractions were established by MALDI-
TOF-MS. The fractions contained peptides with identical molecular masses as the
corresponding l-amino acid-containing hexapeptides.
The assays performed with extracts of other mouse tissues revealed no isomer-
isation. It is likely that the substrate hexapeptides were degraded by peptidases in the
tissue extracts during the assay procedure.
In this work, mouse hearts from both male and female mice were positive for the
activity, and this occurred across different ages from 3 weeks to 18 months.
In the assays with platypus venom peptide isomerase, l- to d-isomerisation
occurred rapidly on incubation at 378. At ca. 4 h, the reaction has almost reached
equilibrium (quasi-steady state; data not shown here). However, the standard reaction
(same relative volumes of extract and peptide) with the mouse heart extracts was much
slower, and clear evidence of the isomerisation was only seen after several hours of
incubation. This occurred even though the extracts had been concentrated. This slow
isomerisation by the mouse heart extracts may have resulted from either a lower
absolute concentration of the enzyme in the tissue, or different kinetic parameters such
as the binding affinity, Michaelis–Menten parameters, or susceptibility to inhibitory
effector molecules, both endogenous and added to the assay medium. This aspect of the
investigation is encompassed by our future plans.
Substrate Specificity. It has been proposed that isomerisation involves at least three
amino acid residues, where the second amino acid residue must have a long

1606
CHEMISTRY & BIODIVERSITY – Vol. 7 (2010)
Fig. 2. RP-HPLC Chromatograms from the assay of peptide isomerase activity with the synthetic
hexapeptide I(Nle)Fsrs, catalysed by platypus venom peptide isomerase. The peptide substrate was
incubated with a partially purified platypus venom-gland extract at 378; HPLC Analysis was carried out
at 0, 2, and 4 h. The elution of I(nle)Fsrs is indicated by the arrow; peak complexes marked with *
corresponded to the elution of salts, and # coincided with the programmed abrupt change in elution-
solvent concentration.
hydrophobic chain and the third amino acid residue has to be aromatic [9]. Peptide
isomerase does not act on short peptides derived from OvCNPa and OvCNPb. This
forms an important basis of our isomerase assays by using synthetic short peptides as
substrates because they can be purchased cheaply or easily made in the laboratory.
However, it does mean that other peptide isomerases could be missed if they require
larger peptide substrates.

CHEMISTRY & BIODIVERSITY – Vol. 7 (2010)
1607
Fig. 3. RP-HPLC Chromatograms from the assay of peptide isomerase activity in mouse heart extract with
the synthetic hexapeptide IMFsrs, incubated at 378. HPLC Analysis was carried out at 0, 2, 4, and 24 h. The
appearance of the ImFsrs peak is indicated by the arrow; peak complexes marked with * corresponded to
the elution of salts.
Protease and Peptidase Inhibition. In the assays using I(Nle)Fsrs as the substrate,
the isomerisation to I(nle)Fsrs by platypus venom peptide isomerase was seen to be
faster than the isomerisation of IMFsrs to ImFsrs [9]. As for all enzymes, peptide
isomerase catalyses the reaction in both directions, and the fact that the l- to d-
isomerisation took a shorter time to reach the equilibrium (in reality a quasi-steady
state) implies a bias of the equilibrium in favour of the d-form. On the other hand, since
the conversion of the d-form to the l-form yields a product that is more susceptible to
attack by aminopeptidase(s) in the tissue extract, this impression may be artefactual. In
other words, the assay medium contains a ꢀcocktailꢁ of protease and peptidase
inhibitors, but it is known that at least one of the inhibitors (amastatin) is an inhibitor of

1608
CHEMISTRY & BIODIVERSITY – Vol. 7 (2010)
Fig. 4. RP-HPLC Chromatograms from the assay of peptide isomerase activity in mouse heart extract,
with the synthetic hexapeptide I(Nle)Fsrs incubated at 378. HPLC Analysis was carried out at 0, 2, 4, and
24 h. The appearance of the I(nle)Fsrs peak is indicated by the arrow; peak complexes marked with *
corresponded to the elution of salts.
the platypus isomerase [2]. Therefore, there was a trade-off between invoking sufficient
inhibition of the peptidases and proteases and not inhibiting the peptide isomerase. It
may be true that in tissues other than the heart the peptidases and proteases ꢀwin the
raceꢁ to the probe peptide and degrade it before significant amounts are isomerised.
Peptide Isomerase Size. The tissue extracts used in the assays were not purified
extracts; they were fractionated by centrifugal ultrafiltration. The fractions with
molecules >30 kDa were used as the source of peptide isomerase because the
molecular size of the enzyme has been estimated at ca. 55–65 kDa [2]. Most of the
peptidases and proteases should have been separated from the fractions >30 kDa

CHEMISTRY & BIODIVERSITY – Vol. 7 (2010)
1609
because of their smaller molecular sizes but presumably there would have been some
retention with the peptide isomerase retentate.
d-Amino Acids in Mammals. d-Amino acids in mammals are not rare although
their occurrence was thought to be a result of nonenzymic processes. d-Serine and d-
aspartic acid have long been known to be present at detectable concentrations in the
mammalian nervous system [12][13]. In mouse brain, high concentrations of d-serine
are detected in the forebrain [14][15]; it acts as an agonist for the N-methyl-d-aspartate
(NMDA) receptor [16]. Serine racemase synthesises d-serine from l-serine in mouse
brain; it is highly specific for l-serine and does not act on any other amino acid [17]. d-
Serine in mouse is degraded by d-amino acid oxidase (DAO) that is involved in the
oxidative deamination of neutral amino acids such as d-serine, d-proline, and d-alanine
[18]. The distributions of serine racemase and DAO correlate with the level of d-serine
in the system [15]. However, the presence of DAO is not restricted to the nervous
system, and it has been reported in mouse kidney, but not in mouse liver or heart [19]. It
is not clear whether DAO has other physiological functions than degrading d-amino
acids [15].
Role of Peptide Isomerase. It is possible that peptide isomerase in mouse heart has a
similar function to serine racemase in the nervous system. However, instead of
synthesising free d-amino acids from l-amino acids, peptide isomerase may catalyse
the l- to d-conversion of particular amino acid residues in peptides. In turn, such
modified peptides could be involved in intracellular signalling pathways, which imply
their binding to d-specific receptors, or alternatively their non-binding when the l-form
does.
Thus, our finding of peptide isomerase activity in mouse heart opens up a new area
of investigation of mammalian biochemistry. It is not known what role peptide
isomerase plays in mouse heart. The heart of the adult mouse weighs only ca. 0.2 g, and
yet it is a highly specialised structure composed of four chambers built up from
involuntary striated cardiomyocytes, connective tissues, blood vessels, and electrically
conductive Purkinje fibres [20].
Closing Comments. In the present work, we showed peptide isomerase activity in
mouse heart that converts the second l-amino acid residue in from the N-terminus into
the d-form. How peptides that contain d-amino acids are involved in maintaining
normal heart function requires further investigation. On the other hand, free d-amino
acids have been found in the heart [21], and it remains to be determined whether they
are formed from the products of peptide isomerisation coupled with peptidase-
mediated degradation, or from other racemisation pathways. The absence of DAO in
heart suggests that these free d-amino acids, no matter how they arise, will be
transported to other tissues for degradation.
The RP-HPLC method for assaying platypus venom peptide isomerase provides a
quick and reliable way to monitor l-/d-amino acid residue interconversion in other
systems whether animate or inanimate; and the general approach is ready to ꢀgo fishingꢁ
for peptide isomerase activities with isomerising substrates that are present in species
other than the platypus.

1610
CHEMISTRY & BIODIVERSITY – Vol. 7 (2010)
This work was funded by an ARC Discovery Project Grant to P. W. K. and Assoc. Prof. J. I.
Vandenberg. Dr. A. Torres, and Dr. P. Bansal are thanked for vital contributions to the work on platypus
venom peptide isomerase. Mr. A. Hallen is thanked for valuable discussions on enzyme purification.
J. M. S. K. was supported by an Australian Postgraduate Award.
Experimental Part
Synthetic Hexapeptides. Customer-specified synthetic hexapeptides were purchased from EZBiolab
(Indiana, USA). To make up the peptide soln., 1 mg of peptides was dissolved in 1 ml of 50 mm
phosphate buffered saline (PBS; 100 mm NaCl, 50 mm Na2H2PO4, pH adjusted with NaOH to 7.4).
Mouse Tissue Extracts. Eight tissues were dissected from wild type C57/BL6 mouse; these tissues
were brain, heart, kidney, liver, lung, small intestine, spleen, and stomach. Each tissue was finely sliced
and then homogenised in 5 ml of ice cold 1% (w/v) trehalose soln. with a Bamix cutter using short bursts
to prevent enzyme denaturation by heating. The homogenate was then centrifuged at 3000g at 48 for
30 min. After centrifugation, the solid material was separated and discarded while the supernatant was
kept and filtered through a 0.2 mm Minisart filter (Sartorius, D-Gçttingen). The filtrate was concentrated
by centrifugal ultrafiltration using a Macrosep 30-kDa molecular weight cut-off membrane (Pall, New
York, USA) and then frozen for later use.
Isomerase Assays. Frozen tissue extract was thawed and further concentrated by centrifugal
ultrafiltration using a Nanosep 30 kDa molecular-weight cut-off membrane (Pall). It was spun at ca.
5000g at 48 for 10 min and then washed with 50 mm PBS (pH 7.4) and 100 mm EDTA (pH 7.0). The
retentate (MW >30 k) was resuspended in 50 mm PBS, pH 7.4. For each assay, 40 ml of the concentrated
tissue extract was mixed with 40 ml of peptide soln. and 40 ml of Complete protease inhibitor cocktail, 5ꢀ
concentration (Roche, D-Mannheim). The mixture was incubated at 378; aliquots of mixture were
withdrawn for RP-HPLC assays at 0, 2, 4, and 24 h.
RP-HPLC. Isomerase assays mixtures were analysed using RP-HPLC with an Agilent Eclipse XDB-
C18 column (4.6 mmꢀ250 mm) on an Agilent Technologies 1200 series system (California, USA). The
solvent system was made up of H2O containing 0.1% (v/v) F3CCOOH (TFA; solvent A) and 90% MeCN
containing 0.1% (v/v) TFA (solvent B). The solvent gradient for the assay was 5–45% B for 15 min, 45–
60% B for 1 min, 60–5% B for 1 min, 5–0% B for 2 min and 0% B for 1 min, at the rate of 1 ml minꢁ1
.
)
Mass Spectrometry. MALDI-TOF-MS was used to determine the mass-to-charge ratio ([MþH]þ
of peptides on a QSTAR XL mass spectrometer equipped with a MALDI source (Applied Biosystems,
California, USA).
REFERENCES
[1] A. M. Torres, M. Tsampazi, C. Tsampazi, E. C. Kennett, K. Belov, D. P. Geraghty, P. S. Bansal, P. F.
Alewood, P. W. Kuchel, FEBS Lett. 2006, 580, 1587.
[2] A. M. Torres, M. Tsampazi, E. C. Kennett, K. Belov, D. P. Geraghty, P. S. Bansal, P. F. Alewood,
P. W. Kuchel, Amino Acids 2007, 32, 63.
[3] A. M. Torres, C. Tsampazi, D. P. Geraghty, P. S. Bansal, P. F. Alewood, P. W. Kuchel, Biochem. J.
2005, 391, 215.
[4] A. M. Torres, G. M. de Plater, M. Doverskog, L. C. Birinyi-Strachan, G. M. Nicholson, C. H.
Gallagher, P. W. Kuchel, Biochem. J. 2000, 348, 649.
[5] S. D. Heck, C. J. Siok, K. J. Krapcho, P. R. Kelbaugh, P. F. Thadeio, M. J. Welch, R. D. Williams,
A. H. Ganong, M. E. Kelly, A. J. Lanzetti, W. R. Gray, D. Phillips, T. N. Parks, H. Jackson, M. K.
Ahlijanian, N. A. Saccomano, R. A. Volkmann, Science 1994, 266, 1065.
[6] C. Ollivaux, D. Gallois, M. Amiche, M. Boscameric, D. Soyez, FEBS J. 2009, 276, 4790.
[7] A. Jilek, C. Mollay, C. Tippelt, J. Grassi, G. Mignogna, J. Mꢃllegger, V. Sander, C. Fehrer, D. Barra,
G. Kreil, Proc. Natl. Acad. Sci. U.S.A. 2005, 102, 4235.
[8] A. Jilek, G. Kreil, Monatsh. Chem. 2008, 139, 1.
[9] P. S. Bansal, A. M. Torres, B. Crossett, K. K. Y. Wong, J. M. S. Koh, D. P. Geraghty, J. I. Vandenberg,
P. W. Kuchel, J. Biol. Chem. 2008, 283, 8969.

CHEMISTRY & BIODIVERSITY – Vol. 7 (2010)
1611
[10] Jmol: an open-source Java viewer for chemical structures in 3D; http://www.jmol.org/.
´
[11] A. Herraez, Biochem. Mol. Biol. Educ. 2006, 34, 255.
[12] D. S. Dunlop, A. Neidle, D. McHale, D. M. Dunlop, A. Lajtha, Biochem. Biophys. Res. Commun.
1986, 141, 27.
[13] A. Hashimoto, T. Nishikawa, T. Hayashi, N. Fujii, K. Harada, T. Oka, K. Takahashi, FEBS Lett.
1992, 296, 33.
[14] A. Hashimoto, T. Nishikawa, T. Oka, K. Takahashi, J. Neurochem. 1993, 60, 783.
[15] L.-Z. Wang, X.-Z. Zhiu, Acta Pharmacol. Sin. 2003, 24, 965.
[16] J. A. Kemp, P. D. Leeson, Trends Pharmacol. Sci. 1993, 14, 20.
[17] H. Wolosker, S. Blackshaw, S. H. Snyder, Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 13409.
[18] M. S. Pilone, Cell. Mol. Life Sci. 2000, 57, 1732.
[19] R. Konno, M. Sasaki, S. Asakura, K. Fukui, J. Enami, A. Niwa, Biochim. Biophys. Acta, Gen. Subj.
1997, 1335, 173.
[20] A. Wessels, D. Sedmera, Physiol. Genomics 2003, 15, 165.
[21] Y. Nagata, T. Akino, K. Ohno, Experientia 1989, 45, 330.