The zinc form of carnosine dipeptidase 2 (CN2) has dipeptidase activity but its substrate specificity is different from that of the manganese form

Accepted Manuscript The zinc form of carnosine dipeptidase 2 (CN2) has dipeptidase activity but its substrate specificity is different from that of the manganese form Nobuaki Okumura, Toshifumi Takao PII:

S0006-291X(17)32078-8

DOI:

10.1016/j.bbrc.2017.10.100

Reference:

YBBRC 38717

To appear in:

Biochemical and Biophysical Research Communications

Received Date: 29 September 2017 Accepted Date: 19 October 2017

Please cite this article as: N. Okumura, T. Takao, The zinc form of carnosine dipeptidase 2 (CN2) has dipeptidase activity but its substrate specificity is different from that of the manganese form, Biochemical and Biophysical Research Communications (2017), doi: 10.1016/j.bbrc.2017.10.100. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

The zinc form of carnosine dipeptidase 2 (CN2) has dipeptidase activity but its

RI PT

substrate specificity is different from that of the manganese form Nobuaki Okumura, and Toshifumi Takao

Institute for Protein Research, Osaka University

To whom correspondence should be addressed: Nobuaki Okumura, Ph.D. Institute for Protein Research, Osaka University 3-2, Yamadaoka, Suita, Osaka, 565-0871, Japan,

FAX: +81-6-6879-4332

TE D

Tel: +81-6-6879-4312

M AN U

SC

Suita, Osaka 5650871, Japan

AC C

EP

E-mail: [email protected]

1

ACCEPTED MANUSCRIPT

Abstract Carnosine dipeptidase II (CN2), a metallopeptidase present in the cytosol of various vertebrate tissues, catalyzes

RI PT

the hydrolysis of carnosine and several other dipeptides in the presence of Mn2+. Although the metal-binding center of mouse CN2 is also able to associate with Zn2+ in vitro, it was not known whether the zinc form of CN2

SC

has any enzymatic activity. In the present study, we show that Zn2+ has a higher affinity for binding to CN2 than Mn2+, as evidenced by native mass spectrometry. The issue of whether the zinc form of CN2 has enzymatic

M AN U

activity was also examined using various dipeptides as substrates. The findings indicate that the zinc form of

CN2 catalyzes the hydrolysis of several different dipeptides including Leu-His, Met-His and Ala-His at a

reaction rate comparable to that for its manganese form. On the other hand, the zinc form of CN2 did not

TE D

catalyze the hydrolysis of carnosine and several other dipeptides that are hydrolyzed by the manganese form of

CN2. Substrate specificity was also examined in HEK293T cells expressing CN2, and the findings indicate that

EP

Leu-His, Met-His, but not carnosine, were hydrolyzed in the cell culture. These results suggest that the zinc form

AC C

of CN2 is an active enzyme, but with a different substrate specificity than that of the manganese form.

Keywords: Metalloprotease, CN2, CNDP2, dipeptide, Native mass spectrometry, Substrate specificity.

Abbreviations: CN2, carnosine dipeptidase II; CN1, carnosine dipeptidase I; ESI-TOF MS, electrospray-ionization time-of-flight mass spectrometry.

2

ACCEPTED MANUSCRIPT

1. Introduction

RI PT

Dipeptides are produced as intermediates in metabolic processes such as protein digestion in the gut, peptide

reuptake in the kidney, and intracellular protein turnover. There are also dipeptides that are enzymatically

SC

synthesized from component amino acids. The latter includes carnosine (β-ala-His), which is stored at high

concentrations in the muscles and the brain [1], and is suggested to have various functions including pH

M AN U

buffering, anti-oxidation, and neuronal regulation [2, 3]. Tissues express a variety of dipeptidases, each of which

hydrolyzes dipeptides with particular specificity and thus may have specific roles in protein and peptide

metabolism. However, characteristics of each enzyme have not fully been elucidated.

TE D

The presence of an enzyme having a Mn2+-dependent carnosine-hydrolyzing activity was first found in

mammalian tissue extracts, and named carnosinase [4, 5]. It was later shown that there are two types of

EP

carnosinases; a Mn2+-dependent bestatin-sensitive one (tissue carnosinase or cytosolic nonspecific dipeptidase),

AC C

and a Mn2+-independent bestatin-insensitive one (serum carnosinase) [6]. In the 2000s, two types of M20 family

metallopeptidases, CN1 and CN2, have been cloned [7, 8]. Their enzymatic properties suggested that CN2 is the Mn2+-dependent one, while CN1 is the Mn2+-independent one.

CN2, also known as CNDP2, is a cytosolic enzyme expressed in various vertebrate tissues [5, 7-9]. In the presence of Mn2+, CN2 hydrolyzes carnosine as well as several other dipeptides such as Leu-His, Met-His and

Gly-Phe [7, 8], indicating that CN2 is involved not only in carnosine metabolism but also in the final step of

3

ACCEPTED MANUSCRIPT

protein breakdown.

A previous X-ray crystallographic study showed that CN2 is a homodimeric protein with each subunit

RI PT

composed of the catalytic domain and the dimerization domain [10]. Each catalytic domain has a catalytic center

with a dinuclear metal-binding site. Interaction of the reaction center with several residues of the dimer

SC

counterpart is essential to the enzymatic reaction as proved by native mass spectrometry [11]. During the

crystallographic study, it was also found that crystals of Zn2+-CN2 complex can be prepared in addition to those

M AN U

of Mn2+-CN2 complex, and that their structures are closely similar to each other (PDB 2ZOF and PDB 2ZOG)

[10]. However, the zinc form of CN2 did not show carnosine-hydrolyzing activity.

In the present study, we first investigated the metal-binding properties of CN2 using native mass

TE D

spectrometry [12, 13]. This method yields mass spectra that reflect the mass values of whole non-covalent

protein complexes in the native state [14, 15], enabling us to analyze protein-protein complexes [11, 16],

EP

protein-ligand interaction [17-19] and protein-metal binding properties [20-22]. Next, we tested whether Zn2+-CN2 had enzymatic activity using various dipeptides as the substrates. As a result, we found that Zn2+-CN2

AC C

has an enzymatic activity whose properties are different from that of Mn2+-CN2.

2. Materials and methods 2.1. Materials

Carnosine was purchased from Peptide Institute Inc. (Osaka, Japan), and other dipeptides were obtained by

4

ACCEPTED MANUSCRIPT

custom synthesis (Genscript, Piscataway, NJ). Anti-CN2 antibody was prepared as reported [8], and anti-CN1

RI PT

antibody was prepared by immunizing a rabbit (Scrum, Tokyo, Japan) with recombinant full-length mouse CN1.

2.2 Preparation of recombinant mouse apo-CN2

SC

Recombinant wild type mouse apo-CN2 and its D132A mutant proteins were expressed in an Escherichia Coli

M AN U

expression system and purified to homogeneity as described previously [11].

2.3. Mass spectrometry

Mass spectrometric analysis was performed by electrospray-ionization time-of-flight mass spectrometry

TE D

(ESI-TOF MS) using an AccuTOF mass spectrometer (JEOL, Akishima, Japan) optimized for detecting high-m/z ions as described previously [11]. Sample protein was prepared at a protein concentration of 10 µM in 100 mM

AC C

of 600 nl/min.

EP

ammonium acetate containing 1 mM DTT, pH 7.4, and introduced into the electrospray ion source at a flow rate

2.4. In vitro dipeptidase assay

Dipeptidase activity against Xaa-His dipeptides was determined by measuring free histidine [23] with some modifications. Standard assay mixture was composed of 20 µg/ml purified recombinant CN2, 10 mM substrate dipeptide, 100 mM NaCl, 1 mM dithiothreitol, 100 µM ZnCl2 or MnCl2, and 100 mM Tris-HCl, pH 7.4. The 5

ACCEPTED MANUSCRIPT

assay mixture was incubated for 30 min at 37 °C, and then a 2 µl aliquot was mixed with 200 µl of 0.3 M NaOH, followed with 40 µl of 10 mg/ml o-phthalaldehyde in methanol. After 5-20 min, absorbance at 405 nm was

SC

RI PT

determined using a microplate reader.

2.5. Expression of CN2 and CN1 in HEK293T cells and dipeptide hydrolysis assay in the cell culture

M AN U

A cDNA encoding full length mouse CN2 [8] or CN1 (clone 5040944, Open Biosystems, Huntsville, AL)

was subcloned into the EcoRI-XhoI sites of pCMV-3Tag1 vector (Stratagene, La Jolla, CA). HEK293T cells were maintained in DMEM containing 10% fetal calf serum, 10 µg/ml gentamycin and 100 µg/ml streptomycin

TE D

under a 5% CO2 humidified air at 37 °C. For transfection, cells were plated on 35 mm culture dishes and allowed to grow to around 80% confluence. The cells were then transfected with an expression vector using

EP

Lipofectamine 3000 regent (Thermo Fisher Scientific, Waltham, MA) in DMEM/Ham’s F12 medium

AC C

supplemented with 10 µg/ml bovine insulin and 10 µg/ml bovine transferrin, and cultured for two days without

changing the medium.

Two days after transfection, medium was changed to 2 ml of Earle’s balanced salt solution (BSS)

supplemented with MEM vitamin solution and 20 mM Hepes-NaOH, pH 7.4. Substrate peptides were then

added to a final concentration of 2 mM and incubated for 60 minutes in 5% CO2 humidified air at 37 °C. After the incubation, a 20 µl aliquot of the culture medium was taken from the dish, and reacted with 10 mM

6

ACCEPTED MANUSCRIPT

o-phthalaldehyde in 30 mM borate-NaOH buffer (pH 9.3) containing 30 mM KCl and 100 mM

2-mercaptoethanol. The resultant reaction mixture was separated on a C18 column (Zorbax Eclipse Plus C18, 2.1

RI PT

x 150 mm, 3.5 micron, Agilent, Santa Clara, CA) operated at a flow rate of 0.2 ml/min on a Hitachi L6000

HPLC system (Hitachi, Tokyo, Japan) with monitoring the absorbance at 335 nm. Elution was carried out by a

SC

gradient created with Solvent A (10 mM sodium phosphate, 10 mM sodium borate, 0.01% sodium azide, pH 8.3)

min: 70 - 100% B, 25 - 28 min: 100% B.

2.6. Western blotting

M AN U

and Solvent B (45% methanol, 45% acetonitrile) as follows; 0- 10 min: 10% B, 10 - 23 min: 10 - 70% B, 23 - 25

TE D

HEK293T cells were lysed in PBS containing 1% Triton X-100 and analyzed by Western blotting [8] with

anti-CN1 and CN2 rabbit antibodies. Immunoreactive bands were detected with a fluorescent secondary

EP

antibody (IRDye 800CW-labelled donkey anti-rabbit IgG, LI-COR, Lincoln, NE) using the 800 channel of an

AC C

Odyssey Fc imager (LI-COR). The membrane was re-probed with mouse anti-FLAG M2 antibody (Agilent),

reacted with Dylight 680-labelled anti-mouse IgG (Cell Signaling Technology, Danvers, MA), and the image

obtained by the 700 channel of the same imager. All antibodies were diluted with Signal Enhancer (Nakalai

Tesque, Kyoto, Japan). Signal intensities were quantified using a software, Image Studio Lite v5.2.5. (Li-COR).

3. Results and Discussion

7

ACCEPTED MANUSCRIPT

3.1. Structure of metal-binding site of CN2 Crystal structure of CN2 in complex with Mn2+ and the competitive inhibitor, bestatin, has been reported

RI PT

previously [10]. CN2 is a 53 kDa protein composed of a catalytic domain and a dimerization domain, the latter

of which provides the dimer interface to form a homodimeric complex (Fig. 1a). The homodimer is stable and

SC

maintained in aqueous solutions as shown by native mass spectrometry [11]. The catalytic domain of each

subunit has a reaction center with a dinuclear metal-binding site composed of H99, D132, E167, D195, and

M AN U

H445 (Fig. 1b). Crystal structure of Zn2+-CN2 complex was also determined, and was closely similar to that of Mn2+-CN2 complex [10].

TE D

3.2. Mass spectrometric detection of apo-CN2 dimer

In order to investigate the metal-binding properties of CN2, we performed native mass spectrometry using an

EP

ESI-TOF mass spectrometer [11]. Fig. 1c shows an overall mass spectrum of apo-CN2 prepared in a

AC C

non-denaturing solvent (0.1 M ammonium acetate, 1 mM DTT, pH 7.4). Peaks detected were predominantly

those derived from CN2 homodimer with charges ranging from 20+ to 24+, suggesting that non-covalent

homodimer of apo-CN2 was successfully detected by this method. A detailed view of the 22+ ion showed that

the m/z value of the peak top (4872.0) was slightly higher than the theoretical value (4862.5) (Fig. 1d). This

difference is possibly due to the presence of adducts with solvent molecules [11].

8

ACCEPTED MANUSCRIPT

3.3. Metal-binding properties of CN2

Next, using this mass spectrometric procedure, we examined the affinity of CN2 to various transition metal

RI PT

ions (Fig. 2). When wild type apo-CN2 was incubated with increasing concentrations of Mn2+ ions (Fig. 2a), the resultant ion peaks of CN2 dimer showed mass shifts to the higher m/z region. In the presence of 10 µM CN2

SC

and 100 µM Mn2+, the shift of the leading edge of the 22+ ion peak was 5.5 m/z at the half-maximum height of this ion peak. This mass shift corresponded to 2.2 Mn2+ ions per CN2 dimer, suggesting that 55% of the

M AN U

metal-binding sites was occupied by Mn2+. On the other hand, the tailing edge of the peaks shifted more than the leading edge especially in the presence of high concentrations of Mn2+. This would probably due to the formation of adducts with Mn2+ ions as well as the solvent molecules.

TE D

In order to exclude the possibility that this Mn2+-induced mass shift was due to non-specific binding of Mn2+

to CN2, we examined a CN2 mutant having a point mutation at D132, which is coordinated to both of the two

EP

metal ions at the metal binding center in wild type CN2-Mn2+ (Fig. 1b) or CN2-Zn2+ complex. This mutant did not show a mass shift even at the Mn2+ concentration of 100 µM (Fig. 2b), suggesting that the Mn2+-induced

AC C

mass shift of wild type CN2 was due to the specific binding of Mn2+ to the metal-binding site. A mass shift of wild type CN2 dimer was also induced by Zn2+ (Fig. 2c). The leading edge of the 22+ ion showed a mass shift of 11.3 m/z in the presence of 100 µM Zn2+.. This shift corresponded to 3.8 zinc ions per CN2 dimer, indicating that Zn2+ occupied 95% of the four metal-binding sites on a CN2 dimer at this concentration. The Zn2+-induced mass shift was abrogated in D132A mutant as in the case with Mn2+ (data not

9

ACCEPTED MANUSCRIPT

shown). On the other hand, Mg2+ did not induce a mass shift in wild type CN2 (Fig. 2d), indicating that Mg2+

does not interact with CN2.

RI PT

Using this system, various divalent metal ions including Ni2+, Fe2+, Zn2+, Co2+, Mn2+ and Mg2+ were tested for its binding activity (Fig. 2e). Among them, Zn2+ showed the highest affinity with the half-maximum binding

SC

concentration of 8 µM. Similar curve was observed with Co2+, but its affinity for CN2 seemed to be slightly lower than that of Zn2+. On the other hand, binding of Mn2+ to CN2 was lower than that of Zn2+ and Co2+. The

M AN U

half-maximal concentration of Mn2+ binding was estimated to be 60 µM. It seems likely that Ni2+ and Fe2+ ions bound to CN2 at concentrations between 5-20 µM. However, at higher concentrations (>50 µM), more than two ions bound to a CN2 monomer, indicating that Ni2+ and Fe2+ ions undergo nonspecific binding to CN2 at high

TE D

concentrations (data not shown). In contrast, Mg2+ ions did not bind to CN2 under the conditions.

Since various metal ions were found to form complexes with CN2, we tested whether these complexes have

EP

carnosine-hydrolyzing activity. However, CN2 complexed with Zn2+, Ni2+, Fe2+, Co2+ or Mg2+ ions showed less than 5% of the activity of Mn2+-CN2 complex (Fig. 2f). When Zn2+ ion was added to Mn2+-CN2 complex, Zn2+

AC C

ions abrogated the Mn2+-induced activity of CN2 in a dose-dependent manner (data not shown), indicating that Zn2+ can be replaced with Mn2+ in vitro.

3.4. Enzymatic activities of Mn2+-CN2 and Zn2+-CN2 against various dipeptides Although carnosine-hydrolyzing activity was prominent only in Mn2+-CN2, we suspected that Zn2+-CN2 also

10

ACCEPTED MANUSCRIPT

had some kind of enzymatic activity, since Zn2+ ions have higher affinity to CN2 and are available in cytosol of most tissues. To examine the possibility, we tested if other Xaa-His dipeptides could be substrates for Zn2+-CN2.

RI PT

As a result, we found that Zn2+-CN2 hydrolyzes Leu-His, Met-His and Ala-His with activities comparable to those with Mn2+-CN2 (Fig. 3a). On the other hand, Ser-His, β-Ala-His, Asn-His, Tyr-His and Pro-His were

SC

hydrolyzed moderately by Mn2+-CN2, while they were hydrolyzed by Zn2+-CN2 with activities less than 10 % compared to those of Mn2+-CN2. This suggests that Zn2+-CN2 and Mn2+-CN2 have different substrate

M AN U

specificities. Activities of Zn2+-CN2 and Mn2+-CN2 against other dipeptides were lower than 5% of their

Met-His-hydrolyzing activities.

In order to examine the mechanisms underlying the differences in their substrate specificity, activities of

TE D

Mn2+-CN2 and Zn2+-CN2 were compared by kinetic analysis. The Vmax values against Met-His were almost similar between the two forms (Zn2+: 100.5 + 2.8 s-1, Mn2+: 106.5 + 7.5 s-1) (Fig. 3b). On the other hand, the Km

EP

value of Zn2+-CN2 was two times higher than that of Mn2+-CN2 (Zn2+: 25.1 + 1.7 mM, Mn2+: 11.5 + 2.8 mM).

Similar characteristics were observed when Leu-His was used as the substrate (Fig. 3c). These results indicate

AC C

that the resistance of several peptides to Zn2+-CN2 may be at least in part due to their low affinity for Zn2+-CN2.

3.5. Dipeptide-hydrolyzing activity of CN2 in HEK293T cells

These differences in substrate specificity indicated that metal ligands have critical roles in the functional

roles of CN2. The affinity of CN2 with metal ions observed in vitro almost conformed to the Irving-Williams

11

ACCEPTED MANUSCRIPT

series (Mn2+ < Fe2+ < Co2+ < Ni2+ < Cu2+ > Zn2+) [24]. However, actual protein metalation is a complex

biological process, which may involve metal transport system, protein folding process and specific metal

RI PT

chaperones [25]. In cyanobacteria, MncA protein conforms the Irving-Williams series in vitro, but it binds to Mn2+ in vivo by folding in Mn2+-rich cellular environment [26]. Therefore, in vitro binding of a protein to metal

SC

ions does not necessary mean the functional protein-metal complexes.

In order to interrogate the activity and metalation state of CN2 under physiological conditions, wild type

M AN U

mouse CN2 was transiently expressed in HEK293T cells, and dipeptide breakdown in the cell culture was

determined (Fig. 4). When carnosine and Leu-His was added to the medium, Leu-His was hydrolyzed to yield

Leu and His, while carnosine was hydrolyzed at a rate lower than 5% of Leu-His hydrolysis (Fig. 4b). On the

TE D

other hand, neither carnosine nor Leu-His was hydrolyzed in mock-transfected cells (Fig. 4a). For comparison,

another M20 family enzyme CN1, which can hydrolyze carnosine in vitro [7], was also examined and found that

EP

CN1-expressing cells hydrolyzed carnosine (Fig. 4c).

AC C

In this assay system, dipeptides added to the culture medium is expected to be incorporated into the cells

through a transporter such as PEPT2 [27], and after hydrolysis, resultant amino acids released to the medium. To

confirm this, conditioned medium obtained from CN2-expressing cells was tested if it has dipeptidase activity.

As shown in Fig. 4d, the activity of condition medium was less the 5% of that observed in cell culture. Therefore,

this assay system would present data that reflect the activity of CN2 in living cells, although it could be affected

by various factors including dipeptide transport and amino acid metabolism.

12

ACCEPTED MANUSCRIPT

Using this assay system, various other dipeptides were also tested if they were hydrolyzed by CN2 in

cultured cells (Fig. 4e). Among them, Met-His and Leu-His were most quickly hydrolyzed in the cells, while

RI PT

carnosine hydrolysis was lower than 5% of Met-His hydrolysis. This pattern of substrate specificity is similar to that observed in Zn2+-CN2 in vitro. On the other hand, there were a few differences between the activity of

SC

CN2-expressing cells and that of Zn2+-CN2 in vitro, i.e., Ala-His, Ser-His and Phe-His were hydrolyzed only by Mn2+-CN2 in vitro, but were hydrolyzed in the cell culture with 20-40% of Met-His hydrolyzing activity.

M AN U

Therefore, the substrate specificity in HEK293T cells might also be affected by some additional factors other than Mn2+ and Zn2+. However, it seems likely that the predominant form of CN2 in this cells is Zn2+-CN2.

Finally, expression levels of CN2 and CN1 in HEK293T cells were confirmed by Western blotting (Fig. 4f).

TE D

The specificities of these antibodies, as well as the relative expression levels of the two proteins were

demonstrated by western blotting with anti-FLAG antibody. Expression levels of CN2 and CN1 was 1:1.1 in

EP

duplicate determinations (data not shown).

In conclusion, the present study demonstrated that Zn2+-CN2 complex has a dipeptidase activity. The

AC C

preference of amino acids at P1 position was more selective in Zn2+-CN2 than in Mn2+-CN2, while P1’ amino

acid preference remains to be determined. Dipeptide hydrolysis in HEK293T cells expressing CN2 showed a substrate preference similar to Zn2+-CN2 complex, indicating that Zn2+-CN2 is actually present in the cells.

These results may advance the understanding of enzymatic properties and physiological functions of CN2.

13

ACCEPTED MANUSCRIPT

Conflicts of interest

RI PT

The authors declare that they have no conflicts of interest.

Funding

SC

This work was supported by Grant-in-Aid for Scientific Research from Japan Society for the Promotion of

M AN U

Science (15K06997).

References

[1] K. Bauer, Carnosine and homocarnosine, the forgotten, enigmatic peptides of the brain, Neurochemical research, 30 (2005) 1339-1345.

[2] A.A. Boldyrev, G. Aldini, W. Derave, Physiology and pathophysiology of carnosine, Physiological

TE D

reviews, 93 (2013) 1803-1845.

[3] K. Nagai, M. Tanida, A. Niijima, N. Tsuruoka, Y. Kiso, Y. Horii, J. Shen, N. Okumura, Role of L-carnosine in the control of blood glucose, blood pressure, thermogenesis, and lipolysis by autonomic nerves in rats: involvement of the circadian clock and histamine, Amino acids, 43 (2012) 97-109.

EP

[4] H.T. Hanson, E.L. Smith, Carnosinase; an enzyme of swine kidney, J Biol Chem, 179 (1949) 789-801. [5] N. Okumura, Carnosine dipeptidase II, in: N.D. Rawlings, Salvesen,G.S. (Ed.) Handbook of proteolytic enzymes, Oxford: Academic Press2013, pp. 1596-1600.

AC C

[6] J.F. Lenney, Separation and characterization of two carnosine-splitting cytosolic dipeptidases from hog kidney (carnosinase and non-specific dipeptidase), Biol Chem Hoppe Seyler, 371 (1990) 433-440. [7] M. Teufel, V. Saudek, J.P. Ledig, A. Bernhardt, S. Boularand, A. Carreau, N.J. Cairns, C. Carter, D.J. Cowley, D. Duverger, A.J. Ganzhorn, C. Guenet, B. Heintzelmann, V. Laucher, C. Sauvage, T. Smirnova, Sequence identification and characterization of human carnosinase and a closely related non-specific dipeptidase, J Biol Chem, 278 (2003) 6521-6531. [8] H. Otani, N. Okumura, A. Hashida-Okumura, K. Nagai, Identification and characterization of a mouse dipeptidase that hydrolyzes L-carnosine, Journal of biochemistry, 137 (2005) 167-175. [9] S. Yamada, Y. Tanaka, S. Ando, Purification and sequence identification of anserinase, Febs J, 272 (2005) 6001-6013.

14

ACCEPTED MANUSCRIPT

[10] H. Unno, T. Yamashita, S. Ujita, N. Okumura, H. Otani, A. Okumura, K. Nagai, M. Kusunoki, Structural basis for substrate recognition and hydrolysis by mouse carnosinase CN2, J Biol Chem, 283 (2008) 27289-27299. [11] N. Okumura, J. Tamura, T. Takao, Evidence for an essential role of intradimer interaction in

science : a publication of the Protein Society, 25 (2016) 511-522.

RI PT

catalytic function of carnosine dipeptidase II using electrospray-ionization mass spectrometry, Protein

[12] P. Lossl, M. van de Waterbeemd, A.J. Heck, The diverse and expanding role of mass spectrometry in structural and molecular biology, The EMBO journal, 35 (2016) 2634-2657.

[13] I. Liko, T.M. Allison, J.T. Hopper, C.V. Robinson, Mass spectrometry guided structural biology,

SC

Current opinion in structural biology, 40 (2016) 136-144.

[14] B. Ganem, Y.T. Li, J.D. Henion, Observation of Noncovalent Enzyme Substrate and Enzyme Product Complexes by Ion-Spray Mass-Spectrometry, J Am Chem Soc, 113 (1991) 7818-7819.

M AN U

[15] D.R. Goodlett, D.G. Camp, 2nd, C.C. Hardin, M. Corregan, R.D. Smith, Direct observation of a DNA quadruplex by electrospray ionization mass spectrometry, Biol Mass Spectrom, 22 (1993) 181-183. [16] C. Schmidt, V. Beilsten-Edmands, C.V. Robinson, Insights into Eukaryotic Translation Initiation from Mass Spectrometry of Macromolecular Protein Assemblies, Journal of molecular biology, 428 (2016) 344-356.

[17] D. Cubrilovic, W. Haap, K. Barylyuk, A. Ruf, M. Badertscher, M. Gubler, T. Tetaz, C. Joseph, J. Benz, R. Zenobi, Determination of protein-ligand binding constants of a cooperatively regulated

TE D

tetrameric enzyme using electrospray mass spectrometry, ACS chemical biology, 9 (2014) 218-226. [18] D. Cubrilovic, R. Zenobi, Influence of Dimehylsulfoxide on Protein-Ligand Binding Affinities, Analytical chemistry, 85 (2013) 2724-2730.

[19] H.Y. Yen, J.T.S. Hopper, I. Liko, T.M. Allison, Y. Zhu, D. Wang, M. Stegmann, S. Mohammed, B. Wu,

EP

C.V. Robinson, Ligand binding to a G protein-coupled receptor captured in a mass spectrometer, Science advances, 3 (2017) e1701016.

[20] Y. Yamazaki, T. Takao, Metalation states versus enzyme activities of Cu, Zn-superoxide dismutase

AC C

probed by electrospray ionization mass spectrometry, Analytical chemistry, 80 (2008) 8246-8252. [21] M. Tomas, J. Domenech, M. Capdevila, R. Bofill, S. Atrian, The sea urchin metallothionein system: Comparative evaluation of the SpMTA and SpMTB metal-binding preferences, Febs Open Bio, 3 (2013) 89-100.

[22] T.W. Rhoads, N.I. Lopez, D.R. Zollinger, J.T. Morre, B.L. Arbogast, C.S. Maier, L. DeNoyer, J.S. Beckman, Measuring copper and zinc superoxide dismutase from spinal cord tissue using electrospray mass spectrometry, Analytical biochemistry, 415 (2011) 52-58. [23] K. Bando, T. Shimotsuji, H. Toyoshima, C. Hayashi, K. Miyai, Fluorometric assay of human serum carnosinase activity in normal children, adults and patients with myopathy, Ann Clin Biochem, 21 ( Pt 6) (1984) 510-514.

15

ACCEPTED MANUSCRIPT

[24] H. Irving, Williams, R.J.P., Order of stability of metal complexes, Nature, 162 (1948) 746-747. [25] A.W. Foster, D. Osman, N.J. Robinson, Metal preferences and metallation, J Biol Chem, 289 (2014) 28095-28103. [26] S. Tottey, K.J. Waldron, S.J. Firbank, B. Reale, C. Bessant, K. Sato, T.R. Cheek, J. Gray, M.J.

RI PT

Banfield, C. Dennison, N.J. Robinson, Protein-folding location can regulate manganese-binding versus copper- or zinc-binding, Nature, 455 (2008) 1138-1142.

[27] D. Zhao, K. Lu, Substrates of the human oligopeptide transporter hPEPT2, Biosci Trends, 9 (2015) 207-213.

M AN U

SC

[28] Schrodinger, LLC, The PyMOL Molecular Graphics System, Version 1.3r1., 2010.

Legends

FIG. 1 Crystallographic and mass spectrometric features of mouse CN2.

TE D

(a) Overall structure of mouse CN2 associated with Mn2+ (magenta) and bestatin (green) (PDB: 2ZOF) [10].

CN2 is present as a homodimer (dark blue and light blue) with each subunit composed of a catalytic domain and

EP

a dimerization domain. Each catalytic domain has a dinuclear metal-binding center where two Mn2+ ions and a

AC C

competitive inhibitor bestatin were associated in this crystal. (b) Structure of the metal-binding site. Two metal

ions were coordinated by H99, D132, E167, D195 and H445. Pictures were drawn using PYMOL 1.3 [28]. (c) Overall mass spectrum of Apo-CN2. Apo-CN2 (10 µM) in 100 mM ammonium acetate (pH 7.4) containing 1

mM DTT was analyzed by ESI-TOF mass spectrometry. Data were collected over a range of 300-8000 m/z.

Arrows indicate the calculated m/z values of apo-CN2 dimer. (d) Magnified view of the 22+ ion peak of

apo-CN2 dimer. The arrow indicates the calculated m/z value of the 22+ ion of apo-CN2 dimer.

16

ACCEPTED MANUSCRIPT

FIG. 2 Mass spectrometric analysis of CN2-metal binding.

RI PT

(a-d) Wild-type apo-CN2 (panel a) or its D132A mutant (panel b) was incubated with indicated concentration of

MnCl2 and analyzed by ESI-TOF mass spectrometry. Data around the 22+ ion peaks were depicted after

SC

smoothing and normalization. Wild-type apo-CN2 was also incubated with indicated concentrations of ZnCl2 (panel c) or MgCl2 (panel d) and analyzed as above. A set of four different length of bars in panels a and c

M AN U

indicates the differences in m/z values corresponding to the binding of 1, 2, 3 and 4 Mn2+ or Zn2+ ions.

Theoretical average mass values are Mn: 54.94, Zn:65.40, recombinant wild type mouse apo-CN2 dimer:

106,954.0 and D132A mutant dimer:106,331.4. Data shown are representatives of three independent experiments.

TE D

Overall spectral features of each spectrum were similar to that shown in Fig. 1c (data not shown). (e) Wild type

apo-CN2 was mixed with metal ions as indicated and analyzed by ESI-TOF mass spectrometry. The number of

EP

metal ions bound to apo-CN2 was calculated from the mass shift observed at the rising ends of 22+ ion peaks.

AC C

Data represent means + SEM of triplicate determinations. (f) Apo-CN2 was incubated with 50 mM carnosine for

30 min at 37 °C in assay mixtures containing divalent metal ions as indicated. Carnosine-hydrolyzing activity

was determined by colorimetric assay of free histidine.

FIG. 3 Substrate specificity of CN2 in the presence of Mn2+ and Zn2+. (a) CN2 was incubated with various Xaa-His dipeptides (10 mM) in the presence of 100 µM MnCl2 or 100 µM 17

ACCEPTED MANUSCRIPT

ZnCl2 for 30 min at 37 °C, and free histidine concentrations were determined. (b-c) CN2 was incubated with various concentrations of Met-His (panel b), Leu-His (panel c) in the presence of 100 µM MnCl2 (open circle) or

RI PT

100 µM ZnCl2 (closed circle) for 30 min at 37 °C, and then free histidine was determined. Data represent means + SEM of triplicate determinations. Regression lines were depicted by fitting the data to the Michaelis-Menten

M AN U

FIG. 4 Dipeptide hydrolysis in HEK293T cells expressing CN2.

SC

equation using the software Origin 9J (OriginLab, Northampton, MA).

(a-c) HEK293T cells were transiently transfected with control, CN2-, or CN1-expression plasmid (panel a, b, c,

respectively), and cultured for 2 days. The cells were then incubated in Earle's BSS containing 2 mM carnosine

TE D

and 2 mM Leu-His for 60 min at 37 °C. Amino acids and dipeptides in the culture media were analyzed by

reversed phase HPLC after pre-column derivatization with o-phthalaldehyde. (d) HEK293T cells expressing

EP

CN2 were incubated in Earle's BSS for 60 min. The medium was then recovered from the culture, filtrated, and

AC C

incubated with 2 mM carnosine together with 2 mM Leu-His for 60 min at 37 °C. Amino acids and dipeptides

were determined as above. (e) Dipeptide hydrolysis activities of CN2-expressing cells were determined as

described above using various dipeptides as the substrates as indicated. Data represented are the ratio of each dipeptide degraded after 60-minute incubation. (f) Western blot analysis of HEK293T cell lysates. Lysates (5 µg

protein each) from cells transfected with control vector (Cont), CN2 expression vector (CN2) and CN1

expression vector (CN1) were analyzed by Western blotting with anti-CN2, anti-CN1 or anti-FLAG antibodies.

18

a

b

RI PT

ACCEPTED MANUSCRIPT

SC

D132

D195

E167

M AN U

H99

c

d

Apo-CN2 dimer

21+

20+

EP

24+

m/z

AC C

22+

Intensity

Intensity

23+

4862.5 (Calculated m/z value)

TE D

22+

m/z

FIG. 1

H445

b D132A Mn (µM) 0 20 50 100

e

m/z

TE D

Mg (µM) 0 5 10 20 50 100

Metal ions per CN2 monomer

WT

AC C

m/z

EP

d Relative intensity

m/z

WT

Zn (µM) 0 5 10 20 50 100

SC

Mn (µM) 0 20 50 100

c

Metal ion (µM)

FIG. 2

f Carnosinase activity (s-1)

WT

M AN U

Relative intensity

a

RI PT

ACCEPTED MANUSCRIPT

m/z

ACCEPTED MANUSCRIPT

a

RI PT

Mn2+

M AN U

SC

Activity (s-1)

Zn2+

Substrate (Xaa-His)

TE D

Activity (s-1)

c

EP

Mn2+ Zn2+

Met-His (mM)

AC C

Activity (s-1)

b

Mn2+ Zn2+

Leu-His (mM)

FIG. 3

b Leu-His

CN1

M AN U

CN2 OD335

OD335

Control Car

c

Car

His

OD335

a

SC

RI PT

ACCEPTED MANUSCRIPT

His

β-Ala

Car

Leu

Leu-His

β-Ala

AC C

Time (min)

Time (min)

anti kDa -CN2 116 97 66 45 31

anti -CN1

anti -FLAG

Cont CN2 CN1

EP

Leu

f

Cont CN2 CN1

Time (min)

TE D

Leu-His

OD335

CN2-CM Car

e

21 14 Xaa-His

FIG. 4

Cont CN2 CN1

Time (min)

Peptide degradation (%)

d

Leu

ACCEPTED MANUSCRIPT

Highlights Carnosine dipeptidase 2 (CN2) forms a high affinity complex with Zn2+. Zn2+-CN2 complex has dipeptidase activity.

RI PT

Substrate specificity of Zn2+-CN2 is different from that of Mn2+-CN2.

AC C

EP

TE D

M AN U

SC

HEK293T cells expressing CN2 hydrolyze dipeptides like Zn2+-CN2.