Báo cáo khoa học: Molecular characterization of a blood-induced serine carboxypeptidase from the ixodid tick Haemaphysalis longicornis docx

Endogenous HlSCP1 is strongly expressed in the midgut and is supposed to localize at lysosomal vacuoles and on the surface of epithelial cells.. Endogenous HlSCP1, identified as a 53 kDa

carboxypeptidase from the ixodid tick Haemaphysalis

longicornis

Maki Motobu1, Naotoshi Tsuji1, Takeharu Miyoshi1, Xiaohong Huang1, M K Islam1,

M A Alim1and Kozo Fujisaki2,3

1 Laboratory of Parasitic Diseases, National Institute of Animal Health, Ibaraki, Japan

2 National Research Centre for Protozoan Diseases, Obihiro University of Agriculture and Veterinary Medicine, Japan

3 Laboratory of Emerging Infectious Diseases, Kagoshima University, Japan

Proteases are known to play essential roles in a wide

range of biological processes, including the

degrada-tion of regulatory proteins, precursor processing,

apop-tosis and digestive processes For both endo- and

ectoparasites, proteases are involved in parasite

inva-sion and survival [1–3] It has been postulated that

exopeptidases may also take part in the proteolytic

cascades for hemoglobin (Hb) degradation [4]

Serine carboxypeptidases (SCPs) belong to the

a⁄ b-hydrolase-fold enzyme superfamily and contain

a conserved amino acid triad, Ser-Asp-His, which

catalyzes hydrolysis of C-terminal residues in peptides

and proteins at acidic pH [5] SCPs are widely distri-buted among fungi, plants and animals Among SCPs, yeast serine carboxypeptidase Y (CPY) has been well studied, and has been shown to participate in the processing of precursors to form secreted mature proteins [6,7] In plants, SCPs have been shown to be involved in growth, apoptosis, brassinosteroid signa-ling and seed development [8–12] In arthropods, however, there is little information available on SCPs Recently, a SCP has been identified in the orange wheat blossom midge, Sitodiplosis mosellana, and has been shown to have dual functions as a digestive

Keywords

blood digestion; haemoglobin; hydrolysis

activity; serine carboxypeptidase; tick

Correspondence

N Tsuji, Laboratory of Parasitic Diseases,

National Institute of Animal Health, National

Agriculture and Food Research Organization,

3-1-5 Kannonndai, Tsukuba, Ibaraki

305-0856, Japan

Fax: +81 29 838 7780

Tel: +81 29 838 7749

E-mail: tsujin@affrc.go.jp

Database

The nucleotide sequence data has been

deposited in the GenBank database under

the accession number AB287330

(Received 7 February 2007, revised 16 April

2007, accepted 1 May 2007)

doi:10.1111/j.1742-4658.2007.05852.x

Ticks feed exclusively on blood to obtain their nutrients, but the gene products that mediate digestion processes in ticks remain unknown We report the molecular characterization and possible function of a serine carboxypeptidase (HlSCP1) identified in the midgut of the hard tick Haem-aphysalis longicornis HlSCP1 consists of 473 amino acids with a peptidase S10 family domain and shows structural similarity with serine carboxypep-tidases reported from other arthropods, yeasts, plants and mammals Endogenous HlSCP1 is strongly expressed in the midgut and is supposed

to localize at lysosomal vacuoles and on the surface of epithelial cells Endogenous HlSCP1, identified as a 53 kDa protein with pI value of 7.5, was detected in the membrane⁄ organelle fraction isolated from the midgut, and its expression was upregulated during the course of blood-feeding En-zymatic functional assays revealed that a recombinant HlSCP1 (rHlSCP1) expressed in yeast efficiently hydrolyzed the synthetic substrates specific for cathepsin A and thiol protease over a broad range of pH and temperature values Furthermore, rHlSCP1 was shown to cleave hemoglobin, a major component of the blood-meal Our results suggest that HlSCP1 may play a vital role in the digestion of the host’s blood-meal

Abbreviations

CPY, yeast serine carboxypeptidase Y; E64, trans-epoxysuccinyl- L -leucylamido-(4-guanidino) butane; Hb, hemoglobin; HlSCP, Haemaphysalis longicornis serine carboxypeptidase; Pyr, L -pyroglutamyl; SCP, serine carboxypeptidase; Suc, succinyl; Z, benzyloxycarbonyl.

enzyme and an exopeptidase involved in degrading

vitellogenin [13]

The host’s blood-meal is the only source of energy in

ticks Unlike blood-sucking insects, ticks make a blood

pool by rupturing blood vessels under the host’s skin

and feed on this blood for a relatively long period,

varying from several days to weeks, depending on the

life stage, the host type and the species of tick involved

[14] Only one blood-meal is taken during each life

stage, and after completion of feeding, ticks can survive

for several months without a further blood-meal [15]

Because ticks have such unique feeding behavior, it is

speculated that they are equipped with an efficient

blood-digestion and nutrient-utilization system for

sur-vival In addition, ticks act as vectors of disease-causing

agents in humans and animals by injecting their saliva,

which contains anticoagulants and other bioactive

com-ponents as well as pathogens, into the blood pool

dur-ing feeddur-ing [16] Suppression of tick vector populations

is thus crucial for controlling diseases transmitted by

ticks However, chemical acaricides, which are

cur-rently used for tick control, have the disadvantages of

causing acaricidal resistance problems [17], leading to

food animals containing chemical residues, which are a

threat to human health Consequently, novel

approa-ches are sought to control tick populations based on

tick-specific potential biochemical pathways

We describe here the cloning and partial

characteri-zation of a cDNA encoding a SCP from the ixoidid

tick Haemaphysalis longicornis, which has a wide

geo-graphical distribution in Russia, eastern Asia,

Austra-lia, and New Zealand, and has the potential to

transmit pathogens including viruses, rickettsia and

protozoan parasites that cause important human and

animal diseases [18–20] The deduced precursor protein

contains amino acids conserved among peptidase S10

family members and SCPs Endogenous H longicornis

serine carboxypeptidase (HlSCP1) was strongly

expressed in the vacuoles of midgut epithelial cells,

and its expression was found to be upregulated by

the blood-digestion process A recombinant HlSCP1

(rHlSCP1) expressed in Phicia pastoris hydrolyzed not

only synthetic peptide substrates for SCP and thiol

proteases, but also bovine Hb These findings suggest

that HlSCP1 may be involved in digestion of the host’s

blood-meal

Results

HlSCP1 cDNA encodes a SCP homolog

Sequence analysis revealed that HlSCP1 cDNA is

1688 bp long The start codon is predicted at

nucleo-tides 146–148 and there is a stop codon at nucleonucleo-tides 1565–1567 HlSCP1 cDNA has an ORF extending from position 146 to position 1567, coding for 473 amino acids with a predicted molecular mass of

53 294 Da (Fig 1) This deduced protein has a poss-ible cleavable signal peptide of 27 amino acid residues and the preprotein has a predicted molecular mass of

50 365 Da The HlSCP1 sequence possesses a single peptidase S10 family domain consisting of the evolu-tionarily conserved regions and three catalytic residues [21] These serine, histidine, and aspartic acid residues that are known to form the catalytic triad of SCPs are found in HlSCP1 at positions 178, 450 and 397, respectively There are three potential sites (residues

357, 416 and 439) for N-glycosylation in the putative polypeptide encoded by HlSCP1 A search of the protein database using the National Center for Bio-technology Information revealed that HlSCP1 has sequence similarity with SCPs from Sitodiplosis mosell-ana (GenBank accession no AAY27740, 27% identity), Arabidopsis thaliana (NP_194790, 32%), Saccharomyces cerevisiae CPY (CAA56806, 28%), chicken cathepsin A (NP_001026662, 39%), mouse cathepsin A (AAA39982, 40%) and human cathep-sin A (NP_000299, 39%)

Endogenous HlSCP1 is localized in midgut epithelial cells

Two-dimensional immunoblotting was performed to identify endogenous HlSCP1 in ticks Mouse anti-rHlSCP1 serum reacted with a protein having a molecular mass of 53 kDa and a pI of 7.5 (Fig 2A), corresponding to the predicted size of the putative pro-tein calculated from the HlSCP1 amino acid sequence

To detect the localization of endogenous HlSCP1, immunohistochemistry was performed in adult female ticks, blood-fed for a total period of 72 h, using mouse anti-rHlSCP1 serum It was found that endogenous HlSCP1 was mainly expressed in the midgut of the female ticks (Fig 2B) Serum from mice prior to immunization, however, did not show any reactivity

Subcellular localization of HlSCP1

To clarify the subcellular localization of HlSCP1 in the midgut, immunoblot analysis of the subcellular frac-tions of midgut tissues obtained from 72-h-fed adult female H longicornis was performed using mouse anti-HlSCP1 serum An immunoreactive band was detected

in the fraction containing membranes and membrane organelles (Fig 3A) To determine the endogenous form of HlSCP1 in the midgut epithelial cells,

Fig 1 Comparison of the deduced HlSCP1 amino acid sequence with several known SCPs S.m., Sitodiplosis mosellana (GenBank acces-sion no AAY27740); A.t., Arabidopsis thaliana (NP_194790); Ye, Saccharomyces cerevisiae carboxypeptidase Y (CAA56806); Mo, mouse cathepsin A (AAA39982); Hu, human cathepsin A (NP_000299); Ch, chicken (NP_001026662) Asterisks, identical or conserved residues; colons, conservative substitutions; periods, semiconservative substitutions Three conserved regions, characteristic of SCPs, are underlined Conserved catalytic triad residues are marked with rhombuses The vertical arrow shows the signal peptide cleavage site Numbers on the right refer to the amino acids within the sequences.

immunofluorescent staining of flat sections of 72-h-fed

adult H longicornis was performed Examination of the

stained sections revealed that HlSCP1 is localized in the

vacuoles of midgut endothelial cells, where the ingested

host’s blood-meal is thought to be degraded by

proteas-es (Fig 3B) Thproteas-ese rproteas-esults suggproteas-est that that HlSCP1 is

localized in lysosomal vacuoles and on the cell surface

Expression of endogenous HlSCP1 is induced

by the blood-feeding process

Expression levels of endogenous HlSCP1 were also

examined in unfed and partially fed adult ticks

Immu-noblot analysis showed that HlSCP1 was expressed

weakly at 24 h and its expression was significantly

increased at 72 and 96 h of blood-feeding (Fig 4A) In

accordance with the results of immunoblot analysis, the fluorescence intensity of HlSCP1 in the midgut epithelial cells also gradually increased as blood-feeding progressed, reaching a maximum at 72 h of blood-feeding (Fig 4B) These results suggest that endogenous HlSCP1 expression is induced by the blood-feeding process

Purity of the recombinant HlSCP1 The ORF of HlSCP1, except for the signal sequence, was subcloned into the pPICZB vector (Invitrogen, Carlsbad, CA) Recombinant HlSCP1 was expressed in

P pastoris and found to migrate as a 54 kDa fusion protein with a hexahistidine tag of 3 kDa by SDS⁄ PAGE (Fig 5, lane 3) The molecular mass of rHlSCP1 protein is  51 kDa, similar to the mass pre-dicted from the amino acid sequence of HlSCP1 exclu-ding the signal sequence rHlSCP1 was purified by metal chelation chromatography under native condi-tions The purified rHlSCP1 was used for enzyme activity and Hb hydrolysis assays

Recombinant HlSCP1 is an active SCP

A series of synthetic substrates for SCPs was used in kinetic analyses to determine the values of Km, kcat and kcat⁄ Km for rHlSCP1 As shown in Table 1, rHlSCP1 had a substrate preference for benzyloxy-carbonyl (Z)-Phe-Leu and Z-Phe-Ala over Z-Glu-Tyr The kcatvalues were comparable among the substrates, but the Km value for Z-Glu-Tyr was 10 times higher than that of the other substrates SCPs, classified as

55

36

66

A

B

31

pI

Fig 2 Endogenous HlSCP1 in adult female H longicornis (A)

Iden-tification of HlSCP1 by 2D-PAGE Fifty micrograms of tick extract

were separated using 2D pH gradient gel electrophoresis, and the

proteins were transferred to a nitrocellulose membrane The arrow

shows endogenous HlSCP1 (B) Immunohistochemical localization

of HlSCP1 Ticks after 72 h of blood-feeding were fixed in

parafor-maldehyde and embedded in paraffin Flat sections of a whole adult

tick were exposed to mouse anti-HlSCP1 serum (upper; scale bar,

1 mm; lower left; scale bar, 100 lm) or preimmune mouse sera

(lower right; scale bar, 100 lm) cu, cuticle; mu, muscle; mg,

mid-gut; sg, salivary gland The area marked by a square is shown at

higher magnification.

B

A

Fig 3 Intracellular localization of endogenous HlSCP1 in midgut epi-therial cells (A) Identification of HlSCP1 by immunoblotting Fraction-ated proteins of midgut epithelial cells were separated by SDS ⁄ PAGE, and the proteins were then transferred to a nitro-cellulose membrane The membrane was reacted with mouse anti-HlSCP1 serum diluted 1 : 150 Cy, cytosolic fraction; Me, mem-branes and organelles; Nu, nuclear fraction; Cs, cytoskeleton The arrow shows endogenous HlSCP1 (B) Immunofluorescence staining

of HlSCP1 Midgut epithelial cells were collected from midguts of

72 h blood-feeding ticks The cells were fixed in paraformaldehyde, permeabilized with Triton X-100, and incubated with mouse anti-HlSCP1 serum followed by visualization with Alexa Fluor 488 (green) Nuclei were stained with DAPI (blue) (scale bar, 50 lm).

carboxypeptidase Cs, have high affinity for

hydropho-bic amino acids at the P1¢ position [5] The affinity of

members of the carboxypeptidase C family, such as

CPY and cathepsin A, for hydrophobic amino acids is

similar to that of rHlSCP1 [22,23] In addition to SCP

activity, it was found that rHlSCP1 hydrolyzed

l-pyro-glutamyl (Pyr)-Phe-Leu-pNA, a substrate for thiol

proteases (Km¼ 1.46 · 10)4m)1, kcat¼ 20.1 s)1) and

showed much lower enzyme activities toward

Z-Ala-Ala-Leu-pNA, succinyl (Suc)-Ala-Ala-pNA and

Bz-(dl)-Arg-pNA, substrates for subtilisin A, elastase and

trypsin, respectively (data not shown) To determine

the optimum conditions for rHlSCP1 activity toward

the substrates Z-Phe-Leu and Pyr-Phe-Leu-pNA, the enzyme activities were assayed at different pH values and temperatures The hydrolysis of Z-Phe-Leu by rHlSCP1 was optimal at pH 6 but significantly decreased at pH values > 7 (Fig 6A) By contrast, enzyme activity toward Pyr-Phe-Leu-pNA had a pH optimum of 7 and weak activity was still seen at pH 9 With increasing temperature, the activity was enhanced, reaching a maximum at 45 C for both sub-strates (Fig 6B) Relatively higher enzyme activities toward Z-Phe-Leu were observed over a broad tem-perature range (37–50C) The same activities were also seen when rHlSCP1 was preincubated at 37–50C

kDa

A

B

Fed (h)

31

55

36 66

21

Fig 4 Induction of endogenous HlSCP1

expression by the blood-feeding process (A)

Expression pattern of endogenous HlSCP1

during the blood-feeding process Soluble

antigens from unfed and partially fed

(24–96 h blood-feeding) adult ticks were

separated by SDS ⁄ PAGE, and the proteins

were transferred to a nitrocellulose

membrane The membrane was reacted

with mouse anti-HlSCP1 serum diluted

1 : 150 The arrow shows endogenous

HlSCP1 (B) Immunofluorescence staining of

HlSCP1 in midgut epithelial cells Ticks were

fixed in paraformaldehyde and embedded in

paraffin Flat sections of a whole adult tick

were exposed to mouse anti-HlSCP1 serum

followed by visualization with Alexa Fluor

488 (green) Nuclei were stained with DAPI

(blue) (scale bar, 25 lm).

for 3 h The effects of protease inhibitors on rHlSCP1

hydrolysis activities were examined (Table 2), and the

inhibition pattern indicated that the recombinant

enzyme belongs to the serine protease family, because

phenylmethylsulfonyl fluoride, a serine protease

inhib-itor, significantly inhibited the hydrolytic activity of

rHlSCP1 toward both substrates (Table 2) It is

nota-ble that pepstatin A, an aspartic protease inhibitor,

showed different inhibitory effects between the

sub-strates Pyr-Phe-Leu-pNA and Z-Phe-Leu: it inhibited

the hydrolysis of Pyr-Phe-Leu-pNA, a thiol substrate,

more potently than cysteine protease inhibitors such

as trans-epoxysuccinyl-l-leucylamido-(4-guanidino) butane (E64) and leupeptin, whereas only a slight inhibitory effect of pepstatin A on the hydrolysis of

kDa

66

55

36

31

3

21

14

Fig 5 SDS ⁄ PAGE analysis of rHlSCP1 expressed in P pastoris.

The proteins expressed by pPICZ B ⁄ HlSCP1 were detected by

sil-ver staining Lane 1, P pastoris lysates before induction; lane 2,

P pastoris lysates 7 h after induction with methanol; lane 3,

rHlSCP1 after purification with a Ni+-chelating column under native

conditions The arrow shows rHlSCP1.

Table 1 Kinetic constants for peptide substrate hydrolysis by

rHlSCP1 HlSCP1 (0.1 mg) was incubated in 25 m M citrate ⁄

phos-phate buffer (pH 6) with 0.15–3 m M of the substrates at 45 C.

Results shown are the means from duplicate experiments.

Substrate

Km

( M )

kcat (s)1)

kcat⁄ K m ( M )1Æs)1)

Temperature (ºC)

0 25 50 75 100

Pyr-Phe-Leu-pNA

Z-Phe-Leu

Pyr-Phe-Leu-pNA

Z-Phe-Leu

0 25 50 75

100

A

B

pH

Fig 6 Effect of pH (A) and temperature (B) on rHlSCP1 activity Enzyme activity was assayed using Z-Phe-Leu or Pyr-Phe-Leu-pNA

as a substrate in 25 m M citrate ⁄ 50 m M phosphate buffer (pH 4–7)

or 50 m M sodium phosphate buffer (pH 8–9) Data are expressed

as the mean percent enzyme activity relative to the maximum activity ± SD (n ¼ 3).

Table 2 Inhibition of rHlSCP1 enzyme activity toward Z-Phe-Leu and Pyr-Phe-Leu-pNA by proteinase inhibitors rHlSCP1 (0.1 mg) was incubated in 25 m M citrate ⁄ 50 m M phosphate buffer (pH 6 for Z-Phe-Leu or pH 7 for Pyr-Phe-Leu-pNA) at 45 C with the protease inhibitors Inhibitory effect is indicated as percentage relative to the maximum hydrolytic activity of rHlSCP1.

Inhbitor

Conc.

(m M )

%inhibition relative to control Z-Phe-Leu Pyr-Phe-Leu-pNA

Phenylmethylsulfonyl fluoride

Z-Phe-Leu was observed Antipain, a trypsin-like

ser-ine protease inhibitor, slightly inhibited the hydrolysis

of both synthetic peptides

Recombinant HlSCP1 hydrolyzes Hb

To assess the hydrolyzing efficiency of HlSCP1, bovine

Hb was incubated with rHlSCP1 for 9 h and was

sub-jected to SDS⁄ PAGE analysis As shown in Fig 7A,

rHlSCP1 hydrolyzed Hb at 45C in a dose-dependent

manner A major portion of Hb was degraded after

6 h of incubation at this temperature (Fig 7B)

rHlSCP1 was shown to hydrolyze Hb efficiently in the

pH range 5–6 (Fig 7C,D) In inhibition experiments,

pepstatin A was shown to inhibit Hb hydrolysis by

rHlSCP1 significantly more potently than

phenyl-methylsulfonyl fluoride, whereas other protease

inhibi-tors showed only weak inhibitory effects (Fig 7E)

Discussion

SCPs are known to play important roles in precursor

processing, growth and apoptosis in bacteria and

plants However, very little is known about SCPs in arthropods, including hematophagous ticks We have isolated a full-length cDNA encoding an SCP from the hard tick H longicornis, whose amino acid sequence shows similarity with mammalian cathepsin A sequences HlSCP1 possesses the catalytic triad and conserved consensus sequence motifs of the SCPs; however, substitution of an alanine for a cysteine was observed in a conserved region (81WLNGGPG-ASS90) In humans, a mutant cathepsin A, in which cysteine was replaced by threonine in the WLNGGPGCSS region, was enzymatically inactive, and accumulated in the rough endoplasmic reticulum, suggesting that the cysteine residue in the conserved region has an important role in creating a proper con-formation for the interaction with substrates and intra-cellular transport [24] However, the same substitution

is observed in SCP of S mosellana and vitellogenic carboxypeptidase of Aedes aegypti, which has similar-ity with SCP [25] These findings indicate that the alan-ine residue in the WLNGGPGASS region is conserved evolutionarily in arthropods and may not be involved

in the enzymatic activity In the case of cathepsin A, a

rHlSCP1 (µg/reaction)

A

Incubation time (h)

B

0.0 0.5 1.0

Temperature (°C)

C

Co

pH

D

0.0

E

0.5 1.0

0.0 0.5 1.0

0.0 0.5 1.0

0.0 0.5 1.0

Fig 7 Effect of rHlSCP1 concentration (A),

incubation time (B), temperature (C), pH (D)

and inhibitor (E) on hydrolysis of bovine Hb.

Bovine Hb (1 lg) was incubated with

rHlSCP1 in 25 m M citrate ⁄ 50 m M phosphate

buffer (pH 4–7) or 50 m M phosphate buffer

(pH 8) Co, reaction buffer containing bovine

Hb without addition of rHlSCP1 and

inhibi-tors.

54 kDa precursor is cleaved into a mature heterodimer

of 32 and 20 kDa subunits, which are linked by

disul-fide bonds [26–28], and a 34 and 20 kDa form also

exists as a transient processing intermediate [29] No

proteolytic cleavage site except for the signal peptide

site was detected at the amino acid sequence level in

HlSCP1 using a cleavage site prediction server (http://

bp.nuap.nagoya-u.ac.jp/sosui) In addition, the

het-erodimer form was not observed when immunoblot

analysis was performed under reducing or nonreducing

conditions (data not shown), indicating that HlSCP1 is

a single-chain enzyme, like CPY

In hematophagous arthropods, it has been suggested

that proteases induced by blood-feeding in the midgut

may play a crucial role in blood digestion [30–32] In

ticks, digestion occurs intracellularly in the midgut

epi-thelium and is accomplished by lysosomal hydrolytic

enzymes in vacuoles [33–35] Various kinds of protease

activities, including exopeptidase, have been described

in the midgut of the cattle tick Boophilus microplus

[36] In this study, we have shown that endogenous

HlSCP1 was strongly expressed in vacuoles of midgut

epithelial cells Furthermore, its expression was

upreg-ulated after blood-feeding These results imply that

HlSCP1 may function as a lysosomal protease in the

process of blood digestion Based on the results of a

search using sosui, a structure prediction program for

membrane proteins [37] (http://bp.nuap.nagoya-u.ac.jp/

sosui/), HlSCP1 is speculated to be a membrane

pro-tein Although endogenous HlSCP1 was detected in

the membrane⁄ organelle fraction extracted from

mid-guts after 72 h of blood-feeding, it has not been

con-firmed whether HlSCP1 exists as a membrane protein

in the vacuoles of midgut epithelial cells Detailed

studies of HlSCP1 localization will be performed in

the future

rHlSCP1 showed optimum enzyme activity at acidic

pH and had a substrate preference for Z-Phe-Leu,

properties which are consistent with those of

cathep-sin A [28,38] In addition to SCP activity, it has been

reported that cathepsin A has deamidase⁄ esterase

activ-ity, which catalyzes the hydrolysis of bioactive peptide

hormones which contain hydrophobic amino acids at

the C-terminus at neutral pH [39,40] It has also been

demonstrated that cathepsin A hydrolyzes

Suc-Leu-Leu-Val-Tyr-AMC [41] and

Suc-Phe-Leu-Phe-thio-benzyl ester [42], suggesting that cathepsin A attacks

N-blocked substrates having an aromatic amino acid at

their C-terminus Therefore, it is thought that

deami-dase⁄ esterase preferentially cleaves peptides where the

P1¢ and ⁄ or P1 position is a hydrophobic amino acid

[39–41] Similar results were observed with rHlSCP1,

which hydrolyzed a thiol substrate, Pyr-Phe-Leu-pNA,

efficiently at pH 7, and this activity was shown to be inhibited by phenylmethylsulfonyl fluoride In addition, rHlSCP1 partially hydrolyzed Z-Ala-Ala-Leu-pNA and Suc-Ala-Ala-pNA, but not Bz-(dl)-Arg-pNA (data not shown), indicating that rHlSCP1 also has a preference for hydrophobic amino acids in the P1 position These results suggest that HlSCP1 has enzyme activity towards various substrates

It is noteworthy that rHlSCP1 has enzymatic activ-ity over a wide range of temperatures (37–50C), and this feature has not been reported for other SCPs except plant SCP having enzymatic activity at 37–

55C [12] This enzymatic property may be related to the feeding season of H longicornis and body tempera-ture of its host In general, the feeding activity of ticks occurs from spring through summer, and the body temperature of hosts such as cattle, dogs and poultry

is higher than 37C To digest blood-meals efficiently under different environmental conditions, a broad tem-perature dependency of the enzyme activity would be required

It was observed that HlSCP1 has potent ability to hydrolyze bovine Hb under the same reaction condi-tions as used for the hydrolysis of synthetic peptides However, it is unclear why the Hb hydrolysis was signi-ficantly inhibited by pepstatin A, an aspartic protease inhibitor Because rHlSCP1 was extracted from P pas-toris producing rHlSCP1 and purified using metal chelation chromatography, the possibility of contamin-ation by aspartic proteases derived from P pastoris is ruled out It has been reported that CPY and human cathepsin A have different preferences for SCP inhibi-tors, implying that nonconserved amino acid residues in the active sites may contribute to the preference of inhibitors [43] Although HlSCP1 possesses the con-served regions and catalytic triad of the SCPs, its over-all sequence similarity with other SCPs is < 50% Therefore, nonconserved regions may be important factors determining the preference of inhibitors

Previous studies have shown that an aspartic prote-ase and serine proteprote-ase derived from H longicornis hydrolyzed rabbit Hb or BSA, suggesting that those proteases are involved in blood digestion in ticks [44,45] Because it is postulated that host Hb is degra-ded by multiple proteases, including aspartic, cysteine and metalloproteases, in blood-feeding parasites [4], the hard tick H longicornis also might have such a Hb-degradation cascade involving multiple proteases

In endoparasites, aspartic, cysteine and

metalloproteas-es are found in the intmetalloproteas-estine, and the analysis of Hb proteolysis by those recombinant proteases indicates that Hb would be initially degraded by aspartic and cysteine proteases, followed by metalloproteases [46]

More detailed studies of the Hb digestion cascade have

been performed in the malaria parasite Plasmodium

falciparum [47,48] Those studies showed that Hb

deg-radation occurs in the food vacuole, where plasmepsin,

a member of the aspartic protease family, initially

cleaves in the conserved hinge region of the Hb alpha

chains [49,50] Falcipain, a member of the cysteine

protease family, degrades the denatured globin [50,51],

and the resultant globin peptides are target molecules

for the zinc metalloprotease falcilysin [52]

Further-more, P falciparum aminopeptidases possess enzymatic

activity to hydrolyze peptides derived from the

endo-proteolytic digestion of hemogloblin to amino acids

[53], supporting the idea that aminopeptidases take

part in the terminal stages of hemoglobin degradation

HlSCP1, having the SCP activity to release amino

acids sequentially from the C-terminus of peptide

chains, may function at the terminal stage of blood

digestion following aspartic and serine proteases

In this study, our results suggested that SCP

identi-fied in the hard tick H longicornis takes part in the

host’s blood-digestion process The protease cascade

for Hb digestion would be crucial for the survival of

ticks, for which the blood-meal from the host is the

only source of energy Elucidation of the function of

HlSCP1 in the hard tick may contribute a better

understanding of the physiology of blood digestion

and development, and thus improved tick control

Experimental procedures

Ticks

Adults of H longicornis obtained from the parthenogenetic

Okayama strain maintained at the Laboratory of Parasitic

Diseases, National Institute of Animal Health (Tsukuba,

Ibaraki, Japan), were bred by feeding on rabbits as

described previously [18]

Animal ethics

All animals used in this study were acclimatized to the

experimental conditions for 2 weeks prior to the

experi-ment Animal experiments were conducted in accordance

with the protocols approved by the Animal Care and Use

Committee, National Institute of Animal Health (Approval

nos 441, 508, and 578)

Cloning and sequencing of HlSCP1

HlSCP1 was identified from expressed sequence tags

con-structed from the midgut cDNA libraries of H longicornis

as described previously [45] Briefly, the plasmids containing

HlSCP1 gene-encoding inserts were extracted using the Qiagen DNA Purification kit (Qiagen, Valencia, CA) The nucleotide sequences of the cDNAs were determined by the big-dye terminator method on an ABI PRISM 3100 automated sequencer (Applied Biosystems, Foster City, CA) BioEdit sequence alignment editor (Isis Pharmaceuticals, Inc, Carlsbad, CA) and the BLAST network server of the National Center for Biotechnology Information (National Institutes of Health, Bethesda, MD) were used to analyze the nucleotide sequence and deduce the amino acid sequences for determining similarities with previously repor-ted sequences in GenBank A primary sequence motif was identified using the PROSITE network server at EMBL Analysis of the signal sequence was performed using signal

ip3.0 at the Center for Biological Sequence Analysis (http://www.cbs.dtu.dk/services/SignalP/)

Generation of anti-HlSCP1 serum Anti-HlSCP1 serum was obtained from a mouse immun-ized with Escherichia coli-expressed recombinant HlSCP1 (rHlSCP1) One set of oligonucleotide primers derived from

an ORF of the HlSCP1 gene was used: a sense primer (5¢-GGGGTACCCCAATGTATACGGTAACCATG-3¢), corresponding to nucleotides 146–163 of the HlSCP1 nucleotide sequence and an antisense primer (5¢-CGGA ATTCCGACTAAAGTGGCTTATTGGC-3¢), correspond-ing to nucleotides 1551–1567 of the HlSCP1 nucleotide sequence The nucleotide sequence of these primers contained KpnI and EcoRI restriction sites, respectively Amplified product was inserted into the pTrcHis B plasmid (Invitrogen, Carlsbad, CA) E coli expression vector after digestion with KpnI and EcoRI The resulting plasmid (pTrcHis B⁄ HlSCP1) was transformed into E coli (Top10F¢, Invitrogen) using standard techniques Expres-sion of the HlSCP1 cDNA in E coli was performed essen-tially as described [54] rHlSCP1 was purified using metal chelation chromatography (Invitrogen) under denaturing conditions as described in the manufacturer’s protocol Proteins eluted with imidazole were concentrated using Centrisart (molecular mass cut-off, 10 000 Da; Sartorius, Goettingen, Germany) and then dialyzed against NaCl⁄ Pi using a Slide-A-Lyzer Dialysis Cassette (Pierce Biotechno-logy, Rockford, IL) Protein concentration was determined using micro-BCA reagent (Pierce) Antisera against rHlSCP1 were generated in BALB⁄ c mice (Japan SLC, Inc., Hamamatsu, Japan) by subcutaneous injection with

100 lg of rHlSCP1 emulsified with complete (first injec-tion) or incomplete (second and later injections) Freund’s adjuvant at 2-week intervals Mice were bled 2 weeks after the fourth injection Antisera from the mice were stored at )20 C until used Animal studies were approved by the Animal Care and Use Committee, National Institute of Animal Health (Approval no 569)

2D electrophoresis

Tick extracts from unfed and partially fed adults were

pre-pared as previously described [55] Tick extracts were treated

with an equal volume of urea mixture composed of 9 m urea,

4% Nonidet P-40, 0.8% ampholine (pH 3.5–10; GE

Health-care, Piscataway, NJ) and 2% 2-mercaptoethanol, and then

subjected to 2D PAGE Non-equilibrium pH gradient

elec-trophoresis was performed in the first dimension using a

rect-angular gel electrophoresis apparatus (AE-6050A; ATTO,

Tokyo, Japan) After electrophoresis at 400 V for 2 h, gels

were incubated in equilibration buffer for 10 min on a

sha-ker Electrophoresis in the second dimension was performed

on 12.5% SDS⁄ PAGE gels under reducing conditions The

proteins were transferred to nitrocellulose membranes

Immunoblot analysis

Immunoblot analysis was performed as previously

des-cribed [44] Tick extracts or rHlSCP1 separated by 1D or

2D electrophoresis were transferred onto nitrocellulose

membranes, and the membranes were incubated for 30 min

with 5% skim milk For the detection of endogenous

HlSCP1 or rHlSCP1, antiserum against rHlSCP1 diluted

1 : 150 was used After the membranes were washed with

Tris-buffered saline containing 0.1% Tween-20 (NaCl⁄

Tris-T), they were incubated with alkaline

phosphatase-conju-gated goat anti-(mouse IgG) (Invitrogen) as a secondary

antibody After the membranes were washed, the proteins

bound to the secondary antibody were visualized with

Nitro Blue Terazolium⁄ 5-bromo-4-chloro-3-indolyl

phos-phate (Invitrogen)

Immunohistochemistry

Immunohistochemistry was performed with

peroxidase-labeled goat anti-(mouse IgG) secondary antibody as

des-cribed previously [44] Flat sections of whole partially fed

adults were exposed to mouse antirHlSCP1 serum diluted

1 : 150 overnight at 4C After washing with NaCl ⁄ Pi, the

sections were reacted with peroxidase-labeled mouse IgG

secondary antibody and the substrate

3¢,3¢-diaminobenzi-dine tetrahydrochloride (Sigma Fast DAB set; Sigma

Aldrich, St Louis, MO)

Preparation of protein fractions

Midguts collected from adult ticks after 72 h of

blood-feeding were subjected to stepwise extraction of cytosolic,

membrane, nuclear and cytoskelton proteins using a

Pro-teoExtract subcellular proteome extraction kit and

follow-ing the manufacturer’s instruction (S-PEK; Calbiochem,

San Diego, CA) Expression of endogenous HlSCP in each

fraction was analyzed by immunoblot analysis as described

above

Immunofluorescence staining For intracellular localization of HlSCP1, midguts were collected from adult ticks after 72 h of blood-feeding, and midgut cells were prepared by teasing midguts through a stainless steel mesh To remove cell debris and host’s eryth-rocytes, the cells were fractionated by centrifugation on a Percoll density gradient (GE Healthcare) Isotonic Percoll solution was made with 10· NaCl ⁄ Pi(pH 7.4) and Percoll (1 : 9 v⁄ v), and diluted in NaCl ⁄ Pi containing 1% fetal bovine serum A Percoll gradient was made by placing 50 and 80% isotonic Percoll in the centrifuge tube, and then the midgut cell suspension was slowly placed on top of the gradient and centrifuged for 20 min at 1700 g at room tem-perature The 50% Percoll fraction was collected and washed with NaCl⁄ Pi, and the cells were attached to a glass slide using Shandon cytospin (Thermo Electron, Walt-ham, MA) After fixation with 4% paraformaldehyde in NaCl⁄ Pi for 20 min at room temperature, the cells were permeabilized in NaCl⁄ Pi containing 0.1% Triton X-100 for 20 min at room temperature After washing with NaCl⁄ Pi, cells were blocked with 10% goat serum (MP Bio-medicals, Irvine, CA) for 30 min at room temperature, and then incubated with mouse anti-rHlSCP1 serum diluted

1 : 150 for 1 h at room temperature The cells were washed three times with NaCl⁄ Pi, then incubated with green fluor-escence-labeled mouse IgG secondary antibody [Alexa Fluor 488 goat anti-(mouse IgG) (H + L); Invitrogen] for 1 h at room temperature Immunofluorescence staining

of flat sections was performed as described previously [56] Flat sections of whole unfed or partially fed adult ticks were exposed to mouse anti-rHlSCP1 serum diluted 1 : 150 overnight at 4C Slides were rinsed thoroughly with NaCl⁄ Piand incubated with Alexa Fluor 488 (Invitrogen) for 1 h at room temperature After washing with NaCl⁄ Pi, slides were mounted with VECTASHIELD mounting medium with DAPI (Vector Laboratories, Burlingame, CA), covered with glass cover slips, and then observed under a fluorescence microscope (Leica, Wetzlar, Ger-many)

Functional expression of rHlSCP1

in Pichia pastoris Expression of rHlSCP1 in Pichia pastoris was conducted using an EasySelect Pichia expression kit (Invitrogen) Primers used to generate rHlSCP1 in P pastoris were: sense primer (5¢-CGGAATTCCGAATAATGTCTCAGGGACC TGCTGAGGAC-3¢), corresponding to nucleotides 227–244

of the HlSCP1 nucleotide sequence, and antisense primer

corres-ponding to nucleotides 1551–1567 of the HlSCP1 nucleotide sequence The nucleotide sequence of these primers con-tained an EcoRI and a KpnI restriction site, respect-ively The amplified product was inserted into pPICZB