Angiotensin-converting enzyme cDNA, obtained by 3¢- to 5¢ RACE of testis RNAs, codes for a predicted one-domain protein similar to the mammalian germinal isoform of angiotensin-convertin
Trang 1angiotensin-converting enzyme isoform in crayfish
testis and evidence for its release into the seminal fluid Juraj Simunic, Daniel Soyez and Ne´dia Kamech
Equipe Biogene`se des Signaux Peptidiques, ER3, Universite´ Pierre et Marie Curie, Paris, France
Introduction
Angiotensin-converting enzyme (ACE; dipeptidyl
car-boxypeptidase; EC 3.4.15.1) is an enzyme that belongs
to the family of M2 peptidases with a zinc chelator
motive HEXXH-23(24)-E Substrate hydrolysis in
ACE is activated by chloride ions, which is a unique
feature among metalloproteases However, the mole-cular mechanism behind this is unclear In vertebrates,
it is present as two isoforms, somatic and testicular, which are both transcribed from the same gene under the control of tissue-specific promotors [1,2] The
Keywords
angiotensin-converting enzyme; crayfish;
Crustacea; spermatogenesis; testis
Correspondence
N Kamech, Equipe Biogene`se des Signaux
Peptidiques, ER3, Universite´ Pierre et Marie
Curie, 7 Quai Saint Bernard, 75251 Paris,
Cedex 05, France
Fax: +33 1 44 27 23 61
Tel: +33 1 44 27 22 58
E-mail: nedia.kamech@upmc.fr
(Received 4 May 2009, revised 18 June
2009, accepted 25 June 2009)
doi:10.1111/j.1742-4658.2009.07169.x
In the present study, an isoform of angiotensin-converting enzyme was characterized from the testis of a decapod crustacean, the crayfish Asta-cus leptodactylus Angiotensin-converting enzyme cDNA, obtained by 3¢- to 5¢ RACE of testis RNAs, codes for a predicted one-domain protein similar to the mammalian germinal isoform of angiotensin-converting enzyme All amino acid residues involved in enzyme activity are highly conserved, and a potential C-terminus transmembrane anchor may be predicted from the sequence Comparison of this testicular isoform with angiotensin-converting enzyme from other crustaceans, namely Carci-nus maenas, Homarus americaCarci-nus (both reconstituted for this study from expressed-sequence tag data) and Daphnia pulex, suggests that membrane-bound angiotensin-converting enzyme occurs widely in crustaceans, con-versely to other invertebrate groups where angiotensin-converting enzyme
is predominantly a soluble protein In situ hybridization and immunohisto-chemistry performed on testis sections show that angiotensin-converting enzyme mRNA is mainly localized in spermatogonias, whereas protein is present in spermatozoids By contrast, in vas deferens, immunoreactivity is detected in the seminal fluid rather than in germ cells Accordingly, angio-tensin-converting enzyme activity assays of testis and vas deferens extracts demonstrate that the enzyme is present in the membrane fraction in testis, but in the soluble fraction in vas deferens Taken together, the results obtained in the present study suggest that, during the migration of spermatozoids from testis to vas deferens, the enzyme is cleaved from the membrane of the germ cells and released into the seminal fluid To our knowledge, this present study is the first to report such a maturation process for angiotensin-converting enzyme outside of mammals
Abbreviations
ACE, angiotensin-converting enzyme; Asl, Astacus leptodactylus; DIG, digoxigenin; EST, expressed-sequence tag; gACE, germinal isoform of angiotensin-converting enzyme; tACE, testicular isoform of angiotensin-converting enzyme.
Trang 2somatic isoform exhibits two catalytic domains and is
present in numerous tissues, such as on the surface of
endothelial cells in the lung, myocardium, liver, intestine
and testis, as well as in the epithelial cells of the kidney
and intestine [3] Its role in the regulation of the renin–
angiotensin–aldosterone system has been well
character-ized [4] The enzyme cleaves angiotensin I to produce
angiotensin II, a powerful vasosuppressor It also
cleaves the vasodilatator peptide bradikinin, and thus
contributes to the augmentation of blood pressure
On the other hand, the role of the testicular isoform
(tACE), also called germinal ACE (gACE), is not so
clear In mice, the testis ACE protein is first detected
in step 10 spermatids, whereas ACE mRNA is first
detected in developmentally younger cells, the
pachy-tene spermatocytes, implying a delay in ACE
transla-tion [5] A similar phenomenon was described in
human testes, with mid-pachytene spermatocytes
expressing the mRNA and stage III spermatids
con-taining the protein, which corresponds to a delay by
one germ cycle [6]
The physiological role of tACE is still a matter of
debate, but numerous studies point to the importance
of this enzyme in male and female reproduction; for
example, mice males with a ‘knockout’ for the Ace
gene show extremely reduced fertility [7]
Analogues of ACE have also been identified in many
invertebrate species, most notably insects, and have
been shown to play a major role in reproduction
Indeed, as in mice, males of Drosophila with a
knock-out for Ace show a dramatic decrease in fertility
because the developing spermatozoids cannot complete
the phase of individualization and demonstrate an
abnormal morphology [8] In Lepidoptera, the
treat-ment of adults with the ACE inhibitor, captopril,
causes a decrease in egg-laying [9] In Haematobia
irri-tans exigua, a blood meal initiates the strong synthesis
of ACE in the testes, but not in the ovaries [10] By
contrast, in female Anopheles stephensi, a dramatic
increase in ACE activity is observed in the ovary after
a blood meal, with a maximum just prior to
egg-lay-ing, and ACE is completely transferred to newly-laid
eggs [11] Similar results were obtained from the
tomato moth Lacanobia oleracea [12] Recently, such a
transfer of ACE from males to females was reported
to take place during copulation in Drosophila
melanog-aster[13]
Even if the implication of ACE in reproduction
appears to be well established, the possible substrates
involved remain to be determined To date, the only
substrate identified in vivo in invertebrates is an 11-mer
peptide (Neb-ODAIF) isolated from the ovaries of the
fly Neobellieria bullata [14]
In previous studies, RT-PCR and northern blotting
on RNAs from several tissues of the crayfish Asta-cus leptodactylus (hepatopancreas, haemolymph and testis) revealed the presence of four different ACE iso-forms, including two from the hepatopancreas [15] Correlatively, an ACE-like activity was demonstrated
in membrane fractions from hepatopancreas and testis,
as well as in haemocytes
In the present study, we present the molecular char-acterization of ACE from the crayfish testes The cellu-lar expression of the enzyme was explored using in situ hybridization and immunohistochemistry on testis and vas deferens sections We have established that, in the testis, the ACE RNAs are detected in germ cells at an early stage of development (spermatogonia), whereas the protein is mainly present in later stages (spermato-zoids) Conversely, in vas deferens, ACE immunoreac-tivity was found in the seminal fluid rather than in cells Accordingly, activity assays have demonstrated that ACE activity shifts from the insoluble (i.e mem-brane) fraction in testis to the soluble fraction in vas deferens To our knowledge, this is the first demonstra-tion of a dynamic maturademonstra-tion process of ACE in inver-tebrates, similar to that already described in mammals
Results
Molecular characterization of A leptodactylus testicular ACE
In our previous studies, the partial cDNA sequence of the region surrounding the testicular ACE active site was obtained [15] To complement this cDNA sequence, 5¢- to 3¢ RACE was performed The exten-sion to the 3¢ end of the cDNA was realized success-fully, which was not the case for the 5¢ end Consequently, new specific primers were designed, based on the ACE cDNA sequence reconstructed from lobster (Homarus americanus) and crab (Carcinus mae-nas) expressed-sequence tags (ESTs) (see below) From the alignment of these two cDNAs, we synthesized two degenerate primers based on the 5¢ region of these two sequences One of those primers (sequence provided in the Experimental procedures) provided a satisfying result and the A leptodactylus (Asl)-tACE cDNA sequence obtained has a length of 2.3 kb, with the first stop codon at 1.9 kb (accession number: FN178630) The deduced amino acid sequence comprised 635 amino acids (Fig 1) This protein had a predicted hydrophobic region of 26 amino acids near the C-ter-minus, suggesting that the enzyme is anchored to the cellular membrane The predicted molecular weight was 73.7 kDa, with an isoelectric point (pI) value of
Trang 3Fig 1 Alignment of the predicted amino acid sequence of the A leptodactylus testicular ACE with the D melanogaster AnCE and human testicular tACE Important residues are indicated as: active site (bold underlined), zinc-binding residues (green), chloride-binding residues (orange for the first chloride ion and blue for the second one), sites of glycosylation (red), and cysteine residues forming disulfide bridges (boxed) The predicted transmembrane anchor is shown in underlined italics.
Trang 46.24 The native protein could have a higher molecular
weight because one N-glycosylation site was predicted
at Asn288 This site is conserved when compared with
the D melanogaster AnCE sequence Eight cysteinyl
residues, probably involved in the formation of four
disulfide bonds, are conserved in both AnCE and
Asl-tACE (Cys110-Cys118, Cys312-Cys330, Cys444-Cys590
and Cys499-Cys517)
The ACE active site motif HEXXH with two
zinc-binding histidines (His343 and His347) was conserved
in Asl-tACE and additional coordination was provided
by the third zinc-binding ligand (Glu371), 24 amino
acid residues downstream, which is also conserved
when compared with the Drosophila AnCE sequence
In silico reconstitution of ACE cDNA sequences
from three crustacean species
Three new ACE sequences have been deduced by
in silico methods The Daphnia pulex ACE sequence
was obtained by the blast of the genome using the
As-tacus sequence (GeneID: NCBI_GNO_452254),
whereas Homarus and Carcinus sequences were
recon-stituted from ESTs (accession numbers: BN001300 and
BN001299, respectively) Comparison of Asl-tACE
with three other crustacean sequences shows that
Asl-tACE has 79% sequence identity with Homarus, 75%
with Carcinus and 53% with Daphnia (Fig 2) Both
the Carcinus and Daphnia sequences contain a
pre-dicted transmembrane region, whereas the EST
assem-bly of Homarus is incomplete in its 3¢-terminus
Furthermore, all cysteines implicated in disulfide
bridge formation are conserved, as well as the residues
involved in the coordination of chloride ions
A leptodactylus testicular ACE expression and
tissue localization
As shown in Fig 3, the A leptodactylus testis is
com-posed of three lobes One vas deferens exits from each
of the two lateral lobes During spermatogenesis,
mature spermatozoids accumulate and are maintained
in vas deferens until fertilization, which results in a
dramatic size increase In the resting period, the vas
deferens are atrophied and are barely visible At the
cellular level (Fig 4A), the testis is composed of acini
that open into collector canals and finally into a vas
deferens The acini contain mesodermal cells and
sper-matogonia in different stages of development, as well
as mature spermatozoids
To provide more detailed information about which
cells in the testes are involved in both Asl-tACE
mRNA synthesis and protein expression, we performed
in situ hybridization and antibody staining For in situ hybridization, an antisense 158 bp long digoxigenin (DIG)-labelled cRNA probe was used Tissue sections (5 lm) were prepared from testes taken from animals
in active spermatogenesis, as indicated by the presence
of a highly developed vas deferens filled with seminal fluid The results show a specific hybridization of mRNAs in cells that morphologically correspond to spermatogonia (Fig 4B) A weak signal was also detected in some spermatozoids (Fig 4C) No signal was observed in negative controls performed using a sense probe (not shown)
To localize the expressed protein, labelling of 5 lm sections was performed using an antibody developed against a synthetic peptide designed from the Asl-tACE sequence obtained by cloning
The protein distribution is the inverse of the expres-sion pattern of mRNAs, namely the strongest signal is displayed on the cytoplasm membranes of spermato-zoids, whereas the spermatogonia exhibit very faint staining only (Fig 4D) The thickness of the staining
is the result of the morphology of spermatozoid in Astacus Indeed, the spermatozoid is almost devoid of cytoplasm, and the cell membrane is highly invagi-nated, forming crests that enter deeply into nucleo-plasm The spermatozoid is surrounded by a periodic acid-Schiff positive casing of finely granular material, suggesting the presence of complex carbohydrates such
as in mucus [16,17] This most likely has resulted in the thick staining of spermatozoid membrane that we observed on our preparations Similarly, on vas deferens sections, some staining was also present on the outer membrane of spermatozoids, although the signal was much weaker than in testis By contrast, a strong signal was found in the seminal fluid itself (Fig 4E, F) Control testis sections incubated with preabsorbed antibodies failed to exhibit any signal (not shown)
A leptodactylus testicular ACE activity assays
To determine the enzymatic activity of the testicular
A leptodactylus ACE, an activity assay with a radioac-tive substrate was performed (see Experimental proce-dures) We tested the activity in testes and in vas deferens separately (Fig 5), which were sampled from animals in the reproductive period and in genital rest
In testes sampled during spermatogenesis, a strong enzymatic activity was found in the insoluble fraction that contains membranes, whereas the soluble fraction showed very little enzymatic activity In the vas defer-ens, the activity was as strong, but, interestingly, it was present mostly in the soluble fraction In animals
Trang 5Fig 2 Alignment of A leptodactylus testicular ACE with the H americanus, C maenas and D pulex ACE sequences Predicted signal pep-tides are shown in bold Active sites are underlined, with zinc coordinating residues shown in green The predicted transmembrane anchor
is shown in underlined italics.
Trang 6sampled during the resting period, no significant
activ-ity was found in testes and, because the vas deferens
almost completely disappears during this period, this
tissue was not tested
Discussion
The present study aimed to provide a detailed
charac-terization of the ACE isoform found in the testes of
the crayfish A leptodactylus (Asl-tACE) By
perform-ing RT-PCR and 5¢- to 3¢ RACE on testis RNAs,
using degenerated primers deduced from crustacean
ACEs, we were able to clone a 2.3 kb cDNA This size
corresponds to the values obtained previously by
northern blotting (i.e in the range 2–2.5 kb for the
ACE mRNA from the crayfish testis) [15]
In silico translation of the cloned cDNA has shown
that the encoded protein of 635 amino acid residues
shares most of the common characteristics of ACEs
from other invertebrate species: all the residues
puta-tively important for the coordination of two chloride
ions (Arg147, Tyr185, Trp445, Arg449 and Arg482)
were conserved, as were the positions of the cysteinyl
residues that are probably involved in the formation of
four disulfide bonds
Based on the crystal structure of AnCE bound to
captopril and lisinopril [18], we found that the residues
implicated in inhibitor binding are highly conserved
(Glu123, Thr127, Gln242, His313, Ala314, Asn336,
Thr340, Glu344, Lys471, His473, Tyr480 and Tyr483),
apart from Asp146 and Asp360 in Drosophila, which
are replaced by Glu123 (as in human ACEs) and
Asn336, respectively
Comparison with the human tACE sequence shows
that Asl-tACE has conserved residues that are
impli-cated in the coordination of two chloride ions The
first chloride ion is coordinated by Arg147, Arg450
and Trp446 and the second one is bound to Tyr186
and Arg483
In addition to these features common to most ACEs, Asl-tACE displays some more specific ones Especially, it appears to be a membrane-bound protein because a potential hydrophobic transmembrane anchor comprising 26 amino acid residues was found
in the C-terminal region of the molecule Conversely, most of the invertebrate ACEs described to date are soluble proteins, with the exception of two Anophe-les gambiae ACEs (AnoACE7 and AnoACE9), which appear to be membrane-bound enzymes [19] However, these forms do exhibit two catalytic domains, such as somatic mammalian ACE, whereas Asl-tACE, with less than 700 residues, is likely to display only one catalytic domain, such as mammalian gACE
Interestingly, data mining and reconstruction of putative ACEs from other crustacean species, namely ESTs from C maenas and H americanus, and the whole D pulex genome, indicate the presence of a sim-ilar transmembrane C-terminal region, in addition to other conserved features (Fig 2) Accordingly, an ACE-like activity was reported in the membranes of
C maenas gills, which may be easily solubilized by detergent application [20] At present, it is not possible
to speculate whether, in crustaceans, in contrast to other groups, ACE isoforms are always membrane-bound proteins because no genome has been sequenced from a crustacean species other than Daphnia In A le-ptodactylus, several different ACE isoforms have been identified [15] and it will prove very informative to determine whether or not every isoform displays a transmembrane region Knowing whether all isoforms
in Astacus are membrane bound proteins is interesting because it raises questions about both the evolution of ACE in different groups of animals and the physiologi-cal significance of membrane bound isoform compared
to the soluble form, which was almost exclusively found in other invertebrate groups
Subsequent to early studies conducted in the rat and pig [21,22], it is well established that ACE is present in germ cells This has been described for several insect species, including Drosophila [8] Accordingly, our
in situ hybridization experiments peformed on testis sections show that ACE mRNA is mainly present in spermatogonia (i.e in the early stages of spermato-genesis), whereas only a small amount or even an absence of RNA was detected in mature spermato-zoids, and no signal was ever observed in mesodermal cells or in vas deferens Immunolocalization of the ACE protein using a hapten-specific antibody revealed
a very different distribution pattern; in the testes, the protein appeared to be present on the external side of the cytoplasmic membrane of spermatozoids, but not on spermatogonia By contrast, in vas deferens,
Vas
deferens
Testis lobes
1 cm
Fig 3 Morphology of A leptodactylus testis during
spermatogene-sis The testis is composed of three lobes Two vasa deferentia are
well developed and contain mature spermatozoids.
Trang 7immunoreactivity on germ cell membranes was much
weaker, with a strong signal being present in the
seminal fluid
The fact that no ACE protein was detectable in
earlier stages of spermatogenesis when corresponding
RNAs are present suggests the existence of a
transla-tional arrest, a phenomenon already known to occur
in the expression of testicular ACE in mice [5] On the
other hand, in Astacus during spermatid
differentia-tion, the nucleus undergoes several stages of
reorgani-zation in which chromatin density and pattern of
distribution change on several occasions In the last
stage, there is actually a decrease in density of nuclear
material [16]
When the ACE activity was assayed in both soluble and insoluble (membrane) fractions from testis and vas deferens homogenates, it was observed that maximal activity is associated with membranes in testis, but with soluble material in vas deferens During genital rest, where no spermatozoids are present in the germi-nal tract, no significant activity was detected in the testis extract
Taken together, the results of in situ hybridization, immunohistochemistry and activity assays strongly suggest a shift of the enzyme from the germ cell mem-brane to the seminal fluid when the spermatozoids migrate from the testes to the vas deferens during sper-matogenesis
Fig 4 In situ hybridization and immunohistochemical localization of A leptodactylus testicular ACE expression in testis and vas deferens Morphology of testis during spermatogenesis: (A) mesodermal cells (mc), spermatogonia (g) and spermatozoids (spz) In situ hybridization: (B) strong mRNA signal in the spermatogonia (g); (C) weak ⁄ no signal in the spermatozoids (spz) Immunohistochemistry: (D) In the testis, staining is present on the membranes of mature spermatozoids (spz), whereas spermatogonia are unstained (E, F) In the cross section of vas deferens, staining is strong in the seminal fluid (sf), with weak staining also being present in the spermatozoid membranes (spz) The walls (w) of the vas deferens are not stained.
Trang 8In mammals, gACE was shown to be anchored in
the cytoplasmic membrane of spermatozoids and to be
released in the epididymal fluid during the transit of
the sperm in the epididymis [23] This cleavage involves
a serine protease (sheddase) [24] Such a maturation
process has never been described for ACE in animal
groups other than mammals until our present study,
which clearly indicates the presence of a similar
mecha-nism of protein ectodomain shedding in crayfish testis
The similarity between the maturation process of
crustacean and mammalian ACE could imply the
pres-ence of an unknown protease in crayfish testis, which
may cleave the Asl-tACE from spermatozoid
mem-brane; however, the cleavage site (-Arg-Leu-) identified
in mammalian tACE [24] was not found in the
Asl-tACE sequence
To date, the physiological role of testis ACE
rem-ains a matter of debate In mammals, experiments with
Ace ‘knockout’ mice have shown that ACE plays a
role in fertilization because the absence of testicular
ACE leads to defects in sperm transport in oviducts as
well as in binding to zonae pellucidae, without
modifi-cation of sperm morphology and counts [7] Because
there is no apparent ACE substrate involved in
fertil-ization, the molecular mechanism for this effect
remains unknown, with one possible explanation being
that tACE could be involved in the distribution of ADAM3, a protein essential for sperm–zonae pelluci-dae interactions [25] The results obtained in other studies [26] indicate that ACE could have a glycosyl-phosphatidylinositolase activity that is unrelated to its peptidase active site, although this hypothesis is strongly debated [27,28]
In invertebrates, ACE has an important role in both reproduction and development In dipteran insects, it has been reported that inhibition of ACE activity affects different aspects of reproduction [29] and that ACE inhibition by dietary administration of inhibitors reduces oviposition in A stephensi female mosquitoes
In males of the same species, inhibitor feeding results
in an 80% loss of fecundity, which is expressed as the reduction in the number of eggs laid by blood-fed females mated with ACE-inhibited males It has been suggested that Drosophila Ance, which is present in secretion vesicles in spermatocytes, may have a func-tion in the maturafunc-tion of bioactive peptides during spermatogenesis [8] In the lepidopteran species L oler-acea, it was shown that ACE is transferred from the male to the female during mating [12] In addition, through activity assays including ACE inhibitors and HPLC analysis, ACE was demonstrated to be an important protease among the peptide-degrading enzymes present in the female spermatophore⁄ bursa copulatrix Regarding its physiological function, it is hypothesized that ACE, along with other peptidases present in the spermatophore⁄ bursa copulatrix, could provide dipeptides or amino acids that are necessary for different metabolic pathways However, to date, no experimental evidence is available to support this hypothesis Such a hypothesis is unlikely in A lepto-dactylus because the spermatozoids lack a flagellum and therefore are not mobile Nevertheless, it cannot
be excluded that Asl-tACE may play a role in sperma-tozoid metabolism Indeed, during mating, the male crayfish deposits sperm near the openings of the female gonoducts (i.e at the base of the third periopods), using the two first pair of pleopods that are modified
to copulatory appendices for guidance of the sperm into the female spermatheca, where it may be kept for
a period of up to several months before fertilization During this period, it is possible that Asl-tACE could play a role in metabolic pathways that are important for spermatozoid maintenance and survival
In conclusion, the results obtained in the present study demonstrate that the ACE maturation process by protein shedding is similar in crayfish and mammalian testis It remains to be elucidated whether, similarly, the function of the testicular ACE, which still remains obscure, is conserved throughout animal evolution
Fig 5 ACE activity in testes and vas deferens Activity (c.p.m per
mg) was determined by an in vitro assay using tritiated
hippuryl-gly-cyl-glycine as substrate (see Experimental procedures) The bars
represent the difference between c.p.m values per mg of protein
of the [ 3 H]hippurate obtained after incubation with and without
10 l M captopril Soluble (SOL) and insoluble (INSOL) fractions were
tested in animals in active spermatogenesis and in resting period.
Three experiments were conducted P-values between the soluble
and insoluble fractions, as calculated using Student’s t-test, were
0.01 and 0.002 for the testis and vas deferens, respectively.
Trang 9Experimental procedures
Animals
Crayfish A leptodactylus were obtained from a commercial
supplier They were kept in the laboratory in recirculated
filtered water and fed twice a week with cat food pellets
(Friskies pellets, Nestle Purina PetCare SAS,
Rueil-Malmai-son, France) Before dissection, animals were anesthetized
in crushed ice and ice-cold water
Male crayfish exhibit only one spermatogenesis cycle per
year, from spring until late summer There are no external
signs to indicate whether the animal is in active
spermato-genesis or in a resting period and, within a tank, the
repro-ductive cycles are not fully synchronized Therefore, the
reproductive status of the animal could only be estimated
accurately after dissection
Males with well-developed vas deferens filled with
semi-nal fluid (Fig 3) were considered to be in active
spermato-genesis, whereas those with vas deferens reduced to a thin
whitish cord were considered as males in the resting period
Molecular characterization
Reconstitution of three crustacean ACE cDNA from
EST and genomic data
A cDNA ACE of the green shore crab C maenas was
reconstituted from four ESTs of multiple tissues [30]
The Genbank accession numbers of those ESTs are:
DW249183, DY657025, DN203050 and DV944439
An ACE cDNA of the lobster H americanus was also
reconstituted from five ESTs of multiple tissue
(corre-sponding Genbank accession numbers: FF277412,
CN854003, EH401278, CN854103 and EG949475)
D pulex sequence was found by blast of D pulex
genome assembly (JGI-2006-09): scaffold_25 position
from 1 040 210 to 1 042 148 Multiple sequence
align-ments were performed with clustalw2 [31] at the
European Bioinformatics Institute (http://www.ebi
ac.uk/Tools/clustalw2/index.html)
Cloning the testicular ACE isoform of crayfish
Specific upper and lower primers were selected from the
partial Asl-tACE sequence previously obtained to
deter-mine the 3¢- and 5¢ regions of the protein Total RNA
used for the amplification was isolated from
spermato-genic testes by the SV Total RNA Isolation System
according to the manufacturer’s instructions (Promega,
Madison, WI, USA) The first strand cDNA for
3¢-RACE was synthesized from 1 lg of total RNA using
a SMARTRACE cDNA Amplification Kit (Clontech,
Mountain View, CA, USA) according to the
manufac-turer’s instructions The 3¢-RACE was performed between the upper primer T3U1 5¢-GGGACTTCTG TAATGGCAAAG-3¢ and the Universal Primer A Mix (UPM), provided with the kit The PCR product was directly sequenced using the T3U1 primer by Cogenics Genome Express (Cogenics, Meylan, France)
To determine the 5¢ region of the mRNA, a degener-ate upper primer was synthesized in the 5¢ region of ACE sequences from H americanus and C maenas, deduced by EST assembly as described above The sequence of this primer was: 5¢-AGGARCTTCCTG MASGAGWTGGAC-3¢ The lower primer had the sequence: 5¢-GGTCTTGTTTGGGAAGGGCAGCTG TGC-3¢
PCR products were purified from 1.5% agarose gels and subcloned into the transcription vector pGEM-T Easy vector (pGEM-T Easy Vector System II; Pro-mega) and propagated in JM-109 Escherichia coli bacteria Recombinant plasmids were purified with WizardPlus SV Minipreps kit (Promega) Sequencing was performed by Cogenics Genome Express using the dideoxy chain termination method Another degenerate primer was synthesized in the signal peptide region, although it failed to produce satisfactory results
In silico analysis of sequences Possible glycosylation sites were identified with the netnglyc 1.0 Server (http://www.cbs.dtu.dk/services/ NetNGlyc/) The pI and molecular weight were cal-culated with the compute pI⁄ Mw Tool (http://www expasy.ch/tools/pi_tool.html)
Prediction of signal peptides was performed using the signalp 3.0 Server (http://www.cbs.dtu.dk/services/ SignalP/) and prediction of transmembrane regions was performed using the sosui engine, version 1.11 (http://bp.nuap.nagoya-u.ac.jp/sosui/)
Tissue preparation for in situ hybridization and immunohistochemistry
Testis and vas deferens tissue were dissected and immedi-ately immersed in Bouin’s fixative solution (75% picric acid, 20% formaldehyde, 5% acetic acid) for 24 h at room temperature After dehydration in graded ethanol solutions (2· 70%, 3 · 95%, 1 · 100%), the tissue was embedded in paraffin wax according to conventional histological proce-dures Five micrometer sections were cut and alternately mounted on poly(l-lysine)-coated slides (Polysine, Menzel-Glaser, Germany) The slides were deparaffinized in EZ-DeWax deparaffinization solution (InnoGenex, San Ramon,
CA, USA), hydrated and used for in situ hybridization or immunohistochemistry
Trang 10In situ hybridization
A 158 bp probe was obtained by amplification of the
frag-ment spanning nucleotides 950–1107 of the Asl-tACE
cDNA with the upper primer 5¢-GGGACTTCTGTAATG
GCAAAG-3¢ and the lower primer 5¢-GAACCCTGGGTT
GGCTCCGCTTCG-3¢ The cDNA was cloned in the
tran-scription vector pGEM-T Easy vector (pGEM-T Easy
Vector System II; Promega) After linearization of the
tem-plate DNA, in vitro transcription reactions were carried out
in the presence of DIG-UTP (DIG RNA labelling Kit;
Roche Diagnostics, Meylan, France), with T7 and SP6
polymerases for antisense and sense probes, respectively
The template was degraded with RNase-free DNase (Roche
Diagnostics) The DIG-labelled RNA probes were purified
by ethanol and sodium acetate precipitation and stored at
)20 C in 0.1% diethyl pyrocarbonate (Sigma-Aldrich, St
Louis, MO, USA)-treated water until used for in situ
hybridization All solutions and glassware were RNase-free
The sections were treated with 0.1% pepsin (Roche
Diag-nostics) in 0.2 m HCl (37C for 10 min) Postfixation was
performed by treating the sections with a fresh solution of
2% paraformaldehyde in NaCl⁄ Pi (10 mm sodium
phos-phate, pH 7.4, 0.1 mm KCl, 0.8% NaCl) for 4 min, and
immersed in 1% hydroxylammonium hydrochloride in
NaCl⁄ Pi for 15 min The sections were then dehydrated
with successive ethanol washings For hybridization, a
humid chamber with 4· SSC (1 · SSC: 150 mm NaCl,
15 mm sodium citrate, pH 7.4) was prepared DIG RNA
probes (antisense or sense) were diluted to a final
concentra-tion of 50 ngÆlL)1in the hybridization mixture [50%
form-amide, 10% dextran sulfate, 4· SSC, 1 · Denhardt’s
solution (0.1% Ficoll 400, 0.1% polyvinylpyrrolidone, 0.1%
BSA in water), 0.1% yeast tRNA and 0.1% salmon sperm
DNA], denatured at 70C for 5 min and cooled on ice
One hundred and twenty microlitres of this mix was placed
on each tissue section Sections were covered with cover
slips and placed in the humid chamber at 45C overnight
(16 h) Post-hybridization washes in consecutive stringency
baths of SSC (reducing concentrations; · 2, · 1, · 0.5,
· 0.2, · 0.1; 20 min each bath) were used to remove
non-specifically bound probes The sections were treated with
anti-DIG alkaline phosphatase-conjugated IgGs (Roche
Diagnostics) for 30 min, washed and, finally, the
phospha-tase substrate (nitro blue tetrazolium⁄
5-bromo-4-chloro-3-indolyl phosphate; Sigma-Aldrich) was added with 1 mm
levamisole to block possible endogenous alkaline
phospha-tase After satisfactory colour development ( 1 h), the
slides were washed carefully in tap water and mounted with
glycerol⁄ gelatin (Sigma-Aldrich) preheated at 42 C
Immunohistochemistry
The sections were washed with NaCl⁄ Pi ⁄ Triton 0.5% ⁄ goat
serum (Sigma-Aldrich) 3% buffer before incubating
overnight with rabbit polyclonal antibodies raised against the synthetic peptide RENYGEEHVSRRGP, located between R217 and P230 of the Asl-tACE sequence During synthesis, the R217-P230 peptide was extended at the C-ter-minus by a cysteine residue to facilitate coupling to keyhole limpet haemocyanin The antibodies were produced in two rabbits by GenScript Corporation (Piscataway, NJ, USA) Negative controls were performed by incubating the sec-tions with antiserum adsorbed with R217-P230peptide
As secondary antibody, Alexa Fluor 568 goat anti-rabbit IgG (H+L) (Molecular Probes, Carlsbad, CA, USA) was used Analyses were performed on a confocal laser-scanning microscope (TCS4D confocal imaging system; Leica, Heidelberg, Germany) with an argon-krypton ion laser They were scanned sequentially at an excitation wavelength
of 568 nm A series of confocal sections (thickness in the range 0.1–2 lm) was collected for each specimen Focal ser-ies were then processed to produce single composite images
or montages, combining high spatial resolution and high field depth (nih image, version 1.63; NIH Image, Bethesda
MD, USA) Micrographs were processed and assembled with Adobe photoshop 8.0 (Adobe Systems Inc., San Jose, CA)
ACE activity assays
Testis and vas deferens tissue from males in active sper-matogenesis were dissected out and weighed One hundred milligrams of each tissue were prepared separately After homogenization in 700 lL of assay buffer (50 mm HEPES-HCl buffer with 300 mm NaCl, pH 8.3), the homogenate was centrifuged (9200 g for 20 min at 4C) The Asl-tACE activity was tested in the supernatant (solu-ble fraction) and in the pellet which was resuspended in
600 lL of assay buffer (insoluble fraction) The protein content of each fraction was estimated by the Bradford method using the Bio-Rad Protein Assay reagent (Bio-Rad Laboratories GmbH, Muenchen, Germany) with BSA as standard Activity assays were performed in accordance with a previously described protocol [15] Briefly, the enzyme activity was determined by incubating tissue samples with the ACE radiolabelled substrate [phenyl-4(n)-3H-hippuryl-glycyl-glycine (282 mCiÆmmol)1; Amersham, Little Chalfont, UK)] 3H-labelled hippurate, the product of ACE hydrolysis, was separated by ethyl acetate extraction and the radioactivity was assayed for
2 min with a b-IV scintillation counter (Kontron Instru-ments, Watford, UK) to obtain the ‘total c.p.m.’ value
To discriminate ACE activity from other peptidase activi-ties, a reaction assay was performed in the same condi-tions, except that 10 lm captopril, a specific ACE inhibitor, was added to the tube, giving the ‘captopril c.p.m.’ value ACE activity was calculated as: (total c.p.m – captopril c.p.m.)⁄ mg protein Each data point was assayed in quadruplicate