Analysis of metallothioneins revealed that certain CvMT-I isoforms showed preferential association either with cadmium or with copper and zinc, even after exposure to cadmium.. CvMT-I an
Trang 1Diversity of metallothioneins in the American oyster,
proteomic approaches
Matthew J Jenny1, Amy H Ringwood4, Kevin Schey2, Gregory W Warr3and Robert W Chapman4
1
Marine Biomedicine and Environmental Sciences Center,2Department of Cell and Molecular Pharmacology and3Department of Biochemistry and Molecular Biology, Medical University of South Carolina, Charleston, SC, USA;4Marine Resources Research Institute, South Carolina Department of Natural Resources, Charleston, SC, USA
Metallothioneins are typically low relative molecular mass
(6000–7000), sulfhydryl-rich metal-binding proteins with
characteristic repeating cysteine motifs (X-Cys or
Cys-Xn-Cys) and a prolate ellipsoid shape containing single
a-and b-domains While functionally diverse, they play
important roles in the homeostasis, detoxification and stress
response of metals The originally reported metallothionein
of the American oyster, Crassostrea virginica showed the
canonical molluscan ab-domain structure Oyster
metallo-thioneins have been characterized as cDNA and as expressed
proteins, and here it is shown that the previously reported
metallothionein is a prototypical member of a subfamily
(designated as CvMT-I) of ab-domain metallothioneins
A second extensive subfamily of oyster metallothioneins
(designated as CvMT-II) has apparently arisen from (a) a
stop mutation that truncates the protein after the a-domain, and (b) a subsequent series of duplication and recombination events that have led to the development of metallothionein isoforms containing one to four a-domains and that lack a b-domain Analysis of metallothioneins revealed that certain CvMT-I isoforms showed preferential association either with cadmium or with copper and zinc, even after exposure
to cadmium These data extend our knowledge of the evo-lutionary diversification of metallothioneins, and indicate differences in metal-binding preferences between isoforms within the same family
Keywords: cadmium; gene expression; MALDI-TOF; met-allothionein; oyster
Metallothioneins (MTs) are a superfamily of ubiquitously
expressed metal-binding proteins that can be upregulated
by metal exposure, oxidative stress and immune challenge
Typical MTs are low relative molecular mass (Mr) (6000–
7000) proteins of high thiol content that lack histidine and
aromatic amino acids [1,2] While they are functionally
diverse, they play major roles in metal homeostasis and
detoxification The defining characteristic of MTs is the
high cysteine content ( 30%) and conserved Cys-Xn-Cys
motifs, where X can be any amino acid other than cysteine
The proteins typically have a one-or two-domain structure
and bind multiple mono-and divalent metal ions The
structure of MTs, and the nature of their metal-binding,
reveal extensive evolutionary diversification While fungi
and early diverged metazoans have small, single-domain
MT proteins capable of binding up to eight monovalent
metal ions [3–6], most MTs are comprised of two domains,
designated a and b, which are capable of binding metals
independently and are separated by a short linker region [7,8] The a-domain typically contains 11 or 12 cysteines, binds four divalent metal cations, and is believed to convey structure and stability to the protein [9] In contrast, the b-domain contains nine cysteines, binds three divalent metal cations and participates in metal exchange reactions invol-ving glutathione-shuttling with zinc- and copper-requiring apoproteins [10–12] Some crustacean MTs deviate from this canonical structure, possessing two b-domains capable
of binding six metal cations [13]
While the induction of MTs by various metals, partic-ularly cadmium, has been established in a variety of metazoan taxa [14–17], to date only one MT, a cadmium-inducible isoform, has been identified from Crassostrea virginica [18], although biochemical studies indicated the presence of two cadmium-binding proteins of 10 and
24 kDa [19,20] Several metal-rich proteins, representing putative MTs, have been identified in control and metals-treated C virginica larvae [21], and the presence of multiple
MT isoforms has been demonstrated in other molluscan species [22–26], including bivalves (the blue mussel, Mytilus edulis[23] and the Pacific oyster, Crassostrea gigas [27]) and gastropods (the terrestrial snail, Helix pomatia [25,26]) It
is clear that MT characteristics, especially amongst the invertebrates, are more varied than previously believed Two metal-specific MTs, a copper-specific isoform isolated from mantle tissue and a cadmium-inducible isoform isolated from the midgut gland, have been found in the
Correspondence to R W Chapman, Marine Resources Research
Institute, South Carolina Department of Natural Resources, Hollings
Marine Laboratory, 331 Fort Johnson Road, Charleston, SC 29412,
USA Fax: + 1 843 762 8737, Tel.: + 1 843 762 8860,
E-mail: chapmanr@mrd.dnr.state.sc.us
Abbreviations: Cv, Crassostrea virginica; IAA, iodoacetic acid;
M r , relative molecular mass; MT, metallothionein.
(Received 21 January 2004, accepted 4 March 2004)
Trang 2snail [25,26], and in the Pacific oyster a MT comprised of
three metal-binding domains (abb), has been reported [27]
Here we report the results of a study, combining
tran-scriptomic and proteomic approaches, designed to increase
our knowledge of the structure and function of oyster MTs
in the American oyster, C virginica, and to shed light on
the evolutionary diversification of this supergene family
Experimental procedures
Collection ofC virginica
Adult C virginica were collected from Lighthouse Creek
and Sweetgrass Creek (Charleston, SC, USA) or St Pierre,
ACE Basin (National Estuarine Reserve, SC, USA) and
maintained in aerated natural seawater (25 ppt salinity,
1 lm filtered) at the Marine Resources Research Institute,
South Carolina Department of Natural Resources (MRRI,
SCDNR) Oysters were allowed to depurate in the
labor-atory for 24–96 h before use and were fed a phytoplankton
suspension consisting of Chaetocerus gracilis Strain
(Bacil-lariophyceae) and Isochrysis galbana Strain
(Prymnesio-phyceae) every 48 h while maintained in the laboratory
cDNA library construction fromC virginica
24 h D-veliger
Gametes were stripped, under sterile conditions, from
four female and three male oysters and mixed to allow
fertilization to occur Fertilized eggs were diluted with
sterile natural seawater to 50 embryos per mL and
incubated for 24 h, under control conditions or conditions
containing metal treatments – either copper (0.16 lM) or
cadmium (0.18 lM) – until the D-veliger developmental
stage was reached Three separate cDNA libraries were
constructed from 200 000 D-veliger larvae from each
treatment RNA was isolated using the RNeasy Miniprep
kits (Qiagen) and cloned using the PCR-based SMART
cDNA Library Construction Kit (Clontech Laboratories,
Inc.) Library construction has been described in detail
previously [28]
Library screening
Each library was differentially screened by plating 3· 105
plaque-forming units, which were transferred to replicate
nitrocellulose filters (Schleicher & Schuell BioScience, Inc.,
NH, USA) Filters were prehybridized at 65C in
Church-Gilbert solution [29] and incubated overnight with probes
generated from the cDNA of oyster cadmium-binding MT
(generously provided by G Roesijadi, Florida Atlantic
University, Boca Raton, FL, USA) [18] Probes were
radiolabeled with random hexanucleotide primers, Klenow
DNA polymerase, and 50 lCi of [32P]dATP[aP]
(Perkin-Elmer, Boston, MA, USA) MT cDNA positive plaques
were purified through subsequent screenings Plasmids were
isolated from positive plaques using Qiagen Turboprep 96
kits on the Qiagen Biorobot 9600, according to the
manufacturer’s instructions All plasmid samples were
sequenced in both directions by the Biotechnology Resource
Laboratory, Medical University of South Carolina, using
the Clontech sequencing 5¢ (5¢-AGCTCCGAGATCTG
GACGAGC-3¢) and 3¢ (5¢-TAATACGACTCACTATA GGGC-3¢) primers for the pTriplEx2 plasmid
Experimental metal challenges for expression analysis Adult oysters were exposed to equimolar concentrations (0.25 lM) of copper, cadmium, or zinc for a period of 96 h Gill and hepatopancreas tissues were dissected for total RNA isolation using the methods previously described For protein analysis, adult oysters were treated with 0.44 lMof cadmium for 96 h before hepatopancreas tissue was dissec-ted, flash frozen in liquid nitrogen and stored at)80 C Typically, the hepatopancreas tissues from two to three oysters were combined to increase the protein yield RT-PCR
Multiple cDNAs identified from library screenings were compared in order to design consensus primers for the concurrent amplification of both CvMT-I and -II isoforms; forward consensus primer (5¢-GCCGAYTGTAYCACAG ACAC-3¢) and reverse consensus primer (5¢-CTCTYATT RGTCGAGCGYTC-3¢) Total RNA was isolated with the RNeasy Miniprep kits (Qiagen) First-strand cDNA was synthesized from 1 lg of total RNA using an oligo (dT) primer and 200 U of M-MLV reverse transcriptase (Promega) Complementary isoforms were amplified with
25 cycles of PCR under the following conditions: denatur-ation at 94C for 30 s, annealing at 55 C for 60 s, and extension at 72C for 60 s
TOPO TA cloning of RT-PCR products RT-PCR products from four separate reactions for each primer set (control, copper-and cadmium-treated C vir-ginica larvae; and cadmium-treated adult hepatopancreas tissue) were cloned into the pCR2.1-TOPO vector accord-ing to the manufacturer’s instructions (TOPO TA Clonaccord-ing Kit; Invitrogen Corporation) The pCR2.1-TOPO con-structs were transformed into chemically competent XL1 Blue MRF¢ Escherichia coli cells Plasmids were isolated using Qiagen Turboprep 96 kits on the Qiagen Biorobot
9600, according to the manufacturer’s instructions Plasmid samples were sequenced by SeqWright, Inc (Houston, TX, USA) using M13 forward and reverse sequencing primers Any unresolved nucleotides were confirmed with additional sequencing by the Biotechnology Resource Laboratory (Medical University of South Carolina), using the internal consensus primers previously described in the RT-PCR protocols as well as two additional internal consensus primers (5¢-CGCCTCTCATTGGTCGAGCGC-3¢) and (5¢-GARCGCTCGACYATTRAGAG-3¢) The sequences were deposited in the NCBI nonredundant database with sequential accession numbers AY331695 to AY331707 Genomic Southern blot analysis
Genomic DNA was prepared from individual oysters using the total tissue remaining after removal of gonadal and hepatopancreatic tissue The frozen tissue was ground with
a mortar and pestle, transferred to lysis buffer (100 mM EDTA, 50 m Tris/HCl pH 8.0, 1% SDS) containing
Trang 320 lgÆlL)1proteinase K and incubated overnight at 55C.
Genomic DNA was extracted with phenol/chloroform/
isoamyl alcohol (25 : 24 : 1) and precipitated with 70%
ethanol Separate restriction digests were performed on
7.5 lg of genomic DNA with one of three enzymes, EcoRI,
AvaII, or BamHI (Gibco BRL) The resulting fragments
were separated on 0.8% agarose gels and transferred to
nitrocellulose membrane (Nytran; Schleider & Schuell),
using an upward transfer technique, in 20· NaCl/Cit (3M
NaCl, 0.3M sodium citrate, pH 7.0) Hybridization was
performed using [32P]dATP[aP]-labelled CvMT-I probes
Because of the strong similarity in DNA sequence, the
probes generated from CvMT-I will hybridize with all
identified CvMT-II isoforms
Northern blot analysis
Total RNA from adult oysters (5 lg) was electrophoresed
in a 1.2% agarose, 0.6% formaldehyde gel Denatured
RNA was transferred to Nytran membrane, using an
upward transfer technique, in 10· SSPE (1.5M NaCl,
100 mMNaH2PO4, 10 mMEDTA, pH 7.4) Hybridization
was performed in Denhardts reagent buffer [50%
forma-mide, 1% SDS, 5· SSPE (750 mM NaCl, 50 mM
NaH2PO4, 5 mM EDTA, pH 7.4], and 2· Denhardts
reagent [0.04% Ficoll 400, 0.04% poly(vinylpyrrolidone),
0.04% BSA)], overnight at 42C, with probes generated
by random priming from the cDNA of CvMT-I After
autoradiography, membranes were stripped by boiling in
0.1% SDS and rehybridized with probes for b-actin
(ACCN_ BG624786)
Fractionation of metal-binding proteins by size-exclusion
chromatography
Hepatopancreatic tissue samples (1 : 2.5 ratio of tissue to
buffer; g/mL) were partially thawed in buffer (30 mM
NH4HCO3, pH 8.2) containing 1 mM dithiothreitol and
1 mMphenylmethanesulfonyl fluoride Samples were
homo-genized under helium gas and centrifuged (32 000 g) for
60 min at 4C Supernatant was removed and centrifuged
(32 000 g) for an additional 30 min at 4C and filtered
through a 0.45 lMmembrane Proteins were first separated
by size-exclusion HPLC on a Superdex 75 PC 3.2/30 column
(Pharmacia Biotech, Inc.) with 30 mMNH4HCO3
contain-ing 1 mM dithiothreitol at a flow rate of 0.5 mLÆmin)1
Fractions were collected every 30 s and monitored for
cad-mium, zinc, and copper using a Perkin-Elmer AAnalyst
Model 700 atomic absorption spectrophotometer
Commer-cially available rabbit MT (Sigma Chemical Co.) was used to
approximate the elution time of comparable oyster MTs
Partial purification of metal-binding proteins by HPLC
Anion-exchange HPLC was used to characterize the
cadmium-rich pools isolated by size-exclusion
chromato-graphy Proteins were separated using an anion-exchange
column (TSKgel DEAE-5PW) with a 35 min linear
gradi-ent of 30–350 mMNH4HCO3containing 1 mM
dithiothre-itol (pH 8.2) Proteins were eluted at a flow rate of
0.65 mLÆmin)1and fractions were collected at 30 s intervals
A 25 lL aliquot was removed from each fraction, dried,
reconstituted in 2% HNO3and analyzed for cadmium, zinc, and copper by atomic absorption spectrophotometry The remaining fraction was frozen at)20 C until analysis by MALDI-TOF
Determination of mass and cysteine content
by MALDI-TOF MS Anion exchange-HPLC fractions, representing individual metal-rich peaks (molecular mass range of 6–22 kDa), were concentrated to 100 lL volumes using Centricon YM-3 filter devices Samples were acidified with 15 lL of trifluoroacetic acid to a pH range of 2–3 and diluted to 1.5 mL with 2.5% trifluoroacetic acid Samples were con-centrated to 100 lL by centrifuging with the YM-3 filter devices and demetallated by washing the concentrate with
1 mL of double distilled H2O through YM-3 filters until
a final volume of 200 lL was achieved These samples were lyophilized and reconstituted in 100 lL of denaturing buffer (6Mguanidine/HCl, 0.5MTris/HCl, 4 mMEDTA;
pH 8.0) A 20 lL sample was stored at)80 C for mass determination of the native proteins The remaining 80 lL was subjected to carboxymethylation with iodoacetic acid (IAA) Briefly, the sample was diluted into 920 lL of denaturing buffer deoxygenated with N2gas The remaining steps were performed under N2 gas with deoxygenated reagents Sixty microlitres of 100 mM dithiothreitol was added to the sample, which was then incubated at 37C for
90 min, after which 120 lL of 0.2MIAA was added and incubated continued at 37C for 120 min in the dark Samples suspected to contain MTs in the 6000–7000 Mr range were concentrated using the YM-3 filter devices and washed with ddH2O Samples believed to represent the high Mr isoforms (> 15 000) were subjected to a buffer exchange by elution through a Superdex 75 PC column in
10 mM NH4HCO3, lyophilization with a speedvac and reconstitution in 25 lL of ddH2O The native and carb-oxymethylated proteins were desalted with ZipTipC18 pip-ette tips (Millipore) and eluted in 0.1% trifluoroacetic acid containing 50% acetonitrile Samples were diluted in three parts sinapinic acid matrix (50 mM 3,5-dimethoxy-4-hydroxycinnamic acid/70% acetonitrile/0.1% trifluoro-acetic acid) and the mass was determined by MALDI-TOF
MS (Voyager-DE STR BioSpectrometry Workstation; Applied Biosystems) Cysteine content was determined by subtracting the mass of the native protein from the carboxymethylated protein and dividing by 58 Da (mass
of the IAA derivative)
The metal-rich fractions believed to correspond to the small Mr MT ( 4000) isoforms were lyophilized and reconstituted in 60 lL of denaturing buffer (6Mguanidine/ HCl, 0.5MTris/HCl, 2 mMEDTA; pH 8.2) YM-3 filters or buffer exchange were likely to result in significant loss
of sample, so the demetallation step was not performed
A 10 lL aliquot of sample was stored at)80 C for mass determination of the native proteins A modified method was used for carboxymethylation of the remaining 50 lL Briefly, the sample was deoxygenated with N2gas and 2 lL
of 100 mMdithiothreitol was added before the sample was incubated at 45C for 60 min, after which 6 lL of 0.2M IAA was added and incubated at 45C for 60 min in the dark The native and carboxymethylated proteins were
Trang 4purified with ZipTipC18pipette tips and eluted in 0.1%
tri-fluoroacetic acid containing 50% acetonitrile Samples were
diluted in three parts a-cyano matrix (50 mM
a-cyano-4-hydroxycinnamic acid/70% acetonitrile/0.1%
trifluoroace-tic acid) and the mass was determined by MALDI-TOF MS
Results
Diversity of oyster MTs at the level of the transcriptome
cDNA libraries constructed from control, cadmium-, and
copper-treated larvae were screened with a probe
represent-ing the oyster MT originally reported [18] From these
respective libraries, 20, 25, and 38 clones were plaque
purified, and 10 distinct isoforms were identified and
com-pletely sequenced (NCBI accession numbers AY331695 to
AY331707; also viewable at http://www.marinegenomics
org) The nucleotide sequences obtained all showed strong
similarity (> 85%) to the known oyster MT that was used
as the probe and which is designated as CvMT-IA
Although this strong sequence conservation indicates that
all of the sequenced MTs belong to the same family, the
10 sequences could be divided into two separate subfamilies
(CvMT-I and CvMT-II), based on their conceptual
trans-lation and the inferred domain structure of the encoded
MTs (Fig 1A,B) Two clones were of the CvMT-I
sub-family and eight clones were of the CvMT-II subsub-family The
designation CvMT-I is used to represent the traditional class
of molluscan MT proteins with a-and b-domains and 21 conserved cysteines [30] In addition to CvMT-IA, a novel isoform of the same subfamily (CvMT-IB) was identified from the control (nonmetals challenged) larval cDNA library (Fig 1B), and showed five amino acid substitutions
in the a-domain, one in the b-domain and conservation of all 21 cysteines The CvMT-II subfamily is distinguished by the presence of only a-domains in its conceptual translation This structure arises from the presence of a mutation (AfiT) which converts a lysine codon (AAG) in the linker region separating the a-and b-domains into a stop codon (TAG) The CvMT-II subfamily is exemplified by two related isoforms (CvMT-IIA and -IIB) which, in conceptual translation, are single a-domain peptides of inferred Mrof
4100 The CvMT-II subfamily contains additional mem-bers (designated CvMT-IIC through CvMT-IIH) in which two, three and four a-domains are encoded (Fig 1A), and have inferred Mr values of 9200, 14 600 and 20 200, respectively
Inferred exon structure of theCvMT-IA gene The sequencing of MT cDNAs revealed several partially spliced CvMT-IA transcripts which, taken together, permit-ted deduction of the intron/exon structure of the CvMT-IA gene, as shown in Fig 2A This deduced gene structure was
Fig 1 Diversity of metallothionein (MT) isoforms from Crassostrea virginica (A) Schematic representation of diversity in the CvMTI/II family, demonstrating the characteristic domain structures of the isoforms characterized as cDNA The cysteine-rich domains are classified as either a or b, based on the number and configuration of the cysteine motifs The noncoding exon region represents the b-domain region of the transcript that has been truncated by the introduction of a stop codon in the linker region (B) Amino acid sequences of representative isoforms from the CvMT-I/II family, as deduced from cDNA sequences Conserved residues are designated by (d) and the linker region (KVK) between the two domains is underlined The alignment demonstrates the presence of the three a-domains present in CvMT-IIC, with each domain being presented as a separate, rather than contiguous, sequence.
Trang 5compared with the structures reported for the C virginica
MTAgene (NCBI accession number AF506977) and the
C gigas MT1gene [22], as shown in Fig 2B The CvMT-I
gene was found to have the same intron phase (1,1) as the
CvMTAgene, whereas the C gigas MT1 gene had a (1,2)
intron phase The three genes differed remarkably in their
intron lengths The first introns in the two CvMT-I genes
were the only ones of similar length The second intron was
much shorter in CvMT-IA than in CvMTA; however, it
shared 86% identity with the last 92 nucleotides of the
second intron of CvMTA The second intron of CvMT-IA
was found to contain a noncanonical donor splice site
(AG/tT, Fig 2A), which has been reported as a rare splice
site variant in mammalian genes [31] Taken together with
the cDNA cloning data, these results strongly suggest that in
C virginicathere is a large family of CvMT-I/II genes In
order to gain further insight into this gene family, genomic
Southern blot analysis of two oysters was performed, using
a probe that detects members of both the CvMT-I and
CvMT-II subfamilies The results (Fig 3) are clearly
compatible with the presence of multiple copies of these
sequences in the oyster genome, and the variability of
intensity between the hybridizing bands suggests that there
may be multiple, closely linked CvMT-I/II sequences The
differences in restriction fragment length between the two
oysters also indicated substantial allelic polymorphism in
this gene family
CvMT-I and CvMT-II gene expression is induced
by cadmium
The expression of CvMT-I/II isoforms in gill and
hepato-pancreas, and their upregulation by exposure to copper,
zinc and cadmium, were examined by Northern blot analysis The CvMT-I and CvMT-II messages were readily distinguishable by their relative electrophoretic mobility,
as indicated in Fig 4 While the results suggested that hepatopancreas has a higher basal expression level of CvMT-I/II isoforms than does gill tissue, it is clear that cadmium exposure strongly upregulated CvMT-I/II expres-sion in both tissues However, there were no significant changes in the levels of CvMT-I/II expression following exposure to copper or zinc at the same concentration as cadmium
Fig 2 A comparison of metallothionein (MT) gene structure in oysters.
(A) Diagrammatic representation of two partially spliced transcripts of
CvMT-IA Exons are represented by boxes Introns are represented by
straight lines The angled lines represent the intron spliced from the
partially processed transcript The second intron contains a
nonca-nonical splice donor site (tTgAG) (B) Comparison of the proposed
exon/intron structure of CvMT-IA with that of two characterized
MT genes from Crassostrea virginica and C gigas (aAF506977,
b AJ242657) Exons are designated by boxes with the number of
nucleotides in each open reading frame Introns are designated by
lines, with the length (number of nucleotides) shown in parentheses.
Fig 3 Southern blot analysis of the CvMT-I/II gene family in Cras-sostrea virginica Genomic DNA from two individual oysters was digested with EcoRI, AvaII and BamHI, and analyzed, after electro-phoresis and blot transfer, with a probe for the CvMT-IA cDNA Each lane contains 7.5 lg of genomic DNA.
Fig 4 Northern blot analysis of control and metal-treated tissues from adult Crassostrea virginica Adult oysters were exposed to copper, cadmium and zinc for 96 h, and 5 lg of total RNA from gill and hepatopancreas tissues was analyzed by Northern blot using a probe for the CvMT-IA cDNA.
Trang 6Diversity of oyster MTs at the level of the proteome
Studies of oyster MT proteins were undertaken, first, to
confirm that the diversity of CvMT-I/II sequences seen at
the cDNA level was reflected at the level of the expressed
proteins and, second, to test the possibility that different
isoforms of the CvMT-I subfamily preferentially associate
with cadmium In the initial characterization of oyster MTs,
extracts of hepatopancreas from control and
cadmium-exposed oysters were separated by gel filtration
chromato-graphy The eluted proteins were monitored at three
wavelengths The relative absorption at 220/280 nm allowed
the detection of proteins (such as MTs) that are deficient in
aromatic amino acids, while cadmium/thiol interactions
yielded an increased absorbance at 254 nm Comparison of
the elution profiles identified three protein peaks in the
cadmium-treated samples that showed specific increases of
absorption at 220 and 254 nm, but not at 280 nm (Fig 5A,
right panel) These three fractions also corresponded to
peaks of cadmium in the elution profile (Fig 5B) The
elution profile of the three cadmium-rich pools (A, B, C;
Fig 5B) was consistent with the predicted diverse Mrvalues
of the multiple CvMT-I/II isoforms detected at the cDNA
level, but to determine the exact nature of these proteins,
further analysis was undertaken to determine
experiment-ally their Mrand cysteine content Each of the three pools
was fractionated by anion-exchange HPLC, and the metals
elution profile for the high and low Mrpools were measured
for cadmium, while the classic Mrpool was also measured
for copper and zinc (Fig 6) The metals elution profiles for
the high and low Mr pools were determined from single
representative samples of cadmium-exposed oysters The
metals elution profile of the classic Mrpool was a composite
of control and cadmium-treated oysters The proteins in the
fractions identified by the metals elution profiles were
subjected to MALDI-MS analysis before and after
deriva-tization with IAA Typical MALDI-MS traces are shown,
in Fig 7, for the analysis of three of the fractions before and
after derivatization Overall, 10 proteins were identified, by
MALDI-MS, whose Mr and calculated cysteine content were consistent with their identification as members of the CvMT-I and CvMT-II subfamilies Of these, three MTs (peaks e, f and g, Fig 6B), of approximate Mr7242–7375, were characterized in zinc-rich fractions, with peak g identified from a control oyster All the other MTs were found in cadmium-rich fractions after 96 h of exposure of oysters to 0.44 lMof cadmium As summarized in Table 1, putative MTs of the ab-domain structure (CvMT-I) and with one, three and four a-domains (CvMT-II), could readily be identified Sequence diversity (of unknown extent) within the CvMT-I/II family, and uncertainties over post-translational modifications of MTs (such as N-acetylation [18]), probably contribute to the small divergence seen between the conceptual (cDNA translation) and experi-mentally observed Mr values Although the examination
of MT representation in the oyster proteome was not exhaustive, it is clear from the data presented in Table 1 that there is a substantial complexity of the CvMT-I/II family
Discussion
The data presented in this study were obtained by an initial transcriptomic and proteomic study, and reveal a diversity
of oyster MTs that has implications for our understanding
of the evolution of this gene family and for interpreting structure/function relationships in molluscan MTs Diversity of oyster MTs at the transcriptomic level
It is known from previous studies [22–26] that molluscan MTs show a diversity of structure that encompasses not only the canonical ab-domain structure, but also molecular forms in which this structure has been modified, e.g as in the abb MT seen in the Pacific oyster, C gigas [27] The data reported here reveal a structural and functional diversity within the MT family of the American oyster (C virginica) that, while proposed by prior studies at the protein level [19–21], has not previously been documented
Fig 5 Gel filtration profile of
cadmium-exposed hepatopancreas tissue from adult
Crassostrea virginica Extracts of
hepatopan-creas tissue from control and
cadmium-exposed adult oysters were subjected to gel
filtration chromatography with a Superdex 75
PC 3.2/30 column (A) Chromatograms
clearly demonstrate three strong peaks (21,
25.5, and 31.5 min), detectable at 220 nm but
not at 280 nm, in extracts of cadmium-treated
tissues The corresponding absorbance at
254 nm is consistent with cadmium–thiol
interactions expected of metallothionein
pro-teins (B) Cadmium elution profile of the same
samples demonstrates the presence of three
cadmium-rich pools (A, B and C)
corres-ponding to the 220 nm/254 nm absorbance.
The rabbit MT (rMT) standard eluted at
26 min.
Trang 7In particular, it is clear that in this species of oyster, the
family of canonical ab-domain-containing MTs (the
CvMT-I subfamily) has undergone substantial expansion
to include MTs that solely express a-domains (the CvMT-II
subfamily) While cDNA cloning showed the presence
of CvMT-II transcripts encoding MTs with one to four a-domains, analysis at the protein level (discussed below) identified putative expressed molecules corresponding to three of these MTs: those with one, three and four a-domains Analysis of cDNA sequences of CvMT-I/II clones, along with the intronic sequence of the CvMT-IA gene, permitted deduction of the series of events that probably led to the generation of genes encoding CvMT-II family members Initially, the mutation of a lysine codon to
a stop codon in the linker region would have truncated the
MT protein after the a-domain Subsequent tandem duplications of the a-encoding sequence (the first two exons) would then have readily generated the multiple CvMT-IIgenes identified in this study
While the data reported here confirmed that the
CvMT-IAgene had the same pattern of three exons/two introns previously reported for a C virginica MT gene (CvMTA, ACCN_AF506977) and for a C gigas MT gene [22], the variations seen in intron length suggest that molluscan MT genes, while conforming to a basic exon structure, probably show, as predicted, substantial variations in their introns The presence of a rare noncanonical donor splice site in the CvMT-IA gene (Fig 2) suggests that this variation in intronic sequences may have implications for the expression
of the oyster MT genes While Southern blot analysis confirmed CvMT-I/II as a multigene family, it was unable
to distinguish the total or relative genomic representation of the CvMT-I and CvMT-II subfamilies of genes
The expression of oyster MT genes in response to metals exposure was measured by Northern blot, and showed that there was apparent global upregulation of CvMT-I/II transcripts induced by cadmium, but not by comparable concentrations of copper or zinc Variable baseline expres-sion of MTs was observed in unchallenged oysters (Fig 4), but it is not known if this pattern of expression reflects prior exposure to metals or other stresses that may induce MT expression, or is representative of basal expression associ-ated with normal metals homeostasis
Diversity of oyster MTs at the proteomic level Characterization of oyster MTs was undertaken at the protein level in order to confirm and extend the results obtained from cDNA analysis Gel filtration and anion-exchange chromatography, combined with MALDI-MS analysis, resulted in the identification of 10 MTs in extracts
of control and cadmium-treated oysters On the basis of size and cysteine content, these could be identified, with confidence, as members of the CvMT-I/II family (Table 1) Because of the probable size of the CvMT-I/II gene family, uncertainties exist concerning the full range of sequences of the encoded MTs, as well as the potential post-translational modification of MTs These uncertainties make difficult any attempt to correlate the observed oyster MT proteins with the isoforms inferred from the cDNA sequences However, based on cadmium-binding, Mr and cysteine content characteristics, it is highly likely that CvMT-IA, the prototypical oyster metallothionein [32], has been identified (Fig 6B; peak d) In addition, we identified three MTs whose Mr values ( 7200) were not consistent with the predicted characteristics of CvMT-IA or -IB These three MTs were the only ones identified in this study in copper/
Fig 6 Anion exchange HPLC of the cadmium-rich pools The three
cadmium-rich pools identified from gel filtration chromatography
(asterisked in Fig 5A) were subjected to anion exchange
chromato-graphy Metal elution profiles (cadmium and/or copper and zinc) were
determined by spectrometry (PerkinElmer AAnalyst Model 700
atomic absorption spectrophotometer), and eluted proteins were
analyzed by MALDI-MS to determine the M r and cysteine content
(Fig 7, Table 1) All peaks labeled with a lowercase letter (a–i) had a
cysteine content consistent with their identification as metallothionein
(MT) (A) Anion-exchange chromatography of the high molecular
mass pool from gel filtration (Fig 5) identified three cadmium-rich
peaks (a–c) containing MTs (B) Anion-exchange chromatography of
the intermediate molecular mass pool from gel filtration (Fig 5)
identified four peaks (d–g) containing MTs Only one strong
cadmium-rich MT-containing peak (d) was recovered from this pool, but three
MT-containing peaks (e–g) were identified as copper/zinc rich.
(C) Anion-exchange chromatography of the low molecular mass pool
from gel filtration (Fig 5) identified two cadmium-rich peaks (h,i)
containing MTs.
Trang 8Fig 7 Identification of CvMTI/II family
pro-teins by MALDI-TOF MS The mass of the
native and iodoacetic acid (IAA)-derivatized
proteins from the metals-rich peaks (b,d,h;
Fig 6) identified by anion-exchange HPLC
was determined by MALDI-TOF MS Three
representative trace spectra clearly illustrate
the identification of metallothioneins (MTs)
from each of the three M r pools based on gel
filtration analysis (Fig 5).
Table 1 Characteristics of CvMT-I/II family members Predicted, conceptual translation of nucleotide sequence; Observed, measured by MALDI-MS; Length, number of amino acid residues; M r , relative molecular mass calculated from the amino acid sequence; M r acetylated, relative molecular mass calculated assuming N-acetylation of the MT; Fraction, as designated in Figs 6 and 7; M r native, MALDI-MS of underivatized proteins; M r IAA, MALDI-MS of iodoacetic acid (IAA) derivatized proteins; Cys residues, number of cysteine residues calculated from the mass difference between the native and IAA-derivatized proteins measured by MALDI-MS; Subfamily, the exact assignment of isoform was not attempted.
Isoform
Length M r
M r
acetylated
Domain structure
Cys residues Fraction
M r
native
M r
IAA
Cys residues Subfamily
f 7250.4 8470.6 21.0 CvMT-I
g 7375.9 8390.4 17.5 CvMT-I
CvMT-IIB 42 4122 4164 a 12 h, i 4234.8 4933.2 12.0 CvMT-II
CvMT-IID 148 14 758 14 800 (a) 3 38 a 14 638.9 16 879.9 38.6 CvMT-II CvMT-IIE 148 14 592 14 634 (a) 3 38 b 14 640.7 16 856.4 38.2 CvMT-II
CvMT-IIG 203 20 202 20 244 (a) 4 51 b 20 478.9 23 310.9 48.8 CvMT-II CvMT-IH 200 19 777 19 819 (a) 4 51 c 20 461.2 23 342.1 49.7 CvMT-II
Trang 9zinc-rich and cadmium-poor fractions, despite the fact that
two were isolated from cadmium-treated oysters Thus, they
may represent novel CvMT-I isoforms with highly
prefer-ential binding for copper/zinc
Observations presented here support the hypothesis that
the a-domain of molluscan MTs has characteristics similar
to those of other vertebrate and invertebrate species
The analysis of proteins induced by cadmium exposure
(Figs 5–7) identified multiple CvMT-II isoforms, containing
one, three and four a-domains, all associated with
cad-mium-rich fractions Although the data support the
expres-sion and cadmium-inducibility of CvMT-II proteins, the
mass accuracy of the analysis was not adequate to confirm
the identity of specific CvMT-II isoforms identified as
cDNA
Evolution of MT structure
The MT b-domain has been proposed as the ancestral MT
domain with a primary role in the homoeostasis of
physiologically relevant metals, such as copper and zinc
It has been suggested that duplication of the b-domain
gave rise to a two-domain MT, and the subsequent
divergence of the two domains eventually gave rise to the
canonical a/b structure of the MTs [33] The b-domain has
a binding preference for copper [34], whereas the a-domain
has a preference for cadmium and zinc [35] This suggests
that selective pressures may have led to the evolution of
two domain MTs with specific functions carried out by the
two domains, with the b-domain more important for metal
homeostasis and the a-domain more important for metal
storage and detoxification This hypothesis is supported
by the presence of single b-domain, copper-thionein systems
present in Drosophila [4] and fungi [6,36] and the existence
of the crustacean MTs, comprising two b-domains, that
function in copper homeostasis related to the synthesis
and degradation of hemocyanin [13,37] Of interest is the
single-domain MT peptide (containing 41 amino acids and
capable of binding four cadmium ions) that has been
identified in a terrestrial worm, Eisenia foetida [38] This is a
cadmium-inducible MT derived from a two-domain
mole-cule by post-translational cleavage The four-metal-cluster
binding stoichiometry of this MT would suggest functional
analogy to a single a-domain MT This theory of domain
duplication is further supported by the widespread
occur-rence of the ab-and ba-domain structures of many
invertebrate and vertebrate MTs and their roles in zinc
homeostasis and cadmium detoxification This notion can
also explain the presence of the high molecular mass MT
proteins, which may enhance metals-resistance in benthic
and terrestrial organisms experiencing a greater exposure to
metals owing to their ecological niche [27,39,40] It should
be acknowledged that the theory of gene duplication
experiences some difficulties when invertebrate and
verteb-rate MT gene structures are compared: in many
inverte-brates, the a-domain is N-terminally encoded, whereas in
vertebrates the reverse is the case, with the b-domain being
N-terminally encoded [41] Thus, true homology between
the a-and b-domains of invertebrate and vertebrate
domains would require an inversion within the MT gene
of the a and b encoding segments, an event of which there is
no obvious record in the genes
The presence of the three-domain abb MT protein from
C gigas [27] is an interesting contrast to the multiple a-domains of the CvMT-II isoforms While both species of oyster appear to have adopted similar strategies for survival
in environments that can be metals-rich, selective pressures and novel genetic mutations in C virginica appear to have resulted in the unusual structure of the CvMT-II isoforms Thus, under the general hypothesis discussed above, the differences in domain structure between CvMT-I and -II may represent an example of evolutionarily divergent domain functions While the stability and metal-binding affinity of the CvMT-II proteins are not yet known, the cadmium-inducibility and in vivo cadmium-binding proper-ties of these proteins suggests similar roles in metals-detoxification and metals-resistance, as proposed for high molecular mass isoforms of MTs and other cysteine-rich proteins present in other species [27,40]
Acknowledgements
The authors would like to thank Drs Paul Gross and Mats Lundquist, Darlene Middleton, and members of the Marine Genom-ics Program for support and advice with this study We would also like to thank Dr G Roesijadi for the generous donation of the cDNA for CvMT-IA In addition, we would like to thank the staff of the ACE Basin National Estuarine Reserve and members of the South Carolina Department of Natural Resources, Marine Resources Research Institute, for their assistance with oyster collection Additional thanks to the MUSC Mass Spectrometry Facility This paper is the Charleston, SC Marine Genomics Group contribution
#4, #01-04 of the Cooperative Institute of Fisheries Molecular Biology and #537 of the South Carolina Department of Natural Resources Research was supported by the National Oceanic and Atmospheric Administration, National Marine Fisheries Service (NA07FL0498) and National Science Foundation (EPS0083102) Part of the research was conducted under an award from the Estuarine Reserves Division, Office of Ocean and Coastal Resource Management, National Ocean Service, National Oceanic and Atmo-spheric Administration.
References
1 Kagi, J.H & Hunziker, P (1989) Mammalian metallothionein Biol Trace Element Res 21, 111–118.
2 Kagi, J.H & Schaffer, A (1988) Biochemistry of metallothionein Biochemistry 27, 8509–8515.
3 Valls, M., Bofill, R., Romero- Isart, N., Gonzalez- Duarte, R., Abian, J., Carrascal, M., Gonzalez- Duarte, P., Capdevila, M & Atrian, S (2000) Drosophila MTN: a metazoan copper-thionein related to fungal forms FEBS Lett 467, 189–194.
4 Domenech, J., Palacios, O., Villarreal, L., Gonzalez-Duarte, P., Capdevila, M & Atrian, S (2003) MTO: the second member of a Drosophila dual copper-thionein system FEBS Lett 533, 72–78.
5 Beltramini, M & Lerch, K (1986) Primary structure and spec-troscopic studies of Neurospora copper metallothionein Environ Health Perspect 65, 21–27.
6 Winge, D.R., Nielson, K.B., Gray, W.R & Hamer, D.H (1985) Yeast metallothionein: sequence and metal-binding properties.
J Biol Chem 260, 14464–14470.
7 Braun, W., Vasak, M., Robbins, A.H., Stout, C.D., Wagner, G., Kagi, J.H & Wuthrich, K (1992) Comparison of the NMR solution structure and the x-ray crystal structure of rat metallothionein-2 Proc Natl Acad Sci USA 89, 10124– 10128.
Trang 108 Riek, R., Precheur, B., Wang, Y., Mackay, E.A., Wider, G.,
Guntert, P., Liu, A., Kagi, J.H & Wuthrich, K (1999) NMR
structure of the sea urchin (Strongylocentrotus purpuratus)
metallothionein MTA J Mol Biol 291, 417–428.
9 Jiang, L.J., Vasak, M., Vallee, B.L & Maret, W (2000) Zinc
transfer potentials of the alpha-and beta-clusters of
metallothio-nein are affected by domain interactions in the whole molecule.
Proc Natl Acad Sci USA 97, 2503–2508.
10 Brouwer, M., Hoexum-Brouwer, T & Cashon, R.E (1993) A
putative glutathione-binding site in CdZn-metallothionein
identi-fied by equilibrium binding and molecular-modelling studies.
Biochem J 294, 219–225.
11 Suzuki, K.T & Kuroda, T (1994) Direct transfer of copper from
metallothionein to superoxide dismutase: a possible mechanism
for differential supply of Cu to SOD and ceruloplasmin in LEC
rats Res Commun Mol Path Pharmacol 86, 15–23.
12 Jiang, L.J., Maret, W & Vallee, B.L (1998) The glutathione redox
couple modulates zinc transfer from metallothionein to
zinc-depleted sorbitol dehydrogenase Proc Natl Acad Sci USA 95,
3483–3488.
13 Narula, S.S., Brouwer, M., Hua, Y & Armitage, I.M (1995)
Three-dimensional solution structure of Callinectes sapidus
metallothionein-1 determined by homonuclear and heteronuclear
magnetic resonance spectroscopy Biochemistry 34, 620–631.
14 Unger, M.E & Roesijadi, G (1993) Sensitive assay for molluscan
metallothionein induction based on ribonuclease protection and
molecular titration of metallothionein and actin mRNAs Mol.
Mar Biol Biotechnol 2, 319–324.
15 Chan, M.K., Othman, R., Zubir, D & Salmijah, S (2002)
Induction of a putative metallothionein gene in the blood cockle,
Anadara granosa, exposed to cadmium Comp Biochem Physiol.
C Toxicol Pharmacol 131, 123–132.
16 Lemoine, S., Bigot, Y., Sellos, D., Cosson, R.P & Laulier, M.
(2000) Metallothionein isoforms in Mytilus edulis (Mollusca,
Bivalvia): complementary DNA characterization and
quantifica-tion of expression in different organs after exposure to cadmium,
zinc, and copper Mar Biotechnol 2, 195–203.
17 Hensbergen, P.J., Donker, M.H., Van Velzen, M.J., Roelofs, D.,
Van Der Schors, R.C., Hunziker, E & Van Straalen, N.M (1999)
Primary structure of a cadmium-induced metallothionein from the
insect Orchesella cincta (Collembola) Eur J Biochem 259,
197–203.
18 Unger, M.E., Chen, T.T., Murphy, C.M., Vestling, M.M.,
Fenselau, C & Roesijadi, G (1991) Primary structure of
mollus-can metallothioneins deduced from PCR-amplified cDNA and
mass spectrometry of purified proteins Biochim Biophys Acta
1074, 371–377.
19 Engel, D.W (1999) Accumulation and cytosolic partitioning of
metals in the American oyster Crassostrea virginica Mar Environ.
Res 47, 89–102.
20 Fowler, B.A., Engel, D.W & Brouwer, M (1986) Purification and
characterization studies of cadmium-binding proteins from the
American oyster, Crassostrea virginica Environ Health Perspect.
65, 63–69.
21 Ringwood, A.H & Brouwer, M (1993) Expression of constitutive
and metal-inducible metallothioneins in oyster embryos
(Cras-sostrea virginica) Comp Biochem Physiol B Biochem Mol Biol.
106B, 523–529.
22 Tanguy, A., Mura, C & Moraga, D (2001) Cloning of a
metallothionein gene and characterization of two other cDNA
sequences in the Pacific oyster Crassostrea gigas (CgMT1) Aquat.
Toxicol 55, 35–47.
23 Mackay, E.A., Overnell, J., Dunbar, B., Davidson, I., Hunziker,
P.E., Kagi, J.H & Fothergill, J.E (1993) Complete amino acid
sequences of five dimeric and four monomeric forms of
metallothionein from the edible mussel Mytilus edulis Eur J Biochem 218, 183–194.
24 Barsyte, D., White, K.N & Lovejoy, D.A (1999) Cloning and characterization of metallothionein cDNAs in the mussel Mytilus edulis L digestive gland Comp Biochem Physiol C Pharmacol Toxicol Endocrinol 122, 287–296.
25 Berger, B., Dallinger, R., Gehrig, P & Hunziker, P.E (1997) Primary structure of a copper-binding metallothionein from mantle tissue of the terrestrial gastropod Helix pomatia L Bio-chem J 328, 219–224.
26 Dallinger, R., Berger, B., Hunziker, P.E., Birchler, N., Hauer, C.R & Kagi, J.H (1993) Purification and primary structure
of snail metallothionein: similarity of the N-terminal seq-uence with histones H4 and H2A Eur J Biochem 216, 739– 746.
27 Tanguy, A & Moraga, D (2001) Cloning and characterization of
a gene coding for a novel metallothionein in the Pacific oyster Crassostrea gigas (CgMT2): a case of adaptive response to metal-induced stress? Gene 273, 123–130.
28 Jenny, M.J., Ringwood, A.H., Lacy, E.R., Lewitus, A.J., Kemp-ton, J.W., Gross, P.S., Warr, G.W & Chapman, R.W (2002) Potential indicators of stress response identified by expressed sequence tag analysis of hemocytes and embryos from the American oyster, Crassostrea virginica Mar Biotechnol 4, 81–93.
29 Church, G.M & Gilbert, W (1984) Genomic sequencing Proc Natl Acad Sci USA 81, 1991–1995.
30 Binz, P.A & Kagi, J.H.R (1999) Molecular evolution and clas-sification in metallothionein In Metallothionein, Vol IV (Klaas-sen, C., ed.), pp 7–13 Birkhauser, Basel.
31 Burset, M., Seledtsov, I.A & Solovyev, V.V (2000) Analysis of canonical and non-canonical splice sites in mammalian genomes Nucleic Acids Res 28, 4364–4375.
32 Roesijadi, G., Kielland, S & Klerks, P (1989) Purification and properties of novel molluscan metallothioneins Arch Biochem Biophys 273, 403–413.
33 Cols, N., Romero-Isart, N., Bofill, R., Capdevila, M., Gonzalez-Duarte, P., Gonzalez-Gonzalez-Duarte, R & Atrian, S (1999) I n vivo cop-per-and cadmium-binding ability of mammalian metallothionein beta domain Protein Eng 12, 265–269.
34 Nielson, K.B & Winge, D.R (1984) Preferential binding of cop-per to the beta domain of metallothionein J Biol Chem 259, 4941–4946.
35 Nielson, K.B & Winge, D.R (1983) Order of metal binding in metallothionein J Biol Chem 258, 13063–13069.
36 Lerch, K (1980) Copper metallothionein, a copper-binding pro-tein from Neurospora crassa Nature 284, 368–370.
37 Brouwer, M., Syring, R & Hoexum Brouwer, T (2002) Role of a copper-specific metallothionein of the blue crab, Callinectes sapidus, in copper metabolism associated with degradation and synthesis of hemocyanin J Inorg Biochem 88, 228–239.
38 Gruber, C., Sturzenbaum, S., Gehrig, P., Sack, R., Hunziker, P., Berger, B & Dallinger, R (2000) Isolation and characterization of
a self-sufficient one-domain protein: (Cd)-metallothionein from Eisenia foetida Eur J Biochem 267, 573–582.
39 Willuhn, J., Schmitt-Wrede, H.P., Greven, H & Wunderlich, F (1994) cDNA cloning of a cadmium-inducible mRNA encoding a novel cysteine-rich, non-metallothionein 25-kDa protein in an enchytraeid earthworm J Biol Chem 269, 24688– 24691.
40 Tschuschke, S., Schmitt-Wrede, H.P., Greven, H & Wunderlich,
F (2002) Cadmium resistance conferred to yeast by a non-metallothionein-encoding gene of the earthworm Enchytraeus.
J Biol Chem 277, 5120–5125.