Download Genomics Reveal Ancient Forms of Stanniocalcin in Amphioxus and

Integrative and Comparative Biology, volume 50, number 1, pp. 86–97
doi:10.1093/icb/icq010
SYMPOSIUM
Genomics Reveal Ancient Forms of Stanniocalcin in Amphioxus
and Tunicate
Graeme J. Roch and Nancy M. Sherwood1,
*Department of Biology, University of Victoria, Victoria, British Columbia V8W 3N5, Canada
From the symposium ‘‘Insights of Early Chordate Genomics: Endocrinology and Development in Amphioxus, Tunicates
and Lampreys’’ presented at the annual meeting of the Society for Integrative and Comparative Biology, January 3–7,
2010, at Seattle, Washington.
1
E-mail: [email protected]
Synopsis Stanniocalcin (STC) is present throughout vertebrates, including humans, but a structure for STC has not been
identified in animals that evolved before bony fish. The origin of this pleiotropic hormone known to regulate calcium is
not clear. In the present study, we have cloned three stanniocalcins from two invertebrates, the tunicate Ciona intestinalis
and the amphioxus Branchiostoma floridae. Both species are protochordates with the tunicates as the closest living
relatives to vertebrates. Amphioxus are basal to both tunicates and vertebrates. The genes and predicted proteins of
tunicate and amphioxus share several key structural features found in all previously described homologs. Both the
invertebrate and vertebrate genes have four conserved exons. The predicted length of the single pro-STC in Ciona is
237 amino acids and the two pro-hormones in amphioxus are 207 and 210 residues, which is shorter than human
pro-STCs at 247 and 302 residues due to expansion of the C-terminal region in vertebrate forms. The conserved pattern
of 10 cysteines in all chordate STCs is crucial for identification as amphioxus and tunicate amino acids are only 14–23%
identical with human STC1 and STC2. The 11th cysteine, which is the cysteine shown to form a homodimer in vertebrates, is present only in amphioxus STCa, but not in amphioxus STCb or tunicate STC, suggesting the latter two are
monomers. The expression of stanniocalcin in Ciona is widespread as shown by RT-PCR and by quantitative PCR. The
latter method shows that the highest amount of STC mRNA is in the heart with lower amounts in the neural complex,
branchial basket, and endostyle. A widespread distribution is present also in mammals and fish for both STC1 and STC2.
Stanniocalcin is a presumptive regulator of calcium in both Ciona and amphioxus, although the structure of a STC
receptor remains to be identified in any organism. Our data suggest that amphioxus STCa is most similar to the common
ancestor of vertebrate STCs because it has an 11th cysteine necessary for dimerization, an N-glycosylation motif, although
not the canonical one in vertebrate STCs, and similar gene organization. Tunicate and amphioxus STCs are more similar
in structure to vertebrate STC1 than to vertebrate STC2. The unique features of STC2, including 14 instead of 11
cysteines and a cluster of histidines in the C-terminal region, appear to be found exclusively in vertebrates.
Introduction
Stanniocalcin is a large protein hormone that was
originally purified from specialized glands, known
as the corpuscles of Stannius, associated with the
kidneys of teleostean and holostean fishes. Wagner
et al. (1986) isolated and provided a partial aminoacid sequence for the first stanniocalcin (referred to
as ‘teleocalcin’) from sockeye salmon corpuscles.
They determined that it was a dimerized glycoprotein with hypocalcemic activity. Shortly thereafter, an
orthologous hormone was isolated, cloned, and
sequenced in eels (Butkus et al. 1987). Characterization of the hormone in several additional teleosts
followed, including other salmonids; trout (Lafeber
et al. 1988a; Lafeber et al. 1988b), chum (Sundell
et al. 1992; Yamashita et al. 1995) and coho
salmon (Wagner et al. 1988; Wagner et al. 1992),
supporting the initial evidence of calcium downregulation by secretion of stanniocalcin from the corpuscles to effectors in the gills. More recently,
the hormone was isolated from divergent groups
of bony fishes, including the early teleostean
bony-tongued fishes: arawana, butterflyfish, and
Advanced Access publication April 1, 2010
ß The Author 2010. Published by Oxford University Press on behalf of the Society for Integrative and Comparative Biology. All rights reserved.
For permissions please email: [email protected].
87
Stanniocalcin in amphioxus and tunicate
elephantnose (Amemiya et al. 2002; Amemiya et al.
2006), and the non-teleostean gar and bowfin (Amemiya and Youson 2004). All these hormones share
consistent structural features, including a conserved
N-glycosylation site and 10 invariant cysteine residues that form intrachain disulfide bonds. Also, an
11th conserved cysteine residue that is critical to
dimerization (Hulova and Kawauchi 1999; Trindade
et al. 2009) is found in all species with the exception
of the bony-tongued fishes (Order, Osteoglossiformes). The latter secrete a unique, monomeric isoform of stanniocalcin (Amemiya et al. 2002;
Amemiya et al. 2006).
A human ortholog of stanniocalcin was later identified and cloned (Chang et al. 1995; Wagner et al.
1995; Olsen et al. 1996). Mouse stanniocalcin was
also isolated (Chang et al. 1996), and the mammalian
homologs both shared moderate (450%) amino-acid
identity with their fish counterparts, as well as the
conserved structural features. The expression of
mammalian stanniocalcin was widespread throughout both endocrine and non-endocrine tissues, leading to the suggestion that it might serve a more
autocrine or paracrine role (Ishibashi and Imai
2002). Early investigations of function for mammalian stanniocalcins suggested a familiar role in mineral homeostasis, although focused more on
upregulated phosphate reabsorption rather than on
downregulation of calcium (Haddad et al. 1996;
Olsen et al. 1996). As research progressed, several
novel roles for the hormone were implicated, including bone formation, growth effects, angiogenesis and,
neuroprotection (Yoshiko and Aubin 2004). This was
complicated by the discovery of tissue-specific isoforms of stanniocalcin found in the ovaries, adrenals,
and adipocytes, referred to as ‘big stanniocalcin’
(Paciga et al. 2003). The nature of the big isoform
is still unknown, even after studies on its composition and structure (Trindade et al. 2009).
A second stanniocalcin gene was idenitified in
mice and humans, and cloned by several groups
(Chang and Reddel 1998; DiMattia et al. 1998;
Ishibashi et al. 1998). This hormone, known as
STC2, was also expressed across a wide variety of
tissues and possessed the conserved structural features of the original stanniocalcin, now referred to
as STC1. STC2 contained 3 additional conserved
cysteines, and a histidine-rich C-terminal region.
Later, STC2 was isolated from fishes (Luo et al.
2005; Shin and Sohn 2009) and birds, indicating
that it was not a mammalian innovation.
Functional activities of STC2 are not as highly
defined as those of STC1, but recent work has
demonstrated roles in the unfolded protein response
(Ito et al. 2004), ovarian development (Luo et al.
2005), growth (Gagliardi et al. 2005), pancreatic
alpha-cell function (Moore et al. 1999), and cancer
(Bouras et al. 2002; Raulic et al. 2008; Ieta et al.
2009; Tamura et al. 2009). It is clear that the functional portfolio of the stanniocalcin family has
expanded from its roots in early fish studies.
Until now, the body of stanniocalcin research has
focused on species within the vertebrates, with the
most divergent isolated orthologs belonging to the
basal Actinopterygii. Limited investigations of invertebrate stanniocalcins have been performed using fish
antisera on sea snails (Wendelaar Bonga et al. 1989)
and the horse leeches (Tanega et al. 2004).
These results, although tantalizing, leave us without
an isolated hormone that could be sequenced and
confirmed. Questions regarding the origin of
the stanniocalcins have thus remained largely
unanswered beyond the bony fishes.
Recent developments in the genomics of two
model invertebrate chordates (protochordates), the
tunicate Ciona intestinalis (Dehal et al. 2002) and
the amphioxus Branchiostoma floridae (Putnam
et al. 2008) have provided key insights into the evolution of vertebrates. Both animals have genomes
that possess simplified versions of several
multi-membered vertebrate gene families and, by
comparison, inferences regarding the ancestral chordate complement may be drawn. Utilizing genomic
resources, we isolated and cloned a single stanniocalcin ortholog from C. intestinalis and 2 from B. floridae. These orthologs possess key structural features
conserved in their vertebrate counterparts.
Phylogenetic analysis suggests that the protochordate
stanniocalcins are monophyletic and distinct from
both vertebrate STC1 and STC2. Together with
expression studies, our data support the presence
of invertebrate stanniocalcins as part of a hormone
family of ancient mediators of mineral homeostasis,
dating at least to the origin of the chordates.
Methods
Animals
Wild-type adult C. intestinalis individuals were provided by the Ascidian Stock Center (Santa Barbara,
CA, USA). Animals were kept in seawater until dissection, after which their tissues were snap frozen in
liquid nitrogen, and stored at 808C. Wild-type
B. floridae adults were purchased from Gulf
Specimen Marine Laboratory (Panacea, FL, USA).
Whole animals were snap-frozen in liquid nitrogen
and stored at 808C.
88
RNA extraction and cDNA synthesis
Total ribonucleic acid (RNA) was extracted from
individual C. intestinalis tissues and whole B. floridae
adults. Tissues were immersed in lysis buffer and
processed using the Qiagen RNeasy Mini Kit
(Qiagen Inc., Mississauga, ON, Canada) according
to the manufacturer’s instructions. Extracted RNA
was quantitated by absorbance at 260 nm, and 1
mg
samples
were
reverse-transcribed
with
Superscript III (Invitrogen Canada Inc., Burlington,
ON, Canada) as per the manufacturer’s guidelines,
using 200 units of enzyme and 500 ng oligo dT(12–
18) to amplify messenger (m) RNAs. The reaction
was incubated at 508C for 60 min. 50 and 30 rapid
amplification of cDNA ends (RACE)-ready complementary deoxyribonucleic acid (cDNA) was also prepared from C. intestinalis tissues with the
RLM-RACE kit (Applied Biosystems/Ambion,
Austin, TX, USA) using 1 mg total RNA and the
manufacturer’s standard protocol.
PCR amplification and cloning
From genomic resources at the Joint Genome
Institute (JGI) for C. intestinalis (http://genome
.jgi-psf.org/Cioin2/Cioin2.home.html) and B. floridae
(http://genome.jgi-psf.org/Brafl1/Brafl1.home.html),
full or partial stanniocalcin gene models were identified and primers were designed against these
sequences. Primers for full-length, open-reading
frames were designed as follows: C. intestinatlis
stc forward 50 -ATGTACGCTACAAGACATTGTAC
CG-30 ,
reverse
50 -TTAGCTCTGTCTTGGCGAA
0
CT-3 ; B. floridae stca forward 50 -ATGACAGCAGT
GGGAAGACC-30 , reverse 50 -TTAGCCGTACATGGA
GAACC-30 , B. floridae stcb 50 -ATGCAGTTTGTAAGC
TGGATGC-30 , reverse 50 -TCACTTGCCGAGGCCCC
TCA-30 . Sequences were amplified from cDNA samples using 1 unit of Platinum Taq DNA Polymerase
High Fidelity (Invitrogen), with 2 mM magnesium
sulfate, 0.2 mM mixed deoxynucleoside triphosphates
(dNTPs), and 0.4 mM of each primer. Cycling conditions were as follows: 948C for 2 min, followed by 35
cycles of 948C for 15 s, then 55–608C for 30 s, and
688C for 2 min. Polymerase chain reaction (PCR) for
50 and 30 RACE was carried out using RLM-RACE
kit primers, and the respective forward and reverse
primers from C. intestinalis previously mentioned.
Amplicons were separated on a 1.2% agarose gel,
and bands were excised and purified using the
QIAquick Gel Extraction Kit (Qiagen) according to
the manufacturer’s instructions. Purified amplicons
were ligated to the pGEM-T vector (Promega
Corporation, Madison, WI, USA) overnight at 168C
G. J. Roch and N. M. Sherwood
and transformed by electroporation into XL-1 Blue
electrocompetent cells (Stratagene, La Jolla, CA,
USA). Cells were plated under ampicillin selection
and incubated overnight at 378C, and colonies were
picked and placed in Lysogency broth (LB) overnight
at 378C. Plasmid DNA was extracted from broth cultures using the QuickClean 5M Miniprep Kit
(Genscript USA Inc., Piscataway, NJ, USA) using
the manufacturer’s protocol. Plasmids were separated
on 1.2% agarose gels, and clones were sequenced at
the UVic CBR DNA Sequencing Facility (Victoria,
BC, Canada).
Phylogenetics
Predicted full-length peptide sequences from C. intestinalis and B. floridae were aligned with stanniocalcin
orthologs from human and zebra fish listed in the
National Center for Biotechnology Information
(NCBI) Protein database (http://www.ncbi.nlm.nih
.gov/protein). Human STC1 (NCBI accession
NP_003146.1), human STC2 (NP_003705.1), zebra
fish STC1 (NP_956833.1), and zebra fish STC2
(NP_001014827.1) were aligned with protochordate
sequences using ClustalW (Thompson et al. 2002).
Gene diagrams were designed using the JGI genomic
resources for C. intestinalis and B. floridae, as previously mentioned. Additional information on
sequences for the untranslated regions of amphioxus
stcb was procured from the B. floridae cDNA
Database (http://amphioxus.icob.sinica.edu.tw/). The
human STC gene diagram was based on information
located within NCBI’s Gene database (http://
www.ncbi.nlm.nih.gov/gene). An amino-acid identity
matrix was created using sequences lacking their
putative signal peptides in BioEdit (Hall 1999).
A phylogenetic tree based on maximum likelihood
was produced using the previously mentioned peptide
sequences as well as the following, listed with their
respective NCBI accession numbers: silver arawana
STC1 (BAB43868.1), bowfin STC1 (BAC66163.1), butterflyfish STC1 (BAD99601.1), chicken STC1
(XP_425760.2), chicken STC2 (XP_414534.2), eel
STC1
(AAB91483.1),
elephantnose
STC1
(BAD99600.1), flounder STC1 (ABI64157.1), flounder
STC2 (ACJ06521.1), frog STC1 (NP_001086522.1),
frog
STC2
(NP_001016004.1),
fugu
STC1
(NP_001072056.1), fugu STC2 (NP_001072057.1),
gar STC1 (BAC66164.1), mouse STC1 (NP_
033311.3), mouse STC2 (NP_035621.1), opossum
STC1 (XP_001373050.1), opossum STC2 (XP_
001370619.1), platypus STC1 (XP_001506746.1), and
salmon STC1 (AAB26419.1). Sequences were aligned
in ClustalW, trimmed before and after the region
89
Stanniocalcin in amphioxus and tunicate
containing conserved cysteines, and degapped. Trees
were generated in PhyML 3.0 (Guindon et al. 2009)
using the LG (Le and Gascuel amino-acid replacement
matrix) substitution model with the following options:
proportion of invariable sites–estimated, tree topology
search operations–best of NNI and SPR, and bootstrapping with 100 replicates.
Reverse transcriptase and quantitative PCR of
stanniocalcin in C. intestinalis tissues
Individual tissue cDNA samples were analyzed
for relative stc expression by standard reverse transcriptase-PCR (RT-PCR) using the following primers:
forward 50 -GTTTACGAAACAGTCACTGCGATG-30
and reverse 50 -TTGACATAGCAGCCCGACAAAC-30 .
Amplification was carried out using 2.5 units of Taq
DNA Polymerase (Invitrogen) with 2 mM magnesium chloride, 0.2 mM mixed dNTPs, and 0.4 mM
of each primer under the following conditions:
948C for 3 min, followed by 33 cycles of 948C for
30 s, then 558C for 30 s, and 728C for 45 s. Reactions
were loaded on a 1.2% agarose gel and electrophoresed, stained with ethidium bromide and imaged.
Quantitative PCR (qPCR) was carried out to
determine relative stc expression in C. intestinalis as
follows. Primers amplifying cDNAs for stanniocalcin
and the housekeeping reference, TATA-box binding
protein (TBP), were as follows: stc forward 50 -CG
ACAATCCGCTGGTATC-30 , reverse 50 -CGCTGGAA
ATCCGACTTGA-30 ; tbp forward 50 -CAGTCAGTTC
TCAAGTTATGAGCC-30 , reverse 50 -ATATACTTTG
GCACCTGTTAGAACTAC-30 . Genes were amplified
from tissue cDNA samples from three adult C. intestinalis in a Mx3000P Real-Time PCR System
(Stratagene) with 1 unit Platinum Taq DNA
Polymerase (Invitrogen), 3 mM magnesium chloride,
0.25 mM mixed dNTPs, and 0.25 mM of each primer.
Cycling conditions were as follows: 958C for 5 min,
followed by 40 cycles of 958C for 30 s, then 558C for
60 s, and 728C for 30 s. This was followed by a
dissociation melting curve control to ensure fluorescence was due to a single amplified product.
Following amplification and cycle threshold (Ct)
value determination for the transcripts in each
tissue, results were averaged and plotted as the relative ratio of target copies to reference copies.
Results
Isolation, sequencing, and structure of
stanniocalcin from tunicates and amphioxus
Initially, BLAST homology searches were performed
on the draft genome databases of both the tunicate
C. intestinalis (http://genome.jgi-psf.org/Cioin2/
Cioin2.home.html) and amphioxus B. floridae
(http://genome.jgi-psf.org/Brafl1/Brafl1.home.html)
using full-length vertebrate stanniocalcins as queries.
One tunicate and two amphioxus orthologs were
identified, and primers were designed based on
these sequences. Full-length stanniocalcin open reading frames (ORFs) from both amphioxus and tunicates were amplified by PCR from cDNA libraries
derived from at least two individuals of each species,
and the resulting amplicons were sequenced.
The predicted full-length amino-acid sequences
of the tunicate and amphioxus stanniocalcins are
aligned in Fig. 1A with human and zebra fish STC1
and STC2 homologs. Conserved cysteine residues are
highlighted in black and the predicted signal peptides
are shaded gray. As shown, 10 of the 11 invariant
cysteine residues within the vertebrate stanniocalcins
were preserved in the protochordate orthologs as well.
The 11th conserved cysteine residue in vertebrate
stanniocalcin homologs, established as the site of
disulfide linkage between stanniocalcin monomers,
was only present in amphioxus STCa. Although the
conserved N-glycosylation motif of vertebrate
stanniocalcins, boxed in Fig. 1A, was absent from
all of the protochordate sequences, a potential
N-glycosylation motif at an alternate site was identified in amphioxus STCa using the N-glycosylation
prediction algorithm from NetNGlyc 1.0 (http://
www.cbs.dtu.dk/services/NetNGlyc) (Fig. 1A). The
C-terminal histidine-rich motif present in vertebrate
STC2 homologs (shaded gray in Fig. 1A) was not
present in either tunicate or amphioxus stanniocalcins. The putative gene structures of human STC1
and the protochordate stanniocalcins are displayed
in Fig. 1B. As shown, both the tunicate and
amphioxus orthologs have the conserved four-exon
structure found in the vertebrates, indicating that it
is preserved throughout the chordates.
Phylogenetic analysis of chordate stanniocalcins
The amino-acid identity of the tunicate and
amphioxus mature stanniocalcins is compared with
human and zebra fish homologs in Table 1.
Sequences were aligned without signal peptides, and
a matrix identifying identical amino acids between
each pairwise comparison was generated. As shown,
the human and zebra fish STC2 orthologs displayed
the highest homology, at 60% identity. The next
highest homology was found between the human
and zebra fish STC1 orthologs, at 50% identity.
The identity of the human and zebra fish STC1
and STC2 paralogs was significantly lower in any
combination (21–26%). This was comparable with
90
G. J. Roch and N. M. Sherwood
Fig. 1 (A) ClustalW alignment of putative stanniocalcin peptides from tunicates (C. intestinalis), amphioxus (B. floridae), as well
as representative sequences for STC1 and STC2 from humans and zebra fish. The predicted signal peptides are highlighted at
the N-terminus in gray, and conserved cysteine residues are highlighted in black. The C-terminal cysteine residue is also indicated
as the site of dimerization. Predicted N-glycosylation sites are indicated and boxed, and the histidine-rich C-terminal motif in STC2
is highlighted in gray. (B) Gene diagrams for human and protochordate stanniocalcins. Exons are represented as boxes and introns
as lines, with the shaded regions indicating the ORFs. Numbers correspond to the length of each exon in nucleotides.
the identity of the two amphioxus paralogs
at 24%. Tunicate stanniocalcin displayed low homology with both the vertebrate (17–23% identity)
and the amphioxus (19–24% identity) sequences.
Amphioxus STCa displayed slightly higher
identity with the vertebrate homologs than with
91
Stanniocalcin in amphioxus and tunicate
Table 1 Stanniocalcin amino-acid identity matrix
Sequence
Human STC1
Human
STC1 (%)
–
Zebra fish
STC1 (%)
–
Human
STC2 (%)
–
Zebra fish
STC2 (%)
–
Tunicate
STC (%)
–
Amphioxus
STCa (%)
–
Zebra fish STC1
50
–
–
–
–
–
Human STC2
24
21
–
–
–
–
Zebra fish STC2
26
22
60
–
–
–
Tunicate STC
23
21
17
17
–
–
Amphioxus STCa
23
23
17
18
24
–
Amphioxus STCb
18
17
14
15
19
24
Predicted peptide sequences from tunicates (C. intestinalis), amphioxus (B. floridae), humans, and zebra fish were aligned without signal peptides
using ClustalW. An identity matrix was computed with BioEdit Sequence Alignment Editor. Amino-acid identity percentages between all pairwise
comparisons are listed.
Fig. 2 Phylogenetic tree of stanniocalcin sequences. Protochordate protein sequences, along with vertebrate sequences from both
STC1 and STC2 isoforms were aligned using the conserved region between cysteine residues. Maximum likelihood (PhyML) was used
to construct the resulting phylogeny. Bootstrap analysis (100 bootstraps) was employed and the frequency of support is indicated at
each node.
STCb: 17–23% compared with 14–18%. Amphioxus
STCa was also more homologous to tunicate STC;
24% identity versus 19% for STCb.
Stanniocalcin protein sequences from representative
species were aligned between the conserved cysteine
residues and their phylogeny established using maximum likelihood (PhyML) with bootstrap support, as
shown in Fig. 2. The vertebrate STC1 and STC2 proteins were grouped together in distinct clades. Further
nodes established the monophyletic distinctions
between the tetrapod and fish groups of STC2; however, this resolution was not repeated within the STC1
grouping. STC1 sequences within the fishes formed
distinct groups corresponding to their phylogeny,
with the non-teleosts (gar and bowfin) separate from
the teleosts. Within the teleosts the bony-tongued
Osteoglossiformes, including elephantnose, freshwater
butterflyfish, and silver arawana, formed a clade separate from the others. It should be noted that the
Osteoglossiformes also lack the terminal cysteine
92
Fig. 3 RT-PCR analysis of C. intestinalis stanniocalcin was
performed for the indicated tissues using -tubulin as a control.
These results were refined using qPCR, which shows the ratio
of expressed stanniocalcin cDNA against a housekeeping
control, TBP cDNA.
residue that all other vertebrate stanniocalcins possess,
critical to dimerization. The amphioxus and tunicate
stanniocalcins were grouped together, separated from
the vertebrate hormones. As seen in Fig. 2, the protochordate STCs were all highly divergent from vertebrate STC1s and STC2s, as well as from each other,
with amphioxus STCb being significantly more divergent than the others.
Expression of stanniocalcin in tunicates
Expression localization studies on C. intestinalis stanniocalcin cDNA were carried out for a variety of tissues.
As seen in Fig. 3, initial RT-PCR experiments indicated
a high level of expression in a variety of tissues, most
prominently the neural complex, heart, and gonad.
When qPCR experiments were normalized against a
housekeeping gene (Fig. 3), however, expression in
the heart was predominant. Relative expression in the
branchial basket, neural complex, and endostyle was
also measurably higher than in other tissues tested.
Discussion
The sequences characterized in this study represent
the first stanniocalcin gene orthologs cloned and
G. J. Roch and N. M. Sherwood
characterized
from non-vertebrates.
Previous
research has identified stanniocalcins from several
species of tetrapods and major groups of bony
fishes (Fig. 2) including two species from non-teleost
fish, the gar, and bowfin (Amemiya and Youson
2004). The only prior evidence of non-vertebrate
stanniocalcin homologs came from experiments in
which immunoreactivity was detected in sea-snail
(Wendelaar Bonga et al. 1989) and leech tissues
(Tanega et al. 2004) using fish antisera. In leeches,
a salmon STC1 antibody and in situ hybridization
with an RNA probe specific for salmon Stc1 resulted
in staining within adipose and skin cells. Here, we
present a single gene orthologous to vertebrate stanniocalcins from the tunicate C. intestinalis, and two
orthologous genes from the amphioxus B. floridae.
These invertebrate homologs possess conserved
sequence elements found in their vertebrate counterparts, and combined with the pattern of stc gene
expression displayed in tunicate they confirm a role
for stanniocalcin dating back to the early chordates.
The genes and predicted proteins of tunicate and
amphioxus stanniocalcins share several key structural
features found in all previously described homologs.
The invertebrate genes have the conserved four-exon
structure found in the vertebrates (Fig. 1B), although
the invertebrate hormones are not as long as
their vertebrate counterparts. Full-length tunicate
pro-STC is predicted to be 237 amino acids and
the amphioxus paralogs are between 207 and 210
amino acids, whereas human pro-STC1 and
pro-STC2 are 247 and 302 amino acids, respectively.
This is due to the C-terminal expansions in the vertebrate stanniocalcins, shown in Fig. 1A, downstream
of the terminal conserved cysteine residue. The functional significance of this is unclear, although a
C-terminal fragment of eel stanniocalcin representing
this region has been demonstrated to repress calcium
influx on its own (Fenwick and Verbost 1993;
Verbost and Fenwick 1995).
Apart from this C-terminal extension, the invertebrate hormones share the salient characteristics
found in vertebrate stanniocalcins, including a
signal peptide followed by the 10 invariant cysteines
distributed throughout the mature hormone
(Fig. 1A). These conserved cysteine residues are a
hallmark of all STC homologs isolated so far. They
are critical to a series of specific intrachain disulfide
bridges, which have been confirmed by studies on
salmon STC1 (Hulova and Kawauchi 1999) and
human STC1 (Trindade et al. 2009). Vertebrate
STC2 orthologs have an additional three conserved
cysteine residues that the STC1 group and the invertebrates lack; the function of these has been
Stanniocalcin in amphioxus and tunicate
suggested to involve additional intermolecular disulfide bonds (Luo et al. 2005). Another feature unique
to the STC2s is the C-terminal histidine-rich motif,
displayed in Fig. 1A. None of the invertebrate
sequences have these STC2-specific attributes.
The vast majority of vertebrate stanniocalcins also
possess a C-terminal 11th conserved cysteine residue,
with the exception of homologs from teleost fishes
within the order Osteoglossiformes (Amemiya et al.
2002; Amemiya et al. 2006). This cysteine is distinct
from the rest in its function, as it has been demonstrated to be the sole determinant of STC1 dimerization in salmon and humans (Hulova and Kawauchi
1999; Trindade et al. 2009). The dimerization of
native and recombinant stanniocalcins has been
established in several species using reducing and
non-reducing SDS–PAGE analysis, including salmon
STC1 (Wagner et al. 1986; Flik et al. 1990), human
STC1 and STC2 (Moore et al. 1999; Luo et al. 2005),
hamster STC2 (Moore et al. 1999), and turbot STC1
(Shin et al. 2006), among others. The structural analysis of cysteine residues within salmon and human
STC1 confirmed the C-terminal cysteine as the site
of dimerization (Hulova and Kawauchi 1999;
Trindade et al. 2009). Also, the lack of the terminal
cysteine residue within silver arawana revealed a
monomeric form of the hormone under reducing
and non-reducing experiments (Amemiya et al.
2002; Amemiya et al. 2006). Tunicate STC and
amphioxus STCb do not have this dimerizationcritical cysteine, indicating that they are present as
monomers (Fig. 1A). Amphioxus STCa, however,
contains a cysteine at the same relative position as
the C-terminal residue in the vertebrate homologs,
suggesting it may be dimerized. The consequences of
this dimerization are not understood.
The last highly conserved feature present within
vertebrate stanniocalcins is the N-glycosylation
motif present between the third and fourth conserved cysteines in both STC1 (Asn-Ser-Thr) and
STC2 (Asn-Asn-Ser) (Fig. 1A). This site is postulated
to play a critical role in the glycosylation of the
mature hormone, which is the native state of vertebrate stanniocalcins isolated to date, including eel
STC1 (Butkus et al. 1987), trout STC1 (Lafeber
et al. 1988a; Lafeber et al. 1988b), bowfin STC1
(Marra et al. 1992), human STC1 and STC2
(Wagner et al. 1995; Moore et al. 1999; Luo et al.
2005), arawana STC1 (Marra et al. 1995; Amemiya
et al. 2002), and turbot STC1 (Shin et al. 2006).
None of the invertebrate stanniocalcin sequences
have this conserved motif at the same site, although
a N-glycosylation motif at a different site is present
93
in amphioxus STCa (Asn-Asn-Thr) (Fig. 1A), indicating it may be a glycoprotein in its native state.
Given the number of conserved features between
the invertebrate and vertebrate stanniocalcins, it is
interesting they retain such low sequence identity.
As seen in Table 1, the only sequences with more
than 50% homology were the human and zebra fish
STC1 and STC2 orthologs, respectively. The vertebrate paralogs shared significantly reduced identity,
between 21% and 26%. The tunicate and amphioxus
stanniocalcins were comparably distinct from both
the vertebrate hormones and each other, with an
identity of only 14% between amphioxus STCb and
human STC2. Evidently, outside the conserved
sequence features previously mentioned, the primary
sequence of stanniocalcins across the Phylum
Chordata can vary to a large extent. Given the
cross-activating potential of STC-immunoreactive
extracts from animals as distinct as salmon and
leeches (Tanega et al. 2004), the data indicate that
the conserved disulfide bridging and possibly glycosylation are the major determinants in the binding
and functioning of the hormones. Elucidation of the
stanniocalcin binding to a receptor or ion channel
will prove essential to this determination.
RT-PCR and qPCR experiments were carried out on
tunicate stc mRNA to determine its tissue localization.
As shown in Fig. 3, after normalization against a
housekeeping gene, the qPCR displayed predominant
expression of stc in the C. intestinalis heart, with lower
expression in the neural complex, branchial basket,
and endostyle. The branchial basket houses and
includes the gill slits of tunicates, and the endostyle
is a mucus secreting gland found in protochordates
and jawless fishes that shares homology with the vertebrate thyroid gland (Ogasawara et al. 1999;
Ogasawara 2000). Tunicates do not have an organ
homologous to the corpuscles of Stannius, an outgrowth from the kidney and originally thought to be
the region of exclusive stanniocalcin expression in
fishes; thus, no direct comparisons can be made.
Following the isolation of stanniocalcin in humans
and other mammals, however, many other regions of
stanniocalcin expression have been found in both tetrapods and fishes. Stanniocalcins in tetrapods are
reported to be expressed in the nervous system
(Zhang et al. 1998; McCudden et al. 2001; Shin et al.
2006; Tseng et al. 2009) and the thyroid (Chang et al.
1995; Moore et al. 1999). Also, human STC1 and STC2
have been identified in human heart among several
other tissues (Chang et al. 1995; DiMattia et al. 1998;
Ishibashi et al. 1998; Moore et al. 1999). Stanniocalcins
have also been identified in the heart of gar and bowfin
(Amemiya and Youson 2004), flounder (Hang and
94
G. J. Roch and N. M. Sherwood
Table 2 Hormones related to calcium homeostatsis in amphioxus (B. floridae), tunicates (C. intestinalis) and humans.
Hormones
Amphioxus
H
Tunicate
R
H
Human
R
H
STC
1
1
Calcitonin
4
5,6,7
Calcitonin Gene-related
peptide (CGRP)
Parathyroid
hormone (PTH)
R
2,3
5
4
6,7,8
9
10,11
PTH-related protein
(PTHrP)
Calcitriol
(Vitamin D metabolite)
References: (1) Present study, (2) http://www.ncbi.nlm.nih.gov/protein/NP_003146.1, (3)
http://www.ncbi.nlm.nih.gov/protein/NP_003705.1, (4) Holland et al. 2009, (5) Sekiguchi et al.
2009, (6) Sherwood et al. 2006, (7) Cardoso et al. 2006, (8) Kamesh et al. 2008, (9) Schubert et al.
2008, (10) Dehal et al. 2002, and (11) Yagi et al. 2003.
The identification of hormones (H) or their cognate receptors (R) is indicated by a black box.
Balment 2005; Shin and Sohn 2009) and turbot (Shin
et al. 2006).
With predominant expression in the heart, the question of tunicate stanniocalcin function is intriguing.
Most of the studies on hormonal functions thus far
have focused on its canonical role as a hypocalcemic
regulator in fish gills or the pleiotropic roles it plays in
phosphate and calcium homeostasis in mammals (for
reviews see Yoshiko and Aubin 2004; Wagner and
DiMattia 2006). A few recent studies have examined
heart expression. Jiang et al. (2000) found immunoreactivity of STC1 in mouse heart cells at embryonic
Day 8.5, and a few days later predominantly in the
myocardium of the atrium and ventricle, suggesting
an early role in heart development. Embryonic expression of STC2 was also detected in developing chick
hearts (Mittapalli et al. 2006). A study by
Sheikh-Hamad et al. (2003) focused on the role of
STC1 in patients with heart failure. They found upregulation of the hormone in the failing heart, and also
demonstrated that addition of STC1 to cultured cardiomyocytes resulted in inhibition of calcium currents
through L-channels and slowed the cells’ contractions.
Another recent study found high-affinity binding sites
for STC1 on red blood cells (James et al. 2005). Taken
together and considering the important role of calcium
in heart contractions, it is possible that tunicate stanniocalcin plays a role in the developing heart and its
regulation thereafter.
The identification of stanniocalcin homologs in
tunicates and amphioxus provides additional insight
into the evolution of the hormone family in addition
to the calcium-regulatory system at large. The gene
structure of the vertebrate and invertebrate hormones has remained largely unchanged; however,
significant mutation has altered all but the most conserved residues of the amino-acid sequence. It is now
clear that the duplication producing STC1 and STC2
is an innovation that occurred early in vertebrate
evolution with the tunicate and amphioxus hormones phylogenetically distinct from both, as seen
in Fig. 3. Whether or not this divergence is present
within the cyclostomes remains in question, as an
orthologous sequence could not be retrieved from
the lamprey (Petromyzon marinus) draft genome to
date. The two amphioxus paralogs appear to be a
lineage-specific duplication as well, although this
cannot be certain as they are located on separate
scaffolds, reducing the likelihood of tandem duplication. Structurally, the invertebrate stanniocalcins
do share more conserved features with the STC1
orthologs to date.
Stanniocalcin in amphioxus and tunicate
95
In the context of calcium regulation as a whole,
we now have a more complete picture stretching to
the invertebrate chordates provided by the tunicate
and amphioxus genomes. As seen in Table 2, calcitonin genes have been identified in both protochordates, first in B. floridae by ourselves (Holland et al.
2008) and more recently in C. intestinalis (Sekiguchi
et al. 2009). Calcitonin gene-related peptide has not
been detected. Orthologs of the parathyroid hormone receptor have also been identified in tunicates
and amphioxus, but not their cognate ligands. A
nuclear receptor belonging to the same family as
the calcitriol (vitamin D) receptor has been identified
in tunicates but not in amphioxus. Combined, we
now have a patchwork of hormones that regulate
calcium homeostasis to investigate the origin of calcium regulation in invertebrates.
Cardoso JCR, Pinto VC, Vieira FA, Clark MS, Power DM.
2006. Evolution of secretin family GPCR members in the
metazoan. BMC Evol Biol 6:108.
Acknowledgments
Fenwick JC, Verbost P. 1993. A C-terminal fragment of the
hormone stanniocalcin is bioactive in eels. Gen Comp
Endocrinol 91:337–43.
We thank Ryan Liebscher for his help with an early
phase of the research. Also, we thank the National
Science Foundation-USA for funds to attend the
conference, and the Division of Comparative
Endocrinology of the Society for Integrative and
Comparative Biology for partial payment of the
registration fee.
Funding
Natural Sciences and Engineering Research Council
of Canada.
References
Amemiya Y, Irwin DM, Youson JH. 2006. Cloning of stanniocalcin (STC) cDNAs of divergent teleost species:
Monomeric STC supports monophyly of the ancient teleosts, the osteoglossomorphs. Gen Comp Endocrinol
149:100–7.
Amemiya Y, Marra LE, Reyhani N, Youson JH. 2002.
Stanniocalcin from an ancient teleost: a monomeric form
of the hormone and a possible extracorpuscular distribution. Mol Cell Endocrinol 188:141–50.
Amemiya Y, Youson JH. 2004. Primary structure of stanniocalcin in two basal Actinopterygii. Gen Comp Endocrinol
135:250–7.
Bouras T, Southey MC, Chang AC, Reddel RR, Willhite D,
Glynne R, Henderson MA, Armes JE, Venter DJ. 2002.
Stanniocalcin 2 is an estrogen-responsive gene coexpressed
with the estrogen receptor in human breast cancer. Cancer
Res 62:1289–95.
Butkus A, Roche PJ, Fernley RT, Haralambidis J,
Penschow JD, Ryan GB, Trahair JF, Tregear GW,
Coghlan JP. 1987. Purification and cloning of a corpuscles
of Stannius protein from Anguilla australis. Mol Cell
Endocrinol 54:123–33.
Chang AC, Dunham MA, Jeffrey KJ, Reddel RR. 1996.
Molecular cloning and characterization of mouse stanniocalcin cDNA. Mol Cell Endocrinol 124:185–7.
Chang AC, Janosi J, Hulsbeek M, de Jong D, Jeffrey KJ,
Noble JR, Reddel RR. 1995. A novel human cDNA highly
homologous to the fish hormone stanniocalcin. Mol Cell
Endocrinol 112:241–7.
Chang AC, Reddel RR. 1998. Identification of a second stanniocalcin cDNA in mouse and human: stanniocalcin 2.
Mol Cell Endocrinol 141:95–9.
Dehal P, et al. 2002. The draft genome of Ciona intestinalis:
insights into chordate and vertebrate origins. Science
298:2157–67.
DiMattia GE, Varghese R, Wagner GF. 1998. Molecular cloning and characterization of stanniocalcin-related protein.
Mol Cell Endocrinol 146:137–40.
Flik G, Labedz T, Neelissen JA, Hanssen RG, Wendelaar
Bonga SE, Pang PK. 1990. Rainbow trout corpuscles of
Stannius: stanniocalcin synthesis in vitro. Am J Physiol
258:R1157–64.
Gagliardi AD, Kuo EY, Raulic S, Wagner GF, DiMattia GE.
2005. Human stanniocalcin-2 exhibits potent growth-suppressive properties in transgenic mice independently of
growth hormone and IGFs. Am J Physiol Endocrinol
Metab 288:E92–105.
Guindon S, Delsuc F, Dufayard JF, Gascuel O. 2009.
Estimating maximum likelihood phylogenies with PhyML.
Methods Mol Biol 537:113–37.
Haddad M, Roder S, Olsen HS, Wagner GF. 1996.
Immunocytochemical localization of stanniocalcin cells in
the rat kidney. Endocrinology 137:2113–7.
Hall TA. 1999. BioEdit: a user-friendly biological
sequence alignment editor and analysis program for
Windows 95/98/NT. Nucleic Acids Symp Ser 41:95–8.
Hang X, Balment RJ. 2005. Stanniocalcin in the euryhaline
flounder (Platichthys flesus): primary structure, tissue distribution, and response to altered salinity. Gen Comp
Endocrinol 144:188–95.
Holland LZ, et al. 2008. The amphioxus genome illuminates
vertebrate origins and cephalochordate biology. Genome
Res 18:1100–11.
Hulova I, Kawauchi H. 1999. Assignment of disulfide linkages
in chum salmon stanniocalcin. Biochem Biophys Res
Commun 257:295–9.
Ieta K, et al. 2009. Clinicopathological significance of stanniocalcin 2 gene expression in colorectal cancer. Int J Cancer
125:926–31.
Ishibashi K, Imai M. 2002. Prospect of a stanniocalcin endocrine/paracrine system in mammals. Am J Physiol Renal
Physiol 282:F367–75.
96
G. J. Roch and N. M. Sherwood
Ishibashi K, Miyamoto K, Taketani Y, Morita K, Takeda E,
Sasaki S, Imai M. 1998. Molecular cloning of a second
human stanniocalcin homologue (STC2). Biochem Biophys
Res Commun 250:252–8.
Paciga M, McCudden CR, Londos C, DiMattia GE,
Wagner GF. 2003. Targeting of big stanniocalcin and its
receptor to lipid storage droplets of ovarian steroidogenic
cells. J Biol Chem 278:49549–54.
Ito D, Walker JR, Thompson CS, Moroz I, Lin W,
Veselits ML, Hakim AM, Fienberg AA, Thinakaran G.
2004. Characterization of stanniocalcin 2, a novel target
of the mammalian unfolded protein response with cytoprotective properties. Mol Cell Biol 24:9456–69.
Putnam NH, et al. 2008. The amphioxus genome and the
evolution of the chordate karyotype. Nature 453:1064–71.
James K, Seitelbach M, McCudden CR, Wagner GF. 2005.
Evidence for stanniocalcin binding activity in mammalian
blood and glomerular filtrate. Kidney Int 67:477–82.
Jiang WQ, Chang AC, Satoh M, Furuichi Y, Tam PP,
Reddel RR. 2000. The distribution of stanniocalcin 1 protein in fetal mouse tissues suggests a role in bone and
muscle development. J Endocrinol 165:457–66.
Kamesh N, Aradhyam GK, Manoj N. 2008. The repertoire of
G protein-coupled receptors in the sea squirt Ciona intestinalis. BMC Evol Biol 8:129.
Lafeber FP, Flik G, Wendelaar Bonga SE, Perry SF. 1988a.
Hypocalcin from Stannius corpuscles inhibits gill calcium
uptake in trout. Am J Physiol 254:R891–6.
Lafeber FP, Hanssen RG, Choy YM, Flik G, HerrmannErlee MP, Pang PK, Bonga SE. 1988b. Identification of
hypocalcin (teleocalcin) isolated from trout Stannius corpuscles. Gen Comp Endocrinol 69:19–30.
Luo CW, Pisarska MD, Hsueh AJ. 2005. Identification of a
stanniocalcin paralog, stanniocalcin-2, in fish and the paracrine actions of stanniocalcin-2 in the mammalian ovary.
Endocrinology 146:469–76.
Raulic S, Ramos-Valdes Y, DiMattia GE. 2008. Stanniocalcin
2 expression is regulated by hormone signalling and
negatively affects breast cancer cell viability in vitro.
J Endocrinol 197:517–29.
Schubert M, Brunet F., Paris M, Bertrand S, Benoit G,
Laudet V. 2008. Nuclear hormone receptor signaling in
amphioxus. Dev Genes Evol 218:651–65.
Sekiguchi T, Suzuki N, Fujiwara N, Aoyama M, Kawada T,
Sugase K, Murata Y, Sasayama Y, Ogasawara M, Satake H.
2009. Calcitonin in a protochordate, Ciona intestinalis–the
prototype of the vertebrate calcitonin/calcitonin generelated peptide superfamily. FEBS J 276:4437–47.
Sheikh-Hamad D, et al. 2003. Stanniocalcin-1 is a naturally
occurring L-channel inhibitor in cardiomyocytes: relevance
to human heart failure. Am J Physiol Heart Circ Physiol
285:H442–8.
Sherwood NM, Tello JA, Roch GJ. 2006. Neuroendocrinology
of protochordates: insights from Ciona genomics. Comp
Biochem Physiol Part A 144:254–71.
Shin J, Oh D, Sohn YC. 2006. Molecular characterization and
expression analysis of stanniocalcin-1 in turbot
(Scophthalmus maximus). Gen Comp Endocrinol
147:214–21.
Marra LE, Groff KE, Youson JH. 1995. The corpuscles of
Stannius in arawana (Osteoglossum bicirrhosum), an
ancient teleost. Tissue Cell 27:425–37.
Shin J, Sohn YC. 2009. cDNA cloning of Japanese flounder
stanniocalcin 2 and its mRNA expression in a variety of
tissues. Comp Biochem Physiol A Mol Integr Physiol
153:24–9.
Marra LE, Youson JH, Butler DG, Friesen HG, Wagner GF.
1992. Stanniocalcin-like immunoreactivity in the corpuscles
of Stannius of the bowfin Amia calva L. Cell and Tissue
Research 267:283–90.
Sundell K, Bjornsson BT, Itoh H, Kawauchi H. 1992. Chum
salmon (Oncorhynchus keta) stanniocalcin inhibits in vitro
intestinal calcium uptake in Atlantic cod (Gadus morhua).
J Comp Physiol B 162:489–95.
McCudden CR, Kogon MR, DiMattia GE, Wagner GF. 2001.
Novel expression of the stanniocalcin gene in fish.
J Endocrinol 171:33–44.
Tamura K, et al. 2009. Stanniocalcin 2 overexpression in castration-resistant prostate cancer and aggressive prostate
cancer. Cancer Sci 100:914–9.
Mittapalli VR, Prols F, Huang R, Christ B, Scaal M. 2006.
Avian stanniocalcin-2 is expressed in developing striated
muscle and joints. Anat Embryol 211:519–23.
Tanega C, Radman DP, Flowers B, Sterba T, Wagner GF.
2004. Evidence for stanniocalcin and a related receptor in
annelids. Peptides 25:1671–9.
Moore EE, et al. 1999. Stanniocalcin 2: characterization of the
protein and its localization to human pancreatic alpha cells.
Horm Metab Res 31:406–14.
Thompson JD, Gibson TJ, Higgins DG. 2002. Multiple
sequence alignment using ClustalW and ClustalX. Curr
Protoc Bioinformatics Chapter 2:Unit 2.3.
Ogasawara M. 2000. Overlapping expression of amphioxus
homologs of the thyroid transcription factor-1 gene and
thyroid peroxidase gene in the endostyle: insight into evolution of the thyroid gland. Dev Genes Evol 210:231–42.
Trindade DM, Silva JC, Navarro MS, Torriani IC, Kobarg J.
2009. Low-resolution structural studies of human stanniocalcin-1. BMC Struct Biol 9:57.
Ogasawara M, Di Lauro R, Satoh N. 1999. Ascidian homologs
of mammalian thyroid peroxidase genes are expressed in
the thyroid-equivalent region of the endostyle. J Exp Zool
285:158–69.
Olsen HS, Cepeda MA, Zhang QQ, Rosen CA, Vozzolo BL.
1996. Human stanniocalcin: a possible hormonal regulator
of mineral metabolism. Proc Natl Acad Sci USA 93:1792–6.
Tseng DY, Chou MY, Tseng YC, Hsiao CD, Huang CJ,
Kaneko T, Hwang PP. 2009. Effects of stanniocalcin 1 on
calcium uptake in zebrafish (Danio rerio) embryo. Am J
Physiol Regul Integr Comp Physiol 296:R549–57.
Verbost PM, Fenwick JC. 1995. N-terminal and C-terminal fragments of the hormone stanniocalcin show differential effects in eels. Gen Comp Endocrinol
98:185–92.
Stanniocalcin in amphioxus and tunicate
Wagner GF, Dimattia GE. 2006. The stanniocalcin family of
proteins. J Exp Zool A Comp Exp Biol 305:769–80.
Wagner GF, Dimattia GE, Davie JR, Copp DH, Friesen HG.
1992. Molecular cloning and cDNA sequence analysis of
coho salmon stanniocalcin. Mol Cell Endocrinol 90:7–15.
97
system of neuroendocrine peptidergic neurons in the
pond snail Lymnaea stagnalis, with antisera to the teleostean hormone hypocalcin and mammalian parathyroid hormone. Gen Comp Endocrinol 75:29–38.
Wagner GF, Fenwick JC, Park CM, Milliken C, Copp DH,
Friesen HG. 1988. Comparative biochemistry and physiology of teleocalcin from sockeye and coho salmon. Gen
Comp Endocrinol 72:237–46.
Yagi K, Satou Y, Mazet F, Shimeld SM, Degnan B, Rokhsar D,
Levine M, Kohara Y, Satoh N. 2003. A genomewide survey
of developmentally relevant genes in Ciona intestinalis III.
Genes for Fox, ETS, nuclear receptors and NFkB. Dev
Genes Evol 213:235–44.
Wagner GF, Guiraudon CC, Milliken C, Copp DH. 1995.
Immunological and biological evidence for a stanniocalcin-like hormone in human kidney. Proc Natl Acad Sci
USA 92:1871–5.
Yamashita K, Koide Y, Itoh H, Kawada N, Kawauchi H. 1995.
The complete amino acid sequence of chum salmon stanniocalcin, a calcium-regulating hormone in teleosts. Mol
Cell Endocrinol 112:159–67.
Wagner GF, Hampong M, Park CM, Copp DH. 1986.
Purification, characterization, and bioassay of teleocalcin,
a glycoprotein from salmon corpuscles of Stannius. Gen
Comp Endocrinol 63:481–91.
Yoshiko Y, Aubin JE. 2004. Stanniocalcin 1 as a pleiotropic
factor in mammals. Peptides 25:1663–9.
Wendelaar Bonga SE, Lafeber FP, Flik G, Kaneko T, Pang PK.
1989. Immunocytochemical demonstration of a novel
Zhang KZ, Westberg JA, Paetau A, von Boguslawsky K,
Lindsberg P, Erlander M, Guo H, Su J, Olsen HS,
Andersson LC. 1998. High expression of stanniocalcin in
differentiated brain neurons. Am J Pathol 153:439–45.