ral
ssBioMed Cent
BMC Biochemistry
Open Acce
Research article
Comparison of the receptor FGFRL1 from sea urchins and humans
illustrates evolution of a zinc binding motif in the intracellular
domain
Lei Zhuang
1
, Andrei V Karotki
2
, Philip Bruecker
3
and Beat Trueb*
1,4
Address:
1
Department of Clinical Research, University of Bern, 3010 Bern, Switzerland,
2
Department of Biochemistry, University of Zürich, 8057
Zürich, Switzerland,
3
Living Elements Ltd., Vancouver BC V6J 1M7, Canada and
4
Department of Rheumatology, University Hospital, 3010 Bern,
Switzerland
Email: Lei Zhuang - lei.zhuang@dkf.unibe.ch; Andrei V Karotki - karotki@bioc.uzh.ch; Philip Bruecker - livingelements@telus.net;
Beat Trueb* - beat.trueb@dkf.unibe.ch
* Corresponding author
Abstract
Background: FGFRL1, the gene for the fifth member of the fibroblast growth factor receptor
(FGFR) family, is found in all vertebrates from fish to man and in the cephalochordate amphioxus.
Since it does not occur in more distantly related invertebrates such as insects and nematodes, we
have speculated that FGFRL1 might have evolved just before branching of the vertebrate lineage
from the other invertebrates (Beyeler and Trueb, 2006).
Results: We identified the gene for FGFRL1 also in the sea urchin Strongylocentrotus purpuratus
and cloned its mRNA. The deduced amino acid sequence shares 62% sequence similarity with the
human protein and shows conservation of all disulfides and N-linked carbohydrate attachment
sites. Similar to the human protein, the S. purpuratus protein contains a histidine-rich motif at the
C-terminus, but this motif is much shorter than the human counterpart. To analyze the function of
the novel motif, recombinant fusion proteins were prepared in a bacterial expression system. The
human fusion protein bound to nickel and zinc affinity columns, whereas the sea urchin protein
barely interacted with such columns. Direct determination of metal ions by atomic absorption
revealed 2.6 mole zinc/mole protein for human FGFRL1 and 1.7 mole zinc/mole protein for sea
urchin FGFRL1.
Conclusion: The FGFRL1 gene has evolved much earlier than previously assumed. A comparison
of the intracellular domain between sea urchin and human FGFRL1 provides interesting insights
into the shaping of a novel zinc binding domain.
Background
FGFRL1 is the fifth member of the fibroblast growth factor
receptor (FGFR) family. It was originally discovered in a
muscles. Furthermore, all tissues examined so far contain
low, basal levels of FGFRL1 [1-4].
Published: 18 December 2009
BMC Biochemistry 2009, 10:33 doi:10.1186/1471-2091-10-33
Received: 30 July 2009
Accepted: 18 December 2009
This article is available from: http://www.biomedcentral.com/1471-2091/10/33
2009 Zhuang et al; licensee BioMed Central Ltd.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0),
which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.Page 1 of 10
(page number not for citation purposes)
cDNA library prepared from human cartilage [1], but it is
also expressed at relatively high levels in bone and some
The classical FGFRs (FGFR1-FGFR4) are cell surface pro-
teins with a single transmembrane domain, three extracel-

BMC Biochemistry 2009, 10:33 http://www.biomedcentral.com/1471-2091/10/33
lular Ig-like loops and an intracellular protein tyrosine
kinase domain [5-7]. The first Ig-loop is separated from
the second by a stretch of negatively charged amino acids
that are sometimes referred to as "acidic box". The classi-
cal receptors are widely expressed in mammalian tissues
and control a diversity of biological functions, including
proliferation, migration and differentiation of many cell
types. Germline mutations in FGFR genes are able to cause
a number of skeletal disorders such as craniosynostosis
syndromes and chondrodysplasias [8]. Somatic muta-
tions in FGFRs can lead to unrestricted cellular growth and
cancer as observed in bladder carcinomas, multiple mye-
lomas and chronic myeloproliferative diseases [9].
The function of the fifth FGFR is not yet understood in
detail. Similar to the classical FGFRs, it contains three
extracellular Ig-like loops and a transmembrane domain
[1-3]. However, in contrast to the other FGFRs, it lacks the
intracellular protein tyrosine kinase domain that would
be required for signal transduction by trans-phosphoryla-
tion. Instead, it contains an intracellular domain with a
peculiar histidine-rich motif that does not share much
homology with any known protein.
Recombinant FGFRL1 interacts with heparin and FGF2 in
a manner analogous to the classical FGFRs [10]. When
overexpressed in MG63 osteosarcoma cells, it has a nega-
tive effect on cell proliferation [10]. In a luciferase reporter
gene experiment, it is capable of inhibiting the activity of
the FGF inducible responsive promoter element FIRE
[11]. Furthermore, its synthesis is significantly up-regu-
lated during differentiation of myoblasts into myofibers
[12]. We have therefore concluded that FGFRL1 might
function as a decoy receptor that binds FGF ligands and
sequesters them from the other FGFRs. In this way, it
might inhibit cell proliferation and promote cell differen-
tiation.
More information about the function of the novel recep-
tor can be obtained from animal experiments. When the
synthesis of FGFRL1 was down-regulated with mor-
pholino constructs in a zebrafish model [13], the animals
failed to properly form the pharyngeal arches. It is there-
fore likely that FGFRL1 is involved in the development of
the gill cartilage. Recently we generated mice with a tar-
geted disruption of the FGFRL1 gene [12]. These knock-
out mice develop normally to term, but die immediately
after birth due to respiratory distress. The respiratory
problems are explained by a severely reduced diaphragm
muscle that is not strong enough to inflate the lungs after
birth. Another research group that has generated similar
FGFRL1 deficient mice reported on alterations in the skel-
eton and the heart, in addition to the malformed dia-
tification of the first human mutation in a patient who
suffers from Antley Bixler Syndrome [11]. This patient dis-
plays a frameshift mutation in the last exon of the FGFRL1
gene and shows craniosynostosis, radio-ulnar synostosis
and genital abnormalities. As demonstrated by cell culture
experiments, the mutant protein stays for a prolonged
period of time at the plasma membrane, where it interacts
with FGF ligands, while the wild-type protein is rapidly
removed from the plasma membrane and sorted to lyso-
somes [11]. Taken together, these studies suggest that
FGFRL1 controls the proper development of the muscu-
loskeletal system.
The gene for FGFRL1 is found in all vertebrates from fish
to man [1-3,10,15,16]. Teleostean fish have even two
genes, fgfrl1a and fgfrl1b, because they have undergone a
whole genome duplication [17]. Vertebrates represent one
subphylum of the chordates, together with the cephalo-
chordates and the urochordates. Recently, we were able to
identify the gene for FGFRL1 also in the cephalochordate
Branchiostoma floridae, but not in the urochordate Ciona
intestinalis [18]. We therefore concluded that it might
have evolved just before branching of the vertebrate line-
age from the other chordates.
Here we cloned a cDNA for FGFRL1 from the sea urchin
Strongylocentrotus purpuratus. The fact that sea urchins
also possess the gene for FGFRL1 demonstrates that this
gene must be much older than originally thought.
Results
An FGFRL1 related gene from Ciona intestinalis
A search through the completed genome sequence of the
sea squirt Ciona intestinalis allowed the identification of
a newly annotated gene (ID 100185880) that shows strik-
ing similarity to human FGFRL1. The corresponding
mRNA (XM_002125800.1) encodes a protein
(XP_002125836.1) of 658 amino acid residues with three
typical Ig-like domains, a transmembrane domain and an
intracellular domain. This putative protein shares 35%-
36% sequence identity (45% sequence similarity if con-
servative amino acid substitutions are included) with
FGFRL1 from humans and lancelet. The similarity is con-
fined to the three Ig-like domains, which show 45%-49%
sequence similarity with the three Ig-like domains from
vertebrates and lancelet (Table 1). The other parts of the
putative protein do not share much similarity. In particu-
lar, the N-terminal domain of the C. intestinalis protein is
roughly 100 amino acid residues longer than the verte-
brate proteins and cannot be aligned. The intracellular, C-
terminal region is also longer than the vertebrate counter-
part and shows less than 20% sequence identity with the
human protein. No motif related to the histidine-richPage 2 of 10
(page number not for citation purposes)
phragm [14]. The involvement of FGFRL1 in the
formation of the skeleton is in accordance with the iden-
sequence of vertebrate FGFRL1 can be identified at the C-
terminal end of the Ciona protein. Curiously enough, the

BMC Biochemistry 2009, 10:33 http://www.biomedcentral.com/1471-2091/10/33
Page 3 of 10
(page number not for citation purposes)
Table 1: Protein similarities
Human Lancelet Sea Urchin Mouse Chicken Frog Fish A Fish B Sea
Squirt
Human 100 72 62 96 86 83 77 77 49
Lancelet 100 61 73 73 72 69 69 47
Sea Urchin 100 63 62 64 61 61 45
Mouse 100 85 83 78 77 48
Chicken 100 91 78 79 48
Frog 100 78 80 48
Fish A 100 83 46
Fish B 100 46
Sea Squirt 100
Alignment of the FGFRL1 sequences from eight different speciesFigure 1
Alignment of the FGFRL1 sequences from eight different species. Identical residues are boxed. The putative signal
peptidase cleavage site is shown by an arrow. The three Ig-like domains are marked by brackets. The transmembrane domain
is given by a stippled box. Conserved glycosylation sites (NXT) are indicated by asterisks. The relative positions of introns in
the corresponding genes are shown by triangles. The sequence that was used for the preparation of the zinc-binding GST
fusion protein is underlined. The accession numbers are: FishA (Takifugu rubripes A) BN000669, FishB (Takifugu rubripes B)
BN000670, Chicken AJ535114, Frog (Silurana tropicalis) AJ616852, Human AJ277437, Mouse AJ293947, Lancelet (Branchios-
toma floridae) AJ888866, Sea Urchin (Strongylocentrotus purpuratus) FN252817, Sea Squirt (Ciona intestinalis)
XP_002125836. From the sea squirt sequence only the conserved domain (residues 112-456) is included.
FishA
FishB
Chicken
Frog
Human
Mouse
Lancelet
SeaUrchin
SeaSquirt
FishA
FishB
Chicken
Frog
Human
Mouse
Lancelet
SeaUrchin
SeaSquirt
FishA
FishB
Chicken
Frog
Human
Mouse
Lancelet
SeaUrchin
SeaSquirt
FishA
FishB
Chicken
Frog
Human
Mouse
Lancelet
SeaUrchin
SeaSquirt
FishA
FishB
Chicken
Frog
Human
Mouse
Lancelet
SeaUrchin
1
1
1
1
1
1
1
1
112
91
91
111
111
117
113
110
119
205
206
205
223
223
229
225
212
231
301
319
318
336
336
342
338
326
343
420
420
422
426
423
437
433
435
462
90
90
110
110
116
112
109
118
204
205
204
222
222
228
224
211
230
300
318
317
335
335
341
337
325
342
419
419
421
425
422
436
432
434
461
456
475
478
487
484
504
529
499
532
----------------------------G P P R VSARV THRQN A R L G R TMK L P C P V - E G D P P P L I MW T K D G RN I H S GW T R F R V MQHAL R I K E V E T E D A G T Y I C K A T N G F G S VN I N Y T L I V
----------------------------G P P R VSEKV THRQ S A R I GSAIK L P C P V - E G D P P P L I MW T K D G RN I H S GW I R F R ILRMG L K I K E V E A D D S G T Y I C K A T N G F G S VN I N H T L I V
--------MGL QLALL LAGIVALSDSARG P P R IADKV IHRQS V R L G R TIK L L C P V - E G D P P P L T MWM K D G R T I H S GW T R F R ILQQG L K I K E V E S E D A G T Y I C K A T N G F G S TNV N Y T L I V
--------MEL QVAFL IVGMIAFSDSARG P P R ISDKVMHRQ TVR L G R AIK L L C P V - E A D P P P L I MWM K D G R T I H S GW T R F R V YQHGM K I K D V E S E D A G S Y I C K A T N G F G S INV N Y T L I V
--MTPSPLLLL LLPPL LLGAFPPAAAARG P P KMADKV VPRQ V A R L G R TVR L Q C P V - E G D P P P L T MW T K D G R T I H S GWS R F R V LPQG L K V K Q V E R E D A G V Y V C K A T N G F G S LSV N Y T L V V
- -MTRSPALLL ----L LLGALPSAEAARG P P RMADKV VPRQ V A R L G R TVRL QC P V - E G D P P P L T MW T K D G R T I H S GWS R F R V LPQG L K V K E V E A E D A G V Y V C K A T N G F G S LSV N Y T L I I
---------ML WTAL L FFLHVQSTLGARG P P R LSERV ITH Q T V R L G RNIK L P C P V - E G D P P P L T MW T K D L K T I H S GW T R F R V LRQG L K I K N A E A H D A G H Y T C K A T N G F G S INV N Y S L T V
MARVSSALNSL LILWL GVFLRVASAELRG P P KISEQVNGFRKA V QG RSIK L P C P V - T G N P P P L I MW T K D G V T I H S GW E R F R V RSEG L K V NDV VIE D S G S Y I C R A T N G F G S VSV N F T L T I
--------------------------VGN P P R VLKNDVTSVIA FKG ETLT L P C M VMSG K P R P RTQWY K D G V E I G K DW E R F K QQRKG L R I RNV ALE D RG V Y RC E A V N G Y G S EEL EIM L L V
I D D SGS -RGGAGAADGGPD- -GPPELAGKLVPP R F T Q P S K M R K-R V I A R P V G S S V R L K C T A S G N P R P D I VW L K D NRPL LDEQSRAAGEEGRKK RW T L S L K N VTP E Q S G K Y T C H V F N RAG
I D D SASDRTGPAAADGAETERSTDGLSEKLVRP R F T H P T K M R K-R R I E R P V G S S V R L K C M A S G N P R P E I VW L K D DR-L LTAQE--VG-EGRQK KW T L T L RN L T P E Q S G K Y T C R V S N QAG
I D D TSSGKNSQT- -PEGSNGEYEDHSGKQWAQP R F T Q P A K M R R-R V I A R P V G S S I R L K C V A S G N P R P D I TW L K D NKPL MPHEI----GENKKK KWT L N L K N L K P E D S G K Y T C R V F N KVG
I D D SSSGKTNQP- -PDGPNAEVEEFSSKQWARP R F T Q P A K M R K-R V I A R P V G S S V R L K C I A S G N P R P D I TW L K D NKPL TSQEI----GDNKKK KWT L N L K S L K P E D S G K Y T C Q V Y N RVG
L D D ISPGKESLG- -PDSSSGGQEDPASQQWARP R F T Q P S K M R R-R V I A R P V G S S V R L K C V A S G H P R P D I TWM K D DQAL TRPEA----AEPRKK KW T L S L K N L R P E D S G K Y T C R V S N RAG
MD D ISPGKESPG- -PGGSSGGQEDPASQQWARP R F T Q P S K M R R-R V I A R P V G S S V R L K C V A S G H P R P D I MWM K D DQTL THLEA----SEHRKK KWT L S L K N L K P E D S G K Y T C R V S N KAG
I D D SNPARPVGTTPVGG-------------TKP R F T Q P Q K M R R-KI I Q R P MG S S V R L K C S A S GQ P K P E I I WS K D GRTL TDED---LGGAKKGQRWT L K L RN L R P RD S G R Y T C K V F N RIG
T D GSSSPEEDPSRV-DTPTVEERPDRPTSGTMPQ F S E LSK MMKAQK I M K P L N S S V R L K C K A S G H P R P E I VW E K D GTRMVITEGR------QSRQF T L K L S H L K P G D S G T Y MC I V F N KHG
- ------KVKLNTTTDQIESE----------KP I F T E P I K M R RYN - I A K P I G RD V R F S C K A K G K P P P Q I QW F K D GNFTSSPD-----GNPRRQQWT L T L RQL K LTD S G R Y T C R VWN KAG
E I N A T Y K LEV I Q R T N S K P V L T G T H P V N T T V D Y GG T T S F Q C K V RS D V K P V I QW L K R V E PGEE TKY-----N S T I E V GDHHF V V L P T G E VWS R P - D G S Y L N K L L I T R A K E D D A GM Y I C L G A
E I N A T Y YIEV I Q R T S S K P V L T G T H P V N T T V D Y GG T T S F Q C K V RS D V K P V I QW L K R V E YYEESRY-----N S T I E V GDHRF V V L P T G E VWS R P - D G S Y L N K L L I T R A K E E D A GM Y I C L G A
E I N A T Y K V E V I Q R T RS K P I L T G T H P V N T T V D Y GG T T S F Q C K V RS D V K P V I QW L K R V E YGTESKY-----N S T I D V G GQKF V V L P T G E VWS R P - D G S Y L N K L M I T R A K E E D A GM Y I C L G A
Q I N A T Y K V E V I Q R T RS K P I L T G N H P V N T T V D F GG T T S F Q C K V RS D V K P V I QW L K R V E YGTENKY-----N S T I D V G GQKF I V L P T G E VWS R P - D G S Y L N K L I I T R A R E E D A GM Y I C L G A
A I N A T Y K V D V I Q R T RS K P V L T G T H P V N T T V D F GG T T S F Q C K V RS D V K P V I QW L K R V E YGAEGRH-----N S T I D V G GQKF V V L P T G D VWS R P - D G S Y L N K L L I T R A RQDD A GM Y I C L G A
A I N A T Y K V D V I Q R T RS K P V L T G T H P V N T T V D F GG T T S F Q C K V RS D V K P V I QW L K R V E YGSEGRH-----N S T I D V G GQKF V V L P T G D VWS R P - D G S Y L N K L L I S R A RQDD A GM Y I C L G A
S I N A T Y K V D V I A RMTT K P Q L T G V H P V N T T V D F G T T A S F Q C R V RS G V K P H I QW L K R V E EHELHKHP----N T T I D V G G VKF I V L P T G D VW L R P - D G S Y L N K L M I S K A K E E D S GM Y I C L G A
R I N A T Y E V D V VDQVKKK P E L L G E H P V N T T V EFGG T T S F Q C R V RS D I K P H I QW L K R V E --THSTPT----N A T I E MD GQYF V V L P A G D V LPR P - D G S Y L N K L I I SYA T SDD A GM Y I C L G A
A I S F N Y S LRV EGLFQLN P V L V KPH P V N T T V RVG E K T S L Q C R V RS NIV LTV KWM K Q I E PSAQYQYSERLLN N S LRIDGNLY V I L P T QETLYL P E E R T Y L N KHVI V R A T E H D A G R Y L C V A A
N T MG Y S F RS A Y L T V L P D Q-----QP PGN I I P AAG---TPS---L PWP V I I G I P A --G VAF I L G TAFL W F C HSKRHC- - - - -SSSSSASSALPAGQRLPATSR E R AGAGLPPQSASSDKD
N T MG Y S F RS A F L T V L P D T-----KP P ---IP PIF---S PAYNPL PWP V I V G I P A --G IVL I F G AALL W F C QSRKHCP - -PPSAP AAA AQVMQSSHRPPYRER E RGCAAPAS ISSSPDKD
N T MG Y S F RS A F L T V L P D P -----KP PSA- -P VPP- - -S SVS-SL PWP V I I G I P A --G AVF I F G TILL W L C Q TKK -----KPCSP PAA APV----HRP--QPRD R IC-----VSQVPDKD
N T MG Y S F K S A F L T V L P D P -----KP P IA--P V-A---S TNS -SL PW P V I I G I P A --G AVF I F G TILL W L C Q TKK -----KPCSP PAA AAI----HR ---QPA-R VA-----IPQMPDKE
N T MG Y S F RS A F L T V L P D P -----KPQGP - - P VAS - - -S SSATSL PWP V V I G I P A --G AVF I L G TLLL W L C Q AQK -----KPCTP APA PPL - -PGHRPPGTARD RSG-----DKDLPSLA
N T MG Y S F RS A F L T V L P D P -----KP PGP- -PMAS - - -S SSSTSL PWP V V I G I P A --G AVF I L G TVLL W L C Q TKK -----KPCAP ASTLPV- -PGHRPPGTSR E RSG-----DKDLPSLA
N T MG Y S Y K A A F L T V K P D PD-KVEDP AITTKSVMSR- -S SLSSKV P L V V V I G I P A ALG LIL VIGGV- - W F C QQRK RCP-----SP PASVRTPTRQYVPQHDATSHGTTVLSLQRDKDSSG
N T MG Y S F K S A F L E V F P D PNMNWIDP ALQQTP RVNPVVPEAWGHIEQGLLI G I P VSLALLF I V IVIICTIS RKKARSQRRRHRTAPPA PQLPNSS IKKEYNHRDYLNRTNL TNPHMMD I E
N T RG Y R L RG A Y L S V L P A RAKKQDEP P IRSTAKLNTIP
C---LSY----EE Y--VAHQQ LLL SQG---GTGLAPK V Y ----P K I Y T D I H T H T H S H VDGKVH----QH QH I HYQC
CIASMNY----EE Y--LAQQQ LLL SH-----PALPSK V Y ----P K I F T D I H T H T H S H VDGKVH----QH QH I HFQC
CISSINY----EE Y--VAQQQHLL SQGPALAPAMASK M Y ----P K I Y T D I H T H T H S H VEGKVH----QH QH I QYQC
GIASINY----DE Y--LAQQQHLL AQGPVLAPAMASK M Y ----P K I Y T D I H T H T H S H VEAKVH----QH QH I HYQC
ALSAGPGVGLCEEHGSPA APQHLL GPGPVAGP- - - -K L Y ----P K L Y T D I H T H T H T H SHTHSHVEGKV H QH I HYQC
-------VGICEEHGSAMAPQHIL ASGSTAGP----K L Y ----P K L Y T D V H T H T H T H TCTHTLSCGGQGS S TP-A C PLSVLNTANLQALCPEVGIWGPRQQVGRIENNGGRVS
SLSSGMHPPSG-LLSPAASPTGTL SSNSTS----TREKY ----L NRY A D I H T H T H T H TEG- -VSKVYQH QH I HYQC
QCAPLTAAQC-----APMTSQQTL HIYPVPSRENVLNRYDVHSPSSF T D T L N S N GSLTSSRPNMHHHL H Q H TILH C

BMC Biochemistry 2009, 10:33 http://www.biomedcentral.com/1471-2091/10/33
putative protein also lacks a signal peptide that would be
required for insertion of the receptor into the plasma
membrane. Nevertheless, the trace archive of the National
Center for Biotechnology Information (NCBI) contains
17 expressed sequence tags (ESTs) from C. intestinalis that
partially overlap with the entire length of the predicted
mRNA, thus confirming sequence and authenticity of the
predicted protein.
The three Ig-like domains were aligned with the FGFRL1
sequences from five vertebrates and lancelet (Fig. 1). In
the resulting multiple sequence alignment, all the cysteine
residues that participate in disulfide-bond formation of
the three Ig-like loops are conserved. In contrast, the
attachment sites for asparagine-linked carbohydrates are
not conserved. The C. intestinalis sequence contains a
total of 7 glycosylation signals (NXT/S) from which only
two match glycosylation sites in the vertebrate sequences.
The conserved part of the sea squirt sequence with the
three Ig-like domains was used to build a phylogenetic
tree by the neighbour joining method (Fig. 2). In the
resulting unrooted tree, the Ciona sequence is placed on
one branch together with the lancelet sequence as would
be expected for two members of the chordata. However,
the relative distance between the sea squirt and lancelet is
extremely large such that the sea squirt behaves like an
outgroup.
When the predicted mRNA sequence was compared to the
genome sequence of C. intestinalis, 11 coding exons
could be identified (not shown). An inspection of the
splice phases showed that in 6 cases the exon/intron
boundaries disrupt the codons for the amino acids after
the first nucleotide (splice phase 1), in 3 cases after the
second nucleotide (splice phase 2) and in one case after
the third nucleotide (splice phase 0). This gene structure
differs completely from that of the human FGFRL1 gene
which contains only 6 coding exons (and a 5' non-coding
exon) and in which the splice boundaries disrupt the
amino acid codons always after the first nucleotide (splice
phase 1).
It is difficult to speculate on the relationship of the Ciona
gene with the vertebrate FGFRL1 genes. The striking con-
servation of the three Ig-like domains suggests that the
Ciona gene is the orthologue of human FGFRL1. How-
ever, the substantial differences in the gene structure, the
lack of the histidine-rich domain at the C-terminus and
the apparent absence of a signal peptide argue against this
possibility. One way to explain the controversial findings
would be the assumption that the predicted C. intestinalis
gene still contains many sequencing errors. However, the
existence of 17 overlapping EST clones argues against this
possibility. Another way to reconcile the findings would
be that the Ciona gene does no longer serve an active func-
tion and consequently is about to evolve or decay at an
accelerated speed.
Cloning of sea urchin FGFRL1
A similar search through the recently published genome
sequence of the sea urchin Strongylocentrotus purpuratus
(assembly Spur V2.1 [19]) revealed one gene
(LOC578697) that is highly homologous to human
FGFRL1. The predicted mRNA sequence has a length of
1674 nt and codes for a protein of 557 amino acids shar-
ing 48% sequence identity (55% similarity) with human
FGFRL1. However, a detailed comparison of the predicted
mRNA sequence with the S. purpuratus genome indicates
that several portions must have been misinterpreted by
automated gene analysis with the program Gnomon.
First, the N-terminus of the predicted protein does not
include a hydrophobic signal peptide required for inser-
tion of the receptor into the plasma membrane. Secondly,
the sequence corresponding to the linker between Ig
domain I and II (acidic box) is encoded by many short
fragments that repeatedly change the splicing phase.
We therefore tried to verify the predicted mRNA sequence
by comparing it with EST clones. Although the Trace
Archive of NCBI contains numerous EST clones from S.
purpuratus, none of them overlaps with the predicted
sequence. EST databases from related species were there-
fore consulted. In a database compiled at the Max Planck
Institute for Molecular Genetics in Berlin (http://
www.molgen.mpg.de/~ag_seaurchin/[20]) we finally
found two EST clones from the related sea urchin Paracen-
trotus lividus that are highly homologous to human
FGFRL1. One covers the sequence corresponding to
Phylogenetic analysis of the FGFRL1 sequencesFigure 2
Phylogenetic analysis of the FGFRL1 sequences. An
unrooted tree was built by the neighbour joining method.
Bootstrap values from 1000 random replicates are indicated
at the nodes. The length of the branches inversely correlates
with the degree of similarity. Only the sequences from the Page 4 of 10
(page number not for citation purposes)
human amino acid residues 14-212 and shows 47% iden-
tity at the amino acid sequence level, the other covers the
extracellular domains without signal peptides and transmem-
brane domains were used.

BMC Biochemistry 2009, 10:33 http://www.biomedcentral.com/1471-2091/10/33
sequence corresponding to residues 14-69 and shows
51% identity.
In order to establish the correct sequence for S. purpuratus
FGFRL1, we set out to clone the corresponding mRNA.
Total RNA was isolated from a freshly collected specimen
of S. purpuratus and transcribed into first strand cDNA.
Primers were designed according to the 5' sequence from
the P. lividus EST clones and the 3' sequence from the pre-
dicted S. purpuratus mRNA. With the help of these prim-
ers, we were able to amplify a cDNA fragment of 1.6 kb.
Direct sequencing of this fragment revealed an open read-
ing frame of 1599 bp (accession number FN252817) that
displayed 50% sequence identity with the human mRNA
sequence. The open reading frame could be translated
into an amino acid sequence of 532 residues, which
showed 48% sequence identity (55% sequence similarity)
with human FGFRL1.
The corrected mRNA sequence for S. purpuratus FGFRL1
contained 11 ambiguous nucleotides. It is likely that these
sites represent single nucleotide polymorphisms (SNPs)
since none of them caused a change in the derived amino
acid sequence and since they persisted when the cloning/
sequencing experiment was repeated. Since the RNA had
been isolated from a single animal the SNPs must repre-
sent two haplotypes originating from the two different
alleles. Indeed, when the PCR product was subcloned into
the sequencing vector pUC and resequenced from a single
clone, no ambiguous nucleotides were observed. High
heterozygosity has previously been noted during the elu-
cidation of the S. purpuratus genome [19].
Gene structure
When the mRNA sequence was aligned with the S. purpu-
ratus genome, six exons could be identified (Table 2). All
the splice sites of these exons conform to the consensus
rules for donor (GTRNRT) and acceptor (YAG) sequences.
The splice phases and the exact splice sites are fully con-
served between the sea urchin and the human gene [1].
Exon 1 codes for the signal peptide, exon 2 for Ig domain
I, exon 3 for the linker region (acidic box), exon 4 for Ig
domain II, exon 5 for Ig domain III and exon 6 for the
transmembrane segment and the intracellular domain as
previously observed in the human gene.
Homology to mammalian proteins
When the amino acid sequence of FGFRL1 from S. purpu-
ratus was aligned with the FGFRL1 sequences from verte-
brates and lancelet, several regions with high homology
could be observed (Fig. 1). Yet, the degree of sequence
conservation is not uniformly distributed over the total
length of the aligned sequences. Particularly good conser-
vation (>60%) is noted for all three Ig domains, whereas
the signal peptide, the linker between Ig domain I and II
(acidic box) and the majority of the intracellular domain
show negligible conservation (~20%). The location of the
six cysteines that are involved in disulfide bond formation
is fully conserved, as is the location of the glycosylation
signals NXT for asparagine-bound carbohydrates. Yet, the
FGFRL1 sequence from S. purpuratus contains an addi-
tional glycosylation site (NSS) at position 169.
The sequence of the extracellular domain (without signal
peptide and without transmembrane domain) shares 61-
64% identity with FGFRL1 from lancelet and vertebrates
(Table 1). This portion of the FGFRL1 sequence was uti-
lized to build a phylogenetic tree by the neighbour joining
method (Fig. 2). In the resulting unrooted tree, FGFRL1
from sea urchin, sea squirt and lancelet are placed on one
branch, FGFRL1 from fish on a second branch and
FGFRL1 from mammals, chicken and frog on a third. The
confidence for placement of the individual nods as deter-
mined by the bootstrapping method was found to be
highly significant (85%-100%). This phylogenetic tree is
in good agreement with our current understanding of ver-
tebrate evolution with the only exception of the sequence
from C. intestinalis as already discussed above.
Tissue expression of sea urchin FGFRL1
To confirm expression of the FGFRL1 gene in S. purpura-
tus, we prepared a Northern blot using RNA from five dif-
Table 2: Structure of the Fgfrl1 gene from Strongylocentrotus purpuratus
Exon Acceptor Donor Splice phase Exon size Domain encoded
1 cgccgataga ATGGCTCGGG... ...GAACTTCGAGgtaagt 1 >85 bp 5'UTR, signal
MetAlaArgV GluLeuArgG
2 gtcattgcag GGCCACCAAA... ...ACAATTACTGgtgagt 1 273 bp Ig I
lyProProLy ThrIleThrA
3 cttaccatag ATGGATCCAA... ...ACATCAGGAAgtaagt 1 84 bp acidic box
spGlySerSe ThrSerGlyT
4 cttaccatag CTATGCCTCA... ...GATGTAGTAGgtgagt 1 282 bp Ig II
hrMetProGl AspValValA
5 gtttatttag CTATGCCTCA... ...AGATGTAGTAgtgagt 1 281 bp Ig III
spGlnValLy ValPheProA
6 tcccctatag ATCCTAACAT... ...TCTTCATTGCtagcat 1 >521 bp transmembrane, Page 5 of 10
(page number not for citation purposes)
spProAsnMe eLeuHisCys intracellular, 3'UTR

BMC Biochemistry 2009, 10:33 http://www.biomedcentral.com/1471-2091/10/33
ferent tissues and the cloned cDNA as a probe. A single
band was observed on the resulting blot that migrated
with a relative mobility corresponding to 8500 nucle-
otides (Fig. 3). This size suggests that the open reading
frame (1599 nt) is followed by an extremely long 3'
untranslated region (6000 nt). Thus, the size of the S. pur-
puratus mRNA is considerably longer than that of the
mouse (2500 nt [15]), human (3000 nt [1]) and chicken
(6000 nt[10]) mRNA.
The bands on the Northern blot demonstrate that sea
urchin FGFRL1 is expressed in all the five tissues exam-
ined, with particularly high levels found in mouth,
esophagus and intestine (Fig. 3). This is in sharp contrast
to our previous findings with human [1] and mouse [4]
tissues, where high level expression seemed to be
restricted to cartilage, bone and some muscles.
Zinc binding
The only region of the intracellular domain that is con-
served between FGFRL1 from mammals and S. purpuratus
is the histidine-rich motif at the very C-terminal end. In
this region, the human sequence contains 10 histidines
within the last 26 residues, whereas the S. purpuratus
sequence contains 6 histidines within the last 12 residues
(Fig. 1). Histidine-rich sequences are usually found in zinc
finger proteins [21-23] that are involved in protein-pro-
tein interactions and in DNA binding. It is therefore pos-
sible that zinc ions bind to the C-terminal region of
FGFRL1. Since the sea urchin protein contains a signifi-
cantly lower number of histidine residues, it might also
bind less zinc ions than the human protein.
To address this question, we expressed three fragments
from the intracellular domain of FGFRL1 as GST-fusion
proteins in a bacterial expression system, namely an upper
human fragment of 72 residues covering amino acids 400-
471, a lower human fragment of 33 residues covering
amino acids 472-504 and a sea urchin fragment of 41 res-
idues covering amino acids 492-532. On a polyacryla-
mide gel, the purified fusion proteins migrated with
electrophoretic mobilities that are consistent with the cal-
culated molecular weights (Fig. 4). To demonstrate an
interaction with metal ions, the fragments were loaded
onto nickel columns, washed with loading buffer and
eluted specifically with imidazole. Under our experimen-
tal conditions, the lower human fragment bound specifi-
cally to the column, whereas the upper human fragment
as well as the sea urchin fragment barely interacted with
the column matrix (Fig. 4). The binding experiments were
repeated with a column matrix for which the nickel ions
had been exchanged by zinc ions. Again, the lower human
fragment bound to the zinc column and could specifically
protein has considerably lower affinity for zinc ions than
the human homologue.
The actual amount of zinc ions bound to the expressed
fusion proteins was determined by atomic absorption
spectroscopy. We found that the upper human fragment
contained negligible amounts of zinc (<0.5 mole zinc/
mole protein). The lower human fragment bound 2.6
mole zinc/mole protein, the corresponding sea urchin
protein 1.7 mole zinc/mole protein (Fig. 5). Taking into
account that a small proportion of the recombinant
polypeptides might be present in partially degraded or
incorrectly folded form, these numbers may correspond
to 3 zinc ions for the human and 2 zinc ions for the sea
urchin protein, respectively. Neither of the three frag-
ments bound significant amounts of copper ions. Thus,
the C-terminus of FGFRL1 interacts specifically with zinc
ions and the number of bound zinc ions is higher in the
human than in the sea urchin FGFRL1.
Discussion
FGFRL1 is a transmembrane receptor of the FGFR family.
We have previously demonstrated that the FGFRL1 gene
occurs in all vertebrates [1,10,15,17], while invertebrates
such as insects and nematodes do not appear to possess a
related gene. Vertebrates belong to the chordates, a phy-
lum that also includes the subphyla of the cephalochor-
dates and urochordates (tunicates). Recently, we also
found the FGFRL1 gene in the cephalochordate Branchi-
ostoma floridae, but not in the urochordate Ciona intesti-
nalis or in more distantly related invertebrates [18]. We
therefore concluded that the FGFRL1 gene might have
evolved just before branching of the vertebrate lineage
from the other chordates.
This conclusion appears to be wrong. The recently
sequenced genome of the sea urchin S. purpuratus indi-
cates that echinoderms are the closest known relatives of
the chordates [19]. Here we were able to clone a cDNA for
FGFRL1 from S. purpuratus and show that it is closely
related to the vertebrate homologue. The sea urchin pro-
tein shares 61-64% amino acid sequence identity with the
vertebrate protein and displays the same domain structure
with all disulfides and asparagine-linked carbohydrates
conserved. A related gene could also be identified in the
genome of the urochordate C. intestinalis. However, the
function of this gene remains questionable at the moment
since the encoded protein lacks the signal peptide
required for insertion into the plasma membrane. At any
rate, the presence of FGFRL1 in sea urchins demonstrates
that the FGFRL1 gene is much older than previously
assumed.Page 6 of 10
(page number not for citation purposes)
be eluted with 1.2 M imidazole, while the sea urchin frag-
ment as well as the upper human fragment did not bind
to the column. These results suggest that the sea urchin
Not only the protein sequence but also the gene structure
is fully conserved between sea urchins and vertebrates. In
either case there are six protein coding exons and the

BMC Biochemistry 2009, 10:33 http://www.biomedcentral.com/1471-2091/10/33
splice phases and splice sites are preserved. The three
exons encoding the three Ig domains have correctly been
predicted by automated annotation of the S. purpuratus
genome [19]. However, the first and the third exon have
been misinterpreted by the automated gene prediction
software. The first exon codes for the signal peptide that,
apart from its hydrophobic character, does not show
much conservation among different species. In the
human and rodent genome, this exon is found close to the
promoter region, approximately 10 kbp upstream from
the rest of the gene [1,15]. The third exon codes for the
linker between Ig loop I and II. This exon has probably
been missed by automated prediction because its derived
amino acid sequence displays a particularly low degree of
conservation between sea urchins and vertebrates
(<20%). This region might serve as a flexible linker in the
FGFRL1 protein that allows the first Ig domain to fold
back to the second and third Ig domains [6,7]. In the case
of the classical receptor FGFR1, this linker has been
termed "acidic box" because it contains a total of 13 aspar-
tic and glutamic acids within 28 residues. When it folds
back, Ig domain I may shield the sites for interaction with
heparin and FGF ligands. The acidic box might therefore
implicated in the interaction of FGFR1 with the adhesion
molecule NCAM [24]. However, the corresponding region
of human FGFRL1 contains only 6 acidic amino acids
within 27 residues. It is therefore not clear whether it
might serve a similar function to the acidic box of FGFR1.
In contrast to the extracellular domain with the three
highly conserved Ig loops, the intracellular domain dis-
plays a very low degree of conservation between sea
urchins and humans (23%). The only region of this
domain that reveals some similarity is the histidine-rich
motif at the very C-terminal end. Here FGFRL1 from
humans and sea urchins contain nearly 50% histidines
(10 histidines within 26 residues for humans, 6 histidines
within 12 residues for S. purpuratus). In the human
sequence, the histidines alternate with threonine, serine
or glutamine residues, whereas in the sea urchin sequence,
three of the six histidines occur in a row without inter-
spersed residues. Histidine-rich sequences typically occur
in zinc finger domains that form interfaces for protein-
protein and protein-DNA interactions [21-23]. Such zinc
fingers are found in cytoskeletal and focal adhesion pro-
teins (e.g. zyxin) but also in nuclear proteins and tran-
scription factors (e.g. homeobox proteins). Most zinc
fingers conform to the sequence C2H2, where one zinc
ion is coordinated by two cysteines and two histidines
[21]. The sequence of FGFRL1 clearly differs from the clas-
sic C2H2 motif as it contains mainly histidines distrib-
uted over 12-26 residues. By column chromatography and
by atomic absorption spectroscopy we could demonstrate
that this region of the human as well as the sea urchin pro-
tein interacts with zinc ions. Under our experimental con-
ditions, the C-terminal tail of the human protein bound 3
zinc ions, while that of the S. purpuratus protein bound
only 2 zinc ions. The difference in binding capacity could
be confirmed by zinc-column chromatography, where the
human protein interacted strongly with the column, while
the S. purpuratus protein barely interacted at all. Thus, the
C-terminal domain of FGFRL1 contains a novel zinc bind-
ing motif, which has gradually improved its zinc binding
activity during evolution. When we compare FGFRL1
from sea urchins and humans we may therefore observe
the appearance and shaping of a new functional domain.
What is the function of this histidine-rich domain? As
mentioned above, histidine-rich domains are often found
in cytoskeletal proteins where they enable protein-protein
interactions [21-23]. There is indirect evidence that the
histidine-rich tail might in fact be involved in the interac-
tion of FGFRL1 with cytoskeletal proteins. When full-
length FGFRL1 was overexpressed in HEK293 cells and
then subjected to subcellular fractionation utilizing a
commercial kit, the majority of the proteins was found in
Expression of the FGFRL1 gene in different tissues from S. purpuratusFigure 3
Expression of the FGFRL1 gene in different tissues
from S. purpuratus. A radioactively labelled cDNA probe
corresponding to the sequence for amino acid residues 1-374
was hybridized to a Northern blot containing 7.5 Πg RNA
from five different tissues as indicated. The 26S ribosomal
RNA stained with ethidium bromide is included as a loading
control.Page 7 of 10
(page number not for citation purposes)
be involved in the modulation and regulation of heparin
and FGF binding. Moreover, the acidic box has been
the insoluble fraction that is known to contain cytoskele-
tal proteins [[11]; Rieckmann and Trueb, unpublished

BMC Biochemistry 2009, 10:33 http://www.biomedcentral.com/1471-2091/10/33
results]. In contrast, FGFRL1 lacking the histidine-rich
region was primarily found in the Triton X-100 soluble
fraction that contains most of the membrane proteins.
Unfortunately, the complete insolubility of the full-length
protein has so far hampered all our efforts to characterize
the putative cytoskeletal interaction partner(s) of FGFRL1.
Based on the zinc content of the human and the sea
urchin protein, we concluded above that FGFRL1 must
have increased its zinc binding properties during evolu-
tion. It is possible that along with the increased zinc-bind-
ing properties, the affinity for cytoskeletal proteins may
also have improved. In this way, FGFRL1 could gradually
have acquired a novel function.
In addition to the lower zinc binding capacity of the S.
purpuratus protein, we have also noted another striking
difference between human and sea urchin FGFRL1. In S.
purpuratus, FGFRL1 is expressed at relatively high levels
in all tissues examined, including mouth, esophagus,
intestine, gonads and tube feet. In humans and mice,
FGFRL1 is primarily expressed in cartilage, bones and
note that sea urchins do not possess any cartilage and
bone. The particular function that FGFRL1 might play in
cartilage and bone must therefore have been acquired
after evolution of echinoderms, i.e. FGFRL1 must have
taken over additional functions during evolution. A simi-
lar observation has previously been made with teleostean
fish. Bony fish, including pufferfish and zebrafish, possess
two genes for FGFRL1, fgfrl1a and fgfrl1b, because they
have undergone a whole genome duplication [17]. While
most of the duplicated genes were lost again during evo-
lution, some selected genes were preserved in the genome
because they adopted new functions. Notably, the two
copies of FGFRL1 have been preserved in the fish genome
and today they display slightly different expression pat-
terns. Subfunctionalization might have played a decisive
role in maintaining the two FGFRL1 copies during evolu-
tion of bony fish.
Conclusion
The origin of the FGFRL1 gene is much older than previ-
ously assumed. Its ubiquitous expression in sea urchins,
but its relatively restricted expression in vertebrates lend
further support to the notion that FGFRL1 has gradually
taken over specific functions during evolution. This proc-
ess of subfunctionalization is reflected by improvements
of its zinc binding capacity. We might therefore witness
the shaping of a novel functional domain when we com-
pare the C-terminal end of FGFRL1 from sea urchins and
vertebrates.
Methods
Cloning of sea urchin FGFRL1
Sea urchins of the species Strongylocentrotus purpuratus
were collected from the Pacific Northwest by Living Ele-
ments Ltd. The animals were kept in sea water at 12°C
with an artificial current. Tissues were dissected with scal-
pels and homogenized with a Polytron in guanidinium
isothiocyanate buffer. Total RNA was purified with the
help of the GeneElute Total RNA miniprep kit from
Sigma-Aldrich Co. Purified RNA from tube feet was dena-
tured at 65°C, cooled to room temperature and tran-
scribed into first strand cDNA by reverse transcriptase
from Moloney Murine Leukemia Virus (1.5U/ Πl) as sug-
gested by the supplier of the enzyme (Stratagene). Ran-
dom hexamers served as primers. Aliquots of this cDNA
were amplified by PCR through 39 cycles (30 sec 95°C, 1
min 56°C, 1.5 min 68°C) utilizing Pfx polymerase (Invit-
rogen) and the primer pair GATAGAATGGCTCGGGTT-
TCGTC/AGCGTATGCTAGCAATGAAGAATG. The PCR
product was resolved on a 1% agarose gel. The DNA band
of 1600 bp was excised, purified on an Illustra GFX col-
umn (GE Healthcare Europe GmbH) and directly
sequenced by the dideoxy chain termination method with
Interaction of the C-terminal domain from FGFRL1 with zinc and ni kel i nsFigure 4
Interaction of the C-terminal domain from FGFRL1
with zinc and nickel ions. GST fusion proteins were
mixed with nickel or zinc beads, washed and eluted with 1.2
M imidazole. The eluted proteins were resolved on a poly-
acrylamide gel, transferred to a nitrocellulose membrane and
stained with antibodies against GST. Specifically eluted pro-
teins are compared to the starting material (Input). The
fusion proteins comprised amino acid residues 400-471
(Human U), 472-504 (Human L) and 492-532 (Sea Urchin L).Page 8 of 10
(page number not for citation purposes)
some muscles [1,4], while all other tissues express only
basal levels of FGFRL1. In this context, it is of interest to
a cycle sequencing machine (ABI 3730). For subcloning,
the PCR band was re-amplified as above with the primer

BMC Biochemistry 2009, 10:33 http://www.biomedcentral.com/1471-2091/10/33
pair AAGAATTCATGGCTCGGGTTTCGT/AGCTC-
GAGCTAGCAATGAAGAATG. The product was digested
with Eco RI and Xho I and the resulting two bands were
ligated into the sequencing vector pBluescript SK(-).
Bioinformatics
All sequences were analyzed with the GCG computer soft-
ware package of Accelrys (Cambridge, UK). Similarities
were calculated with the program Old Distances using the
Blosum 62 scoring matrix. A multiple sequence alignment
was performed with the program Pileup. For phylogenetic
analysis, only the extracellular domains without signal
peptides were used. An unrooted tree was built by the
neighbour joining method using the program Paup-
Search. Bootstrap values were calculated for all nodes
from 1000 random replicates [17].
Northern Blot
Total RNA from different tissues was separated on a 1%
agarose gel in the presence of formaldehyde and trans-
ferred to a Nylon membrane by vacuum blotting [1,10].
The blot was hybridized overnight at 42°C with a radioac-
tively labelled cDNA probe in a buffer containing 50%
formamide. This probe had been labelled beforehand
with [ ∆-
32
P]-dCTP by the random primed oligolabelling
method. After washing with standard saline citrate (SSC),
the blot was exposed to an X-ray film.
Fusion proteins
Three fragments derived from the intracellular domain of
human and S. purpuratus FGFRL1 were prepared in a
prokaryotic expression system as fusion proteins with glu-
tathione S-transferase (GST). To this end, the cDNA
sequences for human residues 400-471 (human up),
the Bam HI/Xho I restriction site of the expression vector
pGEX-5X-2 (GE Healthcare) downstream of the GST gene.
The resulting plasmids were transfected into competent
bacteria (E. coli BL21). Fusion proteins were expressed
after induction with isopropylthio- Ε-galactoside as sug-
gested by the supplier of the GST gene fusion system (GE
Healthcare). The bacteria were collected by centrifugation,
resuspended in 50 mM Tris, pH 8.0, 150 mM NaCl, 1 mM
phenylmethanesulfonyl fluoride, 0.3 mM ZnCl
2
, 1% Tri-
ton X-100 and lysed by sonication. Fusion proteins were
purified from the lysates by affinity chromatography on
GSH Sepharose [25] according to the instructions of the
supplier (GE Healthcare). The purified proteins were dia-
lyzed at 4°C against 50 mM Tris, pH 8.0, 150 mM NaCl,
1% Triton X-100 (4 changes over a total of 4 days). In
order to remove any traces of metal ions, the dialysis
buffer had been passed beforehand over a column (1 cm
× 10 cm) of Chelex A100 (BioRad).
Western blots
Proteins were analyzed by SDS polyacrylamide gel electro-
phoresis on gels containing a 5% stacking and a 15% run-
ning gel. After transfer to nitrocellulose by electroblotting,
the polypeptides were detected with the GST detection
module (GE Healthcare) using goat anti-GST antibodies,
followed by alkaline-phosphatase conjugated secondary
antibodies (Sigma). The colour reaction was performed
with bromochloroindolyl phosphate and nitroblue tetra-
zolium as substrate.
Affinity chromatography
Recombinant proteins dissolved at 70 Πg/ml in phosphate
buffer (50 mM sodium phosphate, pH 8.0, 300 mM NaCl,
20 mM imidazole, 1% Triton X-100) were mixed with pre-
equilibrated His-Select™ Nickel affinity beads (Sigma).
The suspension was incubated for 30 min at room tem-
perature, before the beads were washed with the same
buffer. Specifically bound proteins were eluted with 1.2 M
imidazole, 50 mM sodium phosphate, pH 8.0, 300 mM
NaCl, 1% Triton X-100 and analyzed by Western blotting.
To prepare zinc affinity columns, the His-Select™ Nickel
affinity gel was treated with 100 mM EDTA, followed by
extensive washing and re-loading with 40 mM zinc chlo-
ride. The exchange of the nickel ions could be detected by
a specific colour change of the gel from light blue to white.
Atomic absorption
The determination of the zinc and copper content of the
fusion proteins was performed in triplicates using a Spec-
trAA-110 flame atomic absorption spectrophotometer
(Varian Corp., Australia). Prior to metal analysis, the sam-
ples were diluted into 15 mM HNO
3
matrix solution. Zn-
Comparison of the C-terminal FGFRL1 sequences from hu ans, lancele and sea urchinFigure 5
Comparison of the C-terminal FGFRL1 sequences
from humans, lancelet and sea urchin. Histidine resi-
dues are highlighted in bold. Atomic absorption spectroscopy
showed that the human protein bound 2.6 mole zinc/mole
protein, while the sea urchin protein bound 1.7 mole zinc/
mole protein. The standard deviation from three different
measurements was < 2%. The zinc content of the lancelet
protein was not determined (n.d.), but this sequence was
included for reasons of comparison.Page 9 of 10
(page number not for citation purposes)
human residues 472-504 (human low) and S. purpuratus
residues 492-532 (sea urchin low) were subcloned into
and Cu-standards were prepared from commercially avail-

Publish with BioMed Central and every
scientist can read your work free of charge
"BioMed Central will be the most significant development for
disseminating the results of biomedical research in our lifetime."
Sir Paul Nurse, Cancer Research UK
Your research papers will be:
available free of charge to the entire biomedical community
peer reviewed and published immediately upon acceptance
cited in PubMed and archived on PubMed Central
BMC Biochemistry 2009, 10:33 http://www.biomedcentral.com/1471-2091/10/33
able 1000 ppm standards (Fluka, Buchs, Switzerland)
using the same matrix solutions.
Abbreviations
FGF: fibroblast growth factor; FGFR: fibroblast growth fac-
tor receptor; FGFRL1: fibroblast growth factor receptor-
like 1; GST: glutathione S-transferase.
Authors' contributions
LZ and BT designed the study, LZ carried out the biochem-
ical experiments, AVK performed atomic absorption spec-
troscopy, PB collected the animals, BT drafted the
manuscript. All authors read and approved the final
paper.
Acknowledgements
This study was supported by grants from the Swiss National Science Foun-
dation (31003A-127046), the Swiss Foundation for Research on Muscular
Diseases and the Olga Mayenfisch Foundation.
References
1. Wiedemann M, Trueb B: Characterization of a novel protein
(FGFRL1) from human cartilage related to FGF receptors.
Genomics 2000, 69(2):275-279.
2. Kim I, Moon S-O, Yu K-H, Kim U-H, Koh GY: A novel fibroblast
growth factor receptor-5 preferentially expressed in the
pancreas. Biochim Biophys Acta 2001, 1518(1-2):152-156.
3. Sleeman M, Fraser J, McDonald M, Yuan S, White D, Grandison P,
Kumble K, Watson JD, Murison JG: Identification of a new fibrob-
last growth factor receptor, FGFR5. Gene 2001,
271(2):171-182.
4. Trueb B, Taeschler S: Expression of FGFRL1, a novel fibroblast
growth factor receptor, during embryonic development. Int
J Mol Med. 2006, 17(4):617-620.
5. Itoh N, Ornitz DM: Evolution of the Fgf and Fgfr gene families.
Trends Genet 2004, 20(11):563-569.
6. Eswarakumar VP, Lax I, Schlessinger J: Cellular signaling by fibrob-
last growth factor receptors. Cytokine Growth Factor Rev 2005,
16(2):139-149.
7. Beenken A, Mohammadi M: The FGF family: biology, pathophys-
iology and therapy. Nat Rev Drug Discov 2009, 8(3):235-253.
8. Wilkie AO: Bad bones, absent smell, selfish testes: the pleio-
tropic consequences of human FGF receptor mutations.
Cytokine Growth Factor Rev 2005, 16(2):187-203.
9. Coumoul X, Deng CX: Roles of FGF receptors in mammalian
development and congenital diseases. Birth Defects Res C
Embryo Today 2003, 69(4):286-304.
10. Trueb B, Zhuang L, Taeschler S, Wiedemann M: Characterization
of FGFRL1, a novel FGF receptor preferentially expressed in
cartilage. J Biol Chem 2003, 278(36):33857-33865.
11. Rieckmann T, Zhuang L, Flück CE, Trueb B: Characterization of
the first FGFRL1 mutation identified in a craniosynostosis
patient. Biochim Biophys Acta 2009, 1792(2):112-121.
12. Baertschi S, Zhuang L, Trueb B: Mice with a targeted disruption
of the Fgfrl1 gene die at birth due to alterations in the dia-
phragm. FEBS J 2007, 274(23):6241-6253.
13. Hall C, Flores MV, Murison G, Crosier K, Crosier P: An essential
role for zebrafish Fgfrl1 during gill cartilage development.
Mech Dev 2006, 123(12):925-940.
14. Catela C, Bilbao-Cortes D, Slonimsky E, Kratsios P, Rosenthal , Te
Welscher P: Multiple congenital malformations of Wolf-Hir-
schhorn syndrome are recapitulated in Fgfrl1 null mice. Dis
Model Mech 2009, 2(5-6):283-294.
15. Wiedemann M, Trueb B: The mouse Fgfrl1 gene coding for a
novel FGF receptor-like protein. Biochim Biophys Acta 2001,
1520(3):247-250.
pus development resemble those of planarian nou-darake
and Xenopus FGF8. Dev Dyn 2004, 230(4):700-707.
17. Trueb B, Neuhauss SCF, Baertschi S, Rieckmann T, Schild C, Taesch-
ler S: Fish possess multiple copies of fgfrl1, the gene for a
novel FGF receptor. Biochim Biophys Acta 2005, 1727(1):65-74.
18. Beyeler M, Trueb B: Fgfrl1, a fibroblast growth factor receptor-
like gene, is found in the cephalochordate Branchiostoma
floridae but not in the urochordate Ciona intestinalis. Comp
Biochem Physiol B Biochem Mol Biol 2006, 145(1):43-49.
19. Sea Urchin Genome Sequencing Consortium: The genome of the
sea urchin Strongylocentrotus purpuratus. Science 2006,
314(5801):941-952.
20. Poustka AJ, Kühn A, Radosavljevic V, Wellenreuther R, Lehrach H,
Panopoulou G: On the origin of the chordate central nervous
system: expression of onecut in the sea urchin embryo. Evol
Dev 2004, 6(4):227-236.
21. Laity JH, Lee BM, Wright PE: Zinc finger proteins: new insights
into structural and functional diversity. Curr Opin Struct Biol
2001, 11(1):39-46.
22. Ladomery M, Dellaire G: Multifunctional zinc finger proteins in
development and disease. Ann Hum Genet 2002, 66(5-
6):331-342.
23. Dawid IB, Breen JJ, Toyama R: LIM domains multiple roles as
adapters and functional modifiers in protein interactions.
Trends Genet 1998, 14(4):156-162.
24. Sanchez-Heras E, Howell FV, Williams G, Doherty P: The fibroblast
growth factor receptor acid box is essential for interactions
with N-cadherin and all of the major isoforms of neural cell
adhesion molecule. J Biol Chem 2006, 281(46):35208-35216.
25. Rieckmann T, Kotevic I, Trueb B: The cell surface receptor
FGFRL1 forms constitutive dimers that promote cell adhe-
sion. Exp Cell Res 2008, 314(5):1071-1081.yours — you keep the copyright
Submit your manuscript here:
http://www.biomedcentral.com/info/publishing_adv.asp
BioMedcentral
Page 10 of 10
(page number not for citation purposes)
16. Hayashi S, Itoh M, Taira S, Agata K, Taira M: Expression patterns
of Xenopus FGF receptor-like 1/nou-darake in early Xeno-