JOURNAL OF CELLULAR PHYSIOLOGY 208:23–38 (2006)

SEL1L a Multifaceted Protein Playing a Role in

Tumor Progression
IDA BIUNNO,1,3 MONICA CATTANEO,1 ROSARIA ORLANDI,2 CRISTINA CANTON,1
LAURA BIAGIOTTI,1 STEFANO FERRERO,4 MASSIMO BARBERIS,5 SERENELLA M. PUPA,2
ALDO SCARPA,6 AND SYLVIE MÉNARD2*
1
Istituto di Tecnologie Biomediche, CNR, Segrate-Milano, Italy
2
Department of Experimental Oncology, Molecular Targeting Unit,
Istituto Nazionale Tumori, Milan, Italy
3
BioRep, Milano, Italy
4
Department of Pathology, University of Milan School of Medicine,
AOS Paolo, Milano, Italy
5
MultiLab, Gruppo MultiMedica, Milano, Italy
6
Dipartimento di Patologia, Università di Verona, Verona, Italia

Since the cloning in 1997 of SEL1L, the human ortholog of the sel-1 gene of C. elegans, most studies have focused on its role in cancer
progression and have provided significant evidences to link its increased expression to a decrease in tumor aggressiveness. SEL1L
resides on a ‘‘Genome Desert area’’ on chromosome 14q24.3-31 and is highly conserved in evolution. The function of the SEL1L
encoded protein is still very elusive although, several evidences from lower organisms indicate that it plays a major role in protein
degradation using the ubiquitin-proteosome system. SEL1L has a very complex structure made up of modules: genomically it
consists of 21 exons featuring several alternative transcripts encoding for putative protein isoforms. This structural complexity
ensures protein flexibility and specificity, indeed the protein was found in different sub-cellular compartments and may turn on a
particular transcript in response to specific stimuli. The overall architecture of SEL1L guarantees an exquisite regulation in the
expression of the gene. J. Cell. Physiol. 208: 23–38, 2006. ß 2005 Wiley-Liss, Inc.

SEL1L, (sel-1 like gene, Table 1), the human ortholog
of the C. elegans sel-1 gene, opens a new line of research
aimed to define the role of the encoded protein in the
processes of normal cell fate decision, cell differentia-
tion, and transformation. SEL1L may be the first of a
new class of molecules associated with cell fate determi-
nation and involved in cancer progression. While much
effort has been devoted to define the genetic changes
that can lead a cell to proliferate, little is known about
the role that cell determinant molecules play in cancer
progression. Most of the genes found to be relevant in
cell differentiation and cell fate determination are
evolutionarily conserved from simple organisms such
as C. elegans and Drosophila to higher primates, SEL1L
may be one of them. SEL1L shows a high degree of cross-
species conservation in its nucleotide and protein
sequence (Biunno et al., 2002) indicating the remark-
able selective evolutionary pressure to keep the gene
fairly intact. To date, neither functional mutations nor
genomic alterations have been reported for SEL1L,
making it an ‘‘essential’’ protein whose alteration(s) may
be lethal for the survival of an organism (Larsen et al.,
2001; Saltini et al., 2004).
SEL1L is an intriguing gene whose function is still at
a speculative or embryonic stage, but displays several
interesting characteristics. The C. elegans gene, sel-1
(suppressor-enhancer-lin), is considered to be a negative
regulator of the lin-12/Notch family of proteins (Grant
and Greenwald, 1996), probably by acting on their
turn over. The Notch pathway is involved in cell
fate decisions in a variety of tissues and in multiple
organisms, and in the maintenance of the proliferative/
differentiative balance in many cell lineages (Artavanis-
Tsakonas et al., 1999); in human can cause T-cell
lymphomas when introduced into mouse bone marrow
stems cells (Pear et al., 1996) and can transform rat
ß 2005 WILEY-LISS, INC.
kidney cells in co-operation with the adenovirus
E1A oncogene (Capobianco et al., 1997; Gallahan and
Callahan, 1997). No genetic lesions of the Notch
locus have been described in human tumors with the
exception of a rare translocation in T cell malignancies
(Ellisen et al., 1991).
SEL1L shares similarities with the yeast Hrd3p
protein, a member of endoplasmic-reticulum (ER)-
associated protein degradation (ERAD)-specific ubiqui-
tin ligase complex that functions in the detection,
targeting, and degradation of malfolded and normal
endoplasmic reticulum (ER) resident proteins (Hamp-
ton et al., 1996; Hampton, 2002). The human and plant
SEL1L genes belong to the so-called unfolded protein
response genes which are induced in response to the
accumulation of unfolded and misfolded proteins in the
ER stress (Kaneko and Nomura, 2003; Kamauchi et al.,
2005). ER stress has been implicated in the pathogenesis
of a variety of human disorders, including neural
degenerative diseases, diabetes, viral pathogenesis,
and cancer (Kaufman, 2002; Dissemond et al., 2004). It
has been postulated that in cancer, misfolding of key
proteins can cause the cells to lose tumor suppression
Monica Cattaneo and Rosaria Orlandi have equal merit for the
work.
*Correspondence to: Dr. Sylvie Ménard, Department of Experi-
mental Oncology, Molecular Targeting Unit, Istituto Nazionale
per lo Studio e la Cura dei Tumori, Via Venezian 1, Milan, Italy.
E-mail: sylvie.menard@istitutotumori.mi.it
Received 13 October 2005; Accepted 20 October 2005
DOI: 10.1002/jcp.20574
24 BIUNNO ET AL.

TABLE 1. SEL1L gene card

Human SEL1L

Official symbol SEL1L

Gene description Sel-1 suppressor of lin-12 (C. Elegans)-like
Gene alias IBD2, SEL1-LIKE
GeneID 6400
SwissProt Q9UBV2 Sel-1 homolog precursor
Chromosomal location 14q24.3-31
Transcripts Three alternative splicing forms experimentally confirmed
Isoform coding the complete SEL1L gene is isoform A
Ref. Seq. mRNA NM_005065
Protein 794 amino acids; 88755 Da
Ref. seq. protein NP_005056
Domains and motifs Fibronectin II domain
Sel1-like repeats (three clusters)
Hrd3-like motif
Proline-rich motif
Subcellular localization Cytoplasmic and nuclear
Protein expression
Fetal tissues Ubiquitously expressed
Normal tissues Highly expressed in pancreatic cells and other protein secreting cells,
including B cells
Tumor tissues Differentially expressed in pancreatic, gastric, breast, and lung cancer
Function . May have a role in the ER-associated protein degradation (ERAD)
system (similarity with Hrd3)
. Negative regulator of the Notch pathway in C. elegans
. May play a role in TGF-beta signaling
. In breast and pancreatic tumor decrease tumor growth and
aggressiveness, possibly involving cell–matrix interactions
Regulation . Upmodulated in ER-stress
. Upmodulated in early stages of cancer progression
. Downmodulated in comparison to normal tissues in the poor
prognosis breast tumors

functions (Qu and Koromilas, 2004; Stavridi and
Halazonetis, 2004), and this is a novel and very
interesting aspect of tumor biology (Stefani, 2004).
A survey of the expression of SEL1L mRNA as
well as its encoded protein, on a series of cancerous
and pre-neoplastic lesions, revealed, at least partially,
the role of SEL1L in cancer progression (Granelli et al.,
2004; Barberis et al., 2005; Ferrero et al., 2005,
submitted); furthermore, its expression in breast cancer
correlated with patient’s survival (Orlandi et al., 2002a).
Taking a step over, in vitro studies indicated that SEL1L
protein affects those pathways which regulate signaling
(cell–cell and/or cell–matrix) interactions. Presently,
there is not much literature on this gene, which results
in the difficulty in generating a precise theory on
the function of the protein. However, considering the
available data derived from several organisms, an
hypothesis can be forwarded in that SEL1L may
function in the protein degradation processes through
ubiquitin-proteosome system and perhaps in regulating
important pathways, such as Notch and TGF-beta.
Since the protein structure consists of at least three
blocks of highly conserved domains (SEL-1-like repeats
and Hrd3 motif) and of two modules acquired later
during taxonomic evolution (fibronectin type II domain
and proline-rich region), the complexity of its specificity
is highlighted.
SEL1L is not a member of a vast family of proteins but
the several described isoforms (over four) give the
appearance of belonging to a multifamily of molecules
having perhaps redundant functions.
The architectural complexity of the SEL1L gene and
protein is an example of its flexibility and reflects how in
some genes, different combinations of exons can become
active at different times and each combination can yield
a different protein with a different sub-cellular location,
this phenomenon allows the control of gene expression
in an exquisite manner.
Journal of Cellular Physiology DOI 10.1002/jcp
SEL1L GENE AND PROTEIN STRUCTURE
Genomic structure
The SEL1L gene consists of 21 exons spanning
62.24 Kb of genomic DNA (Fig. 1A), the coding region
consists of 2,385 Kb (GenBank accession no.
NM_005065) encoding for a protein of 794 amino acids
with a calculated molecular weight of 88,750 KDa.
Upstream of the starting methionine, a genomic
sequence of 1,271 Kb (GenBank accession no.
AF157516), contains the essential features of the
promoter region. The C-terminal tail consists of
over 5.0 Kb untranslated sequences likely containing
fundamental regulatory elements. Two polymorphic
(CA)n repeats are located in intron 2 and intron 20,
useful in linkage studies (Biunno et al., 2000).
Chromosomal location
SEL1L resides on chromosome 14q24.3-q31 from
81069886 to 81007646 (NCBI build 35.1, August 2005),
on the reverse strand (Fig. 1A), in the same region
as IDDM11, a locus showing evidence for linkage to type
1 diabetes (Pociot et al., 2001). However, no association
was found between SEL1L microsatellite markers and
type 1 diabetes nor with the Graves’ disease also
residing in the same chromosomal band (Ban et al.,
2001; Pociot et al., 2001). Extensive computer aided
analysis of this chromosomal interval indicates that the
7.3 Mb of DNA stretch harboring SEL1L, is rather
scarce in disease causing genes. Indeed, only five protein
coding genes (neurexin 3, deiodinase iodothyronine type
II, thyroid stimulating hormone receptor, general
transcription factor IIA, and fibronectin leucine-rich
transmembrane protein 2), four hypothetical EST
and ten pseudogenes surround the SEL1L gene
(Fig. 1B). The chromosomal interval between SEL1L
and fibronectin leucine-rich transmembrane protein
2 genes belongs to the so-called ‘‘Gene Desert area’’ or
 SEL1L AND ITS STRUCTURAL AND FUNCTIONAL COMPLEXITY
Fig. 1. A: SEL1L genomic structure: SEL1L gene resides on chro-
mosome 14q24.3-q31 from 81069886 to 81007646 on the reverse
strand (NCBI build 35.1, August 2005) spanning 62,240 Kb of genomic
DNA. The coding region consists of 2,385 Kb and contains 21 exons
(NM_005065). B: Simplistic representation of chromosome 14 (mod-
ified from Genomic Cartography, http://acg.media.mit.edu/people/fry/
genocarto.html). This figure engenders a very compact, insightful
visual for chromosome 14, the DNA stretch harboring SEL1L is
enlarged in the botton. The empty gray (yellow) wireframe boxes
signify a gap between genes, an area of unused data. The dark floor
(blue) areas represent parts where genes exist. The dark floor (blue)
wireframe boxes are proportional in size to the start and end point of
the gene within the sequence of letters, usually a few thousand letters
apiece. The solid (blue) boxes within these frames represent the coding
portions of a gene, shown in proportion to the start and end of the gene
itself. By using cubes to represent the data, it allows the map to be re-
weighted, such that the much smaller percentage of coding informa-
tion still remains relevant in scale. Similar to a logarithmic scaling, it
scales using a cube root in the linear direction, resulting in a more
balanced visualization. NRXN3, neurexin 3; DIO2, deiodinase
iodothyronine type II; TSHR, thyroid stimulating hormone receptor;
GTF2A1, general transcription factor IIA; FLRT2, fibronectin leucine-
rich transmembrane protein 2; GALC, galactosylceramide. C: SEL1L
promoter: nucleotide positions are numbered to the left with respect to
Journal of Cellular Physiology DOI 10.1002/jcp
25
the translation initiation site (ATG is in bold and red characters).
Potential cis-acting elements are predicted with MatInspector
program (Jan, 2005), blue and yellow boxes correspond to regulators
involved in ER stress and tissue-specific factors, respectively. The
CAAT and the four GC boxes are boxed in celestial and pink,
respectively. The multiple transcription start sites are blue under-
lined. The lowercase letters indicate part of the first intron sequence.
The two polymorphisms (
366 and
354) are in bold and black
underlined. The CpG island located between nucleotide
500 and the
start codon is highlight in yellow. D: SEL1L protein structure: SEL1L
is a multimodular protein containing several structural and functional
domains as well as signal sequences. The signal peptide (from 1 to 22
amino acid residues) and the Pest sequence (from 80 to 102 amino acid
residues) are represented by red and pink rectangles. The fibronectin
type II domain (from 120 to 168 residues) is symbolized by the hexagon
(FN2), the SEL-1-like repeats are represented by rhombi and are
distributed in tandem along the central portion of the protein in three
large clusters (I cluster: 183–326, II cluster: 373–554, and III cluster:
664–675 residues). The Hrd3 like motif is located within the last SEL-
1-like repeat (664–675 residues) and is represented by a circle. The
transmembrane region (TM) (739–761 residues) and the proline-rich
tail (770–793 residues) are symbolized by a blue and pink rectangle.
The N-linked glycosilation is also underlined.
26 BIUNNO ET AL.
Fig. 1.
‘‘Genome Deserts’’ (http.www.chr7.org/March2003/
GeneDeserts.php: table S7d) since consists in no genes
in >500 Kb genomic stretch. The functional importance
of these areas is undetermined, it has been reported that
megabace deletions of gene deserts result in viable
mice, supporting the hypothesis of the existence of
disposable DNA in mammalian genomes (Nobrega et al.,
2004). A genome-wide analysis of those chromosomal
features able to repress the transcription of the human
immunodeficiency virus identified the integration of the
viral particle in gene deserts areas. This suggests
the presence of a negative transcription regulatory
elements able to maintain the integrated virus in a
latency state (Lewinski et al., 2005). Residing in this
desert area, SEL1L may protect itself from the common
recombination events that occur in the immunoglobin
heavy (IGH) chain locus, a gene cluster of immunologi-
cal importance located in the telomere of chromosome
14 and frequently involved in chromosomal transloca-
tions in lymphocyte malignancies. This isolated location
may ensure SEL1L integrity.
SEL1L does not appear to be a member of a vast family
of proteins; a SEL1L-like protein has been annotated by
NCBI; the gene encoding for this protein is located on
chromosome 20p12.1 and contains structural features
similar to SEL1L (Cattaneo et al., 2004).
Journal of Cellular Physiology DOI 10.1002/jcp
(Continued)
Promoter characteristics
SEL1L contains a TATA-less promoter with multiple
transcription initiation start sites (Fig. 1C). The basal
core of the promoter consists of four SP1 binding sites
and a CAAT box, the sequence of 900 bp extending
upstream, harbors several putative binding sites for
known transcription factors including tissue-specific
regulators such as hepatocyte nuclear factor-3 beta
(HNF3b), homeobox Nkx2-5, hepatic nuclear factor
1 (HNF1), GATA binding factor 1, prostate-specific
homeodomain protein Nkx3.1, Pdx1 (IDX1/IPF1) pan-
creatic and intestinal homeodomain TF, myelin tran-
scription factor 1 like, Myt 1 zinc finger involved in
primary neurogenesis as well as factors involved in the
ER stress such as CHOP and C/EBP alpha hetero-
dimers, and the hypoxia inducible factor bHLH/PAS
(HIFF/HIF1).
Computer-assisted analysis of the entire promoter
region predicted a CpG island located between nucleo-
tide
500 and the start codon, suggesting that SEL1L
may be transcriptionally regulated through changes in
the methylation status.
The SEL1L promoter results highly active in the
pancreatic beta and in the embryonic kidney cells
(Cattaneo et al., 2001b). The tissue-specific regulation
 SEL1L AND ITS STRUCTURAL AND FUNCTIONAL COMPLEXITY
of SEL1L promoter is also underlined by Odom
and his collaborators who demonstrated, by chromatin
immunoprecipitation (ChIP) assay and genomic micro-
array hybridization, that SEL1L promoter interacts
with two transcriptional regulators, the HNF1a HNF4a
in human pancreatic islet cells as well as in hepatocytes
(Odom et al., 2004). Both regulators are required for
the normal function of liver and pancreatic islets.
Mutations in these factors have been associated
with the type 3 form of maturity-onset diabetes of
the young (MODY3 and MODY1). The characteristics of
the SEL1L promoter make it a useful tool to efficiently
drive transcription of a therapeutic gene; indeed
two laboratories reported on the use of the SEL1L
minimal promoter to drive the expression of the cytosine
deaminase suicide gene for gastric cancer gene
therapy (Mathlouthi et al., 2003; Aberle et al., 2004).
Although these findings are very interesting and
promising, they require further in vitro studies in
addition to tumor transplants prior to any clinical
application.
Protein structure
SEL1L is a multimodular protein consisting of several
domains and signal sequences that confer the multi-
faceted specificities to the molecule.
The main domains are schematized in Figure 1D.
(i) The fibronectin type II domain is a small compact
two-disulphide-bond module of about 40 amino acid
residues located in the amino-terminal region (from
120 to 168 residues). It was initially identified in the
collagen-binding region of the matrix protein
fibronectin, and in some seminal fluid proteins
such as bovine seminal plasma proteins PDC-109
and BSP-A3. Related domains are found in the
extracytoplasmic parts of membrane-associated
proteins, such as members of the mannose recep-
tor-phospholipase A2 receptor family, mannose-6-
phosphate receptors. FNII modules are also pre-
sent in matrix metalloproteinases MMP-2 and
MMP-9, as well as serine protease factor XII
and hepatocyte growth factor activator. There is
evidence that high affinity interaction with col-
lagen is stabilized by the presence of multiple FNII
domains in fibronectin and MMP2 proteins (Pick-
ford et al., 2001). Despite their widespread occur-
rence in diverse vertebrate proteins, FNII modules
are absent in invertebrates, including C. elegans
and D. melanogaster. This domain is missing from
the SEL1L invertebrate orthologs arising only in
the chordate lineage, representing an interesting
aspect of the evolutionary history of SEL1L and
indicating a specific lineage function. This suggests
that in addition to the acquisition of an ulterior
function in higher organisms, SEL1L can exert its
function independently of the fibronectin domain,
and its presence may redirect or increase its
specificity.
It is worth mentioning that among the chordates,
only the mouse and rat SEL1L proteins show a
variant form consisting in the absence of the
FNII domain, likely due to alternative splicing
(NP_035474 and NP_808794, respectively).
(ii) The SEL-1-like repeats represent a subtype of
the tetratricopeptide repeat (TPR) sequence dis-
tributed in tandem along the central portion of the
protein and assembled in three large clusters
(see figure) (from 183 to 326, from 373 to 554,
Journal of Cellular Physiology DOI 10.1002/jcp
27
from 627 to 699 residues). The TPR repeats are
protein–protein interaction modules found in mul-
tiple copies in a wide variety of proteins having
different biological functions, the majority of them
are involved in cell cycle control, transcription and
splicing events, protein transport and import,
regulatory phosphate turnover, and protein folding
(Blatch and Lassle, 1999). They are often found in
protein complexes and may form amphipathic a-
helices that mediate protein–protein interactions
as well as the assemblage of multiprotein com-
plexes. The repeat consists of a degenerate
34 amino-acid motif that forms two amphipathic
a-helice subdomains (Das et al., 1998). SEL-1-like
repeats differ from standard TPRs in possessing a
variable number of amino acids between the two a-
helices (Ponting, 2000). They are highly conserved
during SEL1L taxonomic evolution reinforcing
their functional and structural importance. The
SMART’s nrdb database annotates 2258 SEL-1 like
domains in 500 proteins from bacteria to eukar-
yotes. Seventeen of the annotated proteins are
human SEL-1 like multi repeat proteins; the
majority of them do not have as yet a precise
function, with the exception of the Tg737 gene and
the Megalin binding protein that harbor three and
five repeats, respectively. The Tg737 is a candidate
polycystic kidney disease gene with tumor suppres-
sion activity in liver cancer (Isfort et al., 1997;
Richards et al., 1997). The Megalin is an adapter
protein with receptor signaling complex scaffold
activity involved in the signal transduction cascade
and cell communication. This protein probably,
through the SEL-1-like repeats, binds the receptor
tail, the transcriptional regulators, and compo-
nents of signal transduction cascades (Petersen
et al., 2003).
In the Gram-negative human pathogen Helico-
bacter pylori, the secreted cysteine-rich proteins
(HcpC) belong to the family of Sel1-like proteins
(Luthy et al., 2004). Hcp structure has been
proposed as a template for modeling SEL1-like
proteins because the conserved repeat geometry is
highlighted by the consensus sequences. If the
structural homology with the SEL1L protein need
to be proved, it is interesting to note that the overall
fold of Hcp proteins support the hypothesis of
possible multiple protein–protein interaction
among the TPR proteins and confirmed the peptide
binding site seen in eukaryotic TPR protein, such
as Hsp70/Hsp90 and PEX5 (Richter and Buchner,
2001). In this regard, Ponting also predicted
that the sel-1 repeats in Hrd3p of S. cerevisae may
possess a peptide-binding function similar to that of
TPRs in the Hop protein, a co-chaperone binding
heat-shock cognate protein 70 (Hsc70) (Ponting,
2000). In yeast, the Ydl203c protein, a substrate
for the E3 ubiquitin protein ligase Rsp5, contains
the PY motif, a potential target for the E3
ligase and seven terminal SEL1-like repeats (Kus
et al., 2005).
(iii) The Hrd3-like motif was originally identified in the
yeast Hrd3p protein, a component of ER membrane
associated ubiquitin ligase complex (Hampton
et al., 1996). It is the most highly conserved motif
consisting of 12 aa residues (from 664 to 675
residues) being shared by several divergent species
including mammals, insects, plants, and yeast. It
maps within a SEL1L region that plays a crucial
28 BIUNNO ET AL.
role in the transmission of anti-proliferative signals
(Cattaneo et al., 2004).
(iv) The proline-rich motif is an ‘‘adopter’’ system
that brings together proteins and mediates pro-
tein–protein interactions (Kay et al., 2000),
it is located in the COOH-terminal end of the
SEL1L protein (from 770 to 793 residues). These
sequences are found in situations requiring the
rapid recruitment or interchange of several pro-
teins, such as during initiation of transcription,
signaling cascades, and cytoskeletal rearrange-
ments. Thus, its role is not to provide a structurally
defined complex but rather to bring proteins
together in such a way that subsequent interactions
are more probable. The best-known examples of
modules that bind proline-rich region are repre-
sented by SH3, WW domains, and several new
protein interactions. The SEL1L proline stretch
may be better recognized by SH3 domains whose
optimal ligand preference varies around the PxxP
core, where x denotes any amino acids. This
terminal tail is an evolutionary acquisition of the
SEL1L protein since it is present only in the
chordates with the except of chicken and zebrafish.
SEL1L protein harbors several additional signal
sequences:
(i) A cleavable signal peptide consisting in 20 hydro-
phobic amino acids present at the N terminus and
followed by a cleavage site indicating that signal
peptide may not be retained.
(ii) An XXRR-like ER membrane retention signal
(RVRI) located in the first five aa residues.
(iii) A PEST sequence present (from aa 80 to 102
residues) known to target proteins for rapid
degradation.
(iv) A consensus sequences for nuclear localization
(from aa 107 to 113 residues) and export signals
(at position 9–18 and 747–755). These nuclear
signals mediate the binding to karyophein/importin
and karyophein/exportin proteins allowing for large
protein to shuffle in and out of the nucleus, and are
considered extremely important regulators of the
proteins sub-cellular location having a strong
impact on important processes related to cancer,
cell cycle, cell differentiation, and other fundamen-
tal aspects of cell viability (Kressel and Schmucker,
2002; Qu and Koromilas, 2004).
Using a novel strategy for quantitative glycoprotein
profiling, Zhang and his collaborators, reported that
SEL1L is a N-linked glycoprotein since SEL1L-N-
glycosylated peptide was identified in the crude mem-
brane fraction from the LNCaP prostate cancer epithe-
lial cell line (Zhang et al., 2003). The SEL1L glycosylated
peptide contains the motif NSSQ at position 272, three
additional glycosylated motifs are predicted at the N-
terminus: (NGSN), (NHTK), and (NETY). N-linked
glycosylation is prevalent in proteins destined for
extracellular environments, including extracellular
side of the plasma membrane, secreted proteins, and
proteins contained in body fluids, such as blood serum,
cerebrospinal fluid, urine, breast milk, saliva, lung
lavage, fluid, or pancreatic juice, and changes in the
extent of glycosylation and the carbohydrate structure
of these proteins have been shown to correlate with
cancer and other diseases, highlighting the clinical
importance of this modification as an indicator or
Journal of Cellular Physiology DOI 10.1002/jcp
effector of pathologic mechanisms. Although there is
no evidence indicating extracellular location of SEL1L,
computational, and experimental data documented
the existence of two protein isoforms, one of each
is predicted to be secreted, the other is indeed found
to be secreted by Clark et al. (2003); see Section
2.1), highlighting their relevant biological and clinical
implications.
SEL1L cross-species conservation
Comparative sequence analysis across different
regna, including metazoa, fungi, viridiplantae, and
bacteria, revealed the remarkable conservation of
its primary sequence, although the gene structural
complexity increased in evolution, suggesting differen-
tiation of its function (Table 2) (Biunno et al., 2002).
However, several components remain perfectly con-
served suggesting that the SEL1L protein exerts a very
important biological function and may belong to that
class of proteins considered to be ‘‘essential.’’ The
increasing availability of genome sequence data from a
variety of model organisms as well as higher vertebrates
strengthens the strong constrain to keep SEL1L. Among
mammals, SEL1L shares strict amino acid identity with
chimpanzee (99%), dog (97%), hamster (92%), mouse
(93%), and rat (92%). It also shows a good similarity with
the model organisms such as xenopus (82%), chicken
(83%), zebrafish (73%), Drosophila melanogaster (51%),
and C. elegans (46%) (Table 2). Arabidopsis thaliana and
Saccharomyces cerevisiae display lower similarity (34%
and 28%, respectively) with the high amino acid identity
in the C-terminal region containing and flanking the
Hrd3 motif.
SEL1L ALTERNATIVE ISOFORMS AND
EXPRESSION PATTERNS
SEL1L alternative splicing events
Beside the 7.5 kb SEL1L mRNA species, several
smaller alternated transcripts have been reported only
in pancreatic cells (Donoviel et al., 1998; Harada et al.,
1999). The Aceview program (August 2005) (http://
www.ncbi.nlm.nih.gov/AceView) supports the presence
of splice variants and predicts at least four different
transcripts putatively encoding for four protein isoforms
(Fig. 2A,B) and five shed variants with exon contact, but
not shared intron boundary; they are ignored in the
assessment of alternative features of the gene since
partial or unspliced variants. The two isoforms (B and C:
genebank accession no BM312853 and AY358651,
respectively) are a truncated version of SEL1L, due to
splicing events that occur within the ninth and eighth
exons, respectively, differing dramatically in the extent
of their 30 terminal ends and in the transcription
initiation start sites. These were experimentally
detected using specific primers by RT-PCR analysis
(Cattaneo et al., personal observations). Both isoforms
loose close to half of the SEL1L protein at the COOH-end
containing the middle and the final clusters of SEL-1
like repeats, the Hrd3 like motif, the transmembrane
region, and the proline tail. Although Aceview program
predicts a deletion in the first 64 nucleotides of the
fourth exon splicing out the beginning of the type II
fibronectin domain, sequence analysis performed on
transcript B extracted from the pancreatic neoplastic
cell line Suit-2 does not confirm this deletion (Cattaneo
et al., personal observations), suggesting the existence
of other tissue-specific variants. The splicing event
 SEL1L AND ITS STRUCTURAL AND FUNCTIONAL COMPLEXITY 29
TABLE 2. SEL1L cross-species conservation

Phylum Speciesa Identities aa Human and subject region Protein length (aa) Acc. number

Chordata SEL1L 100% 1–794 794 NP_005065

Homo sapiens 794/794 1–794
SEL1L 99% 17–794 817 XP_510102
Pan troglodytes 772/778 40–817
SEL1L 97% 1–769 794 XP_537530
Canis familiaris 753/769 1–751
SEL1L 92% 1–794 794 BAB12403
Mesocricetus auratus 736/794 1–794
SEL1L 93% 19–794 790 BAB23750
Mus musculus 726/776 20–790
SEL1L 92% 19–794 794 Q80Z70
Rat rattus 721/776 19–794
SEL1L 83% 24–780 798 XP_421303
Gallus gallus 642/766 25–783
SEL1L 73% 13–767 776 XP_697044
Danio rerio 558/755 16–750
Unknown protein 82% 76–767 822 AAH95916
Xenopus laevis 575/701 90–787
Arthropoda ENSANGP 57% 182–761 596 XP_310466
Anopheles gambiae 333/581 4–579
ENSANG 58% 166–761 659 XP_392802
Apis mellifera 354/601 52–647
CG10221 51% 169–763 819 NP_651179
Drosophila melanogaster 311/606 125–721
GA10167 51% 182–793 771 EAL27774
Drosophila pseudoobscura 315/614 137–739
Nematode Hyp. protein CBG23527 47% 182–768 685 CAE75512
Caenorhabditis briggsae 285/604 92–685
SEL1L 46% 182–768 685 NP_506144
Caenorhabditis elegans 281/604 92–685
Ascomycota Hyp. Protein FG00909.1 34% 206–705 821 XP_381085
Gibberella zeae 192/563 126–681
Hyp. Protein 32% 206–712 1307 XP_331768
Neurospora crassa 185/569 142–705
Hyp. Protein SPAC1B3 24% 202–692 700 NP_594794
Schizosaccharomyces pombe 140/568 86–629
Hrd3p 28% 413–726 833 NP_013308
Saccharomyces cerevisiae 97/335 373–690
Streptophyta SEL1L 35% 236–788 670 BAE02648
Glycine max 206/575 102–666
SEL1L 34% 212–712 678 NP_564049
Arabidopsis thaliana 178/517 83–588
a
The species with relevant biological importance and sharing the high identity with the human SEL1L protein are elencated in the table.

occurring in the eight exon of isoform C generates a
peroxisomal domain (SRL) at position 299–301.
Isoform D (genebank accession no. BE000519) is
reported to be truncated at the 30 end, posses three
exons (18, 19, and 20) encoding for a 72 amino acid
protein containing the functional Hrd3-like domain and
an ER membrane signal.
Supports on the presence of several SEL1L variants
arise also from recent reports and additional computer-
assisted programs.
Clark and collaborators through the secreted protein
discovery initiative strategy (SPDI), a large-scale
method that utilizes both biological and computational
methods, identified the SEL1L C variant (Genebank
AY358651) having the Type II fibronectin domain, a
transmembrane region, and secreted subcellular loca-
tion (Clark et al., 2003). This variant is also predicted by
Swiss-prot program (Q9UBV2-2) (Fig. 2A).
Recently, a new isoform derived from high throughput
cloning of full length human cDNAs from placenta
libraries optimized for large and rare transcripts, is
annotated in the Genbank databases (DR005068.1).
This isoform shares with SEL1L the common exons 1–6
(Fig. 2A).
Overall all these data provide evidences that SEL1L
gene does not encode for one unique expressed protein
but rather for several proteins, increasing the complex-
ity of its biology. Actually, the functional role of all these
isoforms is yet unknown, but it can be hypothesized
Journal of Cellular Physiology DOI 10.1002/jcp
that they may contribute to creation of the diversity
around one single gene in order to allow tissue and
temporal-dependent combinatorial patterns of protein
expression. The role of isoforms B and C may be
restricted to the protein–protein interaction and/
or binding since they retain only the first cluster of
SEL-1-like repeats as well as the fibronectin type II
domain; moreover the predicted extracellular location of
these two isoforms highlights their clinical importance
as useful biomarkers of body fluids, such as blood serum
and urine. The incomplete isoform D may represent the
only variant retaining the functional Hrd3 motif.
Multimodal approaches are essential for the full
characterization of these multiple isoforms in relation
to the cell physiology and SEL1L, since their function
may be restricted to a specific tissue, sub-cellular
compartments, or temporal conditions performing
related roles in various cellular contexts, such as
stimulations, phase of cell cycle, levels of interacting
proteins.
SEL1L expression in healthy adult human tissues
Previously works generated an expression archive of
SEL1L which determined the ubiquitous expression of
the gene only in fetal and neoplastic tissues (Cattaneo
et al., 2000). Northern blot analysis revealed that the
7.5 kb SEL1L transcript is almost always under-
represented in normal and healthy adult cells with the
exception of the pancreas (Biunno et al., 1997). SAGE
30
Fig. 2. A: A graphical representation of SEL1L isorforms. The black numbered rectangles correspond to
the exons, while the white rectangles correspond to the intronic sequence which is retained in the
alternative isoforms. The SEL1L domains are indicated on the top of the isoforms. B: A schematic
summary of SEL1L isoforms features.
analysis using a complete series of tissue sections from
normal and neoplastic cells (www.ncbi.nlm.nih.gov/
projects/SAGE/index.cgi?cmd¼accsearch tags: TAA-
GTTGAGT and TAAGTTGAGTGGAATGT), and the
expression profile documented in the mouse tissues
(Donoviel et al., 1998) confirm our previous observations
which strongly suggest that the function of the encoded
protein needs to be found within the pancreas. At
protein level, SEL1L immunohistochemical expression
is restricted to protein secreting cells with a typically
diffuse granular cytoplasmic positivity. These included
normal breast, pancreas, prostate, salivary glands, and
serous/mucinous glands associated with the gastroin-
testinal and ovary. In the pancreas, the protein is
abundantly present in the acini and in the islet of
Langherans, but completely absent in ductal cells
(Cattaneo et al., 2003). Among endocrine organs,
thyroid, adrenal, and insulae of Langherans are posi-
tive. Plasma cells and megacaryocytes also show intense
staining. Representative examples of immunostaining
of SEL1L are shown in Figure 3.
Immunohistochemical analysis on human tissues
reported in this review has been performed using the
murine monoclonal antibody MSel1 that has been raised
against the N-terminus of human recombinant SEL1L
protein (Orlandi et al., 2002b). The growth arresting
function exerted by the C-terminal region of SEL1L on
bacterial cells prevented the possibility to produce the
Journal of Cellular Physiology DOI 10.1002/jcp
BIUNNO ET AL.
entire recombinant SEL1L protein and select mono-
clonal antibodies directed against the Hrd3 motif
containing C-terminus domains of SEL1L protein.
MSel1 then recognizes all the possible SEL1L isoforms
listed in Figure 2A and the studies performed with this
antibody could not address the question of different
tissue expression or cellular sub-localization of the
SEL1L isoforms. The MSel1 expression pattern in fetal
and adult tissues is concordant with the expression
pattern reported in C. elegans (Grant and Greenwald,
1997) and mouse (Donoviel et al., 1998) using anti sel1-l
conventional polyclonal antisera.
SEL1L expression in human fetal tissues
SEL1L immunoreactivity varies in different fetal
tissues and at different gestational ages. Fetal tissues,
from 6 to 24 weeks (w) of gestational age, obtained
from 12 therapeutic or spontaneous abortions, were
submitted for routine histological analysis. In summary,
SEL1L shows a nucleo-cytoplasmic pattern of reactivity
in the early embryo and in most immature tissues;
at later ages of development the pattern of reactivity
is restricted and present also in glandular structures
with a pattern almost exclusively cytoplasmic. Repre-
sentative examples of immunostaining of SEL1L are
shown in Figure 3.
 SEL1L AND ITS STRUCTURAL AND FUNCTIONAL COMPLEXITY 31

Fig. 3. Representative examples of SEL1L immunostaining in adult and fetal tissues.

Early embryos (6–7 w), show SEL1L immunostaining
in the majority of cells in all fetal layers, with a more
intense reactivity in the neuroepithelial structures; the
pattern of staining is both nuclear and cytoplasmic. At
8–12 weeks, SEL1L expression is still widely present in
several tissues. In gastrointestinal tract, strong nuclear
and cytoplasmic staining is present in the mucosa; non-
epithelial cells of the muscular layer are also labeled,
although with weaker intensity. Ductal pancreatic
structures are identified in a 10 weeks fetus and show a
weak SEL1L staining with scattered cells more inten-
sely stained. Hepatocytes are intensely stained but
reactivity is only cytoplasmic; hematopoietic cells in the
liver are negative with the exception of megakariocytes.
In the respiratory tract, intense staining is observed in
bronchial epithelium showing a nucleo-cytoplasmic
Journal of Cellular Physiology DOI 10.1002/jcp
pattern of reaction. In the urogenital tract, epithelial
structures are weakly stained with glomerular epithe-
lial cells showing strong perinuclear reactivity. Gonads
are diffusely stained but germinal cells are negative.
Neural structures display SEL1L intense cytoplasmic
staining with a nucleo-cytoplasmic pattern of reaction in
more immature cells. Epithelial cells of the choroid
plexus are positive. Most mesenchymal cells are immu-
nostained. Nucleo-cytoplasmic reactivity is observed in
cartilage, skeletal, smooth muscle, and endothelial cells;
myocardial cells are only weakly positive. In more
mature fetal tissues (13–24 weeks), strong SEL1L
immunoreactivity is confined to selected tissues includ-
ing gastrointestinal mucosa and pancreas, while liver
cell positivity is very weak. Immunostaining is also
localized in newly formed structures, such as skin
32 BIUNNO ET AL.
glands, salivary and mucous mucosal glands, thyroid,
endometrium, thymic squamous epithelium, renal
tubular epithelium, some lymphoid element, but usually
only with a cytoplasmic pattern of reaction. In the
pancreas acini and Langherans cells are intensely
stained whereas ductal cells are only occasionally
positive.
In the placenta, intense staining is observed in the
syncitiotrophoblast, whereas cytotrophoblast is weakly
stained or negative; decidual cells and endometrial
glands are positive.
In agreement, RNA in situ analysis of developing
mouse embryos indicated that mSEL-1L was moder-
ately expressed throughout the neural tube and dorsal
root ganglia; particular high levels were observed in the
floor plate of the neural tube beginning at E10.5 and
increased at E11.5. High levels of expression were
observed at E14.5 and E17.5 in the acini of the pancreas
and moderate in the epithelial cells of the gut villi
(Donoviel et al., 1998).
In summary, both in fetal and adult tissues,
SEL1L immunohistochemical expression is restricted
to protein secreting cells indicating that SEL1L
may play a crucial role in secretory/trafficking/quality
control checkpoint processes occurring in proliferative
cells.
SEL1L POSSIBLE FUNCTION(S)
Although the real function of the SEL1L protein in
mammalian cells is still unknown, much can be learned
and deduced from the available nucleotide and protein
sequence, as well as from functional data obtained in
homologous proteins, such as sel-1 in C. elegans and
Hrd3p in S. cerevisiae.
SEL1L and the endoplasmic reticulum (ER)
Human SEL1L was indicated to be the human
candidate homolog of HRD3 (Kaneko and Nomura,
2003) suggesting that it may have a role in the ER-
associated protein degradation (ERAD) system.
ER facilitates the synthesis and processing of
almost all the secretory and membrane proteins,
and is susceptible to various forms of stress that
provoke the accumulation of unfolded or misfolded
proteins (Kaufman, 2002). ER stress can be induced by
a variety of pathophysiological conditions including
expression of mutant proteins, inhibition of asparagine
(N-linked glycosylation), overload of viral proteins
during virus replication (Kaufman, 2002).
Alterations in the homeostasis of the ER by various
forms of stress lead to the accumulation of unfolded
proteins and protein aggregates that are detrimental to
cell survival. Eukaryotic cells can adopt to ER stress by
activating specific signaling pathways and mechanisms
(upregulation of unfolded protein response [UPR] genes
especially those involved in the ERAD processes), whose
primary purpose is to limit the accumulation of unfolded
proteins in the ER. UPR entails the coordinated
transcriptional induction of genes encoding for ER
resident proteins (glucose regulated proteins Grps),
ER chaperones (Kaufman, 2002), and Hrd (Hrd1p and
3p) proteins (Kikkert et al., 2004). In order to decrease
the protein overload in ER, another pathway developed
by the cells is to inhibit protein synthesis (Ron, 2002).
The final pathway undertaken is to eliminate the
unfolded or misfolded proteins by proteasome-depen-
dent proteolysis (Kostova and Wolf, 2003). If cell
adaptation is not sufficient, then the stressed cells are
eliminated by apoptosis by activating the JNK pathway
Journal of Cellular Physiology DOI 10.1002/jcp
and caspases 7, 12, or 3 (Kaufman, 2002). The UPR
and ERAD genes have well been defined in yeast and in
C. elegans, these under normal growth conditions are not
essential for cell viability but are of paramount
importance during ER stress or when UPR is blocked.
One main function of Hrd3p in the ERAD system is
to affect the stabilization of Hrd1p, preventing the
cytosolic RING-H2 domain from programming Hrd1p
degradation possibly through auto-ubiquitination. Stu-
dies with truncated alleles and overexpression of
Hrd3p indicated that it is also required to modulate
Hrd1p RING-H2 ubiquitin ligase activity by physically
interacting with the Hrd1p transmembrane domain.
In the absence of subsrate, Hrd3p stabilizes Hrd1p
by binding to its transmembrane domain thus suppres-
sing the cytosolic RING-H2 domain ubiquitin ligase
activity. When the substrate is present, Hrd3p
senses the requirement for Hrd1p transmembrane
domain to activate Hrd1p function in a correct temporal
and spatial manner. It remains however to be estab-
lished if the human SEL1L interacts directly with
HRD1 to confer stability to the complex. Kikkert
showed that both endogenous and transfected human
HRD1 are relatively stable hence it is unlikely that
SEL1L ensures the stability of HRD1 in a manner
similar to that observed in yeast (Kikkert et al., 2004).
Soon, at least a partial map of all human protein–
protein interactions, ‘‘interactome,’’ will be available
making possible the understanding of how parts work
together.
Actually, studies performed on several organisms,
documented the SEL1L involvement in the UPR/ERAD
pathway.
(i) In C. elegans, sel-1 mediated interference (RNAi)
resulted in marked upregulation of the ER stress
indicator hsp-4::gfp (BIP) in a xbp-1 dependent
manner. Inactivation of sel-1 had no impact on the
viability of wild-type animals and only modestly
reduced the viability of xbp-1 mutants, but when
combined with the inactivation of abu-1 it increased
the lethality of xbp-1 mutants (Urano et al., 2002).
(ii) In pancreatic islet cells from Akita mouse, sub-
jected to ER-stress by the production of misfolded
insulin, was found an upregulation of the resident
molecular chaperone Bip as well as Hrd1 and Sel1l
proteins (Allen et al., 2004).
(iii) In Arabidopsis thaliana, by functional DNA micro-
array and real time PCR, it was observed an
increased expression of the At-SEL1L trancript in
tunicamycin-treated plants (Kamauchi et al.,
2005).
(iv) Studies on human embryonic kidney 293 cells,
demonstrated that treating the cells with ER-stress
inducing agents such as thapsigargin or tunicamy-
cin, resulted in an increase of HRD1 and SEL1L
mRNA levels, and both upregulation contribute to
protect the cells from ER-stress induced death by
degradating unfolded proteins accumulated in the
ER (Kaneko and Nomura, 2003).
(v) In humans, a possible role of SEL1L in ER-stress
processes, may be found in the proteomic and
microarray results obtained while studying the
breast cancer cell line MCF-7 containing the entire
SEL1L gene stably transfected and deliberately
induced to be expressed. The ectopic expression of
SEL1L revealed changes in the level of proteins and
transcripts involved in protein folding, and cell
stress response such as protein disulfide isomerase
 SEL1L AND ITS STRUCTURAL AND FUNCTIONAL COMPLEXITY
A3 (a major protein in the ER lumen, multifunc-
tional folding catalyst, and molecular chaperone),
proteasome b6 subunit, heat shock protein
60, and peroxiredoxin 1 (antioxodant enzyme
that regulates the cellular redox state). Indeed
seven transcript of proteasoma subunits were found
upmodulated in SEL1L overexpressing cells (Bian-
chi et al., 2005).
(vi) In a genome-wide analysis study of UPR response,
in fibroblasts from congenital disorders of glycosy-
lation type-1 patients, it was recently documented
that SEL1L expression is markedly increased along
the PERK inhibitor DNAJC3/P58 (IPK) genes
(Lecca et al., 2005).
SEL1L AND CANCER
Several studies have focused on the role of SEL1L in
many aspects of malignant transformation and tumori-
genic processes, providing significant in vitro and in vivo
evidences to link its increased expression to a decrease
in tumor aggressiveness.
Chromosome 14q24.3-31 deletions
and amplifications
Changes in DNA copy number contributes to cancer
pathogenesis, but alterations of the chromosomal region
14q24.3-q31 in cancer is not a very common event.
Recent studies reported that loss of heterozygosity
(LOH) on chromosome 14q is common in astrocytomas,
oligodendrogliomas, glioblastomas, and meningiomas
(Tse et al., 1997; Hu et al., 2002; Dichamp et al., 2004).
Two distinct minimal regions 14q22.3-14q24.3 (between
markers D14S276 and D14S74) and 14q31.3-14q32.1
(between markers D14S74 and D14S280) have been
reported to be deleted in two subgroups of gliomas
(astrocytomas, oligodendrogliomas, oligoastrocytomas)
and glioblastomas. The last region spans 13.5 Mb (from
77,728,449 to 91,272,543 ) contains, among 23 known
genes, SEL1L gene, 14 pseudogenes and 17 hypothetical
genes, and open reading frames. To our knowledge, no
tumor suppressor genes associated to this chromosomal
region have as yet been identified in these neoplasms.
Studies are in progress to determine whether SEL1L is
lost in oligodendroglial astrocytic tumors (Dichamp
et al., 2004). Allelic imbalance and loss of chromosome
14q were also found in head and neck squamous cell
carcinomas (HNSCCs), in primary colorectal carcino-
mas with metastatic ability as well as metastases, in the
renal cell carcinomas associated with neuroblastoma, in
the Korean ovarian carcinomas, and in the oral
squamous cell carcinomas (Lee et al., 1997; Medeiros
et al., 1999; Thorstensen et al., 2001; Bruder and Moch,
2004).
Mutation analysis and promoter polymorphisms
Mutations in the SEL1L gene were searched in
human normal and lung, pancreatic, and insulinoma
neoplastic tissues by direct sequencing but neither
causative nor functional mutations were found except
for the presence of two base substitutions in the minimal
promoter region in two well differentiated lung adeno-
carcinoma that led to a significant increase in the
transcription of the gene (Cattaneo et al., 2001) (Fig. 1C).
A polymorphic base substitution was reported in the
fibronectin type II domain of the gene in childrens
affected by persistent hyperinsulinemic hypoglycemia
(insulinoma) of infancy, which induces a major change
in the amino acid composition (Saltini et al., 2004). The
Journal of Cellular Physiology DOI 10.1002/jcp
33
absence of SEL1L mutations in cancer and normal cells
suggests that SEL1L must retain its integrity.
SEL1L differential expression in cancer
SEL1L is differentially expressed in cancer tissues
compared to normal counterpart. Analysis of a series of
human primary breast carcinomas and pancreatic
adenocarcinoma revealed downmodulation or absence
of SEL1L protein in about two-thirds of the tumors as
compared to normal epithelial cells (Orlandi et al.,
2002a; Cattaneo et al., 2003). The overall survival
analysis of breast carcinoma patients indicated a
statistically significant correlation between SEL1L
downmodulation and poor prognosis (Orlandi et al.,
2002a). All in ‘situ’ breast tumors show very strong
SEL1L positivity (Orlandi, personal communication). It
has been documented that during the transition from
hyperplasia to ‘in situ’ breast carcinoma, cancer cells
show higher levels of genomic instability and a higher
number of aberrations (Chin et al., 2004). If the genomic
instability is paralleled with the maximum production
of mutated proteins, then it can be hypothesized that the
increase in SEL1L transcription and translation is
required because the increase of intracellular protein
degradation.
In agreement, immunohistochemical analysis per-
formed on oesophageal, prostate, and non-small cell
lung cancer documented that SEL1L protein becomes
consistently expressed in the initial stages of cancer-
ogenesis and persists in neoplastic cells, making it a
useful biomarker of cell transformation with relevant
biological and clinical implications (Granelli et al., 2004;
Barberis et al., 2005; Ferrero et al., submitted). It is
worth noting the dual sub-cellular localization (nucleus/
cytoplasm/nucleo-cytoplasmic) of the SEL1L protein in
the majority of non-small cell lung carcinomas where it
is associated with tumor histotype, indeed the cytoplas-
mic immunostaning is preferentially observed in squa-
mous cell carcinomas while nuclear expression is
significant associated to adenocarcinomas (Ferrero
et al., submitted). Nucleocytoplasmic transfer of the
SEL1L protein may result from specific alternative
splicing exons, in response to several forms of genotoxic
insults. The differential subcellular location is evident
not only in neoplastic cells but also in fetal tissues,
reflecting a common regulatory mechanism and com-
mon SEL1L variant activation. However in healthy
adult tissues, the cytoplasmic location predominates
over the nucleus, indicating that different isoforms are
activated to elicit perhaps the same function in different
cell compartments. There are well known examples of
protein isoforms, i.e., the E3-ligase BRCA1, showing
completely different localization in the cytoplasm versus
nucleus between tumors and normal tissue, due to
different relative expression levels of isoforms (Wilson
et al., 1997). Overall, tissue integrity may be dependent
on the balanced expression level of different isoforms in
tissue and development-specific pattern.
Since no causative nor functional alterations were
found in the SEL1L genomic gene, the differential
representation of the gene product in different
cells and in particular in cancer may be due to both
post-transcription (methylation) and post-translation
(glycosilation or others) regulations. Significant is the
increasing levels of SEL1L protein in those cells under-
going transition from a benign to tumors stage as
reported for breast, lung, esophagus, and prostate
cancers (Orlandi et al., 2002a; Ferrero et al., submitted;
34 BIUNNO ET AL.
Granelli et al., 2004; Barberis et al., 2005). Unpublished
results show different methylation status of the single
CpG island located in the minimal promoter region of
the gene (Cattaneo, personal observations) among a
small series of pancreatic cancer cell lines.
SEL1L possible function in cancer
Two biological systems were used to initially evaluate
the role of SEL1L in cancer: the human breast MCF-7
and pancreatic Suit-2 cancer cell lines both transfected
with the entire SEL1L cDNA driven by an inducible
promoter (Orlandi et al., 2002a; Cattaneo et al., 2003). A
drastic reduction in anchorage-dependent growth
and colony formation in soft agar was observed in
both biological systems and this involved cell-matrix
communication, in pancreatic cancer cells was also
documented a decreased invasive ability to penetrate a
matrigel coated-filter, an inhibition of tumor growth in
immunodeficient mice and an alteration in the cell-cycle
progression (Cattaneo et al., in press).
The reversion of the pancreatic phenotype correlates
with the modulated expression of mediators involved in
the remodeling of the extra-cellular matrix, such as the
matrix metalloproteinases MMP1, MMP7, the inhibi-
tors of MMPs TIMP1, TIMP2 as well as the tumor
suppressor gene PTEN and members of TGF-b signal-
ing, such as Smad4, Activin A, and Activin receptor II
genes (Cattaneo et al., in press).
The proteomic approach and global expression screen-
ing performed on breast system identified in response to
SEL1L signaling, the modulation of the expression of
several proteins, and transcripts operating in different
signaling pathways (TGF beta and Notch pathway), in
cytoskeletal reorganization as well as tumor associated
proteins (Bianchi et al., 2005).
A deeper description of the Notch and TGF-b path-
ways in relation to SEL1L is outlined in Sections 4.4.1
and 4.4.2.
SEL1L and the NOTCH1 pathway. The Notch
pathway, although originally identified in fruit flies, is
now among the most heavily studied in mammalian
biology; it is best characterized as a mediator of the cell-
cell signaling between adjacent cells to generate cellular
heterogeneity. Considerable effort has been placed in
trying to dissect out the role of the proteins in signaling
pathways mediated by Notch. Genetic screens based
on phenotype or suppressor/enhancer activity have
identified a number of genes that influence lin-12/Notch
activity both in C. elegans and in Drosophila (Green-
wald, 1998). Sel-1 (suppressor-enhancer of Lin12/
Notch) is considered to be the C. elegans ortholog of
the human SEL1L. It was identified as an extragenic
suppressor of mutations that reduces lin-12, it behaves
as a negative regulator of the receptor activity
by controlling its turnover (Grant and Greenwald,
1996).
Notch activity and signaling is a highly conserved
evolutionary pathway known for decades to develop-
mental biologists. It is used not only by metazoans to
control their cell fate, but also by humans to influence
differentiation, proliferation, and apoptotic events
throughout all the developmental stages (Miele and
Osborne, 1999; Iso et al., 2003). Indeed there are only
few embryonic tissues that are not influenced by Notch
signaling. Four NOTCH genes are known in rodents and
humans differing in the number EGF-like repeats and
the length of the intracellular domain, however a
comprehensive discussion of Notch signaling is beyond
the scope of this review.
Journal of Cellular Physiology DOI 10.1002/jcp
In the mammalian central nervous system Notch
signaling is implicated in processes ranging from neural
stem cell regulation (to maintain the stem cells in a
progenitor state) to learning and memory (Yoon and
Gaiano, 2005). During the organogenesis of the pan-
creas, Notch signaling ‘‘locks’’ the pancreatic epithelial
cells in an undifferentiated progenitor like state while
maintaining their proliferative capacity through
Fgf10 signaling, and this appears to be critical for the
decision between the progenitor and endocrine fates.
Hart and collaborators reported that the persistent
expression of Fgf10 in the embryonic pancreas of
transgenic mice decreases sel-1 expression, suggesting
that reduced sel-1 expression may result in an atypical
maintenance of activated Notch thus impairing cell
differentiation (Hart et al., 2003).
Over the last 12 years, evidence has steadily accumu-
lated in relating deregulation of Notch signaling in
several malignancies, and it is being considered as a
potential target for therapeutic intervention (Nickoloff
et al., 2003). The first link was the 9:7 chromosomal
translocation associated with about 10% of T-cell
lymphoblastic leukemias, the translocation produces a
truncated Notch-1, resulting in the constitutive activa-
tion of the gene. The truncated version of the Notch
isoforms have transforming activity both in vitro and in
various animal model systems. Several solid tumors and
hematological malignancies, a subset of acute myeloid
leukemias and B-cell chronic lymphoid leukemias have
deregulated expression of wild-type Notch receptors,
ligands, and targets.
The role of human SEL1L in Notch signaling is
still largely unexplored, the only published data are
restricted to two expression studies. One of these studies
examined the transcriptional levels of Notch1, HES1,
and SEL1L genes in leukemia and lymphoma cell lines,
while the other reported the upmodulation of the jagged
2 and Notch 3 transcripts in breast mammary cancer
cells MCF-7 expressing ectopic SEL1L (Chiaramonte
et al., 2002; Bianchi et al., 2005). The first work reported
that SEL1L does not exert a negative regulatory
influence on Notch signaling since was not found an
inverse relationship between SEL1L expression and the
status of Notch signaling (Chiaramonte et al., 2002).
Despite this preliminary evidence, the possible func-
tional link between SEL1L and Notch activity may be
deeper investigated in vitro using more sophisticated
biological models (gain/loss of function genetic screens)
and in vivo during sequential stages of carcinogenesis
since SEL1L expression reflects the different phases of
cancer progression.
The results obtained from the breast model system
prompt to investigate the SEL1L involvement in Notch
signaling in breast tumor. Studies of mammary tumor-
igenesis induced by Notch in mouse and in vitro models
provide evidence that Notch activation is a causal
factor in human breast cancer (Politi et al., 2004). It
was reported that the subversion of Numb/Notch
biological antagonism through specific Numb ubiquiti-
nation and proteasomal degradation contributes to
human mammary carcinogenesis (Pece et al., 2004). It
is becoming increasing clear that the activity of Notch
and several receptor systems, is mediated by the
capacity of a cell to clear itself from the surface
receptors, their ligands and/or other extracellular
soluble molecules. Endocytosis plays an important
role and is required in both the signal-generating and
signal-receiving cell for appropriate Notch activation
(Polo et al., 2004). The ubiquitin-proteosome pathway
 SEL1L AND ITS STRUCTURAL AND FUNCTIONAL COMPLEXITY
appears to be a crucial role in the degradation of such
proteins and is therefore becoming a very important
therapeutic target for cancer (Orlowski and Dees, 2003;
Rajkumar et al., 2005).
It is then very important to explore the function of
SEL1L in the NOTCH pathway, as far as its capacity to
modulate Notch activity.
SEL1L and transforming growth factor-beta
(TGF-b) pathway. Studies performed on mammals
documented the involvement of SEL1L in the TGF-b
pathway. The first line of evidence was reported
by Furue et al. (2001) while working with the rat
RSMG-1 cells (a sub-mandibular gland epithelial cell
line); in this system SEL1L mRNA expression was
induced by Activin A (Furue et al., 2001). Second, it was
reported that SEL1L ability to reduce the aggressive
behavior of human neoplastic cells correlates with the
transcriptional modulation of genes belonging to the
TGF-b pathway: the inducible expression of SEL1L in
the pancreatic cancer cell line Suit-2 upmodulates the
transcriptional levels of endogenous Activin A and
Smad4 (Cattaneo et al., 2003), while SEL1L inactivation
by RNAi resulted in the downmodulation of Activin A
and Activin receptor II transcripts (Cattaneo et al., in
press).
Similar results were obtained in the MCF-7 cells in
which the ectopic expression of SEL1L resulted in the
modulation in the transcription of Activin A receptor IB
and the four signal transducers SMADs (Bianchi et al.,
2005).
The TGF-b superfamily consists of a disparate group
of polypeptide cytokines that regulate a plethora of
biological processes. TGF-b signaling proceeds from the
cell membrane to the nucleus through the cooperation of
type I and II serine/threonine kinase receptors, and
their downstream SMAD effectors. When TGF-b relays
team of signalers fails to function properly, the resulting
imbalance can lead to a variety of diseases, including
cancer, heart disease, and asthma. Too much of TGF-b
actually turns the tumor suppressor pathway into a
tumor-promoting pathway by: (i) promoting angiogen-
esis and (ii) suppressing T cells and other components of
the immune system. As favor to suppressor pathway,
Activin A exerts its effect on a wide range of cellular
targets by modulating cell differentiation and prolifera-
tion (Chen et al., 2002) appearing as a potent cell growth
inhibitor in the human breast cancer cell lines MCF-7
and T47D (Liu et al., 1996; Cocolakis et al., 2001).
Moreover, disruption or mutations in the TGF-b
components are involved in a high percentage of colon,
breast, and pancreatic cancers (Lei et al., 1996; Schutte
et al., 1996), and SMAD4 is inactivated by mutation or
homozygous deletion in about one-half of all pancreatic
cancers (Rozenblum et al., 1997). Furthermore, DPC4/
Smad4 re-expression in pancreatic carcinoma cell line
induced the cell to shift from potently angiogenic to
an antiangiogenic phenotype in vitro and in vivo by
modulating regulators of angiogenesis (Schwarte-
Waldhoff et al., 2000).
The intensity and duration of the TGF-b-Smad
response are important determinants for signaling
specificity, thus the activity of receptors and Smads
are carefully regulated. One mechanism of regulation is
endocytic clearance.
Recent work has shown that TGF-b signaling path-
ways can be compartmentalized through different
internalization routes of the TGF-b receptors. Whereas
clathrin-dependent TGF-b receptor internalization
into SARA-containing early endosomes promotes Smad
Journal of Cellular Physiology DOI 10.1002/jcp
35
signaling, internalization via lipid raft-caveolar com-
partments containing receptor bound to Smad7-Smurf2
results in accelerated receptor turnover by promoting
poly-ubiquitination (Di Guglielmo et al., 2003).
Considering the pancreatic biological system, SEL1L
may play a crucial role in the regulation of Smad7 and/or
Smuf2 levels, since Smad7-Smurf2 are involved in the
TGF-b pathway termination, SEL1L might be impor-
tant as an amplifier of TGF-b signaling.
Conclusion and prospective
This review has posed several questions concerning
the complexity of SEL1L in terms of its structure,
physiology, and function in normal developmental
processes as well as in cancer biology. Considering its
structure, we know it has a multimodular architecture
raising the important issue such as: (i) which and how
many proteins can interact with SEL1L. Are the three
main protein–protein interacting domains (fibronectin
type II, sel-1-like repeats, proline-rich motif) adept
to bind-specific substrates? These questions can be
addressed by assaying the entire protein as well as each
domain in yeast two-hybrid system, co-immunoprecipi-
tation, and pull down experiments. (ii) Are the nuclear
signals sufficient to shuffle the protein in and out of the
nucleus? By tagging the SEL1L nuclear signals with the
green fluorescent protein or peptide epitope tags will
help to dissect out the mechanisms that regulate the
nuclear/cytoplasmic transport.
As for the several computer predicted protein isoforms
as well as those experimentally isolated, it will be
important to understand their biological function in
relation to the complete SEL1L. Are these isoforms
restricted to specific sub cellular compartments? Are
they regulated by specific stimuli or temporal cell
conditions (developmental stages vs. adults; normal vs.
tumor)? Is there one or more substrates/templates per
isoform? Functional studies based on the use of fusion
protein isoforms with peptide epitope tags will aid to
clarify these biochemical and functional aspects.
Considering the overall results published on SEL1L
by various investigators working in different organisms,
it can, perhaps, safely be deduced that this gene plays a
fundamental role in eukaryotic intracellular protein
degradation. However, it still remains to be established
in which step of the ubiquitin proteosoma pathway
SEL1L plays part as well as which domain retains this
function. Since the C-terminal region contains the
hrd3 motif (the most conserved motif) than it is plausible
to hypothesize that it preserves the biological function.
Both yeast and C. elegans complementation experi-
ments have the potential to address this issue.
Arguably, the greatest difficulty is to understand how
human SEL1L works in the protein degradation
processes. The control of protein degradation is highly
sophisticated mechanism which controls the organisms
well being. Since it is a highly delicate and precise
process, the cell engages different kind of proteins (E3-
ligase, chaperones and so on) to work in the same
pathway finalized in the protein degradation. The type
of multi-protein complex that gets formed depends on
the kind of polypeptide, which needs to be degraded.
Thus, does SEL1L interact with Hrd1p or with the other
several E3 ligases described? Does the increased protein
complexity confer to SEL1L the multi-features to
participate to several chaperone/E3 ligase pathways?
Which are the classes of proteins that SEL1L target for
degradation (Notch, TGF-b pathways)? Is SEL1L regu-
lated by ubiquitination?
36 BIUNNO ET AL.
Once these questions will be answered (or at least a
few of these) than we will be able to tackle the issue of its
implication(s) in cancer. The fundamental question
raised by the observation that it gets upmodulated
during the early steps of tumor transformation is of
paramount importance for early diagnosis. Currently, it
is only possible to hypothesize that the increased SEL1L
levels are required in order to meet the advent of genetic
and/or genomic structural alterations acquired during
cancer initiation or to influence intra-cellular signaling.
Its presence may be important in protecting cellular
homeostasis from genetic mutations. This may be
explained by previous studies which revealed that
SEL1L specific knock-down by RNA interference
resulted in severe perturbation of bTC-3 growth and
metabolic activity (Diaferia et al., 2004), and by the
unsuccess in obtaining a SEL1L knock-out mouse
(Biunno personal communication), thus highlighting
the notion that the constitutive expression of SEL1L
may be essential for achieving and maintaining homeo-
static balance.
Malignant cells exhibit an exceptional ability to
maintain homeostasis in an hostile environment,
upregulation of proteins involved in the stress response
results in an increase in the apoptosis resistance, and
in diminishing cell senescence or permanent arrest of
cell division. Tumors are often subjected to hypoxia or
acidosis, conditions in which stress-response proteins
(UPR, PI3K/Akt, MEK1/ERK) enhance cell survival by
maintaining the normal protein-folding environment.
Acute activation of endogenous PI3K/Akt governs
cell survival by directly counteracting ER stress
induced cell death (Hu et al., 2004). By modulating
PTEN expression, a tumor suppressor which inhibits
PI3K-dependent signaling, SEL1L may elicit its anti-
tumor response by affecting survival mechanisms. In
addition to facilitating cell survival in stressfull
environmental challenges, UPR proteins also allow
tumor cells to tolerate alterations from within, includ-
ing mutation of critical signaling molecules that would
otherwise be lethal. Thus, UPR proteins can serve as
biochemical buffers for the genetic instability that is
characteristic of many human cancers, facilitating
and maintaining the transformed phenotype.
The increased expression of chaperone proteins
(Hps70,Hps90, Hps27) over the level seen in normal
tissues seems to be a common finding in many human
cancers, both solid and hematologic malignancies
(Conroy et al., 1998; Vargas-Roig et al., 1998; Ciocca
and Calderwood, 2005).
Protein degradation is becoming a central theme in
cancer biology and recently therapeutic approaches that
use inhibitors of proteins belonging to ubiquitin-proteo-
soma pathway have been developed in solid tumors and
hematological diseases (Orlowski and Dees, 2003;
Bagatell and Whitesell, 2004; Rajkumar et al., 2005).
SEL1L is a new entry in this process and although its
precise allocation within this pathway is still largely
unknown, of paramount importance is to identify the
networks controlled by SEL1L as well as the signals that
regulate its activity in order to develop specific ther-
apeutic approaches both in cancer as well as other
pathological disorders.
LITERATURE CITED
Aberle S, Schug N, Mathlouthi R, Seitz G, Kupper JH, Schroder K, Blin N.
2004. Promoter selection for the cytosine deaminase suicide gene constructs in
gastric cancer. Eur J Gastroenterol Hepatol 16:63–67.
Journal of Cellular Physiology DOI 10.1002/jcp
Allen JR, Nguyen LX, Sargent KE, Lipson KL, Hackett A, Urano F. 2004.
High ER stress in beta-cells stimulates intracellular degradation of misfolded
insulin. Biochem Biophys Res Commun 324:166–170.
Artavanis-Tsakonas S, Rand MD, Lake RJ. 1999. Notch signaling: Cell
fate control and signal integration in development. Science 284:770–776.
Review.
Bagatell R, Whitesell L. 2004. Altered Hsp90 function in cancer: A unique
therapeutic opportunity. Mol Cancer Ther 3:1021–1030. Review.
Ban Y, Taniyama M, Tozaki T, Yanagawa T, Tomita M, Ban Y. 2001. SEL1L
microsatellite polymorphism in Japanese patients with autoimmune thyroid
diseases. Thyroid 11:335–338.
Barberis M, Roz E, Biunno I. 2005. SEL1L expression in prostatic intraepithelial
neoplasia and adenocarcinomas: An immunohistochemical study. Histopatho-
logy (in press).
Bianchi L, Canton C, Bini L, Orlandi R, Menard S, Armini A, Cattaneo M, Pallini
V, Bernardi LR, Biunno I. 2005. Protein profile changes in the human breast
cancer cell line MCF-7 reponse to SEL1L gene induction. Proteomics 5:2433–
2442.
Biunno I, Appierto V, Cattaneo M, Leone BE, Balzano G, Socci C, Saccone S,
Letizia A, Della Valle G, Sgaramella V. 1997. Isolation of a pancreas-specific
gene located on human chromosome 14q31: Expression analysis in human
pancreatic ductal carcinomas. Genomics 46:284–286.
Biunno I, Bernard L, Dear P, Cattaneo M, Volorio S, Zannini L, Bankier A, Zollo
M. 2000. SEL1L, the human homolog of C. elegans sel-1: Refined physical
mapping, gene structure and identification of polymorphic markers. Hum
Genet 106:227–235.
Biunno I, Castiglioni B, Rogozin IB, DeBellis G, Malferrari G, Cattaneo M. 2002.
Cross-species conservation of SEL1L, a human pancreas-specific expressing
gene. OMICS 6:187–198.
Blatch GL, Lassle M. 1999. The tetratricopeptide repeat: A structural
motif mediating protein–protein interactions. Bioessays 21:932–939. Review.
Bruder E, Moch H. 2004. Pediatric renal cell carcinoma. Pathologe 25:324–
327.
Capobianco AJ, Zagouras P, Blaumueller CM, Artavanis-Tsakonas S, Bishop JM.
1997. Neoplastic transformation by truncated alleles of human NOTCH1/TAN1
and NOTCH2. Mol Cell Biol 17:6265–6273.
Cattaneo M, Orlandi R, Ronchini C, Granelli P, Malferrari G, Menard S, Biunno
I. 2000. The expression of SEL1L and TAN-1 in normal and neoplastic cells. Int
J Biol Markers 15:26–32.
Cattaneo M, Zollo M, Malferrari G, Orlandi R, D’Angelo A, Menard S, Biunno I.
2001a. Allelic polymorphisms in the transcriptional regulatory region of
human SEL1L. Mutat Res 458:71–76.
Cattaneo M, Sorio C, Malferrari G, Rogozin IB, Bernard L, Scarpa A, Zollo M,
Biunno I. 2001b. Cloning and functional analysis of SEL1L promoter region, a
pancreas-specific gene. DNA Cell Biol 20:1–9.
Cattaneo M, Orlandini S, Beghelli S, Moore PS, Sorio C, Bonora A, Bassi C,
Talamini G, Zamboni G, Orlandi R, Menard S, Bernardi LR, Biunno I, Scarpa
A. 2003. SEL1L expression in pancreatic adenocarcinoma parallels SMAD4
expression and delays tumor growth in vitro and in vivo. Oncogene 22:6359–
6368.
Cattaneo M, Canton C, Albertini A, Biunno I. 2004. Identification of a region
within SEL1L protein required for tumour growth inhibition. Gene 326:149–
156.
Cattaneo M, Fontanella E, Canton C, Delia D, Biunno I. SEL1L affects human
pancreatic cancer cell cycle and invasiveness through modulation of PTEN and
genes related to cell–matrix interactions. Neoplasia (in press).
Chen YG, Lui HM, Lin SL, Lee JM, Ying SY. 2002. Regulation of cell proliferation,
apoptosis, and carcinogenesis by activin. Exp Biol Med (Maywood) 227:75–87.
Review.
Chiaramonte R, Calzavara E, Basile A, Comi P, Sherbet GV. 2002. Notch signal
transduction is not regulated by SEL1L in leukaemia and lymphoma cells in
culture. Anticancer Res 22:4211–4214.
Chin K, de Solorzano CO, Knowles D, Jones A, Chou W, Rodriguez EG, Kuo WL,
Ljung BM, Chew K, Myambo K, Miranda M, Krig S, Garbe J, Stampfer M,
Yaswen P, Gray JW, Lockett SJ. 2004. In situ analyses of genome instability in
breast cancer. Nat Genet 36:984–988.
Ciocca DR, Calderwood SK. 2005. Heat shock proteins in cancer: Diagnostic,
prognostic, predictive, and treatment implications. Cell Stress Chaperones
10:86–103.
Clark HF, Gurney AL, Abaya E, Baker K, Baldwin D, Brush J, Chen J, Chow B,
Chui C, Crowley C, Currell B, Deuel B, Dowd P, Eaton D, Foster J, Grimaldi C,
Gu Q, Hass PE, Heldens S, Huang A, Kim HS, Klimowski L, Jin Y, Johnson S,
Lee J, Lewis L, Liao D, Mark M, Robbie E, Sanchez C, Schoenfeld J, Seshagiri
S, Simmons L, Singh J, Smith V, Stinson J, Vagts A, Vandlen R, Watanabe C,
Wieand D, Woods K, Xie MH, Yansura D, Yi S, Yu G, Yuan J, Zhang M, Zhang
Z, Goddard A, Wood WI, Godowski P, Gray A. 2003. The secreted protein
discovery initiative (SPDI), a large-scale effort to identify novel human secreted
and transmembrane proteins: A bioinformatics assessment. Genome Res
13:2265–2270.
Cocolakis E, Lemay S, Ali S, Lebrun JJ. 2001. The p38 MAPK pathway is
required for cell growth inhibition of human breast cancer cells in response to
activin. J Biol Chem 276:18430–18446.
Conroy SE, Sasieni PD, Fentiman I, Latchman DS. 1998. Autoantibodies to the
90 kDa heat shock protein and poor survival in breast cancer patients. Eur J
Cancer 34:942–943.
Das AK, Cohen PW, Barford D. 1998. The structure of the tetratricopeptide
repeats of protein phosphatase 5: Implications for TPR-mediated protein–
protein interactions. EMBO J 17:1192–1199.
Di Guglielmo GM, Le Roy C, Goodfellow AF, Wrana JL. 2003. Distinct endocytic
pathways regulate TGF-beta receptor signalling and turnover. Nat Cell Biol
5:410–421.
Diaferia G, Cattaneo M, Saltini G, Proverbio MC, Monferini E, Malferrari G,
Albertini A, Biunno I. 2004. RNA-mediated interference indicates that
SEL1L plays a role in pancreatic beta-cell growth. DNA Cell Biol 23:510–
518.
Dichamp C, Taillibert S, Aguirre-Cruz L, Lejeune J, Marie Y, Kujas M, Delattre
JY, Hoang-Xuan K, Sanson M. 2004. Loss of 14q chromosome in oligoden-
droglial and astrocytic tumors. J Neurooncol 67:281–285.
 SEL1L AND ITS STRUCTURAL AND FUNCTIONAL COMPLEXITY
Dissemond J, Busch M, Kothen T, Mors J, Weimann TK, Lindeke A, Goos M,
Wagner SN. 2004. Differential downregulation of endoplasmic reticulum-
residing chaperones calnexin and calreticulin in human metastatic melanoma.
Cancer Lett 203:225–231.
Donoviel DB, Donoviel MS, Fan E, Hadjantonakis A, Bernstein A. 1998. Cloning
and characterization of Sel-1l, a murine homolog of the C. elegans sel-1 gene.
Mech Dev 78:203–207.
Ellisen LW, Bird J, West DC, Soreng AL, Reynolds TC, Smith SD, Sklar J. 1991.
TAN-1, the human homolog of the Drosophila notch gene, is broken by
chromosomal translocations in T lymphoblastic neoplasms. Cell 66:649–
661.
Furue M, Zhang Y, Okamoto T, Hata RI, Asashima M. 2001. Activin A induces
expression of rat Sel-1l mRNA, a negative regulator of notch signaling, in rat
salivary gland-derived epithelial cells. Biochem Biophys Res Commun
282:745–749.
Gallahan D, Callahan R. 1997. The mouse mammary tumor associated gene INT3
is a unique member of the NOTCH gene family (NOTCH4). Oncogene 14:1883–
1890.
Granelli P, Cattaneo M, Ferrero S, Bottiglieri L, Bosari S, Fichera G, Biunno I.
2004. SEL1L and squamous cell carcinoma of the esophagus. Clinical Cancer
Res 10:5857–5861.
Grant B, Greenwald I. 1996. The Caenorhabditis elegans sel-1 gene, a negative
regulator of lin-12 and glp-1, encodes a predicted extracellular protein.
Genetics 143:237–247.
Grant B, Greenwald I. 1997. Structure, function, and expression of SEL-1, a
negative regulator of LIN-12 and GLP-1 in C. elegans. Development 124:637–
644.
Greenwald I. 1998. LIN-12/Notch signaling: Lessons from worms and flies. Genes
Dev 12:1751–1762. Review.
Hampton RY. 2002. ER-associated degradation in protein quality control and
cellular regulation. Curr Opinion Cell Biol 14:476–482.
Hampton RY, Gardner RG, Rine J. 1996. Role of 26S proteasome and HRD
genes in the degradation of 3-hydroxy-3-methylglutaryl-CoA reductase,
an integral endoplasmic reticulum membrane protein. Mol Biol Cell 7:2029–
2044.
Harada Y, Ozaki K, Suzuki M, Fujiwara T, Takahashi E, Nakamura Y, Tanigami
A. 1999. Complete cDNA sequence and genomic organization of a human
pancreas-specific gene homologous to Caenorhabditis elegans sel-1. J Hum
Genet 44:330–336.
Hart A, Papadopoulou S, Edlund H. 2003. Fgf10 maintains notch activation,
stimulates proliferation, and blocks differentiation of pancreatic epithelial
cells. Dev Dyn 228:185–193.
Hu J, Pang JC, Tong CY, Lau B, Yin XL, Poon WS, Jiang CC, Zhou LF, Ng HK.
2002. High-resolution genome-wide allelotype analysis identifies loss of
chromosome 14q as a recurrent genetic alteration in astrocytic tumours. Br J
Cancer 87:218–224.
Hu P, Han Z, Couvillon AD, Exton J. 2004. Critical role of endogenous Akt/IAPs
and MEK1/ERK pathways in counteracting endoplasmic reticulum stress-
induced cell death. J Biol Chem 279:49420–49429.
Isfort RJ, Cody DB, Doersen CJ, Richards WG, Yoder BK, Wilkinson JE, Kier LD,
Jirtle RL, Isenberg JS, Klounig JE, Woychik RP. 1997. The tetratricopeptide
repeat containing Tg737 gene is a liver neoplasia tumor suppressor gene.
Oncogene 15:1797–1803.
Iso T, Kedes L, Hamamori Y. 2003. HES and HERP families: Multiple effectors of
the notch signaling pathway. J Cell Physiol 194:237–255.
Kamauchi S, Nakatani H, Nakano C, Urade R. 2005. Gene expression in response
to endoplasmic reticulum stress in Arabidopsis thaliana. FASEB J 272:3461–
3476.
Kaneko M, Nomura Y. 2003. ER signaling in unfolded protein response. Life Sci
74:199–205.
Kaufman RJ. 2002. Orchestrating the unfolded protein response in health and
disease. J Clin Invest 110:1389–1398.
Kay BK, Williamson MP, Sudol M. 2000. The importance of being proline: The
interaction of proline-rich motifs in signaling proteins with their cognate
domains. FASEB J 14:231–241. Review.
Kikkert M, Doolman R, Dai M, Avner R, Hassink G, van Voorden S, Thanedar S,
Roitelman J, Chau V, Wiertz E. 2004. Human HRD1 is an E3 ubiquitin ligase
involved in degradation of proteins from the endoplasmic reticulum. J Biol
Chem 279:3525–3534.
Kostova Z, Wolf DH. 2003. For whom the bell tolls: Protein quality control of the
endoplasmic reticulum and the ubiquitin-proteasome connection. EMBO J
22:2309–2317. Review.
Kressel M, Schmucker B. 2002. Nucleocytoplasmic transfer of the NF2 tumor
suppressor protein merlin is regulated by exon 2 and a CRM1-dependent
nuclear export signal in exon 15. Hum Mol Genet 11:2269–2278.
Kus B, Gajadhar A, Stanger K, Cho R, Sun W, Rouleau N, Lee T, Chan D, Wolting
C, Edwards A, Bosse R, Rotin D. 2005. A high throughput screen to
identify substrates for the ubiquitin ligase Rsp5. J Biol Chem 280:29470–
29478.
Larsen ZM, Angelo AD, Cattaneo M, Nerup J, Biunno I, Zollo M, Pociot F. 2001.
Complete mutation scanning of the human SEL 1L gene: A candidate gene for
type 1 diabetes. Acta Diabetol 38:191–192.
Lecca MR, Wagner U, Patrignani A, Berger EG, Hennet T. 2005. Genome-wide
analysis of the unfolded protein response in fibroblasts from congenital
disorders of glycosylation type-I patients. FASEB J 19:240–242.
Lee DJ, Koch WM, Yoo G, Lango M, Reed A, Califano J, Brennan JA, Westra WH,
Zahurak M, Sidransky D. 1997. Impact of chromosome 14q loss on survival in
primary head and neck squamous cell carcinoma. Clin Cancer Res 3:501–
505.
Lei J, Zou TT, Shi YQ, Zhou X, Smolinski KN, Yin J, Souza RF, Appel R, Wang
S, Cymes K, Chan O, Abraham JM, Harpaz N, Meltzer SJ. 1996. Infrequent
DPC4 gene mutation in esophageal cancer, gastric cancer and ulcerative colitis-
associated neoplasms. Oncogene 13:2459– 2462.
Lewinski MK, Bisgrove D, Shinn P, Chen H, Hoffmann C, Hannenhalli S, Verdin
E, Berry CC, Ecker JR, Bushman FD. 2005. Genome-wide analysis of
chromosomal features repressing human immunodeficiency virus transcrip-
tion. J Virol 79:6610–6619.
Journal of Cellular Physiology DOI 10.1002/jcp
37
Liu QY, Niranjan B, Gomes P, Gomm JJ, Davies D, Coombes RC, Buluwela
L. 1996. Inhibitory effects of activin on the growth and morpholgenesis of
primary and transformed mammary epithelial cells. Cancer Res 56:1155–
1163.
Luthy L, Grutter MG, Mittl PR. 2004. The crystal structure of helicobacter
cysteine-rich protein C at 2.0 A resolution: Similar peptide-binding sites in TPR
and SEL1-like repeat proteins. J Mol Biol 340:829–841.
Mathlouthi R, Aberle S, Schug N, Kupper JH, Schroder K, Seitz G, Blin N. 2003.
Assessing optimal promoter activity for constructs in gastrointestinal gene
therapy. Anticancer Res 23:4011–4015.
Medeiros LJ, Palmedo G, Krigman HR, Kovacs G, Beckwith JB. 1999. Oncocytoid
renal cell carcinoma after neuroblastoma: A report of four cases of a distinct
clinicopathologic entity. Am J Surg Pathol 23:772–780.
Miele L, Osborne B. 1999. Arbiter of differentiation and death: Notch signaling
meets apoptosis. J Cell Physiol 181:393–409. Review.
Nickoloff BJ, Osborne BA, Miele L. 2003. Notch signaling as a therapeutic target
in cancer: A new approach to the development of cell fate modifying agents.
Oncogene 22:6598–6608. Review.
Nobrega MA, Zhu Y, Plajzer-Frick I, Afzal V, Rubin EM. 2004. Megabase
deletions of gene deserts result in viable mice. Nature 431:988–993.
Odom DT, Zizlsperger N, Gordon DB, Bell GW, Rinaldi NJ, Murray HL, Volkert
TL, Schreiber J, Rolfe PA, Gifford DK, Fraenkel E, Bell GI, Young RA. 2004.
Control of pancreas and liver gene expression by HNF transcription factors.
Science 303:1378–1381.
Orlandi R, Cattaneo M, Troglio F, Casalini P, Ronchini C, Menard S, Biunno I.
2002a. SEL1L expression decreases breast tumor cell aggressiveness in vivo
and in vitro. Cancer Res 62:567–574.
Orlandi R, Cattaneo M, Troglio F, Campiglio M, Biunno I, Menard S. 2002b.
Production of a monoclonal antibody directed against the recombinant SEL1L
protein. Int J Biol Markers 17:104–111.
Orlowski RZ, Dees EC. 2003. The role of the ubiquitination-proteasome
pathway in breast cancer: Applying drugs that affect the ubiquitin-
proteasome pathway to the therapy of breast cancer. Breast Cancer Res
5:1–7. Review.
Pear WS, Aster JC, Scott ML, Hasserjian RP, Soffer B, Sklar J, Baltimore D.
1996. Exclusive development of T cell neoplasms in mice transplanted
with bone marrow expressing activated Notch alleles. J Exp Med 183:2283–
2291.
Pece S, Serresi M, Santolini E, Capra M, Hulleman E, Galimberti V, Zurrida S,
Maisonneuve P, Viale G, Di Fiore PP. 2004. Loss of negative regulation by
Numb over Notch is relevant to human breast carcinogenesis. J Cell Biol
167:215–221.
Petersen HH, Hilpert J, Militz D, Zandler V, Jacobsen C, Roebroek AJ, Willnow
TE. 2003. Functional interaction of megalin with the megalinbinding protein
(MegBP), a novel tetratrico peptide repeat-containing adaptor molecule. J Cell
Sci 116:453–461.
Pickford AR, Smith SP, Staunton D, Boyd J, Campbell ID. 2001. The hairpin
structure of the (6)F1(1)F2(2)F2 fragment from human fibronectin enhances
gelatin binding. EMBO J 20:1519–1529.
Pociot F, Larsen ZM, Zavattari P, Deidda E, Nerup J, Cattaneo M, Chiaramonte
R, Comi P, Sabbadini M, Zollo M, Biunno I, Cucca F. 2001. No evidence for
SEL1L as a candidate gene for IDDM11-conferred susceptibility. Diabetes
Metab Res Rev 17:292–295.
Politi K, Feirt N, Kitajewski J. 2004. Notch in mammary gland development and
breast cancer. Semin Cancer Biol 14:341–347. Review.
Polo S, Pece S, Di Fiore PP. 2004. Endocytosis and cancer. Curr Opin Cell Biol
16:156–161. Review.
Ponting CP. 2000. Proteins of the endoplasmic-reticulum-associated degradation
pathway: Domain detection and function prediction. Biochem J 351:527–535.
Qu L, Koromilas AE. 2004. Control of tumor suppressor p53 function by
endoplasmic reticulum stress. Cell Cycle 3:567–570.
Rajkumar SV, Richardson PG, Hideshima T, Anderson KC. 2005. Proteasome
inhibition as a novel therapeutic target in human cancer. J Clin Oncol 23:630–
639. Review.
Richards WG, Yoder BK, Isfort RJ, Detilleux PG, Foster C, Neilsen N, Woychik
RP, Wilkinson JE. 1997. Isolation and characterization of liver epithelial cell
lines from wild-type and mutant TgN737Rpw mice. Am J Pathol 150:1189–
1197.
Richter K, Buchner J. 2001. Hsp90: Chaperoning signal transduction. J Cell
Physiol 188:281–290.
Ron D. 2002. Translational control in the endoplasmic reticulum stress response.
J Clin Invest 110:1383–1388. Review.
Rozenblum E, Schutte M, Goggins M, Hahn SA, Panzer S, Zahurak M, Goodman
SN, Sohn TA, Hruban RH, Yeo CJ, Kern SE. 1997. Tumor-suppressive
pathways in pancreatic carcinoma. Cancer Res 57:1731–1734.
Saltini G, Proverbio MC, Malferrari G, Biagiotti L, Boettcher P, Dominici R,
Monferini E, Lorenzini E, Cattaneo M, Antonello D, Moore PS, Zamproni I,
Viscardi M, Chiumello G, Biunno I. 2004. Identification of a novel polymorph-
ism in the fibronectin type II domain of the SEL1L gene and possible relation to
the persistent hyperinsulinemic hypoglycemia of infancy. Mutat Res 554:159–
163.
Schutte M, Hruban RH, Hedrick L, Cho KR, Nadasdy GM, Weinstein CL, Bova
GS, Isaacs WB, Cairns P, Nawroz H, Sidransky D, Casero RA, Jr., Meltzer PS,
Hahn SA, Kern SE. 1996. DPC4 gene in various tumor types. Cancer Res
56:2527–2530.
Schwarte-Waldhoff I, Volpert OV, Bouck NP, Sipos B, Hahn SA, Klein-Scory S,
Luttges J, Kloppel G, Graeven U, Eilert-Micus C, Hintelmann A, Schmiegel W.
2000. Smad4/DPC4-mediated tumor suppression through suppression of
angiogenesis. Proc Natl Acad Sci USA 97:9624–9629.
Stavridi ES, Halazonetis TD. 2004. p53 and stress in the ER. Genes Dev 18:241–244.
Stefani M. 2004. Protein misfolding and aggregation: New examples in medicine
and biology of the dark side of the protein world. Biochim Biophys Acta 1739:5–
25. Review.
Thorstensen L, Qvist H, Nesland JM, Giercksky KE, Lothe RA. 2001.
Identification of two potential suppressor gene regions on chromosome arm
14q that are commonly lost in advanced colorectal carcinomas. Scand J
Gastroenterol 36:1327–1331.
38 BIUNNO ET AL.
Tse JY, Ng HK, Lau KM, Lo KW, Poon WS, Huang DP. 1997. Loss of
heterozygosity of chromosome 14q in low- and high-grade meningiomas.
Hum Pathol 28:779–785.
Urano F, Calfon M, Yoneda T, Yun C, Kiraly M, Clark SG, Ron D. 2002. A
survival pathway for Caenorhabditis elegans with a blocked unfolded protein
response. J Cell Biol 158:639–646.
Vargas-Roig LM, Gago FE, Tello O, Aznar JC, Ciocca DR. 1998. Heat shock
protein expression and drug resistance in breast cancer patients treated with
induction chemotherapy. Int J Cancer 79:468–475.
Journal of Cellular Physiology DOI 10.1002/jcp
Wilson CA, Payton MN, Elliott GS, Buaas FW, Cajulis EE, Grosshans D, Ramos
L, Reese DM, Slamon DJ, Calzone FJ. 1997. Differential subcellular
localization, expression and biological toxicity of BRCA1 and the splice variant
BRCA1-delta11b. Oncogene 14:1–16.
Yoon K, Gaiano N. 2005. Notch signaling in the mammalian central nervous system:
Insights from mouse mutants. Nat Neurosci 8:709–715. Review.
Zhang H, Li XJ, Martin DB, Aebersold R. 2003. Identification and quantification
of N-linked glycoproteins using hydrazide chemistry, stable isotope labeling
and mass spectrometry. Nat Biotechnol 21:660–666.