Domestic Animal Endocrinology 44 (2013) 139–144
Contents lists available at SciVerse ScienceDirect
Domestic Animal Endocrinology
journal homepage: www.domesticanimalendo.com
Relevance of sodium/glucose cotransporter-1 (SGLT1) to diabetes
mellitus and obesity in dogs
D.J. Batchelor a, *, A.J. German b, S.P. Shirazi-Beechey a
a
Epithelial Function and Development Group, Department of Functional and Comparative Genomics, Institute of Integrative Biology, Faculty of Health
and Life Sciences, University of Liverpool, Liverpool L69 7ZJ, UK
b
Small Animal Teaching Hospital, University of Liverpool, Leahurst, Neston, Wirral CH64 7TE, UK
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 1 September 2012
Received in revised form 18 December 2012
Accepted 29 December 2012
Glucose transport across the enterocyte brush border membrane by sodium/glucose
cotransporter-1 (SGLT1, coded by Slc5a1) is the rate-limiting step for intestinal glucose
transport. The relevance of SGLT1 expression in predisposition to diabetes mellitus and to
obesity was investigated in dogs. Cultured Caco-2/TC7 cells were shown to express SGLT1
in vitro. A 2-kbp fragment of the Slc5a1 50 flanking region was cloned from canine genomic
DNA, ligated into reporter gene plasmids, and shown to drive reporter gene expression in
these cells above control (P < 0.001). To determine the effect of the 3 known SNPs in this
region on promoter function, new promoter/reporter constructs (all permutations of these
3 SNPs) were created by site-directed mutagenesis. No significant differences in promoter
function were seen, suggesting that these SNPs do not have a significant effect on the
constitutive transcription of SGLT1 mRNA in dogs. A search for novel SNPs in this region in
dogs was made in 2 breeds predisposed to diabetes mellitus (Samoyed and cairn terrier), 2
breeds that rarely develop diabetes (boxer and German shepherd), and 2 breeds predisposed to obesity (Labrador retriever and cocker spaniel). The Slc5a1 50 flanking region was
amplified from 10 healthy individuals of each of these breeds by high-fidelity PCR with the
use of breed-labeled primers and sequenced by pyrosequencing. The sequence of the
Slc5a1 50 flanking region in all individuals of all breeds tested was identical. On this
evidence, variations in Slc5a1 promoter sequence between dogs do not influence the
pathogenesis of diabetes mellitus or obesity in these breeds.
Ó 2013 Elsevier Inc. All rights reserved.
Keywords:
Genetic variation
Predisposition
Intestinal function
Transporter
1. Introduction
Understanding the genetic background to disease
improves our understanding of disease pathogenesis and
helps identify molecular targets for treatment. Diabetes
mellitus (DM) and obesity are prevalent and important
diseases in humans and dogs, but their genetic background
is extremely complex and mostly unknown, despite enormous research effort [1,2]. Purebred dogs are useful in
genetic studies because they exist in extremely isolated
populations (breeds). The susceptibility of some breeds to
* Corresponding author. Tel.: þ44 1517944255; fax: þ44 1517944244.
E-mail address: [email protected] (D.J. Batchelor).
0739-7240/$ – see front matter Ó 2013 Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.domaniend.2012.12.001
certain diseases, with much lower incidence in other
breeds, can be exploited when looking for genes involved in
complex diseases, because genes conferring risk are concentrated in the predisposed breeds [3].
Diabetes mellitus is common in dogs; most cases are
insulin deficient and canine DM often shares many
features with human latent autoimmune diabetes of
adults (LADA) [4,5]. Breed predispositions are recognized
in canine DM [6–10], evidence that there is a genetic basis
for it as in the human disease, and many of the genes
associated with human type 1 diabetes have also been
associated with DM in the dog [11–14]. No canine equivalent of human type 2 diabetes exists, but obesity is
another common disorder of dogs in which intestinal
140
D.J. Batchelor et al. / Domestic Animal Endocrinology 44 (2013) 139–144
glucose transport could be important [15,16]. Breed associations with obesity are recognized in dogs [16–21], but
no studies investigating the genetic basis for these associations have been reported.
The intestinal tract has been relatively overlooked in
diabetes and obesity research, which is unfortunate
because the site of nutrient absorption is clearly a key part
of energy balance. Glucose and galactose are transported
across the brush border membrane (BBM) of enterocytes
by sodium/glucose cotransporter-1 (SGLT1), coded for by
Slc5a1. Sodium/glucose cotransporter-1 is the sole route
for entry of glucose into the body, and the level of SGLT1
expression dictates BBM transport capacity for glucose. So
far, genes for intestinal sugar transporters have not been
implicated in DM or obesity, but they are good candidate
genes for DM because absorbed dietary carbohydrate
places a burden on the b cell and for obesity because much
of the energy absorbed from the diet is in the form of
glucose.
In this study, the role played in predisposition to DM or
obesity by polymorphisms in the promoter region of the
gene coding for SGLT1 was investigated in dogs. The aims
were to 1) clone the 50 flanking region of canine Slc5a1 to
create a promoter/reporter gene construct and to determine the activity of this wild-type promoter by measuring
reporter gene activity in cultured cells expressing SGLT1
in vitro; 2) identify known SNPs in the canine Slc5a1
50 flanking region; 3) assess the effect of these SNPs on
promoter function, alone or in combination, by performing
site-directed mutagenesis to create new promoter/reporter
constructs; and 4) search for novel SNPs in well-defined
samples of dogs with various risk of diabetes or obesity.
2. Materials and methods
2.1. Culture of cells in vitro
Cells (Caco2/TC7 cells, a human cell line) were maintained at 37 C in 75-cm2 flasks (Appleton Woods, Birmingham, UK) in a humid environment of air supplemented
with 5% CO2, in Dulbecco’s modified Eagles Medium (Sigma,
Dorset, UK; Cat. no. D6546), with added heat-inactivated
fetal bovine serum (10%), nonessential amino acids (1%;
Sigma), L-glutamine (2 mM) and penicillin/streptomycin
(100 U/mL; 100 mg/mL). Passage numbers 35 to 38 were
used.
2.2. Preparation of BBMV from cultured cells
Brush border membrane vesicles (BBMV) were prepared
with the use of a combination of cation precipitation and
differential centrifugation [22]. Frozen harvested cells were
thawed in buffer that contained 100 mM mannitol and 2
mM HEPES/Tris pH 7.1. Magnesium chloride was added to
a final concentration of 10 mM, and the homogenate was
stirred, on ice, for 20 min. Samples were centrifuged at
3,000 g, for 10 min, at 4 C. The pellet was discarded, and
the supernatant fluid was centrifuged at 38,000 g for 45
min at 4 C. The pellet was resuspended in buffer containing
300 mM mannitol, 20 mM HEPES/Tris pH 7.4, 0.1 mM
MgSO4, and 0.02% (wt/vol) NaN3.
2.3. SDS-PAGE and Western immunoblot analysis
Protein components of BBMV were separated by electrophoresis on 8% acrylamide gels at constant current of
12 mA per gel and electrotransferred to polyvinyl difluoride
membrane. Membranes were blocked in PBS that contained 0.5% weight/volume skimmed milk protein (Oxoid,
Basingstoke, UK) and 0.05% volume/volume Tween 20
(GE Healthcare, Buckinghamshire, UK). The antibody to
SGLT1 was custom raised in rabbits to a synthetic peptide
(STLFTMDIYTKIRKKASEK) that corresponded to amino
acids 402 to 420 of SGLT1, an intracellular loop region that
is conserved among various species [23]. Membranes were
incubated in a 1:1,000 dilution of rabbit anti-SGLT1 serum
at 20 C for 1 h, rinsed with PBS, then incubated in a 1:2,000
dilution of horseradish peroxidase-conjugated pig antirabbit IgG (Dako, Ely, UK) for 1 h. Membranes were
developed with the enhanced chemiluminescence system
(GE Healthcare). Bands from Western blot analysis were
quantified with scanning densitometry (Hewlett Packard
Precision Scan, HP2700, and Phoretix 1D Quantifier,
Nonlinear Dynamics, Newcastle, UK).
2.4. Isolation of genomic DNA from canine intestinal tissue
Frozen mucosal scrapings (0.5 g) from the jejunum of
a dog were ground into fragments and defrosted in nuclei
isolation buffer that contained 60 mM KCl, 15 mM NaCl,
15 mM Tris/HCl pH 7.5, 0.5 M sucrose, 0.5 mM EGTA,
2.0 mM EDTA, 0.5 mM spermidine, 0.15 mM spermine,
0.5 mM b-mercaptoethanol, 5 mM dithiothreitol, and 1 mM
phenylmethylsulfonyl fluoride. Tissue fragments were disrupted with a 2-mL Dounce homogenizer, filtered through
2 layers of nylon gauze, overlaid onto a 10-mL cushion of
nuclei isolation buffer/20% (vol/vol) glycerol, and centrifuged at 3,500 g for 30 min at 4 C. The pellet was washed
with nuclei isolation buffer and centrifuged at 12,000 g
for 2 min at 4 C. The pellet was suspended in digestion
buffer that contained 100 mM NaCl, 10 mM Tris/HCl, pH 8.0,
25 mM EDTA, 0.5% weight/volume SDS, and proteinase
K (final concentration 0.1 mg/ml) and incubated at 50 C for
16 h, with shaking. Nucleic acids were extracted with
1:24:25 isoamyl alcohol/chloroform/buffered phenol,
pH 8.0, with RNAse A digestion at 10 mg/mL at 37 C for 1 h.
DNA was precipitated with 1/10 volume 3 M sodium
acetate and 2 volumes 100% ethanol. The sample was
centrifuged at 3,000 g for 15 min at 4 C, washed in 70%
ethanol and centrifuged at 3,000 g for 5 min. The DNA
pellet was air-dried, then dissolved in 10 mM Tris/HCl at
65 C to a concentration of w1 mg/mL.
2.5. Cloning the canine Slc5a1 promoter and ligation into
pGL3 Basic plasmid vector
Amplification of DNA for cloning was performed with
PCR primers designed with the 50 adaptor GCGCGTCGAC,
containing the SalI recognition sequence G/TCGAC. Polymerase chain reaction was performed in 50-mL volumes in
a thermocycler with the use of 2 U of Velocity DNA polymerase (Bioline, London, UK), dNTP mix (final concentration 200 mM each), dimethylsulfoxide 5% volume/volume,
D.J. Batchelor et al. / Domestic Animal Endocrinology 44 (2013) 139–144
sense and antisense primers (2 nM each), and 50 ng of
genomic DNA. Cycling conditions comprised an initial
denaturation step (95 C, 2 min) followed by 25 to 30 cycles
of (95 C for 30 sec, 60 C for 30 sec, 72 C for 60 sec) and
a final extension step of 72 C for 5 min. Amplicons were
gel-purified and digested with SalI to create phosphorylated 50 overhangs. pGL3 Basic plasmid vector (Promega,
Southampton, UK) containing lucþ, a luciferase reporter
gene, was linearized by digestion with XhoI, creating 50
overhangs (TCGA) compatible with the SalI overhangs. To
prevent re-ligation of pGL3 plasmid, the 50 overhangs
created by digestion with XhoI were dephosphorylated by
treatment with calf intestinal alkaline phosphatase. Purified DNA fragments bearing 50 overhangs were ligated into
dephosphorylated, linearized pGL3 with the use of T4 DNA
ligase at 4 C for 16 h.
2.6. Transformation of Escherichia coli cells
Freshly thawed competent JM109 E coli cells (50 mL)
were mixed with 50 ng of ligated plasmids, incubated on
ice for 20 min, subjected to heat shock at 42 C for 45 sec,
then incubated on ice for 5 min. Growth medium (450 mL)
was added to the tube, followed by incubation at 37 C for
60 to 90 min, with shaking at 200 oscillations/min. Bacterial suspension (100 mL) were added to agar/ampicillin
plates and incubated overnight at 37 C. Single colonies
were inoculated into 5 mL of growth medium, incubated at
37 C for 16 h, with shaking at 200 oscillations/min. Clones
containing the DNA insert of interest were identified by
PCR. Glycerol stock cultures were created by adding 800 mL
of bacterial suspension to 200 mL of sterile glycerol in
a 2-mL cryotube before storage at 80 C.
2.7. Transfection
Bacterial clones in glycerol stock culture were inoculated into 5 mL of LB medium that contained 50 mL of
10 mg/mL ampicillin, and incubated for 16 h at 37 C, with
shaking at 200 oscillations/min. Plasmids were extracted
by alkaline lysis of bacterial cells, followed by adsorption of
plasmid DNA onto a silica membrane in the presence of
high salt with the use of the Qiaprep Spin Minikit (Qiagen,
Crawley, UK).
Caco-2/TC7 cells were seeded at a density of 5.3 105
cells/cm2 in 75-cm2 flasks on day 0. On day 7, cells were
trypsinized and seeded into 24-well plates at a density of
8 105 cells per well in 1 mL of complete medium. On day 8,
transient transfection was performed with 2 mg of plasmid
DNA per 1-cm2 well, 0.5 mg of Renilla reniformis luciferase
plasmid DNA (PRL-TK; Promega) per well, and 3 mL of
cationic lipid transfection reagent (Lipofectamine 2000;
Invitrogen, Paisley, UK) per 1 mg of DNA, according to the
manufacturer’s instructions. Medium was replaced on days
9 and 10, and luciferase activity was measured on day 11.
2.8. Measurement of luciferase activity in transfected cells
Transiently transfected cells were washed with 1 mL
warmed PBS, 200 mL 1 Passive Lysis Buffer (Promega) was
added to each well, and the plates were incubated for
141
30 min at 37 C. Cell lysates were transferred to fresh
Eppendorf tubes. Luciferase activity was measured in
a luminometer with the use of Luciferase Assay Reagent II
(Promega) and Stop & Glo reagent (Promega) according to
the manufacturer’s instructions.
2.9. Site-directed mutagenesis
Site-directed mutagenesis was performed in 50-mL
volumes in a thermocycler with the use of 2 U of Velocity
DNA polymerase (Bioline), dNTP mix (final concentration
200 nM each), dimethylsulfoxide 5% volume/volume,
125 ng of each mutagenesis primer, and 50 ng of PGL3 basic
plasmid construct to be mutated. Cycling conditions
comprised an initial denaturation step (95 C, 2 min) followed by 18 cycles of (95 C for 30 sec, 55 to 65 C for 30 sec,
72 C for 4 min) and a final extension step of 72 C for 7 min.
Polymerase chain reaction products were treated with DpnI
at 37 C for 4 h to digest remaining parental (ie, bacterial)
DNA, gel-purified, and used to transform competent JM109
E coli cells as above. Overnight cultures of resultant colonies
were performed as above, with clones of interest identified
by PCR. Plasmids were extracted from positive clones as
described above and submitted for custom sequencing
(Eurofins).
2.10. Search for new SNPs in the canine Slc5a1 promoter
breed selection
To maximize the possibility of finding a SNP with a relevant effect on SGLT1 expression, certain dog breeds were
chosen: 2 breeds predisposed to DM, 2 breeds that rarely
develop DM, and 2 breeds predisposed to obesity. The
Samoyed and cairn terrier were chosen as breeds at high risk
of developing DM, and the boxer and German shepherd were
chosen as breeds apparently protected from development
of DM [9,12,24]. The Labrador retriever and cocker spaniel
were chosen as breeds at increased risk of developing
obesity [16–18]. Boxers and German shepherds are reported
to have low relative risk of developing obesity [17,18].
2.11. Creation of an amplicon library for sequencing the
Slc5a1 promoter in 6 dog breeds
Polymerase chain reaction primers were designed to
allow amplification of overlapping w400-bp regions of the
canine Slc5a1 promoter. Breed-specific primers were
created by assigning each breed a unique 10-base multiplex
identifier (MID) code. These 10 base codes were added to
the 50 end of all primers so that all PCR amplicons would be
labeled with a code according to breed. Genomic DNA from
60 healthy dogs (10 individuals of each of 6 breeds) was
obtained from the UK Companion Animals DNA Archive.
High-fidelity PCR was performed on each dog’s DNA to
amplify separately each of 6 overlapping regions of the
Slc5a1 promoter. Each PCR product was gel-purified individually and quantified. The resulting 360 purified PCR
products, now labeled with MIDs according to breed, were
mixed in equimolar amounts in a single tube. Five micrograms of this mixture was reduced to 10-mL volume in
a vacuum centrifuge and high-throughput sequencing
142
D.J. Batchelor et al. / Domestic Animal Endocrinology 44 (2013) 139–144
(454 sequencing; GS FLX Titanium; Roche, Branford, CT,
USA) was performed at the School of Biological Sciences,
University of Liverpool.
2.12. Statistical analysis
Commercial software (Prism 5; GraphPad Software, San
Diego, CA) was used for statistical analysis. Continuous
variables were shown to be normally distributed with the
use of the Shapiro-Wilk method. Comparison between
groups was performed with the Student t test or one-way
ANOVA, as appropriate. The level of statistical significance
was set at P < 0.05.
3. Results
3.1. Caco-2/TC7 cells express SGLT1
Confluent Caco-2/TC7 cells were harvested, frozen, and
used to prepare BBMV. Protein components of BBMV were
separated by SDS-PAGE and transferred to polyvinyl
difluoride membranes, and subsequent Western blot
analysis confirmed expression of SGLT1 protein (Fig. 1).
Brush border membrane vesicles prepared from the intestine of a rabbit (Dr J. Dyer, University of Liverpool) were
used as a positive control for these blots.
3.2. Identification of known SNPs in the canine Slc5a1 50
flanking region
The dog Slc5a1 gene has the National Center for
Biotechnology Information unique gene code 492299. The
gene itself occupies 71,445 bases of dog chromosome 26,
from chromosome position 27914321 to 27985765. Known
SNPs in the 10-kbp 50 flanking region of the canine
Slc5a1 gene were identified with the following search
in the Single Nucleotide Polymorphism Database of
Nucleotide Sequence Variation (dbSNP): 26[CHR] AND
27909321:27914320[CHRPOS] AND Canis lupus familiaris
[ORGANISM]. This search identified 3 SNPs in the 5-kbp
region immediately upstream of the Slc5a1 gene, at positions 1744, 1752, and 1780 relative to the transcription
start site (Table 1). All these SNPs are unvalidated. To
investigate the effect of these SNPs on promoter function, it
was decided to clone the 2-kbp fragment of the canine
Slc5a1 50 flanking region from 1,974 to þ25 relative to the
transcription start site.
3.3. Wild-type canine Slc5a1 promoter/reporter construct
drives luciferase expression in cultured Caco-2/TC7 cells
The wild-type canine Slc5a1 50 flanking region 1,974/
þ25/PGL3 basic construct (construct 1) was used for transient transfection of Caco-2/TC7 cells, and firefly luciferase
activity was measured 72 h later. Nontransfected cells and
cells transfected with empty PGL3 basic acted as controls.
The wild-type canine Slc5a1 promoter was shown to be
able to drive luciferase production significantly above
control (P < 0.001; Fig. 2).
3.4. The effect of SNPs on canine Slc5a1 promoter function
To determine the effect of promoter SNPs on promoter
function, new promoter constructs were created by sitedirected mutagenesis. For the 3 known SNPs, with 2
possible alleles at each site, 8 possible combinations are
possible. These were named constructs 1 to 8 as shown in
Table 2, where construct 1 is the wild type.
Transient transfection of Caco-2/TC7 cells was performed with all 8 constructs, again with the use of empty
PGL3 basic as a control, and with the use of Renilla luciferase as an internal control. No significant differences in
promoter function were seen between constructs 1 to 8.
3.5. Results of 454 sequencing
The sequence of the Slc5a1 50 flanking region was
determined by high-throughput sequencing in 2 breeds
predisposed to DM, 2 breeds that rarely develop DM, and 2
breeds predisposed to obesity. The sequence was identical
in all individuals of all breeds tested (no polymorphisms
present).
4. Discussion
The genetic architecture of most common diseases is
extremely complex, and these complex, polygenic diseases
are much more difficult to study than rare, monogenic
disorders [25]. Slc5a1 was selected as a candidate gene in
this study because of its importance to glucose handling in
the animal, SGLT1 being the sole route for glucose and
galactose transport across the BBM. To our knowledge, this
Table 1
Known SNPs in the 5-kbp region immediately upstream of the canine
Slc5a1 gene.
Fig. 1. Western blot analysis for SGLT1 in brush border membrane vesicles
(BBMV) prepared from Caco-2/TC7 cells. Lanes 1 to 3: BBMV from Caco-2/
TC7 cells. C, confluent; Cþ3, confluent þ 3d; Cþ6, confluent þ 6 d. Lane 4:
Positive control (rabbit intestinal BBMV). An immunoreactive band is visible
in confluent and postconfluent Caco-2/TC7 cells.
Position on chromosome 26
Position relative to TSS
Alleles
27912515
27912543
27912551
1780
1752
1744
C/A
A/G
C/T
Abbreviation: TSS, transcription start site.
D.J. Batchelor et al. / Domestic Animal Endocrinology 44 (2013) 139–144
***
Relative light units/mg
60000
40000
20000
0
Fig. 2. Results of reporter gene assays showing that the wild-type canine
Slc5a1 promoter is functional in cultured Caco-2/TC7 cells. The graph shows
the results of luciferase assay, reported in relative light units per milligram of
protein, after transient transfection of Caco-2/TC7 cells with a wild-type
canine Slc5a1 promoter/PGL3 basic construct or empty PGL3 basic (control).
Results are shown as mean SD. Luciferase activity in nontransfected cells are
shown for information only and excluded from statistical analysis. The canine
Slc5a1 promoter drives a significant increase in luciferase activity compared
with control. ***Significant difference, P < 0.001.
is the first time that SGLT1 has been assessed as a potential
predisposing factor in diabetes or obesity.
In this study, firefly luciferase was used as a reporter
gene to study promoter function in canine Slc5a1. Canine
enterocytic cell lines are not available, but Caco-2, a human
colonic adenocarcinoma cell line, shows the phenotype of
an enterocyte once the cells reach confluence, with a wellorganized BBM [26–28]. Caco-2 cells have previously been
used in transfection studies of the Slc5a1 promoter in
humans [29]. The subclone Caco-2/TC7 used in this study
expresses more SGLT1 after confluence and is more
homogeneous than the parental strain [30–32].
The 3 known SNPs in the dog Slc5a1 50 flanking region
were used to create promoter/reporter constructs bearing
all possible permutations of the SNPs, and each promoter
had identical function, suggesting that these 3 SNPs do not
have a significant effect on the constitutive transcription of
SGLT1 mRNA in dogs.
High-throughput sequencing was then performed to
detect novel SNPs in this region in dogs: this technique
allows deep sequencing of genomic regions of interest and is
Table 2
Canine Slc5a1 1,974/25 reporter constructs created by site-directed
mutagenesis to investigate the effect of SNPs on Slc5a1 promoter
function.
Construct
Construct 1
(wild-type)
Construct 2
Construct 3
Construct 4
Construct 5
Construct 6
Construct 7
Construct 8
Alleles
SNP 3 (1780)
SNP 2 (1752)
SNP 1 (1744)
C/A
A/G
C/T
C
A
C
C
C
C
A
A
A
A
A
G
G
A
A
G
G
T
C
T
C
T
C
T
143
suitable for detection of novel variants [33]. Ten individuals
(20 chromosomes) of each breed were assessed, giving
a reasonable chance of finding a polymorphism with a minor
allele frequency of 10% or higher. The breeds were chosen to
maximize the chances of discovering a SNP relevant to diabetes or obesity in dogs. Some breeds were chosen by virtue
of their relative risk (RR) for DM [12]. Samoyeds and cairn
terriers had the highest RR for DM of all (17.3 and 6.8,
respectively), and boxers and German shepherds had a RR of
0.07 and 0.15, respectively. This range of RRs across breeds
resembles the situation in humans, whereby some ethnic
groups have similarly high diabetes risk [11]. Labradors and
cocker spaniels were judged to be the dogs at greatest risk of
developing obesity, based on published literature [16–21].
Boxers and German shepherds are among the least likely
breeds to become obese [17,18].
The fact that no SNPs were detected in the Slc5a1
50 flanking region in the 60 dogs examined may indicate
that 1) polymorphisms in the Slc5a1 50 flanking region are
not relevant to the pathogenesis of DM or obesity in dogs;
2) polymorphisms that have an effect may be present, but
appear at low frequency and were not present in the
sampled dogs, by chance; or 3) polymorphisms in the
Slc5a1 50 flanking region are not relevant to DM or obesity
in these breeds, but may be important in other breeds, in
which the pathogenesis of DM or obesity may be different.
These possibilities could be investigated by analyzing
more dogs and including dogs from other predisposed
breeds. Time constraints meant that this could not be
performed as part of this study. Single nucleotide polymorphisms in various T-cell cytokine genes and in Ctla4
have been associated with disease risk in canine DM, but
the implicated SNPs vary between breeds [13], supporting
the idea that the pathogenesis of DM could be different
between dog breeds. If this is true, then the dogs classified
as “high risk” from the available data may have a high RR
for DM because they carry risk alleles or haplotypes for
genes other than Slc5a1, such as major histocompatibility
complex class II genes [11]. This would make it harder to
detect an effect of variations in Slc5a1. Ideally, an association of the Slc5a1 locus on chromosome 26 with DM or
obesity would be confirmed by a linkage study or association study before embarking on extensive further tests.
5. Conclusions
This study assessed the effect of SNPs in the canine
Slc5a1 50 flanking region on gene promoter function, in an
attempt to determine the relevance of these variations to
DM and obesity. The SNPs were shown not to affect
promoter function, and a search for novel SNPs in dogs with
various risk for DM or obesity showed that 10 individuals
each of 6 breeds all had identical Slc5a1 50 flanking region
sequences. On this evidence, variations in Slc5a1 promoter
sequence between dogs do not influence the pathogenesis
of DM or obesity.
Acknowledgments
Caco-2/TC7 cells were a kind gift from Dr A. Zweibaum
and Dr M. Rousset, INSERM, France. We thank Prof Bill
144
D.J. Batchelor et al. / Domestic Animal Endocrinology 44 (2013) 139–144
Ollier, Dr Andrea Short, Dr Lorna Kennedy, and Simon
Rothwell at the Centre for Integrated Genomic Research at
the University of Manchester for their advice and provision
of DNA from the UK Companion Animal DNA Archive. We
thank Prof Neil Hall, Dr Margaret Hughes, and Dr Christiane
Hertz-Fowler at the School of Biological Sciences, University of Liverpool for performing 454 sequencing and Dr
Luca Lienze and Dr Kevin Ashelford for assistance with data
analysis. These studies were funded by the Biotechnology
and Biological Sciences Research Council (BBSRC) and
Pfizer Limited, who had no involvement in study design or
data collection or interpretation. A.J.G.’s senior lectureship
is financially supported by Royal Canin.
References
[1] Pomp D, Mohlke KL. Obesity genes: so close and yet so far. J Biol
2008;7:36.
[2] Prokopenko I, McCarthy MI, Lindgren CM. Type 2 diabetes: new
genes, new understanding. Trends Genet 2008;24:613–21.
[3] Sutter NB, Ostrander EA. Dog star rising: the canine genetic system.
Nat Rev Genet 2004;5:900–10.
[4] Rand JS, Fleeman LM, Farrow HA, Appleton DJ, Lederer R. Canine and
feline diabetes mellitus: nature or nurture? J Nutr 2004;134:2072S–80S.
[5] Catchpole B, Ristic JM, Fleeman LM, Davison LJ. Canine diabetes
mellitus: can old dogs teach us new tricks? Diabetologia 2005;48:
1948–56.
[6] Marmor M, Willeberg P, Glickman LT, Priester WA, Cypess RH,
Hurvitz AI. Epizootiologic patterns of diabetes mellitus in dogs. Am J
Vet Res 1982;43:465–70.
[7] Hess RS, Kass PH, Ward CR. Breed distribution of dogs with diabetes
mellitus admitted to a tertiary care facility. J Am Vet Med Assoc
2000;216:1414–7.
[8] Guptill L, Glickman L, Glickman N. Time trends and risk factors for
diabetes mellitus in dogs: analysis of veterinary medical data base
records (1970-1999). Vet J 2003;165:240–7.
[9] Davison LJ, Herrtage ME, Catchpole B. Study of 253 dogs in the
United Kingdom with diabetes mellitus. Vet Rec 2005;156:467–71.
[10] Fall T, Hamlin HH, Hedhammar A, Kampe O, Egenvall A. Diabetes
mellitus in a population of 180,000 insured dogs: incidence,
survival, and breed distribution. J Vet Intern Med 2007;21:1209–16.
[11] Kennedy LJ, Davison LJ, Barnes A, Short AD, Fretwell N, Jones CA,
et al. Identification of susceptibility and protective major histocompatibility complex haplotypes in canine diabetes mellitus.
Tissue Antigens 2006;68:467–76.
[12] Short AD, Catchpole B, Kennedy LJ, Barnes A, Fretwell N, Jones C,
et al. Analysis of candidate susceptibility genes in canine diabetes. J
Hered 2007;98:518–25.
[13] Short AD, Catchpole B, Kennedy LJ, Barnes A, Lee AC, Jones CA, et al.
T cell cytokine gene polymorphisms in canine diabetes mellitus. Vet
Immunol Immunopathol 2009;128:137–46.
[14] Short AD, Saleh NM, Catchpole B, Kennedy LJ, Barnes A, Jones CA,
et al. CTLA4 promoter polymorphisms are associated with canine
diabetes mellitus. Tissue Antigens 2010;75:242–52.
[15] German AJ. The growing problem of obesity in dogs and cats. J Nutr
2006;136:1940S–6S.
[16] Lund EM, Armstrong PJ, Kirk C, Klausner JS. Prevalence and risk
factors for obesity in adult dogs from private US veterinary practices. Int J Applied Res Vet Med 2006;4:177–86.
[17] Mason E. Obesity in pet dogs. Vet Rec 1970;86:612–6.
[18] Edney AT, Smith PM. Study of obesity in dogs visiting veterinary
practices in the United Kingdom. Vet Rec 1986;118:391–6.
[19] Kronfeld DS, Donoghue S, Glickman LT. Body condition and energy
intakes of dogs in a referral teaching hospital. J Nutr 1991;121:
S157–8.
[20] Colliard L, Ancel J, Benet JJ, Paragon BM, Blanchard G. Risk factors for
obesity in dogs in France. J Nutr 2006;136:1951S–4S.
[21] Weeth LP, Fascetti AJ, Kass PH, Suter SE, Santos AM, Delaney SJ.
Prevalence of obese dogs in a population of dogs with cancer. Am J
Vet Res 2007;68:389–98.
[22] Shirazi-Beechey SP, Davies AG, Tebbutt K, Dyer J, Ellis A, Taylor CJ,
et al. Preparation and properties of brush-border membrane
vesicles from human small intestine. Gastroenterology 1990;98:
676–85.
[23] Dyer J, Al-Rammahi M, Waterfall L, Salmon KS, Geor RJ, Boure L,
et al. Adaptive response of equine intestinal Naþ/glucose cotransporter (SGLT1) to an increase in dietary soluble carbohydrate.
Eur J Physiol 2009;458:419–30.
[24] Catchpole B, Kennedy LJ, Davison LJ, Ollier WE. Canine diabetes
mellitus: from phenotype to genotype. J Small Anim Pract 2008;49:
4–10.
[25] Gregersen PK. Closing the gap between genotype and phenotype.
Nat Genet 2009;41:958–9.
[26] Pinto M, Robine-Leon S, Appay MD, Kedinger M, Triadou N,
Dussaulx E, et al. Enterocyte-like differentiation and polarization of
the human colon carcinoma cell line Caco-2 in culture. Biol Cell
1983;47:323–30.
[27] Grasset E, Pinto M, Dussaulx E, Zweibaum A, Desjeux JF. Epithelial
properties of human colonic carcinoma cell line Caco-2: electrical
parameters. Am J Physiol 1984;247:C260–7.
[28] Engle MJ, Goetz GS, Alpers DH. Caco-2 cells express a combination
of colonocyte and enterocyte phenotypes. J Cell Physiol 1998;174:
362–9.
[29] Martin MG, Wang J, Solorzano-Vargas RS, Lam JT, Turk E,
Wright EM. Regulation of the human Na(þ)-glucose cotransporter
gene, SGLT1, by HNF-1 and SP1. Am J Physiol Gastrointest Liver
Physiol 2000;278:G591–603.
[30] Mahraoui L, Rodolosse A, Barbat A, Dussaulx E, Zweibaum A,
Rousset M, et al. Presence and differential expression of SGLT1,
GLUT1, GLUT2, GLUT3 and GLUT5 hexose-transporter mRNAs in
Caco-2 cell clones in relation to cell growth and glucose consumption. Biochem J 1994;298:629–33.
[31] Mesonero J, Mahraoui L, Matosin M, Rodolosse A, Rousset M,
Brot-Laroche E. Expression of the hexose transporters GLUT1GLUT5 and SGLT1 in clones of Caco-2 cells. Biochem Soc Trans
1994;22:681–4.
[32] Caro I, Boulenc X, Rousset M, Meunier V, Bourrie M, Julian B, et al.
Characterisation of a newly isolated Caco-2 clone (TC7), as a model
of transport processes and biotransformation of drugs. Int J Pharmaceutics 1995;116:147–58.
[33] Ingman M, Gyllensten U. SNP frequency estimation using massively
parallel sequencing of pooled DNA. Eur J Hum Gen 2009;17:
383–6.