International Journal of Obesity (2012) 36, 1017 - 1024
& 2012 Macmillan Publishers Limited All rights reserved 0307-0565/12
www.nature.com/ijo
REVIEW
Causes and consequences of obesity: the contribution of recent
twin studies
J Naukkarinen1,2,3, A Rissanen1, J Kaprio2,4,5 and KH Pietiläinen1,2,4
Obesity is a genetically complex disorder that produces a myriad of health problems. Most of the recognized complications of
obesity are not only strongly influenced by lifestyle factors, but also present with independent genetic predispositions that are
notoriously difficult to disentangle in humans. Most studies on the causes and consequences of acquired obesity are encumbered
by the incomplete ability to control for genetic influences. However, utilizing a unique experiment of nature, namely monozygotic
twins (MZ) discordant for obesity as ‘clonal controls’ of obese and non-obese individuals has enabled the fine characterization of
the effects and possible antecedents of acquired obesity while controlling for the genetic background, as well as pointed to novel
obesity predisposing candidate genes. This review is a distillation of the findings from more than 10 years of research done in an
exceptionally well-characterized collection of MZ and dizygotic (DZ) twins, based on the Finnish Twin Cohorts. Topics covered
include the nature of development of obesity from the childhood onwards, the role of exercise in modifying the genetic
susceptibility, the resulting inflammatory, prediabetic and preatherosclerotic changes in whole body and adipose tissue physiology,
as well as the newest insights provided by the omics revolution.
International Journal of Obesity (2012) 36, 1017 -- 1024; doi:10.1038/ijo.2011.192; published online 11 October 2011
Keywords: adipose tissue; inflammation; liver fat; insulin resistance; BMI; metabolic syndrome
INTRODUCTION
Across time people have been fascinated by monozygotic (MZ)
twins and their striking similarities in appearance, behavior as well
as aspects of health and disease. In mythology, twins are often
described as two halves of a whole, or alternatively as polar
opposite representations of good and evil. MZ twins result from
the early splitting of a single zygote and thereby are identical at
the level of DNA sequence (precluding somatic mutations,
epigenetic modifications and other genetic changes), whereas
dizygotic (DZ) twins result from two different ova being fertilized
by two separate sperm and share on average 50% of their
segregating genes---just as ordinary siblings in a family. Most of the
early genetic research into the heritability and variance of human
traits was based on measuring the differences in within-pair
correlations of various phenotypic traits between pairs of MZ and
DZ twins. In addition to complex, continuous phenotypes such as
height, weight and aspects of personality where MZ twins exhibit
a high degree of similarity, MZ twins are nearly always concordant
for Mendelian genetic diseases. Studying cases where MZ twins
present with differences in the manifestation (onset, severity,
response to treatment, and so on) of Mendelian diseases has
helped to uncover disease mutations as well as broadened the
clinical spectrum of diseases previously thought to be more
uniform in presentation (Reviewed by Zwijnenburg et al., 2010).1
Recently discordant MZ twins have also been utilized in the hunt
for polymorphisms contributing to complex traits such as obesity.2
Twin studies have a long history in the Finland. A large cohort of
adult twins was established in 1974,3 and was extended to two
birth cohorts in the 1990s (FinnTwin12 and Finntwin16 Studies).
These longitudinal cohort studies include both repeated surveys
of the participants and follow-up from national medical registries.
The majority of the studies cited in this review are based upon
data from participants recruited from the FinnTwin16 Cohort,3,4 a
population-based longitudinal study of five consecutive birth
cohorts (1975 -- 1979 of Finnish twins, their siblings and parents
(N ¼ 2453 families). Of a total of 658 MZ twin pairs at 25 years, only
18 healthy pairs of MZ twins significantly discordant for obesity
(intra-pair body mass index (BMI) difference X3 kg m2) were
identified, 14 of which participated in the clinical studies.5--12 The
majority of MZ twin pairs are very similar for BMI (intraclass
correlations 0.8 in FinnTwin16 at age 25)13 and the difficulty in
identifying distinctively discordant pairs underscores the strong
genetic component to BMI and the predisposition to obesity. This
has been demonstrated for adults in adoption, family and twin
studies,14,15 but less so for childhood obesity16 and rate of weight
gain.16,17
Here, we review the discordant twins’ growth patterns and
development of obesity, measures of physical activity and
metabolic rates, and the physiological consequences of acquired
obesity as can be identified through clinical and imaging studies
(Figure 1). In addition, we have included insights from global
analysis of adipose tissue transcriptomics and lipidomics to further
elucidate the metabolical and physiological changes associated
with acquired obesity (Figure 2).
The discordant MZ twin experimental setup allows researchers
to control for not only genetic background, but also for (by
1
Obesity Research Unit, Department of Medicine, Division of Internal Medicine and Department of Psychiatry, Helsinki University Central Hospital, Helsinki, Finland; 2FIMM,
Institute for Molecular Medicine Finland, University of Helsinki, Helsinki, Finland; 3Public Health Genomics Unit, National Institute for Health and Welfare, Helsinki, Finland;
4
Finnish Twin Cohort Study, Department of Public Health, University of Helsinki, Helsinki, Finland and 5National Institute for Health and Welfare, Helsinki, Finland.
Correspondence: Dr KH Pietiläinen, Obesity Research Unit, Department of Medicine, Division of Internal Medicine and Department of Psychiatry, Helsinki University Central
Hospital, Helsinki, Finland.
E-mail: kirsi.Pietiläinen@helsinki.fi
Received 9 March 2011; revised 29 August 2011; accepted 4 September 2011; published online 11 October 2011
Causes and consequences of obesity in twins
J Naukkarinen et al
1018
Figure 1.
Study design for the obesity-discordant MZ twins.
Figure 2. Consequences of acquired obesity in BMI discordant MZ twins. Fold changes in obese as compared with non-obese co-twins’
BMI, subcutaneous abdominal fat, intra-abdominal fat, liver fat, insulin sensitivity (M-value: glucose uptake in mg per kg fat-free mass in
minute), fasting serum insulin, subcutaneous abdominal fat cell size, mitochondrial DNA copy number in adipose tissue, adipose tissue
transcriptomics pathways describing fat cell differentiation (GO: 0045444), BCAA catabolism (GO:009083) and inflammatory response
(GO:0006954). N ¼ 14 obesity-discordant MZ pairs. Non-obese co-twins are set as 1.0, and the obese co-twins’ values are the ratio of obese
to non-obese co-twins’ values. Data are derived from Pietilainen et al.5; Mustelin et al.10; Pietilainen et al.12 fS-insulin, fasting serum insulin;
Ia, intra-abdominal; mtDNA, mitochondrial DNA; Sc fat, subcutaneous abdominal fat. *Po0.05, **Po0.01, ***Po0.001. Statistical test used:
Paired, ranked Wilcoxon’s test.
definition) gender, age and early shared environmental exposures.
In addition, other common confounders of epidemiological
studies such as cohort effects can be minimized using this
approach. The study of multifactorial, complex phenotypes such
as obesity is perhaps most benefitted from such a design.
DEVELOPMENT OF OBESITY
Detailed information about growth throughout childhood and
early adulthood were initially analyzed for the first 10 pairs,6 and
inclusion of an additional 4 pairs for the purposes of this review
confirmed the previously published results:6 a slightly higher birth
weight (193 g) was observed for the twin that developed obesity
during early adulthood, but this difference disappeared by age 6
months and the growth patterns of both twins were virtually
identical until the age of 18 years, after which BMI differences
between the co-twins became statistically significant (Figure 3). As
is the case in twin births, the birth weights of the twins were
International Journal of Obesity (2012) 1017 - 1024
below that of the reference population, but this was followed by a
catch-up growth; both males and females had reached the
population reference weights for age and sex by the age of
2 years. After age 8, the pairs who later became discordant for
obesity were heavier than the population mean, raising the
possibility that genetic or environmental factors predisposing to
obesity may be present in both co-twins of the discordant pairs.
It therefore remains an open question as to whether the lean or
the obese co-twin actually is more closely following the genetic
predisposition. Based on interviews of these twins, there is an
indication that both the conscious efforts to prevent weight gain
in the lean and the more unhealthy lifestyle in the obese co-twins
explains the weight disparities, which appeared at 18 years of age,
at a time when environmental influences may also start to diverge
as the young adults move out of their parental home and establish
more individualistic life styles. The twins themselves pointed out
environmental factors in young adulthood such as living alone,
irregularity of student life, sedentary work, getting a driver’s
& 2012 Macmillan Publishers Limited
Causes and consequences of obesity in twins
J Naukkarinen et al
1019
Figure 3. Development of BMI from full-term birthday to 23 years
in the leaner (dotted lines) and heavier (solid lines) co-twins in MZ
twin pairs discordant for obesity in early adulthood; both sexes
combined after age- and sex-specific calculation of standardized
z-scores based on the British growth reference population.65 Growth
curves of weight-concordant MZ co-twins were completely
overlapping (data not shown).
license (at age 18 in Finland), and marital and societal relationships as provocateurs of weight gain as compared with more
regular eating and physical activity habits in childhood.
Intra-pair correlations of BMI in MZ twins generally tend to
decline during the adult years, perhaps indicating the increasingly
divergent environmental exposures.18 However, twin-to-twin
correlations in BMI are as high as 0.79 in adulthood even in MZ
twins who have been reared apart and thus exposed to different
environments for most of their lives.19 The highly similar response
of MZ twins to environmental exposures has also been shown by
the classic overfeeding20 and weight-loss21 - 23 studies. Hence, it
may be that our collection of obesity-discordant twins represents
an atypical set of MZ pairs being able to vary greatly in their body
weight, a phenomenon itself worth discussion. It has been
suggested that discordant MZ twins may harbor so called
‘variability genes’ that instead of directly determining the adult
weight, may rather make the individuals carrying these variability
genes more susceptible to environmental variation.24--27 One
possible such gene is GHRL that codes for the hormone ghrelin.
Twin pairs concordant and discordant for obesity in middle-age
differed in the Leu72Met genotype frequency, suggesting that
ghrelin, known to be important in the regulation of food intake,
may represent one of the first such variability genes that
predispose carriers to more weight variability in response to
environmental factors.28,29
Relationships of physical activity and weight trajectories in
youth have been inconsistent in previous population and
intervention studies.30 This may be in part because of the
recognized difficulty in measuring levels of physical activity over
long periods of time, or because the cause and effect relationship
between physical activity and tendency to become obese are not
as clear cut as often times represented. The old ‘calories in,
calories out’ thinking is a gross simplification of human
metabolism, and as has been pointed out, the two variables
(calories taken in, calories expended) are not independent
variables, but rather both have an effect on one another;
increasing calorie expenditure also increases hunger for more
calories (‘working up an appetite’) and vice versa.
However, the beneficial effects of physical activity in the
prevention of long-term weight gain have been clearly shown in
the FinnTwin16 Study where 4240 twin individuals were evaluated
for self-perceived physical fitness and physical activity in
adolescence. An additional 10 obesity-discordant MZ twin-pairs
were further evaluated for current physical activity (using triaxial
& 2012 Macmillan Publishers Limited
accelerometers), total energy expenditure (by means of doubly
labeled water) and basal metabolic rate (by indirect calorimetry).
The results suggested that physical inactivity in adolescence
strongly predicted the risk for obesity (OR 3.9) and abdominal
obesity (OR 4.8) at age 25, even after adjusting for baseline and
current BMI. In the MZ obesity-discordant pairs, physical inactivity
was both causative and secondary to the development of
obesity.31 Physical activity was significantly reduced already at
age 16, and it was only after this that the weight differences
began to appear. This was interpreted to support the causal role of
physical inactivity in the young on the subsequent development
of obesity.
At age 25, the obese co-twins were only half as active compared
with their lean co-twin as demonstrated in the 7-day accelerometer measurements.31 However, the total energy expenditure
and activity-induced energy expenditure from the doubly labeled
water did not differ between the co-twins. This discrepancy may
be explained by the fact that the obese twins, while moving on
average less, do expend more energy when they do because of
their higher body weight. The basal metabolic rates (as measured
by calorimetry) were considerably higher in the obese co-twins,
presumably for the same reason.
Physical activity and fitness are associated with beneficial
metabolic effects, for example, on blood lipids and insulin
sensitivity.30 In line with this, a 28-year follow-up of 8182 Finnish
twin pairs showed significant reductions (HR 0.54; 95% CI
0.37 -- 0.78) in the risk for Type 2 diabetes in twin pair members
that had been physically active during the whole of their adult
lives compared with consistently inactive co-twins.32 Intensive
substudies of these physical activity discordant twins further
revealed that although the physically inactive (79.5±18.4 kg) and
active twins (72.9±11.9 kg) had only small differences in body
weight, the inactive twins had 170% more liver fat and 54% more
intramuscular fat than the active twins.33 Thus, the value of
physical activity should not be judged only by its’ effects on body
weight.
A few recent studies have shed some interesting light on the
interplay of physical exercise and its modulatory effects on an
individual’s genetic predisposition to weight gain and obesity. The
prospective Norfolk study of 20 000 men and women showed
physical activity to attenuate the genetic predisposition to
common obesity by 40%, as estimated by the number of risk
alleles carried for 12 recently identified obesity predisposing loci.34
In the same study, the genetic risk score was positively associated
with weight gain in inactive subjects, but negatively associated in
physically active subjects. Similar findings in a twin study of the
United States, Finnish and Danish twins lend support to the notion
that physical exercise can indeed counteract the genetic
predisposition to obesity; estimates of the heritability of waist
circumference, percentage of body fat and BMI significantly
declined in those individuals classified as having ‘high’ physical
activity.13,35,36 One of these studies36 also showed that the protein
content of the diet had no appreciable effect on the heritability of
the same measures of adiposity.
Several other studies have investigated the effect of environmental factors on the apparent heritability of other obesityassociated phenotypes such as insulin sensitivity and lipid
profiles37 and blood pressure,38 strengthening the view that
genetic predisposition can indeed be modified by environmental
factors such as physical exercise and dietary habits.
In addition to differences in physical activity in all its forms
(‘calories out’), the composition and patterns of nutrition
(‘calories in’) have been investigated among obesity-discordant
MZ twins. Both facets are difficult to measure in practice, with the
former suffering from a clear over-reporting and the latter from
under-reporting/under-eating pattern by the obese.39 Measures of
food intake and physical activity were investigated utilizing food
and exercise diaries as well as by having each twin to compare
International Journal of Obesity (2012) 1017 - 1024
Causes and consequences of obesity in twins
J Naukkarinen et al
1020
Figure 4. Within-pair comparison of eating patterns and physical
exercise among obesity-discordant MZ twin pairs (n ¼ 14). The
co-twins were asked to rate themselves in relation to their co-twins
by questions: ‘Which one of you, you or your co-twiny’. Adapted
from Pietilainen et al.39
their own eating with that of their co-twin. The accuracy of the
self-reported data on energy intake and energy expenditure was
then validated by doubly labeled water. The self-reported analyses
could show no differences in energy intake or the amount of
physical activity between the co-twins. However, the more
objective measures via doubly labeled water revealed a substantial reporting bias by the obese co-twins: the under-reporting
of energy intake (3.2±1.1 MJ per day) and over-reporting of
physical activity (1.8±0.8 MJ per day) in the obese twins equaled
to as much as one Big Mac hamburger, a 16-oz bottle of soft drink
and almost 90 min of walking (3 m.p.h.), respectively. Interestingly,
however, when asked to compare their own eating habits and
physical activity to those of their co-twin, both co-twins openly
reported that the obese co-twin had an unhealthier lifestyle with
overeating, snacking and an irregular eating pattern as well as less
physical exercise (Figure 4). Similarly, in our earlier study of
middle-aged weight-discordant MZ twins, both co-twins reported
that the heavier twin consumed more calories, more high-fat and
sweet food as well as alcohol.40 Importantly, such dietary patterns
had arisen already in early adulthood (age 20 -- 30), before the
development of weight discordance. These results were further
confirmed in the FinnTwin16 Cohort, where 713 MZ and 698
same-sex DZ twin pairs aged 22 -- 28 years were asked to evaluate
aspects of their dietary and exercise habits. The co-twin for whom
both members of the twin pair in agreement reported eats more,
snacks more, eats more fatty foods, and sweet and fatty delicacies,
chooses less healthy foods, eats faster and exercises less, had a
significantly higher BMI (0.6 -- 2.9 kg m2) and waist circumferences
(1.5 -- 7.5 cm).41 Multivariate regression analysis identified co-twin
differences in the reported amount of food consumed as
the strongest independent non-genetic predictor of intra-pair
differences in BMI and waist circumference.41
FAT DISTRIBUTION IN ACQUIRED OBESITY
With increasing obesity, excessive fat accumulates in different
depots around the body; subcutaneously, intra-abdominally
and in the ectopic sites such as liver and muscles. The relative
distribution of subcutaneous and visceral fat is under hormonal
control,42 differing between men and women, as well as between
individuals of the same sex according to their genetic predispositions.43,44 The strong role of genetics in the distribution of fat
accumulation is demonstrated by the fact that even among the
discordant MZ twins with considerable differences of total body
fat, the relative proportions of fat measurable subcutaneously,
intra-abdominally and in the liver are strongly correlated.
The distribution of fat in different depots was measured via
magnetic resonance spectroscopy, liver spectra and dual energy
International Journal of Obesity (2012) 1017 - 1024
x-ray absorptiometry in the 14 BMI discordant MZ twins.5 The
heavier co-twin was found to have 64% more abdominal
subcutaneous fat, 93% more intra-abdominal fat and 284% more
liver fat than the lean co-twin. Environmental influences
independent from acquired obesity on liver fat were evaluated
based on questionnaires and food diaries. Alcohol consumption
from detailed questionnaires of the obese (3.7±0.9 doses per
week) and non-obese (3.9±1.1 doses per week) co-twins did not
differ and intra-pair differences in alcohol intake did not
significantly correlate with differences in liver fat (r ¼ 0.30,
P ¼ 0.14). Analysis of data from food diaries showed that the
percentage of energy from fat (r ¼ 0.37, P ¼ 0.02) and saturated fat
(r ¼ 0.38, P ¼ 0.005) did correlate with liver fat.5 The effects of
acquired obesity, however, varied considerably between the twin
pairs; although in all cases the obese co-twin had an increased
amount of subcutaneous and visceral fat, some of the pairs were
clearly more prone to hepatic fat accumulation, regardless of the
weight difference---presumably because of strong genetic influences on this fat depot.45 Calculations of the intra-pair differences
in fat accumulation across different depots revealed that
subcutaneous fat accumulation explains roughly twice as much
of the variation in intra-abdominal (42 -- 67%) than liver fat
(12 -- 33%). Increasing liver fat is reflected in an increased
concentration of serum alanine aminotransferase (S-ALAT), which
correlated with the amount of liver fat (r ¼ 0.70 for women and
r ¼ 0.50 for men) in a larger cohort of MZ and DZ pairs.45 Intra-pair
correlations of S-ALAT were significantly higher between MZ twins
than DZ-twins, providing heritability estimates for S-ALAT at
60%.45 Overall, the results suggest that liver fat may be unevenly
influenced by acquired obesity (not all obese co-twins developed
fatty livers), perhaps because of genetic influence on amounts
of liver fat.
PREDIABETIC CHANGES IN ACQUIRED OBESITY
A number of prediabetic and preatherosclerotic changes were
observed in the obese co-twins in the studies here reviewed. The
observations and their proposed links to these pathological
conditions have been summarized in Figure 5. The most classical
laboratory measurements such as fasting blood glucose did not
differ between these young, obese and lean co-twins.5 Measurements of fasting insulin levels, however, were markedly different
and reflected a developing insulin resistance, as evidenced by the
significantly lower M-values.5 The M-value is mostly a measure of
insulin sensitivity in muscle, and the intra-pair differences
correlated much better with measures of overall obesity (%body
fat, abdominal subcutaneous fat, BMI, intra-abdominal fat with
corresponding r-values 0.80, 0.72, 0.67, 0.55) than with liver
fat (r ¼ 0.20, P ¼ 0.40).5 The accumulation of fat intrahepatically,
however, is clearly correlated with fasting insulin levels independently of accumulation of fat in other depots, underscoring the
importance of liver fat in the development of insulin resistance
and other features of Type 2 diabetes. It should be pointed out
that all the twins studied were considered somatically and
psychiatrically ‘healthy’ by physician clinical examination and
structured psychiatric interview, with no regular use of medication
(precluding oral contraceptives). The changes observed are an
early indication of clearly prediabetic changes in the obese
co-twin already in early adulthood.
PREATHEROSCLEROTIC CHANGES
Dyslipidemia
A number of dyslipidemic changes predisposing to the development of atherosclerosis are known to be associated with acquired
obesity, including increased levels of circulating low-density
lipoprotein, decreased high-density lipoprotein and the associated
& 2012 Macmillan Publishers Limited
Causes and consequences of obesity in twins
J Naukkarinen et al
1021
Figure 5. Proposed mechanisms of the development of preatherosclerotic and prediabetic changes in acquired obesity based on findings
from the obesity-discordant MZ twins. Circles denote changes in adipose tissue, boxes with solid lines changes in serum concentrations
and boxes with dotted lines changes in imaging or metabolic studies in the obese as compared with the non-obese MZ co-twins. CRP,
C-reactive protein; Ia, intra-abdominal; mtDNA copy nro, mitochondrial DNA copy number; Sc, subcutaneous.
hypertriglyceridemia. Despite the considerable difference in BMI,
these ‘classical’ lipid measurements were not much different
between the young, ‘healthy’ co-twins. In these twins, the obese
as compared with the lean twin had on average slightly higher
low-density lipoprotein (mean 2.8±0.2 vs 2.6±0.2 mmol l1),
lower high-density lipoprotein (1.4±0.1 vs 1.5±0.1 mmol l1)
and higher concentrations of triglycerides (1.3±0.1 vs
1.0±0.1 mmol l1). Much clearer pathological changes, however,
were observable by new lipidomics technology,11 shedding new
light on previously unrecognized changes in different lipid
species. Among the most elevated lipid species in acquired
obesity were found to be the proinflammatory lysophosphatidylcholines that also are shown to increase the permeability of
vascular endothelium.46 In addition, ether phospholipids that are
known to have antioxidant properties were decreased in obese
co-twins.11 lysophosphatidylcholines and ether phospholipids had
opposite associations with insulin sensitivity and interestingly
both have been suggested to act as signaling molecules in the
insulin cascade.11 Intra-abdominal (but not subcutaneous) fat and
serum insulin also correlated positively with medium chain
sphingomyelin species, suggesting the role of these lipids in
acquired obesity and insulin resistance.
Endothelial dysfunction
Just as decreased insulin sensitivity precedes the development of
clinical diabetes, endothelial dysfunction predates the development of clinically obvious atherosclerosis by years.47 Various
substances, some circulatory, some produced by the vascular
endothelium itself, control the actions of vasodilation and
vasoconstriction. An imbalance between these factors, broadly
defined as endothelial dysfunction, has been associated with
obesity in many studies.48 Numerous explanations have been
suggested to explain this association, including endothelial insulin
resistance, impairments in circulating mediators such as adiponectin and C-reactive protein, or the presence of genetic factors
that independently influence both the predisposition to obesity
and vascular function.49 Endothelial function can be measured by
assessing the changing resistance of forearm vessels in response
to endothelium-dependent and independent vasodilators.
& 2012 Macmillan Publishers Limited
Studying endothelial function in the obesity-discordant MZ twins
provided new evidence for the effects of acquired obesity,
independent of genetics, on endothelial dysfunction.9 Intraabdominal rather than subcutaneous or overall adiposity was
correlated with endothelial dysfunction. Circulating levels of the
adipocyte-derived hormone adiponectin was the only remaining
factor explaining obesity-related endothelial dysfunction in multiple linear regression analysis, independent of intra-abdominal fat
and C-reactive protein. Adiponectin is secreted by adipocytes and
shown to be reduced in obese individuals.50 The expression of
adiponectin mRNA in adipocytes from the obese co-twins was also
shown to be significantly downregulated.7 The findings in these
obesity-discordant MZ twins support the view that adiponectin
deficiency is indeed an acquired trait associated with obesity.
Adiponectin has a number of other modulatory effects on
metabolic processes, including decreasing gluconeogenesis,
increasing glucose uptake and utilization of fatty acids in muscle
tissue.51 As such, it seems to be in a pivotal position in the
development of obesity-associated pathologies.
ADIPOSE TISSUE
The adipocyte is a remarkably dynamic cell, the current view being
in stark contrast to a belief long held considering fat tissue to be a
relevantly inert storage tissue. The ‘omics revolution’ has enabled
the fine-scale dissection of adipocyte metabolism through studies
of global transcript profiles. The analysis of changes in the
adipocyte morphology, transcriptome and associated cell populations in obesity-discordant MZ twins has provided new information about the pathological events that take place in acquired
obesity, independent of genetic background. Following increased
influx of triglycerides for storage, the adipocyte swells and
becomes engorged with a triglyceride droplet, displacing the
nucleus into the cells periphery. Hypertrophy of the cell is
accompanied by reduced expression of, among other transcription factors PPARG---an important regulator of adipocyte differentiation.7 In the adipose tissue of the obese co-twin a number of
differentiation-associated pathways and individual gene products
were also downregulated, pointing to an impaired differentiation
of the swelling adipocytes.12
International Journal of Obesity (2012) 1017 - 1024
Causes and consequences of obesity in twins
J Naukkarinen et al
1022
Mitochondrial dysfunction is beginning to be recognized as an
important pathological consequence of developing obesity,
especially insulin resistance. A novel finding of great interest in
our obesity-discordant MZ pairs was the dramatic reduction of
copies of mitochondrial DNA in the adipose tissue of the obese
co-twin.12 Although the sequence of mitochondrial sequence was
identical between the MZ twins (no evidence of heteroplasmy),
the copy number of mitochondrial DNA in the obese co-twin’s
adipose tissue was only 53% of that of the lean co-twin. This was
also associated with a clear reduction in the pathway responsible
for the mitochondrial catabolism of branched-chain amino acids
(BCAAs) that act as insulin secretagogues. Interestingly, the
reduced expression of adipose tissue BCAA catabolism was
significantly associated not only with increased fasting serum
insulin but also with increased liver fat. The concentrations of
BCAAs were also measured from the plasma and shown, as
expected, to be increased (especially leucine) in the circulation
and to correlate strongly with fasting insulin levels---providing one
direct link from obesity (adipocyte hypertrophy) to hyperaminoacidemia (specifically of the BCAAs) and hyperinsulinemia,
hypothesized in late 1960s.52 A similar finding in a study of the
Finnish same-sex twins discordant for physical activity for almost
30 years showed a clear downregulation of BCAA catabolism in
both adipose and skeletal muscle tissue among the less active
co-twins.53 Together these findings strengthen the view that
mitochondrial dysfunction and specifically BCAA catabolism has a
key role in the development of obesity-related insulin resistance
and that exercise may counteract these effects.
More detailed study focused on the mitochondrial functions of
the discordant twins revealed that genes involved in the
mitochondrial oxidative phosphorylation pathway were significantly downregulated in the obese co-twins adipose tissue.10
Earlier studies have shown similar results in the muscle tissues of
Type 2 diabetics and their first degree relatives, suggesting that
genetics also have a role.54 As a measure of overall cardiorespiratory fitness, maximal oxygen uptake (VO2max) and working
capacity (Wmax) in the obese co-twins were significantly reduced
compared with the lean co-twins.10 The oxidative phosphorylation
pathway activity was significantly correlated with VO2max and the
M-value in the twins, even after adjusting for % body fat in
multiple regression analysis. The results indicate that acquired
obesity, independent of genetic background decreases physical
fitness and whole body insulin sensitivity with these effects in part
related to decreased mitochondrial capacity for oxidative phosphorylation and decreased copy number of mitochondrial DNA.
As the adipocytes enlarge, they begin to secrete proinflammatory cytokines and attract macrophages into the adipose tissue,
thereby initiating a low-grade chronic inflammation.55,56 This has
been associated with increased insulin resistance, and a vicious
cycle of weight gain followed by a worsening insulin resistance is
initiated.57 In fact, the single most upregulated transcript (5.9-fold
upregulated) in the adipose tissue of the obese co-twin was an
inflammatory cytokine called osteopontin (SPP1, aka OPN).12
Osteopontin is secreted by the swelling adipocyte and recruits an
increasing number of macrophages that in turn recruit more
macrophages, by also secreting osteopontin themselves. The
predominant change clearly observed in the global transcript
profiles was an upregulation of several different proinflammatory
pathways in the adipose tissue of the obese twin. Previously
unrecognized dimensions of the inflammatory pathways were
upregulated, with involvement not only of macrophages, but
antigen presentation via MHC class II molecules, activation of
lymphocytes and of the complement system.12 In addition to local
inflammation of the adipose tissue and concomitant insulin
resistance, circulating markers of inflammation could be observed:
elevated C-reactive protein levels9 as well as newly recognized
proinflammatory lipid species were discovered in large-scale
lipidomics studies.11
International Journal of Obesity (2012) 1017 - 1024
The most proximal physiological mechanism responsible for the
development of inflammation in the adipose tissue has remained
elusive. Our most recent study utilizing obesity-discordant MZ
twins, however, suggested a plausible mechanism for this
initiation of inflammation: as the adipocytes’ increase in size,
changes in the lipid composition of the cell membrane occur to
protect its physical properties and fluidity. These protective changes
occur with a cost, however: the kinds of lipids accumulating in the
‘obese’ adipocytes are known to be precursors for compounds that
can initiate and aggravate inflammatory cascades. We therefore
propose that membrane lipids in the adipocytes have a key role in
adipose tissue inflammation. Interestingly, these same lipids were
associated with increased levels of liver fat.58
Another interesting observation from the obesity-discordant MZ
twin pairs was a significant overexpression of 11b-hydroxysteroid
dehydrogenase type 1 (11-b-HSD-1) in the adipose tissue of the
obese co-twins.8 11-b-HSD-1 catalyzes the interconversion of
the biologically inactive glucocorticoid (cortisone) to the active
glucocorticoid (cortisol) and is thus regarded as a marker of
cellular stress. In our study, 11-b-HSD-1 was closely related to
accumulation of subcutaneous and intra-abdominal fat, as well as
features of insulin resistance, illustrated by the positive relationships of 11-b-HSD-1 gene and protein (50 kDa) expression to
fasting serum insulin and serum-FFA concentrations during
hyperinsulinemic conditions as well as the inverse correlation
between 11-b-HSD-1 expression and whole-body insulin
sensitivity.8
OBESITY CANDIDATE GENES
Family and genetic epidemiological studies (by means of linkage
and GWA studies, respectively) have shed light on the genetic
predispositions to obesity with a fervent search for ‘obesity genes’.
Classical linkage studies have been plagued by a lack of
replication,59 but together with animal models have uncovered
some rare obesity causing genes, such as leptin and its receptor,
as well as mutations in the melanocortin 4 receptor gene.60 The
genome-wide association studies massive study populations
utilizing have uncovered some 40 novel single-nucleotide
polymorphisms that while robustly associating with BMI, even
together only explain less than five percent of the observed
variation.61 Models accounting for all autosomal single-nucleotide
polymorphisms estimate that 17% of the variance in BMI can be
explained using sequence variants in the genome.62 Obesitydiscordant MZ twins have provided a novel approach to dissecting
the massive amount of data provided by the GWA studies to
uncover novel obesity candidate genes missed by earlier methods
of analysis. We recently published a new study design whereby
adipose tissue transcript profiles of obesity-discordant MZ twins
are contrasted with transcript profiles of unrelated individuals
differing in BMI.2 Although most of the transcripts differing
between the obese and lean co-twin were also found to be
differentially regulated in the sample of unrelated individuals, a
collection of transcripts were uncovered that were differentially
regulated only in the unrelated individuals. By definition all genes
differently regulated in the MZ twins should be ‘reactive’ to the
obese state, whereas the differentially expressed transcripts
among the unrelated individuals should be a mix of ‘reactive’
genes (the majority) as well as some genes that may have a causal
role in the variation of BMI. Testing for association of
single-nucleotide polymorphisms in and around the potentially
‘causative’ genes revealed a total of 13 novel genes associating
with BMI in the large ENGAGE consortium study sample
composed of B21 000 individuals. The top finding, also replicated
in a smaller GWA was the F13A1 gene encoding for the blood
coagulation factor 13 involved in forming crosslinks between
fibrin molecules in a forming clot, thereby stabilizing it.63 Variants
in F13A1 have previously been implicated in predisposition to
& 2012 Macmillan Publishers Limited
Causes and consequences of obesity in twins
J Naukkarinen et al
1023
deep-vein thrombosis and with obesity known to be a risk factor
for the development of thrombi,64 this provided a new potential
link between obesity and thrombosis.
FUTURE QUESTIONS
With an ever increasing amount of information concerning factors
influencing the development of obesity and its pathological
sequelae, whole new questions have opened up for research.
New technological advances such as the large-scale microbiomic,
epigenetic, transcriptomic, metabolomic, lipidomic and proteomic
platforms also enable scientists to ask more detailed questions
about the process of obesity, development of insulin resistance,
Type 2 diabetes and atherosclerosis. New insights into the
regulation of gene transcription by epigenetic mechanisms,
the role of brown fat and different nutritional factors are all fields
of study where utilization of rare, discordant MZ twins have much
to offer as a model of acquired obesity, irrespective of genetic
background, age and gender. Specific research questions that are
in need of investigation include the observation that a proportion
of the obese population seems to be protected from the adverse
health effects of obesity (dyslipidemia, diabetes, and so on). Study
of discordant MZ twins that exhibit this ‘metabolically healthy,
obese’ phenotype would shed light on the possible protective
mechanisms---including ones that are responsible for the protection
from liver fat accumulation. Related to this question, more detailed
analyses is needed of the different lipid species originating from
the various fat depots in the body and how they are related to liver
fat, mitochondrial dysfunction, inflammation, and the concomitant
dyslipidemia and insulin resistance. The twin pairs here described
present with numerous different environmental and genetic
factors predisposing to obesity, as well as varying physiological
responses to the challenge brought about by excess body fat. It
can therefore be hypothesized that not all individuals will benefit
from similar interventions or treatments, and twin research will
undoubtedly provide the best human experimental setup for
answering these kinds of questions in the future.
ACKNOWLEDGEMENTS
The study was supported by the Helsinki University Hospital Research Funds
(AR, KHP), grants from Novo Nordisk, Gyllenberg, Yrjö Jahnsson, Biomedicum Helsinki,
Jalmari and Rauha Ahokas Foundation, and the Finnish Foundation for Cardiovascular
Research (KHP), the Academy of Finland Centre of Excellence in Complex Disease
Genetics, EU FP5 project GenomEUtwin, and EU FP7 projects DIOGENES, ENGAGE and
TORNADO (JK). We wish to thank professors Hannele Yki-Järvinen, Matej Oresic,
Leena Peltonen, the TwinFat team and the Finnish twins for their invaluable help in
the metabolic and genetic studies described in this review.
REFERENCES
The authors declare no conflict of interest.
1 Zwijnenburg PJ, Meijers-Heijboer H, Boomsma DI. Identical but not the same: the
value of discordant monozygotic twins in genetic research. Am J Med Genet B
Neuropsychiatr Genet 2010; 153B: 1134 - 1149.
2 Naukkarinen J, Surakka I, Pietilainen KH, Rissanen A, Salomaa V, Ripatti S et al. Use
of genome-wide expression data to mine the ‘Gray Zone’ of GWA studies leads to
novel candidate obesity genes. PLoS Genet 2010; 6: e1000976.
3 Kaprio J, Koskenvuo M. Genetic and environmental factors in complex diseases:
the older Finnish Twin Cohort. Twin Res 2002; 5: 358 - 365.
4 Rose RJ, Dick DM, Viken And RJ, Kaprio J. Gene-environment interaction in
patterns of adolescent drinking: regional residency moderates longitudinal
influences on alcohol use. Alcohol, Clin Exp Res 2001; 25: 637 - 643.
5 Pietilainen KH, Rissanen A, Kaprio J, Makimattila S, Hakkinen AM, Westerbacka J
et al. Acquired obesity is associated with increased liver fat, intra-abdominal fat,
and insulin resistance in young adult monozygotic twins. Am J Physiol 2005; 288:
E768 - E774.
6 Pietilainen KH, Rissanen A, Laamanen M, Lindholm AK, Markkula H, Yki-Jarvinen H
et al. Growth patterns in young adult monozygotic twin pairs discordant and
concordant for obesity. Twin Res 2004; 7: 421 - 429.
7 Pietilainen KH, Kannisto K, Korsheninnikova E, Rissanen A, Kaprio J, Ehrenborg E
et al. Acquired obesity increases CD68 and tumor necrosis factor-alpha and
decreases adiponectin gene expression in adipose tissue: a study in monozygotic
twins. J Clin Endocrinol Metab 2006; 91: 2776 - 2781.
8 Kannisto K, Pietilainen KH, Ehrenborg E, Rissanen A, Kaprio J, Hamsten A
et al. Overexpression of 11beta-hydroxysteroid dehydrogenase-1 in adipose
tissue is associated with acquired obesity and features of insulin resistance:
studies in young adult monozygotic twins. J Clin Endocrinol Metab 2004; 89:
4414 - 4421.
9 Pietilainen KH, Bergholm R, Rissanen A, Kaprio J, Hakkinen AM, Sattar N et al.
Effects of acquired obesity on endothelial function in monozygotic twins. Obesity
(Silver Spring, Md) 2006; 14: 826 - 837.
10 Mustelin L, Pietilainen KH, Rissanen A, Sovijarvi AR, Piirila P, Naukkarinen J et al.
Acquired obesity and poor physical fitness impair expression of genes of
mitochondrial oxidative phosphorylation in monozygotic twins discordant for
obesity. Am J Physiol 2008; 295: E148 - E154.
11 Pietilainen KH, Sysi-Aho M, Rissanen A, Seppanen-Laakso T, Yki-Jarvinen H,
Kaprio J et al. Acquired obesity is associated with changes in the serum lipidomic
profile independent of genetic effects - a monozygotic twin study. PLoS One 2007;
2: e218.
12 Pietilainen KH, Naukkarinen J, Rissanen A, Saharinen J, Ellonen P, Keranen H et al.
Global transcript profiles of fat in monozygotic twins discordant for BMI:
pathways behind acquired obesity. PLoS Med 2008; 5: e51.
13 Mustelin L, Silventoinen K, Pietilainen K, Rissanen A, Kaprio J. Physical activity
reduces the influence of genetic effects on BMI and waist circumference: a study
in young adult twins. Int J Obes (2005) 2009; 33: 29 - 36.
14 Maes HH, Neale MC, Eaves LJ. Genetic and environmental factors in relative body
weight and human adiposity. Behav Genet 1997; 27: 325 - 351.
15 Schousboe K, Willemsen G, Kyvik KO, Mortensen J, Boomsma DI, Cornes BK et al.
Sex differences in heritability of BMI: a comparative study of results from twin
studies in eight countries. Twin Res 2003; 6: 409 - 421.
16 Silventoinen K, Rokholm B, Kaprio J, Sorensen TI. The genetic and environmental
influences on childhood obesity: a systematic review of twin and adoption
studies. Int J Obes (2005) 2010; 34: 29 - 40.
17 Silventoinen K, Kaprio J. Genetics of tracking of body mass index from birth
to late middle age: evidence from twin and family studies. Obes Facts 2009; 2:
196 - 202.
18 Pietilainen KH, Kaprio J, Rissanen A, Winter T, Rimpela A, Viken RJ et al.
Distribution and heritability of BMI in Finnish adolescents aged 16y and 17y:
a study of 4884 twins and 2509 singletons. Int J Obes Relat Metab Disord 1999;
23: 107 - 115.
19 Allison DB, Kaprio J, Korkeila M, Koskenvuo M, Neale MC, Hayakawa K. The
heritability of body mass index among an international sample of monozygotic
twins reared apart. Int J Obes Relat Metab Disord 1996; 20: 501 - 506.
& 2012 Macmillan Publishers Limited
International Journal of Obesity (2012) 1017 - 1024
SUMMARY
To summarize 10 years of the Finnish twin research into obesity,
adult BMI is a highly heritable trait, but environmental influences
have a major role in the eventual development of obesity and
associated comorbidities. Higher levels of physical activity in
adolescence can prevent the development of obesity in adult age,
and the genetic predisposition to weight gain can be counteracted by an active lifestyle. Even at an early age, several
prediabetic and preatherosclerotic changes can be observed
associated with obesity: insulin resistance, accumulation of fat in
the liver, proatherogenic changes in the lipid profile and lowgrade chronic inflammation of the fat tissue that may be initiated
by proinflammatory changes in the lipid composition of the cell
membrane, depletion of mitochondrial DNA copy number and
concomitant downregulation of the BCAA catabolism and energy
generating oxidative phosphorylation in mitochondria (Figure 5).
Simultaneously physical fitness declines and the ability to
participate in the very activities that promote healthy cardiovascular and endocrine functions are compromised, leading to a
vicious cycle of more weight gain and an escalation in the
pathological processes described above. The studies here
reviewed continue to give more information on these processes
and the ongoing follow-up of these rare, discordant twins is sure
to provide even more insight into the continuing process of
weight gain over an individual’s lifetime.
CONFLICT OF INTEREST
Causes and consequences of obesity in twins
J Naukkarinen et al
1024
20 Bouchard C, Tremblay A, Despres JP, Nadeau A, Lupien PJ, Theriault G et al. The
response to long-term overfeeding in identical twins. N Eng J Med 1990; 322:
1477 - 1482.
21 Bouchard C, Tremblay A, Despres JP, Theriault G, Nadeau A, Lupien PJ et al. The
response to exercise with constant energy intake in identical twins. Obes Res
1994; 2: 400 - 410.
22 Hainer V, Stunkard AJ, Kunesova M, Parizkova J, Stich V, Allison DB. Intrapair
resemblance in very low calorie diet-induced weight loss in female obese
identical twins. Int J Obes Relat Metab Disord 2000; 24: 1051 - 1057.
23 Hainer V, Stunkard A, Kunesova M, Parizkova J, Stich V, Allison DB. A twin study
of weight loss and metabolic efficiency. Int J Obes Relat Metab Disord 2001; 25:
533 - 537.
24 Berg K, Kondo I, Drayna D, Lawn R. Variability gene effect of cholesteryl ester
transfer protein (CETP) genes. Clin Genet 1989; 35: 437 - 445.
25 Kirk EA, Moe GL, Caldwell MT, Lernmark JA, Wilson DL, LeBoeuf RC. Hyper- and
hypo-responsiveness to dietary fat and cholesterol among inbred mice: searching
for level and variability genes. J Lipid Res 1995; 36: 1522 - 1532.
26 Zhu M, Ji G, Jin G, Yuan Z. Different responsiveness to a high-fat/cholesterol diet
in two inbred mice and underlying genetic factors: a whole genome microarray
analysis. Nut Metabol 2009; 6: 43.
27 van der Sluis S, Dolan CV, Neale MC, Posthuma D. A general test for geneenvironment interaction in sib pair-based association analysis of quantitative
traits. Behav Genet 2008; 38: 372 - 389.
28 Leskela P, Ukkola O, Vartiainen J, Ronnemaa T, Kaprio J, Bouchard C et al. Fasting
plasma total ghrelin concentrations in monozygotic twins discordant for obesity.
Metabol: Clin Exp 2009; 58: 174 - 179.
29 Friedlander Y, Austin MA, Newman B, Edwards K, Mayer-Davis EI, King MC.
Heritability of longitudinal changes in coronary-heart-disease risk factors in
women twins. Am J Hum Genet 1997; 60: 1502 - 1512.
30 Fogelholm M, Kukkonen-Harjula K. Does physical activity prevent weight gain - a
systematic review. Obes Rev 2000; 1: 95 - 111.
31 Pietilainen KH, Kaprio J, Borg P, Plasqui G, Yki-Jarvinen H, Kujala UM et al.
Physical inactivity and obesity: a vicious circle. Obesity (Silver Spring, Md) 2008; 16:
409 - 414.
32 Waller K, Kaprio J, Lehtovirta M, Silventoinen K, Koskenvuo M, Kujala UM. Leisuretime physical activity and type 2 diabetes during a 28 year follow-up in twins.
Diabetologia 2010; 53: 2531 - 2537.
33 Leskinen T, Sipila S, Alen M, Cheng S, Pietilainen KH, Usenius JP et al. Leisure-time
physical activity and high-risk fat: a longitudinal population-based twin study.
Int J Obes (2005) 2009; 33: 1211 - 1218.
34 Li S, Zhao JH, Luan J, Ekelund U, Luben RN, Khaw KT et al. Physical activity
attenuates the genetic predisposition to obesity in 20 000 men and women from
EPIC-Norfolk prospective population study. PLoS Med 2010; 7: e1000332.
35 McCaffery JM, Papandonatos GD, Bond DS, Lyons MJ, Wing RR. Gene X
environment interaction of vigorous exercise and body mass index among male
Vietnam-era twins. Am J Clin Nutr 2009; 89: 1011 - 1018.
36 Silventoinen K, Hasselbalch AL, Lallukka T, Bogl L, Pietilainen KH, Heitmann BL
et al. Modification effects of physical activity and protein intake on heritability of
body size and composition. Am J Clin Nutr 2009; 90: 1096 - 1103.
37 Wang X, Ding X, Su S, Spector TD, Mangino M, Iliadou A et al. Heritability of insulin
sensitivity and lipid profile depend on BMI: evidence for gene-obesity interaction.
Diabetologia 2009; 52: 2578 - 2584.
38 Greenfield JR, Samaras K, Jenkins AB, Kelly PJ, Spector TD, Campbell LV. Moderate
alcohol consumption, dietary fat composition, and abdominal obesity in women:
evidence for gene-environment interaction. J Clin Endocrinol Metab 2003; 88:
5381 - 5386.
39 Pietilainen KH, Korkeila M, Bogl LH, Westerterp KR, Yki-Jarvinen H, Kaprio J et al.
Inaccuracies in food and physical activity diaries of obese subjects:
complementary evidence from doubly labeled water and co-twin assessments.
Int J Obes (2005) 2010; 34: 437 - 445.
40 Rissanen A, Hakala P, Lissner L, Mattlar CE, Koskenvuo M, Ronnemaa T. Acquired
preference especially for dietary fat and obesity: a study of weight-discordant
monozygotic twin pairs. Int J Obes Relat Metab Disord 2002; 26: 973 - 977.
41 Bogl LH, Pietilainen KH, Rissanen A, Kaprio J. Improving the accuracy of selfreports on diet and physical exercise: the co-twin control method. Twin Res Hum
Genet 2009; 12: 531 - 540.
42 Bjorntorp P. Hormonal control of regional fat distribution. Hum Reprod(Oxford,
England) 1997; 12 (Suppl 1): 21 - 25.
International Journal of Obesity (2012) 1017 - 1024
43 Bouchard C. Genetic determinants of regional fat distribution. Hum Reprod
(Oxford, England) 1997; 12 (Suppl 1): 1 - 5.
44 Heid IM, Jackson AU, Randall JC, Winkler TW, Qi L, Steinthorsdottir V et al. Metaanalysis identifies 13 new loci associated with waist-hip ratio and reveals sexual
dimorphism in the genetic basis of fat distribution. Nat Genet 2010; 42: 949 - 960.
45 Makkonen J, Pietilainen KH, Rissanen A, Kaprio J, Yki-Jarvinen H. Genetic factors
contribute to variation in serum alanine aminotransferase activity independent of
obesity and alcohol: a study in monozygotic and dizygotic twins. J Hepatol 2009;
50: 1035 - 1042.
46 Huang F, Subbaiah PV, Holian O, Zhang J, Johnson A, Gertzberg N et al.
Lysophosphatidylcholine increases endothelial permeability: role of PKCalpha and
RhoA cross talk. Am J Physiol Lung Cell Mol Physiol 2005; 289: L176 - L185.
47 Davignon J, Ganz P. Role of endothelial dysfunction in atherosclerosis. Circulation
2004; 109 (23 Suppl 1): III27 - III32.
48 Meyers MR, Gokce N. Endothelial dysfunction in obesity: etiological role in
atherosclerosis. Curr Opin Endocrinol, Diab, Obes 2007; 14: 365 - 369.
49 Feletou M, Vanhoutte PM. Endothelial dysfunction: a multifaceted disorder
(The Wiggers Award Lecture). Am J Physiol Heart Circ Physiol 2006; 291:
H985 - 1002.
50 Weyer C, Funahashi T, Tanaka S, Hotta K, Matsuzawa Y, Pratley RE et al.
Hypoadiponectinemia in obesity and type 2 diabetes: close association with
insulin resistance and hyperinsulinemia. J Clin Endocrinol Metab 2001; 86:
1930 - 1935.
51 Ahima RS. Metabolic actions of adipocyte hormones: focus on adiponectin.
Obesity (Silver Spring, Md) 2006; 14 (Suppl 1): 9S - 15S.
52 Felig P, Marliss E, Cahill Jr GF. Plasma amino acid levels and insulin secretion in
obesity. N Engl J Med 1969; 281: 811 - 816.
53 Leskinen T, Rinnankoski-Tuikka R, Rintala M, Seppanen-Laakso T, Pollanen E,
Alen M et al. Differences in muscle and adipose tissue gene expression and
cardio-metabolic risk factors in the members of physical activity discordant twin
pairs. PLoS One 2010; 5: e12609.
54 Mootha VK, Lindgren CM, Eriksson KF, Subramanian A, Sihag S, Lehar J et al.
PGC-1alpha-responsive genes involved in oxidative phosphorylation are
coordinately downregulated in human diabetes. Nat Genet 2003; 34: 267 - 273.
55 Weisberg SP, McCann D, Desai M, Rosenbaum M, Leibel RL, Ferrante Jr AW.
Obesity is associated with macrophage accumulation in adipose tissue. J Clin
Invest 2003; 112: 1796 - 1808.
56 Xu H, Barnes GT, Yang Q, Tan G, Yang D, Chou CJ et al. Chronic inflammation in fat
plays a crucial role in the development of obesity-related insulin resistance. J Clin
Invest 2003; 112: 1821 - 1830.
57 Gauthier MS, Ruderman NB. Adipose tissue inflammation and insulin resistance:
all obese humans are not created equal. Biochem J 2010; 430: e1 - e4.
58 Pietilainen KH, Rog T, Seppanen-Laakso T, Virtue S, Gopalacharyulu P, Tang J et al.
Association of lipidome remodeling in the adipocyte membrane with acquired
obesity in humans. PLoS Biol 2011; 9: e1000623.
59 Kettunen J, Perola M, Martin NG, Cornes BK, Wilson SG, Montgomery GW et al.
Multicenter dizygotic twin cohort study confirms two linkage susceptibility loci
for body mass index at 3q29 and 7q36 and identifies three further potential novel
loci. Int J Obes (2005) 2009; 33: 1235 - 1242.
60 Larsen LH, Echwald SM, Sorensen TI, Andersen T, Wulff BS, Pedersen O. Prevalence
of mutations and functional analyses of melanocortin 4 receptor variants
identified among 750 men with juvenile-onset obesity. J Clin Endocrinol Metab
2005; 90: 219 - 224.
61 Speliotes EK, Willer CJ, Berndt SI, Monda KL, Thorleifsson G, Jackson AU et al.
Association analyses of 249 796 individuals reveal 18 new loci associated with
body mass index. Nat Genet 2010; 42: 937 - 948.
62 Yang J, Manolio TA, Pasquale LR, Boerwinkle E, Caporaso N, Cunningham JM et al.
Genome partitioning of genetic variation for complex traits using common SNPs.
Nat Genet 2011; 43: 519 - 525.
63 Ichinose A, Davie EW. Characterization of the gene for the a subunit of human
factor XIII (plasma transglutaminase), a blood coagulation factor. Proc Natl Acad
Sci USA 1988; 85: 5829 - 5833.
64 Wells PS, Anderson JL, Scarvelis DK, Doucette SP, Gagnon F. Factor XIII Val34Leu
variant is protective against venous thromboembolism: a HuGE review and
meta-analysis. Am J Epidemiol 2006; 164: 101 - 109.
65 Cole TJ, Freeman JV, Preece MA. British 1990 growth reference centiles for weight,
height, body mass index and head circumference fitted by maximum penalized
likelihood. Stat Med 1998; 17: 407 - 429.
& 2012 Macmillan Publishers Limited