The
integrated effects of aging on thyroidal function and other hormonal and
regulatory aspects of metabolism are an important element in various aspects of
aging and longevity [1]. As reviewed by
Taylor et al, the quantity of thyroid hormones normally released from the
thyroidal epithelium is determined in large part by the actions of thyroid
stimulating hormone (TSH), which stimulates the thyroid gland to release T4 and
some T3 to the peripheral circulation, in response to metabolic and physiologic
demands including growth and development across the lifespan. The T3 is the
physiologically most active form of the hormone and is produced in peripheral
tissues via the actions of T4-5’ deiodinase activity on the outer phenolic ring
of the T4 moiety [2-4]. Typical effects in early childhood are directed to
genomic expression of neurologic and physical growth and development, while in
adult mammalian species the primary effects may be observed via increases in
resting and catecholamine-stimulated metabolic rates, commonly measured in
rodents via measures of VO2 under conditions of thermal neutrality and
corrected for differences in body size and mass [5,6]. While the progression of
plasma TSH concentrations typically suggest a U-shaped curve, with the highest
plasma concentrations reported at either extreme of the lifespan, the
corresponding plasma T4 or T3 concentrations may not always follow the same
profile [1]. In addition, excursions from the normal plasma concentrations may
become associated with a variety of pathophysiologic conditions including
cardiovascular, musculoskeletal, cognitive and other disorders, in addition to
alterations in diet and environment [1].
Thyroid
hormones initiate their metabolic effects following stereospecific genomic
binding to receptor domains that are highly committed to bind T3, the
physiologically most active form of the iodothyronines [3,7,8]. Thus, it is
instructive to assess the parameters of T3 generation in key tissues and
examine their associated contributions to glycemic and metabolic parameters.
This is especially important as thyroidal and other physiologic factors
contribute to the efficiency of energy metabolism and storage, at a time in
history where the prevalence of obesity, overweight conditions, and metabolic syndrome
(MeTS) are approaching epidemic proportions in many communities [9-13]. The
underlying basis for the changes in the prevalence in overweight conditions in
recent generations is likely secondary to many factors including changes in
common dietary practices linked to the prevailing food supply in modern
society, greater consumption of preprocessed foods of greater energy density, a
more sedentary lifestyle than in previous generations, and likely economic
factors [14].
Some
70 years ago the US Department of Agriculture established the food pyramid, as
a pictorial guide to assist individuals in selecting healthier food selections.
Since the original pyramid was established however, many changes have been
incorporated into the pyramid in attempts to promote decreases in saturated fat
intake, while increasing the proportions of carbohydrate sources in part to
improve palatability, dietary appeal and broader public acceptance. In
addition, the development of manufactured sweeteners including the widely preferred
high fructose corn syrup (HFCS) have now become widely accepted in many
cultures and communities, thereby increasing fructose ingestion up to 5-fold or
more since the introduction of HFCS in the food and carbonated beverage supply
chains and thereby likely further contributing to the burgeoning prevalence of
overweight and obesity and their comorbidities in adolescents and adults [14].
The
independent contributions of aging and obesity or overweight conditions on
sirtuin- mediated thyroidal function have not been fully clarified.
Hyperinsulinemia and insulin resistance are common hallmarks of adiposity, due
in part secondary to disruption of the normal biosynthetic and intracellular
transport of the glucose transporter proteins including GLUT4, essential for
insulin- dependent glucose uptake in muscle and adipose tissues [14-16]. These
two sources are the most prevalent tissue beds where insulin resistance is
expressed, and together they contribute to the multiple comorbidities commonly
associated with obesity [15,16]. The development of the congenic corpulent rat
has now become an important animal model to investigate the independent
contributions of obesity and aging, whereby the only difference between the
obese and the lean phenotypes is the epigenetic expression of early onset
obesity [17-19]. In addition, the background strain is derived from the
longevity-prone LA/N rat maintained in the small animal genetics section of the
NIH, and backcrossed multiple times with the Koletsky rat to acquire the
congenic obesity or corpulent (-cp) trait by Hansen [19].
In
previous studies in the corpulent rat strains have demonstrated that the
capacity for non-shivering thermogenesis (NST) under conditions of thermal
neutrality to be decreased typically by an average of 20% under most
environmental and dietary conditions [17,20]. The decreases in NST were found
to be attributed to a combination of sympathetic and thyroidal actions in the
obese phenotype [11,17,20]. In addition to the impaired thermic responses to
factors of diet and environment, in those studies measures of serum T3
concentrations were also often found to be lower in littermates of the obese
male and female phenotype, suggestive of a physiological subclinical hypothyroid
state [11]. A condition consistent with subclinical hypothyroidism has been
described by several authors, where only modest deficits may account for excess
weight gain [1,12,13]. Adiposity with an onset of middle age or beyond is also
a common observation in aging [10].
Aging
contributes additional impairments in the thermogenic responses to diet and
cold environment in the obese phenotype in this strain [17,20-21]. Although the
physiologic mechanisms that are operative for the impaired thermic responses in
aging remain unclear they do appear to include decreases in mitochondrial
oxidative activity, a primary source of oxygen utilization in mammalian tissues
[22-24]. The biochemical mechanism of the NST component of thermogenesis in
most warm-blooded animals including rodents occurs in large part via
thermogenic activity in brown adipose tissue, where specialized mitochondria
demonstrate a neuroendocrine-mediated activation and uncoupling of oxidative
phosphorylation of ATP to ADP, resulting in the obligatory generation of heat
that can be utilized to maintain body temperature regulation following
alterations in diet and thermal environment [23,24]. Other tissues including
liver and skeletal muscle also contribute to metabolic heat production, albeit
via well- established biochemical mechanisms in intermediary substrate
metabolism rather than the specialized mitochondrial process common to BAT [3,11,21].
Glucose uptake in BAT is essential for BAT thermogenesis to progress, and
Marette et al. demonstrated that insulin resistance is a major factor in the
impaired thermogenic responses in obese rats [25]. In addition, Tatelman et al [22]
reported that mitochondrial activity in liver tissues of lean Sprague Dawley
rats became increased in parallel to serum T3 concentrations following a
thermogenic diet and decreased with advancing age in when followed up to 4
months of age due to adaptations in ?-glycerate phosphate shuttle activity, but
hepatic thermogenic activity responses to a longer duration of aging and to the
obese phenotype remain unclear [20].
Historically
most related studies have been conducted in younger male animals. In previous
studies in the corpulent rat strains, it was reported that factors of diet,
cold exposure, and sympathomimetic responses to norepinephrine were impaired in
the obese phenotype of both the LA/Ntul//-cp, the T2DM-prone SHR/Ntul//-cp and
other obese strains strains [17,18,25-28]. Thus, the purpose of the present
study was to determine the effects of aging and obesity on parameters of T3
generation in a key metabolic tissue and their association with the impaired
thermogenesis in aging and obesity in a congenic female animal model highly
predisposed to obesity without the usual comorbidities of NIDDM or
hypertension. The various roles of sirtuins in mediating thyroidal responses
are undergoing active investigation in several laboratories.
The
sirtuins represent a recently discovered class of NAD+- deacetylases (n = 7 at
last count) that function as silent information transfer factors [29]. The
sirtuins are dependent upon cellular NAD+ and facilitate the deacetylation of
lysine and possibly other amino acid residues of chromatin-based histones and
other cellular proteins that contribute to epigenetic, genomic expression in
tissues following deacetylation [29]. The deacetylation reactions occur in
response to nutritional and environmental signals likely transmitted via neural
signals that likely originate in hypothalamic paraventricular nuclei and result
in their expression in the pituitary and other tissues [29-31]. The NAD+
availability is ultimately derived mostly from the NAD+/NADH ratios mostly
generated via heightened mitochondrial activity, thus are a generalized
reflection of nutritional and environmental status and recent macronutrient
ingestions [22]. In addition, overnutrition is associated with a shift from
non-inflammatory M2, protective ROS generating macrophages toward
proinflammatory M1 (iROS) generating macrophages differentially in visceral vs non-visceral adipose tissue depots,
thereby contributing to the magnitude of systemic IL-6 and TNF?
cytokine-mediated inflammation and their eventual pathophysiologic sequelae
[30-32].
Nutritional
inputs, predominantly from carbohydrate and other metabolizable energy sources,
can bring about increases in insulinogenic activity and a decrease in sirtuin
activity [33,34]. The decreases in sirtuin activity can result in reciprocal
increases in the conversion of T4 to metabolically active T3 and in
accompanying increases in resting metabolic rates. In addition, overnutrition
and sustained positive energy balance contributes to increases in cell cycle
replication and inflammatory M1-macrophage induced DNA damages, accompanied
with a destabilization and shortening of telomeres over extended periods
[35-37]. Since longer telomere length is linked to longevity, the combined
effects of overnutrition on sirtuins thereby may negatively impact potential
longevity [37,38]. In contrast, fasting, caloric deprivation, and starvation
bring about increases in sirtuin availability, improvements in insulin
sensitivity, accompanied by a shifting from outer ring to inner ring deiodinase
activity [3,4,8]. The latter actions increase the diversion of T4 from T3 to
formation of reverse T3 (rT3), a metabolically inactive iodothyronine that
fails to associate with the TR?-subunit domain, with accompanying decreases in
RMR, cell cycle activity, and decreased M1 macrophage inflammatory cytokine
activity including iROS-mediated DNA damage and thus an enhanced, more healthful
longevity when in neutral or negative energy balance [8,38-40].
The
caloric efficiency is notably more favorable in the obese than the lean
phenotype of the corpulent rat strains, while the effects of the
epigenetically-expressed obesity are expressed differentially in different
adipose tissue depots before and following weight gain modulation [41]. Thus,
the purpose of the present study was to determine the effects of aging and
obesity on parameters of T3 generation in a key metabolic tissue and its
association with the impaired thermogenesis often observed in aging and obesity
in female littermates, in a congenic animal model that is highly predisposed to
early onset obesity but without the usual comorbidities of NIDDM or
hypertension.
Methods
Groups
of post-adolescent female, lean and obese littermate LA/Ntul//-cp rats were fed Purina rodent chow
diet throughout the duration of the study to construct a 2 x 3 experimental
design consisting of 2 phenotypes (lean and obese) and 3 age points (4, 14, and
24 months of age). Thus, this design comprised the projected lifespan of the
obese phenotype. The animals were maintained in plexiglass showbox cages lined
with 1 inch of pine shavings, with free access to Purina chow and house water,
and maintained at 20-21 °C, 50% relative humidity under a reverse light cycle
(Dark phase 0800-2000 daily). Animals were routinely studied during the dark
phase and were fasted briefly (approximately 4 hours) prior to measures of
fasting blood glucose, insulin, RMR, or thyroidal parameters. Measures of daily
food consumption were obtained over a 24-hour period as described by Vedula et
al 26 and expressed as kjoules consumed/rat/day based on the manufacturer’s
certificate of analysis and energy density. Measures of live body weight as a
measure of ongoing wellness were obtained periodically with an Ohaus animal
balance and recorded to the nearest gram. Measures of resting metabolic rate
were determined at thermal neutrality (30°C) in fasted, quietly resting animals
via a closed-circuit Collins small animal respiration apparatus fitted with a
4-litre chamber and maintained at 30°C submersed in a circulating water bath
and corrected for factors of altitude and relative humidity [5,6,17]. At 4, 14
and 24 weeks of age groups of lean and obese rats were sacrificed by cervical
dislocation with a small animal guillotine, and truncal bloods collected for
hormone and substrate analysis. The retroperitoneal, dorsal, and interscapular
white adipose tissue depots and the liver were dissected in their entirety and
weighed to the nearest mg. Approximately 100 mg aliquots of liver tissue was
homogenized and prepared for assay of Type II thyroxine 5’-deiodinase activity
in the presence of dithiothreitol (DTT) as described elsewhere [27].
Measures
of tissue and serum T4 and T3 were determined via solid phase RIA [27]. The
plasma half-life of T4 in 4 month-old male biological littermate rats was
determined following the intravenous injection of 1 µCi of 1-131-T4 in the tail
vein within a 1 minute duration, and collection of 100 µl aliquots of tail tip
blood in heparinized microtubes via tail bleeding for up to 8 hours
post-infusion and plotting the rate of decline in plasma radioactivity [28]. Data
were analyzed via standard statistical procedures including descriptive
statistics, ANOVA, Students t test, and Pages L test for detection of trend
analysis [42,43]. The study was approved by the Institutional Animal Care and
use committee.
Results
The
effects of aging on longevity are depicted in (Figure 1) and indicate that the
effects of obesity on longevity are associated with a significant decrease in
typical duration in both phenotypes, with lean females projected survival the
greatest at approximately 50 months of age, while obese male rats exhibit the
shortest projected lifespan, seldom exceeding 25 months of age under typical
laboratory conditions. The effects of obesity on decreased projected longevity
averaged ~28% in both phenotypes. Body weights of the animals of this study at
4, 14 and 24 months of age are exhibited in (Figure 2) and show that body
weights increased in both phenotypes with advancing age, and that the final
body weights of the obese phenotype far exceeded the weights of their lean
littermates at each age surveyed. Adiposity, determined as a percent of final
body weight in the obese phenotype also
greatly exceeded the percent of body weight represented by the sum of three fat
depots of the lean phenotype (Figure 3A,B) and indicated that the sum of the
three fat depots measured in the obese phenotype greatly exceeded those of the
lean phenotype. In addition, the combined fat pad mass increased progressively
with aging in both phenotypes with advancing age with the greatest increases in
the obese phenotype. The effects of aging on RMR are depicted in Figure 4 and
indicate that the RMR of lean rats was greater than the obese rats at each age
studied (p = < 0.05), and that the RMR tended to decrease with advancing age
in both phenotypes with the greatest decrease noted in the obese phenotype
(Figure 4).
The
effects of aging and obesity on daily energy intake are depicted in (Figure 5)
and indicate that energy intake, determined by multiplying the grams of food
intake per day times the energy content per gram of chow (3.34 kcal/13.975
kjoules/g) as greater in the obese phenotype than the lean phenotype, and
increased in both phenotypes with aging at the 14 month age point, with a trend
toward stabilizing at the oldest age studied in both phenotypes. Regardless of
the trend, the consistent greater energy intake of the obese phenotype and is
indicative of simple hyperphagia, albeit it to support a significantly greater
body mass in the obese littermates.
The
effects of ageing on fasting plasma insulin glucose and insulin: glucose ratio
is depicted in Figure 6A and 6B respectively and indicate that fasting plasma
insulin concentrations in the obese phenotype were significantly greater than
were observed in their lean littermates at each age determined (left panel,
Figure 6A). Although there was a modest trend toward greater fasting glucose
concentrations in the obese phenotype, the plasma glucose concentrations
remained within the normal range in both phenotypes at all ages studied.
(Figure 6B, right panel). In addition, plasma insulin concentrations in the
obese phenotype tended to decrease with advancing age, while the corresponding
differences in fasting plasma insulin in the lean phenotype were modest by
comparison (Figure 6A). The insulin to glucose ratios are depicted in (Figure
6B), and indicate the I:G ratios in the obese phenotype were greater than those
in their lean littermates at all ages studied, suggestive of evidence of
chronic insulin resistance in the obese phenotype, albeit it decreasing in
absolute magnitude in spite of the greater relative adiposity depicted in
Figure 3B.
Thyroidal
parameters of aging and obesity on plasma T3, T4 and T4/T3 ratios are presented
in Figures 7 through 9 below. The effects of aging and obesity are depicted in
(Figure 7A and 7B) and indicate that plasma T3 while remaining within the
normal range in both phenotypes and at all three ages studied, illustrated a
modest trend toward an age-related increase. Plasma T4 concentrations were
greater at 4 months of age than at 14 or 24 months of age, but plasma
concentrations were similar in both phenotypes in all age comparisons made. The
T4 to T3 ratios are depicted in Figure 7B and indicate that the ratio decreased
progressively in both phenotypes with aging, suggesting that less T3 may be
being generated per unit of T4 at each age studied, and that the enzymatic
peripheral T3 generation became further decreased with advanced aging in both
phenotypes, with the greatest decreases noted in the obese phenotype. The effects
of aging on liver T4-5’ Type II deiodinase are depicted in Figure 8 and
indicate that the activity of the enzyme was greater in the lean than the obese
phenotype and decreased with advancing age in both phenotypes. The liver is a
major site for generating T3 for subsequent uptake in peripheral tissues via
Type-2 deiodinase. Stereospecific enzymatic conversion of T4 to T3 occurs via
DI-1 a DI-2, which catalyze a deiodination reaction only at the outer ring
5’position to yield metabolically active T3, the physiologically most active
form of the hormone, and which may suggest a decreased receptor-domain binding
affinity for T3.
The
5’-deiodinase activity appears to be substantially greater in the lean than the
obese phenotype, and the 5’-deiodinase activity decreases in both phenotypes
with aging. Tissue T3 concentrations are depicted in the Right panel of (Figure
8) and indicate that despite the changes in deiodinase activity the tissue
concentrations remained similar in both phenotypes with aging. Plasma T4 to T3
ratios are depicted in Figure 7B and indicate that the T4/T3 ratio decreased
with aging in both phenotypes and that the ratio was always greater in the lean
than in the obese phenotype, consistent with the differences in deiodinase
activity and decreases in the rate of deiodinase activity and generation of T3
where progressively less T3 is formed per unit of available T4 throughout most
of the adult lifespan in this strain.
The
plasma half-life of 131I-T4 was determined in a group of lean and obese male
biological littermates and was found to be approximately 40% longer in the
obese vs the lean phenotype, thereby
confirming the delayed peripheral conversion of T4 to T3 in vivo as reflected
above (Figure 9). The observation that the T4 half-life was determined in male
vs female littermate rats of a similar age in this study is deemed of little
consequence, since the onset of and magnitude obesity and the physiologic
characteristics of thermogenesis in response to diet and environment have been
found to be remarkably similar in both genders of this strain.
Discussion
The
various contributions of thyroid hormones to growth, development and aspects of
metabolism have been well documented in man and animals [1,17,20,21]. As
discussed, the regulatory process of thyroid hormone secretion begins in the
thyrotrope cells of the anterior pituitary, where TSH is released in response
to hypothalamic generated thyrotrope releasing hormone TRH, and targets the
thyroglobulin stored in the thyroid gland to release its stored T4 and small
amounts of T3 into the peripheral circulation [1].
Once in circulation, the T4 may be
converted to T3 via D-1 and D-2 deiodinase enzymes that target the 5‘ iodine
position outer phenolic ring to bring about increases in circulating T3 when
needed to support growth, development, and when the effects of nutritional and
environmental conditions dictate such need [3,8,44-47]. In contrast, under
conditions of starvation or dietary privation, the T4 is further inactivated by
T4-5-DI-3 to remove the iodine from the inner ring 5-position, located on the
inner tyrosyl ring, and thereby rendering the T4 to become a physiologically
inactive hormone [8].
In the present study, the T4
journey to metabolic activation has been established and indicates that less T3
becomes available to peripheral tissue receptor domains that are dependent on
an active T3 hormone and have been found to be decreased in the obese phenotype
as the animal undergoes the aging process [31,28,44]. As the peripheral
generation of T3 is conserved and hyperinsulinemia prevails in the obese
phenotype of this strain, the interaction of the two hormones presumably
contributes to an improved efficiency of energy conservation and storage,
decreases in resting metabolic rate, and in at least partial contributions to
the longevity demonstrated in both phenotypes of this strain [19].
The
T4 deiodination reactions have been suggested as a link to survival and
longevity, due at least in part to the broad metabolic functions facilitated by
the iodothyronine hormones, in concert with complimentary actions on aspects of
energy metabolism and storage that are exerted by insulin, catecholamines, and
other entities. Hyperinsulinemia is a hallmark of obesity, and is linked to
energy conservation by decreasing the rates of protein turnover and ATP
utilization in muscle and in lipogenesis in liver and adipose tissues
[3,11,45,46]. Dysregulation of glucocorticoid actions may also contribute to
energy conservation, via impaired endoplasmic reticulum generation and
intracellular transport of insulin-dependent GLUT glucose transporter proteins,
essential factors in cellular glucose uptake in skeletal muscle and adipose
tissues, two of the major locations which constitute the bulk of insulin
resistance in man and animals [15,16,29,32]. Factors of diet and environment
are well established factors in the modulation of thyroidal and insulinogenic
activity in man and animals. In animals, both dietary excursions and cold
exposure have been shown to bring about increases in plasma concentrations of
T3, and an apparent response to increases in thyroidal release of T4 in
combination with increases in peripheral iodothyronine deiodination in liver,
brown adipose tissue and likely other tissues, where selective deiodination of
outer- vs inner-ring iodine moieties can bring about an activation or an
inactivation of thyroidal actions in peripheral tissues. Outer ring deiodination
results in T3 generation, while inner ring deiodination results in the
formation of ‘reverse’ T3, or rT3, a physiologically inactive entity which when
present, fails to associate with the stereospecific genomic receptor domains
that mediate thyroidal actions in most if not all thyroid-receptive tissues
[3,5,33,34].
In
the present study, the peripheral generation of T3 was decreased and plasma
insulin concentrations significantly increased in the obese phenotype at all
ages studied, resulting in energy conservation as indicated by excess lipid
deposition and age associated progressive decreases in resting oxygen
consumption with aging. While the greater daily energy intake in the obese
phenotype also represents an additional factor in the development or obesity in
this strain, the impaired thermic responses to diet and environment in the
obese phenotype reported elsewhere are consistent with impaired physiological
responses to caloric intake in the obese phenotype of this and other
genetically obese strains. Whether the presence of chronic insulin resistance
may play a role in iodothyronine deiodination in the obese phenotype could not
be determined from the present experimental design of this study, but the
association of insulin resistance is likely a confounding and contributing
factor via its biochemical effects on multiple enzymatic stages in intermediary
metabolism, and their likely contributions to obesity and longevity in this
strain [15,16,31-39].
Summary and Conclusions
The
results of this study indicate that the rate of liver T4-5’-deiodinase activity
is greater in the lean than the obese phenotype and decreases progressively
with aging in both phenotypes in parallel to the decreases in fasting plasma
insulin concentrations and in gradually decreasing RMR in both phenotypes. The
significantly greater plasma half-life of T4 in the obese phenotype is
consistent with the deceased hepatic conversion of T4 to T3. A major effect of
T3, the active form of the thyroid hormone is to modulate the rates of
intermediary metabolism and oxidation of substrates, a primary source of
inflammatory ROS generation and of maintaining the rate of resting oxygen
consumption [39]. Thus, the age -related decreases in T4-5’-deiodinase activity
and in vivo peripheral conversion may contribute to the longevity in both
phenotypes via influencing the generation of inflammatory ROS downward, and
thereby minimizing their deleterious and damaging effects on pathogenic
mechanisms in the lean and obese phenotypes when formed [44].
Excess
adiposity contributes to inflammatory ROS generation, and may be a contributor
to the decreased life span in the obese phenotype [39,47]. In addition, the
longevity impact demonstrated in the obese phenotype while resulting in a
significant decrease in longevity when compared to their lean littermates,
still enabled the obese phenotype to outlive the obese of other genetically
obese rat strains, most of which appear to suffer their demise soon after one
year of age. These observations suggest that genetic factors may also be at
play in longevity in this strain, since the background LA/N strain has been
noted for its healthful longevity [17,19]. The autosomal recessive inheritance
of the -cp trait is present in both phenotypes of the strain [17-19]. In
clinical studies, modest decreases in thyroidal activity have been suggested as
a contributing factor in obesity and premature longevity, presumably via
decreasing the rate of age-linked contributors to the pathophysiology of thyroidal-mediated
changes in aging [31-35,44,47].
Acknowledgements
The
author is grateful for the Institutional resources of USAT Montserrat to
develop this manuscript.
USE of AI
No
applications of AI were utilized in the preparation of this manuscript.