Contributors to Metabolic Disease Risk Following Spinal Cord Injury

Abstract

Spinal cord injury (SCI) induced changes in neurological function have significant impact on the metabolism and subsequent metabolic-related disease risk in injured individuals. This metabolic-related disease risk relationship is differential depending on the anatomic level and severity of the injury, with high level anatomic injuries contributing a greater risk of glucose and lipid dysregulation resulting in type 2 diabetes and cardiovascular disease risk elevation. Although alterations in body composition, particularly excess adiposity and its anatomical distribution in the visceral depot or ectopic location in non-adipose organs, is known to significantly contribute to metabolic disease risk, changes in fat mass and fat-free mass do not fully account for this elevated disease risk in subjects with SCI. There are other negative adaptations in body composition including reductions in skeletal muscle mass and alterations in muscle fiber type, in addition to significant reduction in physical activity, that contribute to a decline in metabolic rate and increased metabolic disease risk following SCI. Recent studies in adult humans suggest cold- and diet-induced thermogenesis through brown adipose tissue metabolism may be important for energy balance and substrate metabolism, and particularly sensitive to sympathetic nervous signaling. Considering the alterations that occur in the autonomic nervous system (SNS) (sympathetic and parasympathetic) following a SCI, significant dysfunction of brown adipose function is expected. This review will highlight metabolic alterations following SCI and integrate findings from brown adipose tissue studies as potential new areas of research to pursue.

Introduction

Spinal Cord Injury and Metabolic Disease Risk

Sixty percent of individuals with spinal cord injury (SCI) are younger than 45 years, meaning that more than half of individuals with newly acquired SCI have the potential to live long and healthy lives [1]. While life expectancy is increasing among individuals with SCI who survive the first year after injury, it is still lower than that observed for the able-bodied (AB) population [2]. Lower life expectancy in SCI is thought to result from secondary health conditions associated with neurological impairment as individuals with (SCI) are at increased risk of developing obesity, type 2 diabetes (T2D) and cardiometabolic disease relative to the general population [3–8]. This association has been observed in multiple population cohorts across gender and age groups. A large body of evidence supports an accelerated trajectory of metabolic disorders in the SCI population, such that these health conditions occur at an earlier age vs. AB individuals [3;7–9]. Across studies, a relatively consistent metabolic profile of elevated circulating glucose, insulin resistance and hyperlipidemia is present in subjects with SCI [4;6;10–13]. Explanatory variables including a loss of lean mass (particularly skeletal muscle), reduced physical activity, excess energy intake relative to need, and systemic compensatory responses related to impaired autonomic function, mediated through the above mentioned variables, have been previously studied [14;15]. Although data show reduced energy expenditure in SCI subjects relative to uninjured controls [14;16;17], controlling for the difference in lean body mass and decreased physical activity does not fully account for the decline in energy expenditure and increased disease risk [17], particularly for those individuals with injuries to higher anatomic levels of the spine [5;15;17–20]. While multiple factors may contribute to this differential metabolic-related disease risk in subjects with SCI, this review highlights observations related to the anatomic level of injury, alterations in body composition, skeletal muscle biology, energy expenditure, metabolic substrate metabolism, sympathetic nervous system function, and a potential role for brown adipose tissue in these metabolic-disease risk relationships.

Level of Injury Effects

The site of anatomic injury to the spinal column is variable between individuals with injuries occurring in the upper regions of the spine (high-level injuries - cervical) resulting in greater overall neurologic impairment compared with low level injuries (thoracic and lumbar level). Metabolic impairment and disease risk are greater in individuals with high-level injuries [15;18;21]. Individuals with high-level injuries who have greater neurologic impairment present with significantly lower glucose tolerance, greater insulin resistance and impaired lipid profiles when compared with both uninjured controls and low-level injury cases [4;6;11;16]. Furthermore, individuals with high-level injuries have greater metabolic rate deficits, with thermoregulatory imbalance resulting in periods of hypothermia [22–24]. In addition to the spinal column, the ascending and descending sympathetic chain of ganglia coordinate a sensory/effector response in maintaining general physiologic homeostasis mediated in part through the catecholamine signaling, in particular norepinephrine [25]. Considering its anatomic arrangement, it is not surprising that impairment of SNS relative to the level of injury is observed, with greater impairment and resulting physiologic disruption accompanying a high-level injury [16–18;21;22;26–37]. These physiologic and metabolic alterations with high-level injuries (particularly the deficits in SNS function, metabolic rate and thermogenesis, and glucose-lipid metabolism resulting in glucose-intolerance and ultimately insulin resistance) suggest the presence of an unidentified tissue/organ integrating these phenotypes and mediating the metabolic-related disease risk – which we hypothesize as brown adipose tissue and discuss in the subsequent sections.

Metabolism-Related Alterations Following Spinal Cord Injury

Body Composition

Due to significant neurological impairment, body composition drastically deteriorates, as early as 6 months after SCI with an excessive loss of lean mass below the level of injury and an increased total fat mass [38–40]. Lower metabolic rates as a result of reduced lean mass further accelerate the development of excess adipose tissue (obesity) in individuals with SCI. Obesity is associated with a broad range of metabolic abnormalities that affect several different organ systems in the body via atherogenic, neuro-humoral, and hemodynamic mechanisms in the able-bodied (AB) as well as SCI populations [41–43]. Methods for determining in vivo body composition, lean mass and fat mass, have expanded significantly over the last 30 years. This includes the use of whole body estimates of fat mass and fat free mass or lean mass from anthropometry (e.g. body mass index, waist circumference) and biophysical means (e.g. under water weighing or air displacement plesthymography) to more detailed tissue and regional assessments of fast mass and distribution using in vivo imaging technologies like dual x-ray absorptiometry, computed tomography and magnetic resonance imaging. Despite these advances and the numerous studies in able body populations, detailed and accurate descriptions of the amount and distribution of fat mass and adipose tissue following spinal cord injury remains limited, particularly across population variables like sex, anatomical level and severity of injury, time since injury. Whole body relative fat mass is underestimated by body mass index (BMI) with SCI populations [44–46], indicating BMI categories and disease risk relationships should be carefully considered and possibly revised to lower category values or include additional indices (e.g. waist circumference) in subjects with SCI [46–48]. More recent advances in in vivo imaging modalities open new possibilities for determining organ specific masses, fat infiltration in non-adipose depots and regional distributions of body fat (subcutaneous versus visceral) which have been shown to differentially influence disease risk and will be important for understanding the changes that occur in subjects following SCI [49–51].

Skeletal Muscle Adaptations

In individuals with American Spinal Cord Injury Association Impairment Scale (AIS) category A SCI, the activation and loading levels of the skeletal muscles below the level of the lesion are markedly reduced or absent. Without preventive measures, these muscles rapidly atrophy and show a decline in functional capacity. As soon as 6 weeks after injury, lower limb muscles are 25–45% smaller than in non-injured controls [52]. Muscle protein changes may occur more slowly than fiber atrophy; nevertheless, several months post-SCI, fatigue-resistant and oxidative muscle fibers transform into highly fatigable and glycolytic muscle fibers with impaired oxidative capacity and mitochondrial function and poor fatigue resistance [53–55]. The maintenance of normal glucose homeostasis depends on a finely balanced interaction between muscle sensitivity to insulin and insulin secretion, as up to 70% of glucose utilization occurs at the muscle level [56;57]. Changes in skeletal muscle insulin sensitivity and glucose metabolism are considered to be the initiating defect in type 2 diabetes, evident decades before beta-cell failure. The transformation of muscles from a slow oxidative to a fast glycolytic phenotype following SCI yields a muscle tissue that is insulin resistant and metabolically inflexible. Recent research [58] confirmed that human SCI muscle has histochemical and biochemical properties that are very similar to that of human diabetic muscle, including fewer Type I fibers and a predominance of Type IIax+IIx fibers. These fibers have significantly reduced glucose handling capacity under insulin-stimulated conditions due to lower levels of the insulin receptor, glucose transporter 4 (GLUT 4), hexokinase II, glycogen synthase, and pyruvate dehydrogenase-E1α.

Sympathetic Nervous System & Energy Expenditure

Energy expenditure in the homeothermic organism (e.g. mammals) is related to overall body composition, with energy balance requiring a complex integration of central and peripheral affectors and effectors to balance the processes of metabolism against energy acquisition, dissipation and storage[60]. Each of these metabolic components is variable depending on environmental inputs and is mediated through the CNS and autonomic nervous system, including the parasympathetic and sympathetic nervous systems [25] (see [61–64]). The three main components of daily energy expenditure in the absence of thermal stress include resting metabolic rate (including the basal metabolic rate), activity-related energy expenditure, and the thermic effect of food (including diet-induced thermogenesis). Basal metabolic rate (BMR), referring to the minimal amount of energy expenditure in an unfed, non-active, awake state for a non-reproductive organism within the thermoneutral zone (absence of cold or heat stress), has been measured multiple ways, with most currently technologies relying on the calculation of energy expenditure based on respiratory gas analysis referred to as indirect calorimetry [65;66]. Above the range of ambient temperatures, where core body temperature can be defended by involuntary changes in vascular control and sweating for heat dissipation, energy expenditure rises [64]. Below the lower critical temperature where vascular and behavioral controls are no longer sufficient to preserve core temperature, thermogenesis is required [25]. This includes both shivering (involuntary muscle contractions) and non-shivering thermogenesis. Although not as commonly discussed in the context of daily energy expenditure, in the context of modern society where thermal exposures are modified by controlling ambient temperatures, clothing, environmental surroundings and other factors, there is increasing awareness of the “cold-induced thermogenesis” component including both shivering and non-shivering thermogenesis, that may at times significantly contribute to energy balance and body temperature maintenance [67–71]. This adaptive, facultative cold-induced thermogenesis has been clearly demonstrated in small mammals to be sympathetically controlled and mediated through brown adipose tissue (BAT) [72;73], discussed below in more detail. Basal and resting metabolic rate are generally lower following SCI when compared with non-injured controls, with high-level injuries having a larger metabolic rate deficit, although adjustment for changes in body composition and appropriate covariates may obviate these metabolic deficits at times [65;74–78]. Concomitant with these potential energy expenditure reductions, metabolic substrate utilization based on the respiratory exchange ratio (or respiratory quotient) suggest reduced carbohydrate utilization, particularly after a meal [78;79], in agreement with elevated blood glucose during this same period[80] and the observed glucose intolerance following SCI [5;6;9]. Although making up a smaller overall percentage of daily energy expenditure, diet-induced thermogenesis (or the thermic effect of feeding) and cold-induced thermogenesis have been inconsistently reported as lower in subjects with SCI, with higher level injuries having greater deficits [79;81;82]. Differences between study methodologies including the duration and type of metabolic rate measure, time since injury, standardized conditions (time of day if not whole day, composition of meals used for feeding tests, temperature monitoring of room and clothing utilized, medication usage, SNS function and anatomic level of SCI, etc.) may contribute to some of these discrepancies and should be carefully considered in performing comprehensive energy expenditure measurements [82].

Body Temperature

Given the metabolic rate deficits often observed with SCI, as well as the SNS alterations that contribute to thermal regulation, defense of core body temperature under normal ambient temperature exposures in addition to mild cold challenge might be expected to deviate from normative values of non-injured subjects. Subjects with high-level injury and greater neurologic impairment generally have significantly lower core body temperatures than low-level injuries or high-level SCI with less neurologic impairment under basal conditions [23;24;79;83–85] as well as cold thermal challenges [86–88]. This data agrees with previous observations of a “poikilothermy” phenotype in subjects following SCI [22;24;89], further suggesting alterations in SNS function in thermal regulation which may reflect increased heat loss to the environment as well as possible thermogenesis defects. Whether varying core body temperature and altered thermal balance contributes to differences in metabolic efficiency and propensity to excess weight and adipose gains following SCI remains to be fully elucidated.

Brown Adipose Tissue and Metabolism

Brown Adipose Tissue Anatomical Distribution and Function

As described above, core body temperature is defended by a balance of energy intake, expenditure and dissipation. Adipose tissue plays an important role in energy balance as an energy storage tissue, paracrine/endocrine organ and for physical thermal regulation. While the majority of research in humans over the past half century has focused on the importance of excess white adipose tissue (WAT) in obesity and related metabolic disease, the rediscovery of the presence of metabolically active brown adipose tissue (BAT) in adult humans has reopened questions related to energy balance and thermogenesis [90]. BAT is a type of adipose tissue with high metabolic activity named for its unique color derived from increased mitochondrial content, vascular/blood flow and reduced lipid content relative to the more abundant, storage form of adipose - white adipose tissue [72;91;92]. BAT contains a unique mitochondrial protein called uncoupling protein 1 (UCP1) [72]. As the name implies, this protein dissipates the intermembrane H⁺ gradient thereby “uncoupling” the process of substrate utilization through the electron transport chain from ATP synthesis, resulting in heat generation – a.k.a. thermogenesis [72]. SNS stimulation via release of norepinephrine is essential for BAT activation, with UCP1 levels increasing under chronic or sustained SNS signaling [73;93–96].

Over the last 7 years the presence of metabolically active BAT in humans has been reconsidered thanks in part to advances in medical imaging [90;97;98]. Positron emission tomography (PET) scans are used to identify ‘hypermetabolic’ tissues based on the uptake of a radio-labelled glucose analog. In adult human subjects, cold exposure prior to and/or during the PET scans was found to induce a symmetrical pattern of hypermetabolic tissues anatomically localized to the cervical and upper thoracic regions, particularly in lean subjects and during cold weather months [99;100]. Further investigation revealed the composition of these symmetrical depots to be similar to that of adipose tissue based on computed tomography (CT) densities and containing UCP1 and sympathetic innervation, verifying the presence of metabolically active BAT in at least a portion of healthy, adult humans [101–107]. These sites of metabolically active BAT (particularly the paravertebral depots), as well as other BAT depots in the upper thoracic/cervical regions, anatomically coincide with the SNS anatomic ganglia distribution.

Prior to the “re-discovery” of metabolically active BAT in humans by medical imaging methods (previously described by anatomical dissection in adult humans in the 1970s [108]), animals models had suggested a potential role of defects in BAT function associated with both obesity and glucose/insulin disorders [90;97;109–111]. Since 2009, multiple studies using PET-CT and cold challenges, where the subjects are exposed to a cold room to stimulate non-shivering thermogenesis, have been published showing an inverse association between metabolically active BAT and body fat/BMI in humans [103;112–130]. Furthermore, in line with animal studies which show responsiveness to insulin, leptin, thyroid hormone, beta-3 adrenergic stimulation (norepinephrine), fatty acid uptake and protection from T2D in BAT activated states [90;131;132], human BAT activity is positively associated with lower circulating glucose and improved insulin sensitivity, and inversely associated with T2D [101;112;120;133–137].

Obesity and Sympathetic Alterations

Dysregulation of autonomic nervous function in obese animals is frequently observed in metabolism and obesity research. The primary mediator of this autonomic dysregulation has generally been considered a defect in SNS function, so much so that the acronym MONA LISA (Most Obesities kNown Are Low In Sympathetic Activity) is used to describe it [138–143]. Two common deficits of SNS-mediated metabolism reported in the individuals who are obese is a deficit of diet-induced thermogenesis and non-shivering thermogenesis [144–150]. However, a cause-effect relationship has been difficult to determine given the deficits have most often been measured in the obese state or post-weight loss subsequent to overweight/obese status. Alternatively, patients with adrenal tumors (pheochromocytoma) overproduce epinephrine and norepinephrine with accompanying hypermetabolic status including increased energy expenditure, body leanness and brown adipose tissue hyperplasia [119;151–157]. Considering the necessity of SNS function for BAT stimulation of thermogenesis [93], important and unanswered questions of human, clinical relevance remain to be answered in understanding the relationship between SNS function, BAT function and metabolic-related disease risk.

Intersection of SCI, SNS and BAT

Observations from individuals with SCI demonstrate a decreased SNS activity and obesity development post-injury [13;76;158;159]. Although a HLI model of SCI carries additional neurologic insults outside a singular SNS disruption (reduced physical activity, sensory perception, etc.), the clinical parallel and incomplete understanding of this increased disposition to develop obesity and metabolic dysregulation/disease support a need for additional clarity in this area. Thus, comparisons with LLI and non-injured, able-bodied individuals will aid in assessing the contribution of SNS control of BAT in human metabolism and physiology. Human SCI research involving impaired SNS function (high level injury) has focused on multiple metabolic tissues to the relative exclusion of BAT. This may be in part because of the previously disputed presence of metabolically active brown adipose tissue in adult humans [160], as well as controversy regarding the amount of daily energy expenditure contributed by BAT through diet induced thermogenesis (DIT) and non-shivering thermogenesis (NST) with estimated ranges from 5–15% [161;162] in adult humans. However, in animal models it is well established that sympathetic nervous control is essential for BAT thermogenesis, and that molecular, cellular and tissue level changes in BAT accompany chronic SNS activation as we demonstrated in mice with environmental temperature housing conditions [163]. While pre-clinical research is suggestive of a SNS-mediated BAT deficit [164], significant physiological differences between the rodent research model and humans stress the importance of additional research focused on the contribution (or lack thereof) of BAT to aspects of thermal regulation, metabolic rate, substrate utilization and ultimately metabolic-related disease risk in the SCI population. Given the anatomical distribution of BAT and the SNS, we hypothesize that individuals with high-level SCI (tetraplegia) will have decreased BAT amount and function, relative to the impairment of SNS input.

Conclusions

Metabolic alterations following SCI indicate an increased disease risk for metabolic-related diseases like type 2 diabetes and cardiovascular disease. While reduced energy expenditure reflecting changes in activity and body composition contribute to metabolic and physiologic deficits relative to non-injured subjects, these do not fully account for metabolic dysregulation between subjects with SCI, particularly at higher versus lower anatomic levels of injury. The SNS plays a critical role in metabolic rate and energy balance, and with impaired SNS coincident with the level and severity of SCI. While SNS alterations have previously been observed in multiple models of obesity, the cause-effect relationship has been difficult to ascertain. With the rediscovery of BAT in adult humans, a renewed interest in BAT as a mediator of SNS metabolic activity and metabolic-related disease risk has occurred. Considering the physiologic and metabolic phenotypes observed with SCI, particularly across varying anatomic levels, understanding BAT function and impairment with SCI may help improve clinical practice for subjects with SCI while bringing clarity to a broader understanding of the metabolic significance of BAT in adult humans.

Figure.