A CRH2-CRH1 Theory of Anorexia and Bulimia

1. Introduction

Anorexia nervosa nervosa (AN) and bulimia nervosa (BN) are the two major eating disorders [1,2]. The diagnostic criteria for AN include reduced food intake that leads to significant low body weight, “intense fear of gaining weight” or behaviors that oppose weight gain, mainly purging (intentional vomiting, usage of laxatives and diuretics) and excessive exercise, and “disturbed perception of body weight” (e.g., lack of insight on current low weight). The criteria for BN include recurrent binge eating episodes (eating larger amounts than normal in a short time, with a sense of lack of control over eating), and purging behavior, exercise, or fasting. Many AN patients also show binge eating.

Like almost all psychiatric disorders, the etiology of AN and BN is currently not understood [1,2,3,4]. Patient behavior is known to be influenced by biological factors rather than being a personal choice [3], but there is no theory explaining what these factors are.

Here I present the first theory that explains AN and BN (T*AB). The theory is complete is the sense that it explains their etiology, symptoms, and pathophysiology. T*AB is supported by strong empirical biological evidence, and implies several novel treatment directions.

Theory overview. AN and BN result from chronic stress-induced release of corticotropin-releasing hormone (CRH). This can be due to stress experienced by the patient, or to (epi)genetic parental transmission of the effects of past stress. Moderate levels of CRH activate the CRH1 receptor, whole main role is to respond to energy-deficient states by increasing blood glucose and food intake and stimulating the sympathetic nervous system (SNS). High CRH levels activate the CRH2 receptor, whose role is to support energy excess states by reducing food intake, storing glucose in tissue, and restraining the SNS. CRH2 recruits serotonin (SER) to achieve its function, mainly for suppressing gut motility and food intake.

The effects of CRH on food intake trivially explain the food intake aspects of AN and BN. Regarding purging behavior, it is done in order to reduce the nausea induced by excessive gut SER and CRH2. Excessive exercise in AN is done in order to recruit epinephrine (EPI) and norepinephrine (NEP), both suppressed by CRH2. Patients claim to be too fat because they feel that something is wrong with their adipose tissue, which stems from excessive SER and CRH activity that mimics inflammatory responses. T*AB is supported by strong evidence regarding CRH, SER, and the autonomic nervous system in animal models and patients, and by established patient data.

2. Theory

2.1. CRH, CRH1, CRH2, Serotonin

General. CRH is best known for its role in the hypothalamic-pituitary-adrenal (HPA) axis, where stress induces CRH release from the hypothalamus paraventricular nucleus (PVH), CRH acts on CRH1 receptors in the pituitary to stimulate ACTH release, which induces glucocorticoid (GC, cortisol in humans) release from the adrenals. CRH is also synthesized in other body tissues, including the skin, gut, the reproductive and immune systems [5], brain [6], and adipose tissue [7].

CRH acts via two G protein-coupled receptors, CRH1 and CRH2, which both bind Gs (to stimulate cAMP, PKA) and also Gi/o and Gq [8]. The affinity of CRH to CRH1 is greater. Following ligand binding, the receptors show sustained signaling from endosomes, allowing chronic signaling beyond the original release event [8]. CRH2 is also activated by urocortins, a family of ligands that do not have a major role in AN and BN.

CRH1. The role of CRH1 is well known. Besides its stimulation of GC release, it activates the SNS [9], to increase blood glucose, promote food intake, and induce other sympathetic responses (e.g., increased heart rate, lipolysis, thermogenesis). Note that it is not accurate to view CRH1 as triggering the response to all types of stress, but as responding to deficiency states, including low glucose, low temperature (cold), and low oxygen (hypoxia) [10]. Acute social stress increases the metabolic demands from the body, creating a deficiency stress state.

CRH2. CRH2 is known to oppose many CRH1-induced effects, including appetite, anxiety and depression [9]. In particular, its food intake suppression function is overwhelmingly supported [11]. It also suppresses adrenal EPI release done as part of the counter-regulatory responses to insulin-induced hypoglycemia [12], opposes thermogenesis [13], and promotes glucose uptake by tissue, including skeletal muscle [14] (at least partly insulin-independently), heart [15], brain [16] and adipose tissue [17]. This effect occurs mainly in high glucose levels, and can utilize insulin via direct action on beta cells [18]. In the gut, CRH2 slows gastric transit [19].

Serotonin. SER is present in most body tissues, including the gut, brain, heart, lung, skeletal muscle, adipose tissue, and the reproduction system [20]. The SER system is complex, and for our purposes it suffices to highlight the following facts. First, CRH2 recruits SER, promoting its synthesis [21] and release from dorsal raphe neurons [22] and gut enterochromaffin cells [23]. SER reciprocally activates PVH CRH neurons, via SER2c [24]. Second, SER is well-known to suppress food intake, an effect mediated by SER2c [25]. Eating carbohydrates enhances the synthesis and release of hypothalamus SER, which suppresses carbohydrate intake [26]. Third, the food intake suppression function of CRH2 is done, at least in part, via SER2c [27].

Fourth, SER is a major gut agent. The gut contains most of the body’s SER, and SER is present in both mucosal enterochromaffin cells and in the enteric nervous system [28]. The former support reflexes that increase local motility to enhance absorption after feeding [29], while the latter slow gastric emptying and suppress intestinal motility [28] in response to glucose [30]. Gut glucose induces gut SER release via SER3 on gut vagal sensory neurons [31].

Finally, SER promotes insulin-independent glucose uptake by tissue [32], including adipose tissue [33], skeletal muscle [34] and liver, where it promotes glycogen synthesis [35]. plasma SER and blood glucose levels show a positive correlation [36], and raphe obscurus and raphe pallidus SER activity diminish following insulin-induced hypoglycemia [37]. Most relevant to AN, SER promotes lipogenesis in white adipose tissue (WAT) and opposes its browning and thermogenesis [38,39,40].

SER has 14 different receptors, and can also produce results in the opposite direction to the above, depending on conditions [41]. What is important for our purposes here is that it is clearly recruited by and cooperates with CRH2 in fulfilling its function. A full theory of SER function will be presented elsewhere.

Summary. T*AB posits that the role of CRH2 is to trigger responses to excess stress, including high glucose and heat, but also prolonged social stress. As part of this role, it suppresses the responses to deficiency stress, including EPI release, thermogenesis, and food intake. This account of the role of CRH2 is a novel contribution of the present paper. To achieve its role, CRH2 recruits SER, which suppresses food intake, slows gut motility, responds to high glucose, and promotes its gut absorption and storage in tissue. In particular, SER promotes the growth of white adipose tissue but opposes brown adipose tissue.

2.2. Anorexia Nervosa (AN)

According to T*AB, AN is caused by excessive CRH release that chronically activates CRH2 in the brain and other tissues. Since the affinity of CRH1 to CRH is higher than that of CRH2, there would be tissues in which CRH1 is chronically activated as well. The symptoms induced by these fundamental biological effects are most likely enhanced by modern cultural and media influences. This section explains how stress, including social stress, can cause this state, and shows that this simple premise can explain the defining characteristics of the disorder. Other data are presented in the evidence section.

AN core cause: stress. As described above, CRH is recruited in states of metabolic deficiency or excess. This includes social stress (e.g., maternal separation, conspecific aggression, social defeat, sexual trauma), which produces a deficiency state because the organism initially needs more energy to address the situation, and is well-known to induce CRH release [42]. If the problem is not quickly resolved, CRH release continues and can chronically activate the lower affinity receptor CRH2.

Stress that occurs during sensitive developmental periods, such as early life and adolescence, can yield persistent changes in protein expression via epigenetic marks [43]. Moreover, epigenetic changes can be multigenerational, passed from parents to offspring [44]. It is well-supported that early life stress can persistently increase CRH, e.g., CSF CRH in non-human primates [45], and mRNA in rodents [46].

Thus, prolonged social stress (and other types of stress), especially during early life or adolescence, can create the optimal conditions for AN. Indeed, social pressure and stress are known to be precipitating factors for AN, and social difficulties are very common [4]. Traumatic experiences are common in moderate-severe AN patients, with 50% being at the cutoff threshold for PTSD (without being diagnosed) [47]. In one report, almost half of AN patients have suffered childhood sexual abuse [48].

Food intake. Suppression of food intake is one of the major functions of CRH2, cooperating with SER2c. The AN patient’s brain receives a strong signal of nutrient excess, even if the real state is that of deficiency. CRH2 opposes the main agents that promote food intake, GCs and EPI, and CRH opposes neuropeptide Y, which is elevated in AN but whose effect is not that strong [2]. Adiponectin reflects the status of adipose tissue and is indeed elevated in underweight AN patients, but it is not a strong appetite stimulator [2]. Thus, there is no strong physiological biological signal that can offset the erroneous message conveyed by excessive CRH2 in AN.

Purging behavior. Chronic CRH2 induces chronically high SER in AN. In the gut, SER is known to induce nausea and the vomiting reflex, mainly via SER3 but also via other receptors [49]. This is consistent with SER’s role in suppressing food intake and gut motility. CRH itself has also been hypothesized to be responsible for the cyclic vomiting syndrome [50]. Thus, excessive CRH-induced gut SER explains the urge to vomit in AN.

CRH is produced by colon enterochromaffin cells [51], and triggers inflammation via mast cell activation [52]. Mast cells can uptake, store and secrete SER and CRH [53]. CRH2 specifically increases gut inflammation following stressors [54]. SER also promotes gut inflammation [28]. Since the state of the gut (as that of any other tissue) is continuously reported to the brain via sensory neurons, AN patients experience a constant feeling that there is something wrong in the gut. Since gut carbohydrates induce SER release, this uncomfortable feeling is relieved when the gut is empty. This explains general purging behaviors in AN, and provides another explanation for food avoidance.

Body image. White adipose tissue has a population of mast cells [55], whose activation promotes chronic low inflammation and macrophage accumulation [56]. Since CRH greatly potentiates mast cell activation [57], and mast cells secrete SER and CRH [53], AN adipose tissue involves a continuous inflammation-like state, as in the gut. This state is reported to the brain via interoception pathways [58], a feeling that patients verbalize as “I am too fat” (they could say “something is wrong with my fat stores”, but this sounds strange, as opposed to the former, which is generally accepted by society). Even though patients are not too fat, some processes that make people fat, namely SER-induced glucose uptake and lipogenesis, are indeed chronically active in their WAT.

Binge-eating type. AN patients are defined to be either of the restricting type or the binge-eating/purging type, depending on whether they have engaged in recurrent binge eating or purging episodes during the last 3 months [1].

According to T*AB, the basic active AN state involves CRH levels that are high enough to activate the lower affinity receptor CRH2. If CRH levels diminish for any reason but do not get to full recovery level, or if CRH2 gets desensitized, CRH would strongly activate the CRH1 receptor, which promotes food intake. This, along with other signals stemming from the person’s state of deficiency, explains binge eating episodes. Indeed, more than 40% of the restrictive type show binging and purging at some point [59].

2.3. Bulimia Nervosa

According to T*AB, BN is caused by CRH release that chronically activates the CRH1 receptor. CRH1 strongly promotes food intake, explaining binge-eating episodes. Since the effects of CRH on mast cell inflammation and transit time in the gut can be promoted by both CRH1 and CRH2 [60], chronic CRH1 explains purging behavior in BN in the same manner as in AN.

3. Discussion

This paper presented the first complete biological theory of AN and BN. T*AB is complete in the sense that it explains the etiology, symptoms, and pathology of these disorders. A complete theory does not have to be correct, but being complete is the minimal requirement from a theory.

AN and CRH. It is widely acknowledged that AN is currently not understood, and no biological theory has been formulated to fully explain it. T*AB focuses on stress (mainly social stress) and CRH. Given the well-known link of AN and BN with both of these, and given that the food intake suppression effects of CRH2 have been well-known for more than 20 years [9], it is surprising that a CRH2-based AN theory has not been proposed yet. CRH has been discussed in AN, but this has virtually always been in the context of HPA activity.

A possible explanation for this might be the general view of CRH2 as being anxiolytic, as opposed to the anxiogenic CRH1. Thus, it was perhaps not natural to link CRH2 with AN, which clearly involves anxiety. The new theory of CRH2 presented in this paper resolves this apparent paradox, by noting that deficiency stress is different from excess stress, and by assigning CRH2 to the latter. CRH1 is assigned to the former, and this provides an immediate parallel theory for BN.

The evidence. This paper has brought evidence supporting every aspect of T*AB (CRH, SER, EPI, NEP, gut, skin, epidemiology, various properties) except direct evidence for SER- and/or CRH-induced adipose tissue inflammation. The evidence for AN is larger than that for BN, reflecting the relative interest of researchers in the two disorders.

Other eating disorders. AN and BN are the major eating disorders, but there are additional ones, including binge-eating disorder (BED), pica, rumination, and avoidant/restrictive food intake disorder. These disorders were not dealt with here in order to keep the presentation simple. BED is defined via binge-eating episodes without purging behaviors (as in BN). BED can be explained via chronic activation of CRH1 in areas that control feeding (mainly the brain), without strong activation in the gut. That this can happen is natural, in light of the fact that TPH2 controls SER synthesis in the enteric nervous system and brain, while TPH1 controls it in enterochromaffin cells [28]. These are two different systems.

Serotonin. Depression is treated today by SSRIs, which increase brain SER levels. As a result, SER is commonly associated with positive emotions. Thus, readers may question the negative role SER has in T*AB. The T*AB answer is that SSRIs improve depressive symptoms because they mimic a state of healthy excess in which the organism’s needs (food, sex) have been met. Hence, they reduce deficiency-related stress responses, including CRH activation of CRH1. However, in AN, SER is associated with excess-related stress and CRH2, and is a negative factor. Indeed, SSRIs are known to induce a large variety of gut side effects when used for treating depression.

The negative role of SER in AN has been recognized a long time ago, yielding a proposal that starvation is done in order to decrease SER activity [90]. T*AB agrees with this insight and shows how it fits within a complete theory.

Behavior is rational. According to T*AB, the major drivers for eating and purging behaviors in AN and BN are molecular, not psychological. While sociocultural constructs such as “being thin is better than being fat” and social media-promoted body image probably promote pathogenesis by increasing social stress, only a minor fraction of people are ill with AN. The parsimonious account is that the fundamental cause is biological.

The T*AB account is that metabolic body signals erroneously convey the message that the person has eaten too much, even if the objective state is that of deficiency. The major causal factor in the “intense fear of eating” in AN is not fear of gaining weight. The cause is the stress-inducing activations in the brain, gut and adipose tissue that are exacerbated by eating and by a non-empty stomach. Reduced eating and purging behaviors are a form of self-treatment meant to reduce the highly uncomfortable feeling accompanying these activations. Patient behavior is a rational response to their biological situation. This view can by itself reduce patient stress, and can serve as a basis for novel pharmacological treatments.