Stress, pseudoallergens, autoimmunity, infection and inflammation in chronic spontaneous urticaria

The importance of low molecular weight preservatives, colourings, salicylates and histamine in CSU has waxed and waned over the years. While salicylate mediated inhibition of cyclooxygenase 1 leading to a disturbance in a healthy prostaglandin/leukotriene ratio [15, 16] may underlie an exacerbation of CSU in some patients, the factors other than polymorphisms in centrosomal protein of 68 kDa [17] that predispose to this are unclear. Clearly, the small size of these molecules makes it impossible for them to directly bind IgE and there is no evidence that they act as haptens. Nevertheless, there is compelling evidence that a significant proportion of patients with CSU may benefit from a diet low in these ‘pseudoallergens’ [18‐21]. This was associated with reduced leukotriene E4 excretion in the urine compared to pre-diet levels [18]. However, provocation testing with ‘known food dyes and preservatives’ was positive in only 18% of CSU patients although 73% had originally responded to the pseudoallergen free diet [22]. Interestingly, 71% of these patients reacted to provocation testing with pureed tomato and 41% to a steam extract to tomatoes. In a similar way, Rajan et al. [23] noted only 2 of their 100 patients with CSU reacted to a single blind challenge with 11 of the most frequently used additives. Finally, intestinal permeability was reported to be increased in CSU and this was partly reversed by 24 days on a reduced pseudoallergen diet [24]. Collectively, these results suggest that CSU may in some people be due to a temporary sensitivity to pseudoallergens and natural low molecular weight compounds which enter the body across a leaky gut epithelium. The pseudoallergens may in themselves also contribute to the leaky gut but it is likely that other factors may be responsible for raising mast cell sensitivity to the pseudoallergens. However, once these additional factors have subsided then normal resistance to these compounds may once again prevail.

The presence of a positive autologous serum (AST) or plasma test (APT) in 40% to 60% of patients with CSU has suggested the presence of auto-antibodies capable of stimulating mast cells [25]. Importantly, a positive AST has been shown to increase the possibility of acute spontaneous urticaria progressing to CSU [26] and a reduced chance of resolution within 2 years [27].

Several types of autoimmune diseases are increased in patients with CSU (Table 2). These include particularly hypo and hyperthyroidism but also diabetes, rheumatoid arthritis and Sjogren’s syndrome [28]. IgG anti-thyroid antibodies, in particular, are not infrequent in patients with CSU [29] and especially in females [30]. However, IgE antibodies to TPO have also been detected [31, 32] suggesting a possible mode of action for the benefit of anti-IgE therapy in CSU. Interesting recent work also suggests that a high proportion of the total IgE in those with CSU may be autoreactive and have increased lipophilicity [33]. Regardless, IgG antibodies directly binding to the FcεRI and in consequence capable of activating mast cells have been and found in almost a quarter of patients with CSU with only 3.2% in healthy controls [34]. This has kindled interest in mast cell activating IgE and IgG autoantibodies directed to thyroid and nuclear antigens. Interestingly, passive transfer of IgE anti-TPO antibodies could elicit positive skin prick test response in control subjects and basophils from patients with CSU and anti-TPO antibodies could be activated by incubation with TPO [35]. At a more basic level, auto-antibodies in general may encourage the secretion of Type I interferons by plasmacytoid dendritic cells (pDC) [36] and stimulate recruitment of basophils to lymph nodes and directly activate adaptive immune responses through B and T cells [37]. In an interesting recent report IgE antibodies to IL24 functionally capable of activating mast cells and correlating with disease activity have been reported by Schmetzer et al. [38]. The authors also noted IgE auto-antibodies to over 200 self-proteins from a total of over 9000 proteins screened. The avidity of these antibodies remains unclear and it is possible that they represent non-specifically auto-reactive antibodies generated as part of a polyclonal humoral response to a viral or other infection.

Clinically the increased prevalence of autoimmunity in CSU suggests impaired immune regulation and an imbalance in the ratio of the pro-inflammatory Th17 cells and T regulatory cells (Tregs). Tregs dysfunction, and especially T follicular regulatory cells [39], leading to the survival of autoreactive T and B cells and altered antigen presenting function by dendritic cells may underlie several systemic and organ specific autoimmune diseases [40, 41]. In this regard, it is noteworthy that impaired T regulatory (Treg) cell function has been found in CSU [42, 43] with alterations in other T cell subsets also being evident [44]. From a therapeutic perspective efforts to improve T reg function may therefore be helpful to reduce the severity and duration of CSU. At present this has been utilised predominantly in graft versus host disease after hematopoietic stem cell transplantation [45].

While the precise cause of CSU is often unclear in an individual patient, viral infections and increased stress levels may be important factors in Western countries [46]. Globally, however, parasitic infections may be of greater importance with an average comorbidity of around 10% [47]. In this group anisakis, toxocariasis, fasciolosis, strongyloidiasis and blastocystosis were the most frequent parasitic infections and treatment led to improvement in an average of one-third of patients across several studies [47]. In this regard, empirical anti-helminth therapy was suggested by Nahshoni et al. [48] for people with CSU returning from travel to ‘developing countries’. Earlier work by Hameed et al. [49] suggested that the urticaria could be more properly linked with the amoeboid form of blastocystis hominis. Furthermore, 60% of these patients responded to a single course of metronidazole and 100% to two courses.

More recent work has shown an increased frequency of dysbiosis of the gastrointestinal tract in CSU. As such there was a significantly reduced frequency of Akkermansia muciniphila, Clostridium leptum and Faecalibacterium prausnitzii in the stools of CSU patients while that of Enterobacteriaceae was unchanged [50]. Intriguingly, the number of tryptase and CD117 positive mast cells per high power field were significantly higher in patients with CSU compared to controls and adjunctive treatment with mast cell stabilisers has been suggested [51].

In regards to the role of helicobacter infection the precise significance of this infection in precipitating and/or aggravating CSU has oscillated over the years. More recent work suggests that asymptomatic infection is more frequent in CSU patients than in healthy controls and in helicobacter infection positive patients the severity and extent of the urticaria may be more severe [52].

In terms of CSU and infections outside of the GI tract, there are numerous anecdotal reports or small case series that have purported to show a link with viral, bacterial or fungal infections. Thus, Dreyfus [53] noted serological evidence of human herpes virus 6 infection, sometimes in association with Epstein Barr virus reactivation, to be more frequent in patients with CSU. In their systematic review, Imbalzano et al. [54] considered herpes viruses to be possible aetiologic factors in children and hepatitis viruses were more frequently associated with CSU in adults. The precise role of so-called ‘superallergens’ that can non-specifically stimulate basophils and mast cells and which are produced in viral infections such as hepatitis C and HIV is unclear [55].

In the case of bacterial infections, Calado et al. [56] found streptococcal tonsillar infection to relate to CSU with improvement after antibiotic therapy and more permanent resolution after tonsillectomy. Godse and Zawar [57] found a link with tinea pedis in four patients with significant improvement after anti-fungal therapy. However, Cribier and Noacco [58] did not support a definite link between focal infections such as sinusitis and dental infections and CSU. They also mentioned the importance of curative antibiotic therapy in eliminating CSU as clearly supporting an aetiological link and this was frequently absent in the case of focal infections.

The mechanism by which infection contributes to the onset, perpetuation or worsening of CSU is unclear. It is possible that different components of the immune response to infection may be responsible in varying combinations to these different stages in CSU. As such Kay et al. [59] reported increased cellular expression of the Th2 cytokines IL4 and IL5 as well as IL33, IL25 and thymic stromal lymphopoietin in the lesional but not non-lesional skin of patients with CSU. Notably, the IL25 and IL33 was expressed by multiple cells involved in the innate immune system. In the case of helicobacter infection, a 23 to 35 kDa protein obtained from helicobacter preparations was found to induce the release of histamine, TNF-a, IL-3, IFN-γ, and leukotriene B4 by the LAD2 mast cell line in a dose or time-dependent manner [60]. Additionally, mast cells can be activated by inflammatory cytokines such as IL6 [61] and raised levels of IL6, IL1β and TNFα have been reported in patients with infection and in CSU [62]. It is therefore not inconceivable that infections associated with inflammation may be associated with urticaria. Indeed, some patients with CSU have been improved by the discovery and treatment of occult infection [63]. Overall, however, it is likely that infection acts as a facilitating factor for the initiation and perpetuation of CSU and additional cofactors such as stress may be required for the CSU phenotype to be expressed. This would explain why few patients with severe infections such as a pneumonia, pyelonephritis, abscess etc. develop urticaria.

Urticaria has been reported to be increased in frequency in patients with bipolar disorder [64] which is well known to be accompanied by high levels of stress. CSU is also associated with a significantly higher prevalence of depression, anxiety and poor quality of sleep [3, 65]. In the recent report by Tat [66], depression and anxiety assessed using the Hospital Anxiety and Depression scale was evident in almost half of the 50 CSU patients seen and with a significant positive correlation of both with the urticaria activity score. Regardless, the coexistence of mood disturbance with CSU can have major repercussions for work attendance and efficiency [3, 14]. In children, high rates of various sorts of anxiety as well as social phobia and depression have also been reported [67]. Interestingly, ‘state anger’ measured using the State trait anger expression inventory correlated significantly with levels of pruritis [68]. High levels of anger were also reported by Altınöz et al. [69]. Overall, there is increasing evidence that continued stress can perpetuate and aggravate CSU.

As mentioned previously, patients with CSU have increased levels of emotional distress with an underlying anxiety, depression and somatoform disorder [12]. Conversely, increased levels of stress may perhaps predispose individuals to CSU. As such Yang et al. [70] noted significantly more ‘life events, higher subjective weighting of impacts from life events, more somatic symptoms, more severe insomnia, less family support and more negative coping tendencies’ in 75 consecutively seen CSU patients than 133 control patients with tinea pedis. However, the mechanism by which stress predisposes to CSU is unclear. Interestingly, the basophils of patients with CSU appear to activate more readily with adrenocorticotrophic hormone and corticotrophin releasing factor and to have had a trend to raised cortisol levels compared to healthy controls [71]. Additionally, in mice, stress can lead to a disruption of T regulatory cell function with a shift towards a Th1/Th17 balance leading to exaggerated autoimmunity [72]. In children, stress has been associated with an increased tendency to autoreactive T and B cells and with the production of IgG auto-antibodies [73]. Increased activation of the NLRP-3 inflammasome has also been demonstrated in major depression and stress and this may provide the link between psychological factors and exacerbation of urticaria by changes in mood and especially emotional stress [74].

The link between the brain and mast cells within the skin via C fibre sensory nerves is likely involved in the worsening of skin diseases such as atopic dermatitis and CSU with stress [75]. The increased frequency of CSU in women is likely related to oestrogen and progesterone and the ability of these hormones to stimulate autoimmunity [5]. However, recent work also suggests the importance of pituitary adenylate cyclase activating polypeptide in stress circuits and stress responses and which is modulated by oestrogen [76]. In terms of the chemical link between nerves and mast cells, several neuropeptides, but particularly substance P are able to activate mast cells [77] leading to the release of histamine and related mediators that are involved in CSU. Interestingly neuropeptides such as vasoactive intestinal peptide, α-melanocyte-stimulating hormone and calcitonin gene-related peptide (CGRP) are evident both within the central nervous system and released by cutaneous nerves [78, 79]. In this regards it is noteworthy that stress leading to the release of sensory nerve neuropeptides can alter the behaviour of Langerhans cells within the dermis and skew the cutaneous immune system towards specific T helper cell pathways [80]. CGRP, in particular, appears to encourage Th17 type cells with the potential to increase inflammation by the recruitment of T cells and neutrophils [78, 81].

Recent work suggests that stress induced release of substance P and CGRP by sensory cutaneous nerves leads to itching and mast cell activation via several receptors. These include the Mas-related G protein-coupled receptors (Mrgpr) family as well as transient receptor potential ankyrin 1 (TRPA1) and protease activated receptor 2 (Par2) [78]. The subsequent release of cytokines such as IFN-γ, IL-4, TNF-α, and IL-10 then encourages immune activation and the aggravation of cutaneous inflammation. It is therefore possible that the itch scratch cycle can prolong the cutaneous inflammation that perpetuates CSU and which is also seen in atopic dermatitis.

Many patients with CSU are observed to have mildly raised levels of C-reactive protein. Determining the cause of this abnormality is frequently unrewarding. However, a state of immune readiness and inflammation has been suggested to be present in the peripheral blood of CSU patients by Santos et al. [82]. They reported significantly higher serum levels of several chemokines (CXCL8, CXCL9, CXCL10 and CCL2) in patients with CSU compared to healthy controls and this was irrespective of whether they were ASST positive. Furthermore, the basal secretion of CCL2 by peripheral blood mononuclear cells or induced by Staphylococcus aureus enterotoxin A was higher in those with CSU as indeed was CXCL8 and CCL5 secretion after phytohaemagglutinin stimulation.

In terms of oxidant stress, this has been found to be increased in children with CSU in whom total anti-oxidant activity was also reduced [83]. As such it is interesting that patients with CSU had lower levels of basal cortisol [84] and dehydroepiandrosterone-S (DHEA-S) [85]. Both are reduced in chronic stress and DHEA-S is partly regulated by the nervous system and is known to regulate the immune system. Furthermore, patients with CSU have an increased overall oxidative burden with reduced levels of erythrocyte copper-zinc superoxide dismutase (Cu-ZnSOD) levels [86]. This contrasts the situation in acute urticaria in which increased levels of oxidative stress indicated by raised levels of malonaldehyde and nitrous oxide were associated with increased Cu-ZnSOD and decreased levels of glutathione peroxidase [87].

Chronic low grade inflammation associated with high stress scores, raised C-reactive protein and interleukin 18 [84] may explain the increased frequency of metabolic syndrome in CSU [88, 89]. Hypertension was noted in 18.1% of the 1539 CSU patients that were analysed as part of the German contribution to the AWARE study [90]. The mechanism by which hypertension may prolong CSU is, however, unclear [91] although it should be borne in mind that ACE inhibitors used in hypertension can aggravate and prolong CSU.

Vitamin D3 (cholecalciferol) has a powerful ability to increase T regulatory cell function and reduce Th1 and Th17 type immunity [92]. These immune changes can help ameliorate autoimmunity. In this regard, it is interesting that reduced levels of serum vitamin D3 have been found in several of the systemic autoimmune diseases as well as in asthma and inflammatory bowel disease. In CSU, reduced vitamin D levels are found more frequently than in healthy controls [93] and replacement therapy has been found to reduce the severity and duration of the urticarial wheals [94, 95].

The involvement of the coagulation system in CSU is suggested by the clinical improvement in urticaria when patients are commenced on anticoagulation for other apparently unconnected reasons. Indeed, the present authors have seen this in 5 patients over the years. This is supported by the elevated d-dimer noted by Triwongwaranat et al. [96] in 58 out of 120 patients with CSU and with a correlation between d-dimer levels and severity of urticaria. Prior to this Sakurai et al. [97] observed increased thrombin generation by the peripheral blood cells of patients with CSU compared to healthy controls. In this respect, it is interesting that thrombin injected into the ear skin of mice produced a dose dependent degranulation of mast cells and the mast cells expressed thrombin receptors PARs 1, 3 and 4 [98]. Elevation of other coagulation variables has also been observed by other authors and treatment with anti-coagulant drugs as adjunctive therapy suggested [99]. More recently Yanase et al. [100] have suggested the involvement of the extrinsic coagulation pathway in CSU and highlighted the role of tissue factor expression of endothelial cells as being important. Notwithstanding, the precise mechanism by which coagulation contributes to urticaria is unclear although the interaction between inflammatory factors and coagulation proteins is interesting [101].

Patients with coeliac disease have been shown to have an increased frequency [28] and susceptibility to both acute and chronic urticaria with odds ratios of 1.31; 95% CI 1.12–1.52 and 1.54; 95% CI 1.08–2.18, respectively [102]. The mechanism here is unclear but as coeliac disease is increased in those with IgA deficiency and the latter is associated with autoimmunity this may represent one possible mechanism. Additionally, coeliac disease is associated increased permeability of the upper gastrointestinal tract which may allow the greater absorption of pseudoallergens. Furthermore, vitamin D deficiency is more frequent in coeliac disease which may in turn contribute to impaired T regulatory cell function and raised tendency to autoimmunity.