A Proposal for the Etiopathogenesis of Acquired Cold Urticaria: Role of Substance P, Angiotensin-Converting Enzyme and Mast Cell Chymase.

Background: The etiopathogenesis and cold stimulation mechanism are not fully understood in cold urticaria (CU). Substance P (SP) is released from skin neurons as a result of cold stimulation. It causes mast cell degranulation and therefore causes mast cell chymase (MCC) release. Angiotensin-converting enzyme (ACE) plays a role in removing SP from the environment. ACE also catalyses the conversion of angiotensin I (AT1) to angiotensin II (AT2), like MCC. This study aims to investigate the role of SP, ACE and MCC in the pathogenesis of CU.
Methods: Patients with acquired CU were included in the study. Two punch biopsies were taken from the urticaria plaque resulting from the stimulation and the intact skin without lesions. The samples were evaluated histopathologically. All samples were stained immunohistochemically with SP, ACE and MCC antibodies.
Results: The number of patients included in the study was 21. In the plaque lesion, the presence of dermal neutrophil and eosinophil, neutrophil in the vascular lumen were found to be statistically significantly higher than intact tissue (p = 0.046, P = 0.014, P = 0.014). Strong positive staining was detected in the full thickness of the epidermis, vascular endothelial cells, eccrine and sebaceous glands with ACE. MCC was statistically significantly higher in lesional skin than lesion-free skin samples (p < 0.001). Conclusions: Mast cell maintains its central role in CU pathogenesis. SP, which causes neurogenic inflammation, may not be detected due to its rapid destruction in the tissue. Strong staining of ACE, which takes part in the local renin-angiotensin-aldosterone (RAS) system in the skin, should be documented quantitatively. Copyright:

Introduction

Cold urticaria (CU) is one of the relatively common causes of physical urticaria within the inducible urticaria group. CU is characterized by hives and angio-oedema by releasing histamine, leukotrienes and other proinflammatory mast cell mediators on contact areas after cold exposure.[1] Mast cell-mediated mediators are thought to play a role in the development of plaques, as in patients with chronic spontaneous urticaria (CSU). Still, the initiating event, cold triggering factors and subsequent molecular mechanisms are not clear.[2]

Some of the mechanisms blamed in the etiopathogenesis of CU are autoallergy, autoimmunity, neurogenic inflammation and abnormal temperature sensing.[3] Recent studies have determined that substance P (SP), which is a neuropeptide, causes histamine-mediated vascular permeability, vasodilation, oedema and erythema and is high in urticaria patients.[4] This neuropeptide released after neurogenic stimulation with cold is destroyed in the extracellular space by neutral endopeptidases and angiotensin-converting enzyme (ACE).[5] It is also known that SP stimulates mast cell degranulation.[6] Mast cell chymase (MCC) is a major protein found in mast cell secretory granules from the serine protease family. It is released from the secretory granules by mast cell stimulation. Intradermal injection of chymase leads to increased mast cell accumulation in human skin. Some studies emphasize that MCC plays an essential role in the pathogenesis of allergic diseases.[7] As it is known, ACE causes angiotensin II (AT2) production from angiotensin I (AT1), and AT2 causes vasoconstrictor and cardiovascular effects. MCC also converts AT1 into AT2, such as ACE.[8] The local renin-angiotensin-aldosterone system (RAS) in the skin may play a regulatory role in CU patients.

This study aimed to evaluate the status of SP, MCC and ACE in CU plaques. This may allow us to understand unknown etiopathogenesis and the development of new treatments in CU.

Materials and Methods

This research was carried out with the decision of the Local Clinical Research Ethics Committee, numbered 1551. This is a single-centre, cross-sectional, controlled and prospective study. It was carried out between January 2019 and January 2020 in a tertiary hospital. Written informed consent was obtained from all patients and parents of patients under 18 years of age. Voluntary acquired cold urticaria (ACU) patients diagnosed with an ice cube test and Peltier-effected stimulation test (TempTest®, Courage+Khazaka electronic GmbH, Köln, Germany) were included in the study. The age of patients was between 8 and 65. CU variants excluding AUC (delayed cold contact urticaria, cold-related dermographism, cold-induced cholinergic urticaria), secondary cold contact urticaria associated with serum abnormalities such as cryoglobulinaemia and cryofibrinogenaemia, familial CU, using ACE inhibitors and known granulomatous disease such as tuberculosis, sarcoidosis patients were excluded from the study. The sample size was calculated by predicting that the large effect size dz: 0.8 in the dependent group was considered statistically significant. The total sample size was determined as 19 at 95% Power 0.05 alpha significance level.

Demographic and clinical characteristics of the patients (age, gender, age of onset, disease duration, history of accompanying atopic dermatitis, angio-oedema and anaphylaxis) and laboratory test results (total blood count, sedimentation, C-reactive protein, liver and kidney function tests, hepatitis, syphilis and human immune deficiency virus (HIV) markers, Epstein–Barr virus (EBV) serology, total immunoglobulin E (IgE), antinuclear antibody, cryoglobulin and cryofibrinogen) were recorded.

The antihistamines, non-steroidal anti-inflammatory drugs, topical and systemic steroids used by the patients were discontinued 1 week before the test. Cold stimulation tests, including the ice cube test, were applied to one forearm of the patients and TempTest® to the other forearm. Plastic-coated ice cubes were kept on the forearm for 30, 60, 120 and 300 s and were evaluated after 10 min. Erythematous, oedematous plaque formation was considered a positive test. The other arm of the patient was tested with the TempTest®. After the patient kept his/her arm on the device for 5 min, it was waited for 10 min for evaluation. The erythematous and oedematous plaque that occurred in the patients was accepted as a positive test. Ice cube test results and critical temperature threshold (CTT) values of the patients were recorded. A 4 mm punch biopsy was taken from the erythematous, oedematous plaque formed by stimulation test under local anaesthesia. A 4 mm punch biopsy was taken from another area at least 10 cm away as a control. A single pathologist examined all histopathological and immunohistochemical sections.

The slides stained in Ventana Brench Mark ULTRA fully automated immunohistochemistry stainer device, using ultra view universal 3,3’-diaminobenzidine (DAB) protocol with rabbit polyclonal SP antibody (biorbyt orb13610), rabbit polyclonal MCC antibody (biorbyt orb6398) and rabbit polyclonal ACE antibody (biorbyt orb373680).

The stained slides were examined under a light microscope. For SP, rat brain tissue was evaluated as external control and erector pili muscle as an internal control. Staining in the epidermis and dermis was assessed. The densest five spots (X400) were selected for the MCC, and MCC positive cells were counted and recorded. For ACE, rat brain tissue was used as external control. The presence of staining in the epidermis, vascular endothelium, sebaceous glands and erector pili was recorded.

Statistical analyses were performed with Statistical Package for the Social Sciences (SPSS.15, IBM SPSS Statistics for Windows, Version 15.0. Armonk, NY: IBM Corp.). Descriptive variables were presented as number and percentage or mean and standard deviation or minimum and maximum as appropriate. Differences of numerical variables in dependent groups were performed with paired t-test since they provided a normal distribution condition. Since the normal distribution condition was provided for numerical variables in two independent groups, the two groups were analysed by Student's t-test. One-way analysis of variance (ANOVA) was analysed when a normal distribution condition was achieved in groups of more than two independent groups. When the normal distribution condition was not achieved, the Kruskal–Wallis test was used. The relationship of numerical variables was examined with Pearson correlation analysis when parametric test condition was met and Spearman correlation analysis when not. The rates in independent groups were compared with the Chi-square test, and the rates in dependent groups were compared with the McNemar test. The statistical significance level of alpha was accepted as P < 0.05.

Results

The number of patients included in the study was 21, 5 (23.8%) males and 16 (76.2%) females. The mean age was 35.5 ± 14.8 (Mean ± SD), and the patients’ ages ranged between 8 and 63 (median 40). The mean disease duration of the patients was 98.8 ± 117.0 (Mean ± SD) months. The mean age of onset was 27.5 ± 14.4 (Mean ± SD) months. The mean CTT value obtained according to the TempTest results of the patients was determined as 19.1 ± 5.4 (Mean ± SD)°C.

Simultaneous ice cube test results were found to be positive in all patients in the study. Concomitant angio-oedema was found in eight (38.1%) patients, anaphylaxis in five (23.8%) patients and atopic disease history in six (28.6%) patients.

Antinuclear antibody, cryoglobulin, cryofibrinogen, hepatitis, HIV and syphilis markers were negative in patients with acquired primary CU, which were checked to exclude other possible causes. EBV viral capsid antigen (VCA) IgG and EBV Epstein- Barr Nuclear Antigen (EBNA) IgG were found to be positive in 17 (81%) patients. EBV VCA IgM was negative in all patients.

Histopathological findings in CU patients are summarized in Table 1. In the plaque lesion, the presence of dermal neutrophil and eosinophil, neutrophil in the vascular lumen, oedema in the papillary dermis and orthokeratosis were found to be statistically significantly higher than intact tissue (p = 0.046, P = 0.014, P = 0.014, P = 0.046, P = 0.025) [Figure 1].

Table 1

Histopathological findings in cold urticaria patients (Y: Yes; N: Not)

	Plaque		Control		P
	n	%	n	%
Dermal neutrophil
Y	6	28,6	2	9,5	0,046
N	15	71,4	19	90,5
Dermal melanophage
Y	7	33,3	7	33,3	1,000
N	14	66,7	14	66,7
Interstitial infiltration
Y	3	14,3	1	4,8	0,157
N	18	85,7	20	95,2
Neutrophil in the vascular lumen
Y	6	28,6	-	-	0,014
N	15	71,4	21	100
Endothelial proliferation
Y	4	19,0	1	4,8	0,083
N	17	81,0	20	95,2
Vascular ectasia
Y	9	42,9	8	38,1	0,317
N	12	57,1	13	61,9
Dermal eosinophil
Y	10	47,6	4	19,0	0,014
N	11	52,4	17	81,0
Perivascular mononuclear infiltration
Y	20	95,2	17	81,0	0,083
N	1	4,8	4	19,0
Oedema in the papillary dermis
Y	21	100	17	81,0	0,046
N	-	-	4	19,0
Spongiosis
Y	4	19,0	1	4,8	0,083
N	17	81,0	20	95,2
Acanthosis
Y	5	23,8	2	9,5	0,083
N	16	76,2	19	90,5
Orthokeratosis
Y	19	90,5	14	66,7	0,025
N	2	9,5	7	33,3

Figure 1

Orthokeratosis and mild oedema in the superficial dermis and perivascular mononuclear infiltration in non-lesional intact skin sample (Haematoxylin and Eosin, original magnification: ×200) (a). Orthokeratosis and mild oedema in the papillary dermis, and mononuclear infiltration with neutrophils and perivascular eosinophils in some vascular lumens in the cold urticaria plaque (Haematoxylin and Eosin, original magnification: ×200) (b)

No staining was detected in the epidermis or dermis, either the plaque lesion or controls, with SP. Granular positive staining was seen in the erector pili muscle in all tissue samples [Figure 2a and 2b].

Figure 2

Negative staining in the epidermis and dermis and granular staining in the erector pili muscle in intact skin without lesions (a) and CU plaque (b) (Substance P antibody, original magnification: a, b ×200). Positive staining in intact skin without lesions (c) and CU plaque (d) (Mast cell chymase antibody, original magnification: c, d ×100). Strong positive staining in the full-thickness epidermis and vascular endothelium in intact skin without lesion (e) and CU plaque (f) (Angiotensin-converting enzyme antibody, original magnification: e, f ×100)

Dermal mast cells stained with chymase antibody in all samples [Figure 2c and 2d]. MCC was found to be statistically significantly higher in CU plaques than intact skin samples without lesions (p < 0.001) [Table 2]. There was no statistically significant relationship between tissue MCC level and patients’ age, duration of disease, age of disease onset and CTT values.

Table 2

Tissue mast cell chymase positive cell numbers in cold urticaria plaques and intact skin without lesions

	Mean±SD	Min-Max	P
Mast cell chymase
Plaque	74,3±20,4	42-112	<0,001
Control	55,9±19,7	11-100

Nuclear and cytoplasmic strong positive staining with ACE antibody was detected in keratinocytes, vascular endothelium, eccrine glands, sebaceous glands and erector pili muscle in CU plaques and all control intact skin samples. There was no loss of expression [Figure 2e and 2f].

Discussion

In this study, it was aimed to present a hypothesis in a common context of SP, MCC and ACE [Figure 3]. SP is a neuropeptide commonly found in the skin and peripheral nervous system. It is synthesized in the cell bodies of C-type peripheral neurons, which play a role in the transmission of stimuli such as cold and heat to the central nervous system. It is released from peripheral nerve endings through antidromic depolarization as a result of the stimulation. It causes neurogenic inflammation because of vasodilation and increased vascular permeability. A direct SP-mediated link between nerve fibres and mast cells has been demonstrated. SP stimulates mast cells and mast cell degranulation via neurokinin-1 (NK1) receptors.[9] Tóth-Kása et al.[10] emphasized that in cold and hot urticaria patients, there was no plaque development for 4–7 days in the areas where they applied capsaicin, which causes functional insufficiency in sensitive nerves and prevents SP release, and this neuropeptide may have a role in the development of urticarial plaque. In the study conducted by Metz et al.[4] in 2014, a four-fold increase in serum SP levels was found in CSU patients compared to healthy controls, and a similar increase was seen in CU patients. Based on this, we wanted to investigate the place of tissue SP levels in CU pathogenesis in patients with CU. However, although internal and external control samples were stained, we did not detect any staining in the lesional and non-lesional tissue samples.

Figure 3

The hypothesis for the effect of substance P, chymase and ACE in cold urticaria

Apart from urticaria, SP has also been studied in other dermatological diseases such as psoriasis, vitiligo and atopic dermatitis.[111213] In a study examining the status of neuropeptides, staining was found in the dermis, especially in the perivascular, around nerve fibres and rarely in the epidermis of normal skin. They emphasized that there is relatively less and only staining on the vascular walls in the papillary dermis in patients with urticaria and increased staining in atopic dermatitis patients.[14] The fact that no staining was detected in our study may be due to SP's rapid consumption at the tissue level in CU patients, or it may not be seen due to the difference in the methodological techniques used. The fact that it was detected only in the erector pili muscle may be related to the nerve density in the muscle tissue, or the differences in the degradation mechanism of SP between the tissues may also have been caused. The fact that no comparison could be made with healthy controls is a limitation of our study.

Mast cells have an important role in allergic diseases.[15] In healthy human skin, mast cells are observed in the papillary dermis around blood vessels and epidermal appendages.[16] In addition to similar localization, we found widespread and increased distribution in our patients’ dermis. MCC is one of the two proteases secreted from mast cell granules as a result of the stimulation. It also takes part in the destruction of SP.[6] Its proinflammatory effects are also known in allergic diseases such as atopic dermatitis.[7]

Different results were obtained in studies investigating mast cells in urticaria patients. A study evaluating the histopathology of patients with chronic idiopathic urticaria found that the number of mast cells increased 10 times compared to normal skin. They did not separate mast cells according to their chymase and tryptase content.[17] In another study conducted in patients with chronic idiopathic urticaria, no differences were detected in lesional, non-lesional skin and healthy skin in chymase and tryptase contents, immunohistochemically. They claimed that the increase in histamine in urticaria patients might be due to increased mast cell histamine content rather than an increase in mast cells.[18] In another study conducted on horses with urticaria, they found a decreased mast cell number compared to healthy skin in the urticaria group. They emphasized that this may be due to the loss of staining due to the discharge of the granule contents as a result of stimulation of mast cells.[19] Some reports also show that dermal mast cell numbers are normal in lesional and non-lesional skin of CU patients.[3] These differences in mast cell numbers in urticaria patients may be due to the variety of dyes used in the staining and cell counting techniques. Staining with MCC antibody is highly specific for mast cells, and the risk of staining-related problems is very low.[16] To the best of our knowledge, there is no other study investigating tissue MCC in CU patients. We found a statistically significant increase in MCC in lesional skin than non-lesional skin in our CU patients. We think that the mast cell, which plays a central role in allergic diseases, also plays an important role in CU patients’ pathogenesis.

MCC is also important with its effects on the local renin-angiotensin system. Like ACE, it catalyses the conversion of AT1 to the vasoactive and aldosterone stimulating peptide AT2. It is 20 times more effective than ACE in the cardiac muscle.[20] After realizing that MCC causes more AT2 production than ACE in cardiac tissues, a large number of chymase inhibitors have been developed as an alternative to ACE inhibitors. Its effects have been researched and found effective in many animal models.[21] In an animal model of atopic dermatitis, SUN13834, an orally administered chymase inhibitor, was found to be effective in subjects. It has been suggested as a treatment option for atopic dermatitis.[22] Likewise, chymase inhibitors may have therapeutic efficacy in CU patients. However, further research, including in vivo experiments and clinical studies, is needed to say this.

In recent studies, it has been revealed that there is a local RAS in tissues, unlike the systemic renin-angiotensin system. Steckelings et al.[23] detected the presence of AT1 and AT2 receptors in human skin and also demonstrated the local RAS by showing the presence of ACE, which also plays a role in AT2 production. The ACE expression has been seen in keratinocytes and vessel walls in normal human skin. The physiological and pathophysiological effects of this local RAS on human skin are not fully understood yet.

There are few studies on the role of ACE in dermatological diseases. Tissue ACE levels in psoriasis patients were measured by spectrophotometric method and were found to be higher than healthy skin.[24] In a study examining pathological scar tissues, an increase was measured by chromatography in the samples. In this study, immunohistochemical staining was found in the epidermis of normal intact skin, mainly in the basal and spinous layer and vascular endothelium.[25] In our study, we found full-thickness staining in the epidermis. In another study conducted in patients with alopecia areata, the epidermis was divided into three groups as 0, 1+, 2+ according to immunohistochemical ACE staining levels. Decreased staining was detected in patients with alopecia areata compared to the control group.[26] However, in our patients, we found full-thickness and strong staining in CU plaques and non-lesional skin in the epidermis. To determine this increase quantitatively, it should be supported by further investigations, such as spectrophotometric methods. The role of ACE and local RAS in the normal skin microenvironment seems to be subject to new studies in the future.

Few studies are focussing on the histopathology of CU. Lawlor et al.[27] detected an increase in the number of mast cells in the lesional-skin of CU patients. In this study, in which six patients were evaluated, neutrophilic infiltration was detected in one patient. The authors thought that the exudative process was valid in the pathogenesis of CU, regardless of the duration. In another study, mononuclear infiltration was detected in 10- and 20-min biopsies taken after cold stimulation, while eosinophils and neutrophils were rare.[28] In this study, neutrophilic infiltration was detected in 10/22 patients. Two patients had both dermis and vascular lumen, four patients had the only lumen, four patients had only dermal infiltration. Eosinophils and mononuclear cells were also detected in infiltrates. We think that neutrophils may contribute to the inflammatory process in the etiopathogenesis of CU.

Conclusions

This study is the first to show the status of tissue SP, chymase and ACE in CU plaques to the best of our knowledge. In our study, the absence of SP in the tissue may be due to its consumption due to increased ACE, or it may be secondary to rapid destruction by other molecules in the extracellular area. Also, ACEs causing an increase in SP degradation may cause a shift from the pathway leading to AT2 production and competition, contributing to the decrease in vasoconstrictor effect and vasodilation occurring in CU. A significant increase in MCC plays a fundamental role in the pathogenesis of CU as in CSU. The role of chymase and ACE in local cutaneous RAS is also a subject to be investigated.

Financial support and sponsorship

This research was funded by a grant from Turkish Society of Dermatology (2019/15).

Conflicts of interest

There are no conflicts of interest.