key: cord-1053963-8vk15klh
authors: Wallukat, Gerd; Hohberger, Bettina; Wenzel, Katrin; Fürst, Julia; Schulze-Rothe, Sara; Wallukat, Anne; Hönicke, Anne-Sophie; Müller, Johannes
title: Functional autoantibodies against G-protein coupled receptors in patients with persistent post-COVID-19 symptoms
date: 2021-04-16
journal: J Transl Autoimmun
DOI: 10.1016/j.jtauto.2021.100100
sha: bb210d9828e0fa9d86cd353fa986bb9344bbb5da
doc_id: 1053963
cord_uid: 8vk15klh

Impairment of health after overcoming the acute phase of COVID-19 is being observed more and more frequently. Here different symptoms of neurological and/or cardiological origin have been reported. With symptoms, which are very similar to the ones reported but are not caused by SARS-CoV-2, the occurrence of functionally active autoantibodies ((f)AABs) targeting G-protein coupled receptors (GPCR-(f)AABs) has been discussed to be involved. We, therefore investigated, whether GPCR-(f)AABs are detectable in 31 patients suffering from different Long-COVID-19 symptoms after recovery from the acute phase of the disease. The spectrum of symptoms was mostly of neurological origin (29/31 patients), including post-COVID-19 fatigue, alopecia, attention deficit, tremor and others. Combined neurological and cardiovascular disorders were reported in 17 of the 31 patients. Two recovered COVID-19 patients were free of follow-up symptoms. All 31 former COVID-19 patients had between 2 and 7 different GPCR-(f)AABs that acted as receptor agonists. Some of those GPCR-(f)AABs activate their target receptors which cause a positive chronotropic effect in neonatal rat cardiomyocytes, the read-out in the test system for their detection (bioassay for GPCR-(f)AAB detection). Other GPCR-(f)AABs, in opposite, cause a negative chronotropic effect on those cells. The positive chronotropic GPCR-(f)AABs identified in the blood of Long-COVID patients targeted the β(2)-adrenoceptor (β(2)-(f)AAB), the α(1)-adrenoceptor (α(1)-(f)AAB), the angiotensin II AT1-receptor (AT1-(f)AAB), and the nociceptin—like opioid receptor (NOC-(f)AAB). The negative chronotropic GPCR-(f)AABs identified targeted the muscarinic M(2)-receptor (M(2)-(f)AAB), the MAS-receptor (MAS-(f)AAB), and the ETA-receptor (ETA-(f)AAB). It was analysed which of the extracellular receptor loops was targeted by the autoantibodies.

The pandemic COVID-19 viral infection is often associated with severe respiratory and neurological complications, cardiovascular problems, microvascular and endothelial disorders, and gastrointestinal diseases. Additionally, these symptoms are often observed in patients who have already recovered from the disease and had negative follow-up coronavirus tests. In their Italian study, Carfi et al. [1] indicated that only 12.6% of investigated patients did not develop any persistent symptoms after recovering from COVID-19. Most of the symptomatic post-infection COVID-19 patients suffered from neurological disorders, such as chronic fatigue syndrome, postural orthostatic tachycardia syndrome (PoTS) and dysautonomia [1] . However, other neurological diseases, such as transverse myelitis, acute necrotising myelitis, Guillain-Barré syndrome and others, have also been reported in several recent case reports on patients following SARS-CoV-2 infection [2] [3] [4] [5] [6] [7] [8] [9] . Similar results concerning the extent of post-COVID-19 symptoms were also obtained in a German study that showed only 22% of their investigated COVID-19 patients stayed free of post-disease symptoms [10] .

Besides neurological manifestations, patients who recovered from COVID-19 also often developed cardiovascular implications [11] . The most prevalent abnormalities observed included myocardial inflammation, arrhythmia, tachycardia, bradycardia, and atrioventricular (AV) block [10, [12] [13] [14] [15] [16] [17] [18] . In a large multi-centre study, including the intensive care units of 68 geographically diverse hospitals across the United States, Hayek and co-workers investigated 5019 critically ill COVID-19 patients. They observed that of these 5019 patients, 14% (701/5019) had an in-hospital cardiac arrest which was associated with poor survival, particularly among older patients [19] .

Several authors assumed that autoimmune processes, involving the formation of autoantibodies, may be involved in the pathogenesis [20] and development of a post-COVID-19 syndrome [20] [21] [22] . In an initial study by Zhou et al. [20] , 21 patients critically ill with COVID-19 were investigated for the presence of autoantibodies. It was found that 20% had anti-52 kDa SSA/Ro antibodies (autoantibodies against extractable nuclear antigens), 25% had anti-60 kDa SSA/Ro antibodies and 50% had anti-nuclear antibodies.

Other autoantibodies acting as drivers of the disease have also been reported. It was recently shown by Bastard et al. that over 10% of their investigated COVID-19 patients with a life-threatening pneumonia condition (n = 987) presented with neutralising autoantibodies against interferon-ω (IFN-ω, n = 13), the 13 types of IFN-α (n = 36), or against both (n = 52). A few of their patients also showed autoantibodies against the other three type I IFNs. In contrast, the authors did not see any of these autoantibodies in 663 patients with asymptomatic or mild SARS-CoV-2 infection and they were only present in 4 of 1227 healthy subjects included for comparison [23] . Bastard et al. concluded from their data that the preexistence of neutralising anti-type I IFN autoantibodies was the cause of a critical condition, rather than it being the consequence of the infection [23] .

Novelli et al. concluded from their comprehensive systematic review about chronic inflammatory and autoimmune diseases onset during COVID-19 that "it is likely than the autoimmune manifestations described in COVID-19 represent more the results of the inflammatory cascade and the immune activation triggered by the virus rather than a direct effect of the virus per se" [24] .

In another study, Lyons-Weiler compared immunogenic peptides of SARS-CoV-2 with sequences of human proteins and found a high number of matching homologous sequences [25] . This would explain J o u r n a l P r e -p r o o f the high rate of persisting autoreactivity after SARS-CoV-2 infection. Kreye et al. examined neutralizing anti-SARS antibodies of isolated B-cell clones and observed that some showed "self-reactivity" while others were virus neutralizing only without showing any self-reactivity [26] . The possible impact of autoantibodies on the pathogenesis has most recently been discussed by Khamsi [27] .

It is a proven fact that autoimmune processes and the formation of functional autoantibodies ( f AABs) directed against G-protein coupled receptors (GPCR) play a role in the development of neurological [28] [29] [30] and cardiovascular symptoms [31] . Therefore, in this present study, it was tested if such GPCRf AABs might also be associated with the development of corresponding post-COVID-19 symptoms. We investigated virus-free sera from 31 recovered COVID-19 patients with respect to the occurrence of GPCRf AABs.

Sera were obtained from 31 patients, 29 who were still suffering from post-COVID-19 symptoms, after recovery from acute disease and 2 patients who were symptom-free (all positive tested by PCR). All patients signed a written informed consent form which included giving permission to include the anonymised clinical data in a scientific publication, in agreement with the Declaration of Helsinki. Using the RedCap project for data collection and management [32] with the permission from the ethics commission (no: 295-20B), 6 of the sera were recruited at the University of Erlangen.

As a safety-precaution, the COVID-19 patient sera were heat inactivated for 30 min at 56 °C before use. Afterwards, 0.4 mL of the samples were dialysed against 1 L of dialysing buffer (0.15 M NaCl, 10 mM phosphate buffer, pH 7.4; Membra-Cel MD 44, 14 kDa, Serva) for 24 hours to remove low-molecular weight bioactive compounds and peptides. Finally, 40 µL of the dialysed samples were added to the bioassay (final dilution of 1:50).

For the identification and characterisation of GPCRf AABs, a bioassay was used, as described in great detail by Davideit et al. [33] and Wenzel et al. [34] for GPCRf AABs against the beta1-adrenoceptor, and for other GPCRf AABs by Wallukat et al. [30, 35, 36] . After contact with the respective autoantibodies, a change in basal beating rate of spontaneously beating cardiomyocytes expressing GPCR was used as the measuring signal. The receptor specificity was checked by either subsequent addition of specific receptor blockers, resulting in an annulation of this effect, or by addition of corresponding receptor-epitopecompeting extracellular loop peptides. In detail: for the specification of the β 2f AABs, the receptor antagonist ICI118.551 (0.1µM) was used and also neutralizing peptides corresponding to the first LGYWAFGRVFCN and GWRQPAPEDETICQINEEPGYVLFSAL, respectively. For all peptides 2 µL of a stock solution of 100 µg/mL was added to 40 µL of the corresponding GPCRf AAB sample and incubated for 30min before the mixture was transferred to the cells.

Several different GPCRf AABs were identified in the 31 sera of recovered COVID-19 patients. All 31 investigated patients had between 2 and 7 different GPCRf AAB ( Table I) . This was a surprising unexpected effect. In healthy controls, which are included in many studies, these autoantibodies are only found in a small percentage [37, 38] " Two functionally active autoantibodies, that were seen in almost all investigated former COVID-19 patients, were directed against the β 2 -adrenoceptor (β 2f AAB) and the muscarinic M 2 receptor (M 2f AAB). These f AABs induced a positive and a negative chronotropic response on their targeted receptors, respectively.

Two other f AABs that were also present in 29 (90%) of the 31 investigated post-COVID-19 patients were directed against the angiotensin II AT1 receptor ( f AT1-AAB) and the angiotensin 1-7 MAS receptor (MAS-AAB). These receptors belong to the renin angiotensin system (RAS) and cause a positive and negative chronotropic effect, respectively, when targeted by the respective f AABs.

Post-infection hair loss (alopecia) was experienced by 8 of the recovered patients. In sera of these patients, three additional GPCRf AABs were discovered: the negative chronotropic ETAf AAB (4/8, the positive chronotropic NOCf AAB (5/8), and the positive chronotropic α 1 -AAB (3/8). Not every alopecia patient showed all three of these GPCRf AABs. Instead, their occurrence varied, and a pattern is not yet detectable. As shown in (Table I) In a next step it was analysed at selected ETA-1f AAB, MASf AAB and α 1f AAB, β 2f AAB and NOCf AAB positive samples which of the extracellular loops of the receptors were targeted by the respective autoantibodies by addition of competing loop peptides (Figure 1) . Here it was clearly to see that with four of the tested autoantibodies (ETA-1f AAB, MASf AAB, β 2f AAB, and NOCf AAB) in each of the investigated cases the second extracellular loop of the receptor was targeted by the autoantibody and consequently neutralized by addition of its competing loop peptide. α 1f AAB containing samples targeted the 1 st extracellular loop of the receptor. This analysis provided additional evidence that no nonspecific effects had influenced the outcome.

The astonishing finding of this investigation is the fact that an unusually high number of GPCRf AABs were detected in the serum of recovered COVID-19 patients who mostly suffered from a variety of different post-COVID-19 symptoms. Due to the functionality of such GPCRf AABs, the question of whether these GPCRf AABs may play a role in the development of post-COVID-19 symptoms is raised.

A continuing fatigue-like symptom, persisting long after virus follow-up tests are negative, was a frequently reported impairment in patients of this study (17/31) , and other studies [1] . For patients suffering from a classical coronavirus-independent fatigue syndrome, the occurrence of β 2f AABs, M 2f AABs and, in some cases, also ETAf AABs has already been reported before [39] . Here, with this post-COVID-19 study, almost all investigated sera contained β 2f AABs and M 2f AABs. The combination of β 2f AABs and M 2f AABs have also been identified in sera of patients suffering from PoTS and dysautonomia [40] , both of which are conditions now observed in post-COVID-19 patients (7/31 and 2/31, respectively, not overlapping). Furthermore, this combination of β 2f AABs with M 2f AABs had also been identified before by our group, in patients with complex regional pain syndrome (CRPS) [41] , in patients suffering from narcolepsy type 1, here additionally with the NOCf AAB in 9 of 10 cases [36] and in patients with small fibre diseases.

Two of the identified GPCRf AABs, observed in over 90% of the investigated COVID-19 patient sera (29/31), were directed against receptors of RAS, namely the angiotensin II AT1 receptor and the angiotensin (1-7) MAS receptor. These vasoactive AT1f AABs had been identified before in patients with malignant hypertension, therapy-resistant hypertension, preeclampsia, and kidney diseases [42] [43] [44] . Moreover, Dragun et al. [44] showed that AT1f AABs induced the rejection of kidneys in a subgroup of patients that underwent kidney transplantation. The authors also showed that the transfer of human AT1f AABs to kidney transplanted rats caused the occlusion of the kidney arteries in the recipients.

Given the evidence described above, it may be that vasoactive processes, caused by the occurrence of AT1f AABs and MASf AABs, might also be involved in the pathogenesis of post-COVID-19 symptoms. However, it is highly unlikely that the pathophysiological effects are caused by the f AABs alone. For example, Lukitsch and co-workers [45] already showed that the addition of the human AT1f AABs to isolated kidney arteries induced a contraction of these arteries, but only in ischemic arteries and in arteries that were taken from kidney transplanted rats. Arteries obtained from healthy rats did not respond to AT1f AABs, even though they reacted to their natural agonist, angiotensin II (which confirmed that the receptors were intact). Taken together, these data demonstrated that the AT1f AABs did not act alone but needed ischaemic or inflammatory cofactors to have full effect. With respect to the COVID-19 situation, this is of course an absolutely obvious situation. Thrombo-inflammatory factors which even may become predictive markers for COVID-19 complications have been described by Cremer et al. [46] . Other immune biomarkers, as taken together by Fouladseresht et al. [47] have also been reported.

To date, the evidence suggests that a combination of ischaemic or inflammatory cofactors and autoantibodies can act to maintain a cardiac inflammatory process. Specifically, it has been shown that AT1f AABs and α1f AABs can influence the maturation and degranulation of cardiac mast cells [48] , suggesting that they can contribute to inflammation.

It has also already been reported that COVID-19 induces an imbalance of RAS through viral-occupation of the angiotensin-converting enzyme 2 (ACE2) receptors, which reduces the generation of protectivepeptides angiotensin-(1-7) and (1) (2) (3) (4) (5) (6) (7) (8) (9) . This subsequently decreases the stimulation the MAS-and angiotensin II ATR2-receptors, and is accompanied by an overstimulation of the AT1-receptors due to reduced degradation of angiotensin II by ACE2 [49] . Therefore, Steckelings and Sumners [49] recently suggested that ATR2 receptor agonists could be used to treat COVID-19-induced disorders of various organ systems.

This is in good agreement with the identification of MASf AABs in over 90% of the symptomatic patients examined. Autoantibodies against this receptor have been observed before in a cancer patient after chemotherapy [50] and in patients with multiple sclerosis (MS) (unpublished results). In MS patients we observed the combination of MASf AABs with α 1f AABs. In this context, it is interesting to see that in several case reports it was shown that COVID-19 patients can develop neurological complications like transverse myelitis [2] and Guillain-Barré syndrome [5, 7] . Whether the MASf AABs observed in this COVID-19 study is involved in the development of the reported neurological symptoms should be clarified in further investigations.

Furthermore, our data showed that 2 of the patients with a mild COVID-19 infection developed f AABs but not the symptoms as seen in the other recovered COVID-19 patients. We assume that in both of these patients adaption processes might have prevented the binding of f AABs to the receptor or the receptors are not available for the f AABs as described by Lukitsch and co-worker for angiotensin II AT1 f AAB before [45] .

We strongly assume that the GPCRf AABs play an important role in the development and maintenance of post-COVID symptoms. These GPCRf AABs persistently stimulate their corresponding receptors and the normal, physiological, cell-protective desensitisation of the receptors is inhibited by the f AABs themselves [51] .

It has already been shown in other diseases, such as idiopathic dilated cardiomyopathy (here it is the autoantibody targeting the beta1-adrenoceotor), that GPCRf AABs play a significant role in the pathogenesis of the disease [31] . Additionally, the removal of these f AABs by immunoadsorption led to an improvement in cardiac function and to a significant increase in survival rate [52, 53] . A similar beneficial therapeutic effect of the removal of GPCRf AABs has also been observed in β 2f AAB positive, therapy-refractory, open-angle glaucoma patients. Here, the removal of the f AABs by immunoadsorption resulted in a reduction of the ocular pressure [54] .

With respect to post-COVID-19 symptoms, Masuccio et al. reported a patient who was suffering from post-infection acute motor axonal neuropathy and myelitis. The patient tested positive for the ganglioside anti-GD1b IgM autoantibody, and a partial recovery was achieved through plasma exchange combined with subsequent immunoglobulin substitution [7] .

The Sars-CoV-2 spike protein is a potential epitopic target for biomimicry-induced autoimmunological processes [25] . Therefore, we feel it will be extremely important to investigate whether GPCRf AABs will also become detectable after immunisation by vaccination against the virus.

Our results indicated that all 29 investigated symptomatic post-COVID-19 patients developed f AABs directed against different GPCRs, known to be able to disturb the balance of neuronal and vascular processes. Most of these patients developed an antibody pattern consisting of β 2f AABs, M 2f AABs, AT1f AABs, and MASf AABs. These agonistic f AABs activate their corresponding receptors like classical agonists. The observed specific GPCRf AAB pattern has been observed before in several neurological and cardiac disorders and might also support the development of neurological and/or cardiovascular symptoms after COVID-19 recovery. These results provide valuable clues that are worth pursuing and investigating further.

A major limitation of this study is that it is only a snapshot. Therefore, the causal relationship between the presence of GPCRf AAB and the disease cannot be shown. In order to be able to identify causal relationships, samples from former COVID patients must be systematically collected over a longer period of time for GPCRf AAB detection. Afterwards the data have to be retrospectively assigned to recovering and non-recovering Long-COVID-19 patients. 

Gemelli Against COVID-19 Post-Acute Care Study Group (2020) Persistent Symptoms in Patients After Acute COVID-19

Acute transverse myelitis after COVID-19 pneumonia

Acute transverse myelitis in COVID-19 infection

Acute necrotizing myelitis and acute motor axonal neuropathy in a COVID-19 patient

Guillain-Barré Syndrome Associated with SARS-CoV-2

Transverse myelitis related to COVID-19 infection

A rare case of acute motor axonal neuropathy and myelitis related to SARS-CoV-2 infection

COVID-19-associated acute transverse myelitis: a rare entity

Transverse Myelitis in a Child With COVID-19

Outcomes of Cardiovascular Magnetic Resonance Imaging in Patients Recently Recovered From Coronavirus Disease 2019 (COVID-19)

Cardiovascular Implications of Fatal Outcomes of Patients With Coronavirus Disease 2019 (COVID-19)

Association of Cardiac Infection With SARS-CoV-2 in Confirmed COVID-19 Autopsy Cases

Cardiovascular Considerations for Patients, Health Care Workers, and Health Systems During the COVID-19 Pandemic

COVID-19 and cardiovascular disease: from basic mechanisms to clinical perspectives

Bradycardia in Patients With COVID-19: A Calm Before the Storm? Cureus 12:e8599

Bradyarrhythmias in patients with COVID-19: A case series

Case report: high-grade atrioventricular block in suspected COVID-19 myocarditis

Heart Brake: An Unusual Cardiac Manifestation of COVID-19

In-hospital cardiac arrest in critically ill patients with covid-19: multicenter cohort study

Clinical and Autoimmune Characteristics of Severe and Critical Cases of COVID-19

Covid-19 and autoimmunity

Diverse Functional Autoantibodies in Patients with COVID-19

Autoantibodies against type I IFNs in patients with lifethreatening COVID-19

The JANUS of chronic inflammatory and autoimmune diseases onset during COVID-19 -A systematic review of the literature

Pathogenic priming likely contributes to serious and critical illness and mortality in COVID-19 via autoimmunity

A Therapeutic Non-self-reactive SARS-CoV-2 Antibody Protects from Lung Pathology in a COVID-19 Hamster Model

Rogue antibodies could be driving severe COVID-19

Complex syndromes of chronic pain, fatigue and cognitive impairment linked to autoimmune dysautonomia and small fiber neuropathy

Immunoadsorption to remove ß2 adrenergic receptor antibodies in Chronic Fatigue Syndrome CFS/ME

Functional autoantibodies in patients with different forms of dementia

2020) β1-Adrenoreceptor Autoantibodies in Heart Failure: Physiology and Therapeutic Implications

Research electronic data capture (REDCap)--a metadatadriven methodology and workflow process for providing translational research informatics support

Determination of Agonistically Acting Autoantibodies to the Adrenergic Beta-1 Receptor by Cellular Bioassay

Performance and in-house validation of a bioassay for the determination of beta1-autoantibodies found in patients with cardiomyopathy

Analytics of Functional Autoantibodies in Patients with Chagas Disease

Activating autoantibodies against G proteincoupled receptors in narcolepsy type 1

Autoantibodies directed against α1-adrenergic receptor and endothelin receptor A in patients with prostate cancer

Autoantibodies Activating the β2-Adrenergic Receptor Characterize Patients With Primary and Secondary Glaucoma

Antibodies to β adrenergic and muscarinic cholinergic receptors in patients with Chronic Fatigue Syndrome

Serum Activity Against G Protein-Coupled Receptors and Severity of Orthostatic Symptoms in Postural Orthostatic Tachycardia Syndrome

Autoimmunity against the β2 adrenergic receptor and muscarinic-2 receptor in complex regional pain syndrome

Autoantibodies against the angiotensin receptor (AT1) in patients with hypertension

Agonistic autoantibodies directed against the angiotensin II AT1 receptor in patients with preeclampsia

Angiotensin II type 1-receptor activating antibodies in renal-allograft rejection

Renal ischemia and transplantation predispose to vascular constriction mediated by angiotensin II type 1 receptor-activating antibodies

Elevated markers of thrombo-inflammatory activation predict outcome in patients with cardiovascular comorbidities and COVID-19 disease: insights from the LEOSS registry

Predictive monitoring and therapeutic immune biomarkers in the management of clinical complications of COVID-19

Autoantibodies against G-protein-coupled receptors modulate heart mast cells

Correcting the imbalanced protective RAS in COVID-19 with angiotensin AT2-receptor agonists

Are Agonistic Autoantibodies against G-Protein Coupled Receptors Involved in the Development of Long-Term Side Effects of Tumor Chemotherapy?

Autoantibodies against the beta-adrenergic receptor in human myocarditis and dilated cardiomyopathy: beta-adrenergic agonism without desensitization

Long-term benefits of immunoadsorption in β(1)-adrenoceptor autoantibody-positive transplant candidates with dilated cardiomyopathy

Immunoglobulin adsorption in patients with idiopathic dilated cardiomyopathy

Agonistic Autoantibodies to the β2-Adrenergic Receptor Involved in the Pathogenesis of Open-Angle Glaucoma

The authors thank all the patients and health care workers for their support in this study. Annekathrin Haberland helped to write the manuscript. We thank Sabine Bartel for the supply of the rat cardiomyocytes. 

• Sera from post-COVID syndrome patients contained functionally active autoantibodies which target G-protein coupled receptors • The list included pathogenic autoantibodies targeting the nociception receptor, the β 2 -and α 1 -adrenoceptors, angiotensin II AT1-, muscarinic M 2 -, MAS-, and ETA-receptors • Included syndromes were of neurological and cardiological origin, or a combination of both • This type of pathogenic autoantibody and specific autoantibody pattern were described before the first appearance of SARS-CoV2 in distinct neurological deficits and cardiovascular diseases J o u r n a l P r e -p r o o f

☐ The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.☒The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: