Accepted Manuscript Autoimmunity against a glycolytic enzyme as a possible cause for persistent symptoms in Lyme disease Paolo Maccallini, Serena Bonin, Giusto Trevisan PII: DOI: Reference: S0306-9877(17)30771-5 https://doi.org/10.1016/j.mehy.2017.10.024 YMEHY 8719 To appear in: Medical Hypotheses Received Date: Accepted Date: 20 July 2017 24 October 2017 Please cite this article as: P. Maccallini, S. Bonin, G. Trevisan, Autoimmunity against a glycolytic enzyme as a possible cause for persistent symptoms in Lyme disease, Medical Hypotheses (2017), doi: https://doi.org/10.1016/ j.mehy.2017.10.024 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. Autoimmunity against a glycolytic enzyme as a possible cause for persistent symptoms in Lyme disease Paolo Maccallini1, Serena Bonin2 and Giusto Trevisan2 1: Department of Mechanical Engineering, Sapienza University of Rome-Rome-Italy 2: DSM-Department of Medical Sciences-Unit of Dermatology-University of Trieste-Trieste-Italy Corresponding author: Serena Bonin University of Trieste DSM-Department of Medical Sciences Cattinara Hospital Surgical Pathology BLG Strada di Fiume 447 34149 Trieste Italy Abstract Some patients with a history of Borrelia burgdorferi infection develop a chronic symptomatology characterized by cognitive deficits, fatigue, and pain, despite antibiotic treatment. The pathogenic mechanism that underlines this condition, referred to as post-treatment Lyme disease syndrome (PTLDS), is currently unknown. A debate exists about whether PTLDS is due to persistent infection or to post-infectious damages in the immune system and the nervous system. We present the case of a patient with evidence of exposure to Borrelia burgdorferi sl and a long history of debilitating fatigue, cognitive abnormalities and autonomic nervous system issues. The patient had a positive western blot for anti-basal ganglia antibodies, and the autoantigen has been identified as γ enolase, the neuron-specific isoenzyme of the glycolytic enzyme enolase. Assuming Borrelia own surface exposed enolase as the source of this autoantibody, through a mechanism of molecular mimicry, and given the absence of sera reactivity to α enolase, a bioinformatical analysis was carried out to identify a possible cross-reactive conformational B cell epitope, shared by Borrelia enolase and γ enolase, but not by α enolase. Taken that evidence, we hypothesize that this autoantibody interferes with glycolysis in neuronal cells, as the physiological basis for chronic symptoms in at least some cases of PTLDS. Studies investigating on the anti-γ enolase and anti-Borrelia enolase antibodies in PTLDS are needed to confirm our hypotheses. Keywords: Post-treatment Lyme disease syndrome; autoimmunity; molecular mimicry; enolase; glycolysis. Introduction Borrelia burgdorferi (Bb), the causative agent of Lyme disease (LD), infects humans through Ixodes tick bites. Infection can involve several tissues, including skin, joints, heart and nervous system and can result in arthritis, carditis, and neurological symptoms (1). Involvement of nervous system, known as neuroborreliosis, may manifest as encephalopathy, myelopathy, and peripheral neuropathy, which usually respond to antibiotic treatment (2). Nevertheless, some patients with LD report only partial and transient improvements in cognitive impairment, fatigue and musculoskeletal pain with prescribed courses of antibiotics and develop a chronic condition referred to as posttreatment Lyme disease syndrome (PTLDS) (3). It is currently under debate if a partial response to antimicrobials could be due to persistent infection, damage to the brain, or to some form of 1 acquired immune dysfunction (4). In detail, several studies have supported the hypothesis that some sort of immune dysregulation could occur. Both T and B cell autoimmunity have been detected in drug-resistant Lyme arthritis (5, 6) and several authors have reported on the potential of autoimmunity against neural antigens due to molecular mimicry, in LD: cross-reactivity has been experimentally shown between B. burgdorferi’s flagellin and a component of human peripheral nerve axon (heat shock protein 60) (7), between non-protein antigens of the pathogen and gangliosides (8), and between Borrelia OspA and antigens expressed in human brain, spinal cord and dorsal root ganglia (9). Recently, two studies reported heightened, but apparently non-specific, production of antibodies against brain proteins in about 50% of patients with a history of LD and persistent symptoms (10, 11). Here, we report the detection of anti-γ enolase antibodies in one patient with evidence of Borrelia burgdorferi infection and persistent symptoms of fatigue and cognitive impairment. Mammalian enolase acts as homodimeric or heterodimeric isoenzymes, combining three subunits: α, β, and γ enolase. Alpha-enolase is ubiquitous, β enolase is mainly expressed in muscles, while γ enolase in neurons and neuroendocrine tissues (12). In absence of serum reactivity against the α subunit of the same enzyme, and assuming the enolase of B. burgdorferi as the trigger for the autoantibody, a bioinformatical analysis with an ad-hoc-developed software has been carried out to propose a cross-reactive conformational epitope between γ enolase and Borrelia burgdorferi enolase and to suggest a possible pathogenic mechanism for chronic symptoms in Lyme disease. Materials and methods Patient A man (aged 35 at diagnosis), with a long-lasting history of fatigue and cognitive issues, presented at the University Hospital of Trieste, where he was submitted to a complete clinical and instrumental evaluation. Infectious agents were searched for in the patient’s serum as shown in Table 1. DNA from peripheral blood was submitted to Borrelia detection as previously reported (13). Anti-Borrelia burgdorferi antibodies were detected by Western blot using anti-BorreliaEuroline-RN-AT (Euroimmun, Germany). Anti-basal ganglia antibodies (ABGA) assay was performed as previously described (14). Antigens were derived from sections of human basal ganglia containing caudate and putamen; the sample comes from a donor with no history or evidence of neurological disease. Serum was also analysed for reactivity to α enolase, γ enolase, and pyruvate kinase, which have been linked respectively to the 45 kDa (doublet) and the 60 kDa bands of the ABGA immunoblot (15). Those tests were performed at the Carlo Besta Institute (Neuromuscular and Neuroimmunology Laboratory), except for the anti-α enolase assay, that was performed by Wieslab Laboratory, Sweden. An extensive panel of other autoantibodies (including various anti-neuronal Abs) was also performed (Table 2). Search for unique peptides of γ enolase A software, called EPITOPE, written by one of the authors (MP) in Octave 4.0.0, has been used to identify a set of linear peptides of γ enolase, which are not shared with α enolase. EPITOPE considers each possible peptide of γ enolase with a length of δ amino acids, and calculates the score of the best alignment with every α enolase peptides of the same length. For each comparison, EPITOPE implements the algorithm developed by Needleman and Wunsch (16), with a gap model ܽ + ܾ ⋅ ݔ, where ܽ is the opening gap penalty, ܾ is the extending one, and ݔis the extension of the gap. A penalty for gaps at the end of the alignment was also assumed. As a result, a score is associated to every γ enolase δ-amino-acid peptide. The program then gives as output only those peptides of γ enolase to which a score below S is associated. We regard those peptides of γ enolase as non-cross-reactive with α enolase. Parameters have been chosen according to what follows. As α 2 enolase and γ enolase share about 83% of identity (Figure 1), a possible choice for the substitution matrix is MDM20 (17). Given two random sequences of m and n amino acids respectively, it has been calculated (18) that the probability p that a local alignment attains a score of at least S is given by Eq. 1 = 1 − ݁ ି⋅⋅⋅ షഊೄ To define the parameters λ, K, a, and b, LALIGN provided by the European Bioinformatic Institute was used. If γ enolase is selected as the first sequence, α enolase as the second sequence, MDM20 as substitution matrix, the program gives λ = 0.2298 and K = 0.3234, and selects an opening gap and an extending gap of 22 and 4, respectively. Alignments were considered significant when < 0.05 (17), corresponding to a score ܵ > 60, according to Eq. 1. We calculated δ as the lowest number of amino acids necessary to give a significant score (minimal significant length), which is obtained dividing the minimal significant score S by the relative entropy of the substitution matrix of choice (17). We obtain for δ a value of 7 amino acids. Running EPITOPE, 162 peptides with a score below 60 have been identified. These peptides cluster into 8 subsequences of amino acids of γ enolase which include all the three regions of sequence difference between the human α and γ forms (271-285, 298-316, 416-433), experimentally demonstrated by McAleese et al. to be immunogens that can be used to generate antibodies to γ enolase not reacting with α enolase (19). A threshold score of 50 leads to a narrower selection for unique peptides of γ enolase, which still includes the 3 sequences found by McAleese. Therefore, a score < 50 corresponds to non-cross-reactive peptides, a score > 60 corresponds to cross-reactive peptides, while a score 50 ≤ ܵ ≤ 60 defines borderline results. We obtain 93 peptides scoring < 50 that cluster into twelve subsequences of amino acids of γ enolase (Figure 2, columns 1, 2). Using the experimentally determined 3D structure of γ enolase (20), all twelve subsequences were verified to be displayed on its surface, in agreement with the fact that sites in the protein cores are highly evolutionary conserved (21). Search for cross-reactive epitopes between enolases A second type of software, EPITOPE-LOCAL, has been written to calculate the best local alignment between each peptide of γ enolase found by EPITOPE and the following enolases: human β enolase, enolase from B. afzelii (strain PKo), enolase from B. burgdorferi ss (strain ZS7) and enolase from B. garinii (strain PBr) (Figure 2, columns 3-6). EPITOPE-LOCAL is a simpler version of EPITOPE, which aligns a given peptide of δ amino acids with all possible peptides of the same length from a given sequence, displaying as output the alignment with the highest score. As EPITOPE, this software is based on the algorithm by Needleman and Wunsch and uses the same settings (substitution matrix, gap penalties, value for δ). This analysis aimed to find epitopes of Borrelia enolase that might be a trigger for anti-γ enolase autoantibodies, through a mechanism of molecular mimicry. Protein sequences and 3D structures The following sequences have been used: human α enolase 1 (P06733-1), human β enolase 1 (P13929-1), human γ enolase 1 (P09104-1), Borrelia afzelii (strain PKo) enolase (Q0SNH5-1), Borrelia burgdorferi ss (strain ZS7) enolase (B7J1R2-1), Borrelia garinii (strain PBr) enolase (B7XT07). A three-dimensional structure of γ enolase experimentally determined by Leonard P.G. and colleagues (20) has been considered, the PDB ID of which is 4ZA0. For β enolase, the experimentally determined structure by Vollmar M. and colleagues (22) (PDB ID: 2XSX) has been used. Structures for enolases of Borrelia species have not yet been experimentally determined. Therefore, a theoretical model for B. burgdorferi ss enolase offered by ModBase, which uses as 3 template the experimental 3D structure of enolase from Enterecoccus hirae (PDB ID: 1IYX), has been considered. To verify the surface exposure of candidate epitopes on α, γ and β enolases, downloaded in ASN1 format, the 3D viewer Cn3D 4.3.1 has been used. Results Patient The patient is a 37-year-old man. Since 1999 he has been afflicted by severe fatigue, memory impairment and lack of concentration. At that time, he was 19 and he had never had any neurological symptom before. In the early years, the abovementioned symptoms presented with a cyclic trend of relapses and recoveries. In 2002, a flu-like illness precipitated his condition which has since become chronic. During the last 15 years, he has been experiencing other symptoms, such as muscular and joint pain, acral hypoesthesia, post-exertional malaise, and orthostatic intolerance, resulting in him being house-bound. He fulfils the 2015 diagnostic criteria published by IOM (23) for myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS). In August 2014, the patient presented at Trieste University Hospital, where he resulted positive to Borrelia both by PCR and Western Blot, as shown in Table 1. The patient was also positive to HHV6 (past infection). Current infection by Helicobacter pylori was detected and treated (Table 1). In 2000, the patient presented with a nodule of one centimeter in diameter, located on one nipple. It was investigated with echography and mammography and cancer was ruled out. Its etiology was not further investigated at that time and the lesion resolved spontaneously within months. That lesion was compatible with a nodular form (Spiegler-Fendt type) of lymphadenosis benigna cutis (LABC) of areola mammae, which is a typical second stage manifestation of LD in Europe (24). Patient was submitted to antibiotics, with complete and rapid resolution of acral hypoesthesia and only partial and temporary improvements in the other symptoms. In 2015, an extensive panel of anti-neuronal antibodies was performed (Table 2). Positive anti-Basal Ganglia Antibodies and anti-γ enolase antibody were detected. Cross-reactivity between enolases EPITOPE identifies 93 peptides of γ enolase that do not cross-react with α enolase (Figure 2, column 1, 2). Six of them cross-react with enolases from the three main pathogenic species of B. burgdorferi. With further analyses, we have found that, among these 6, only peptides 49-55, 59-65, 60-66 and 188-194 cross-react with sequences of Borrelia enolases that, in turn, do not cross-react with any peptide of α enolase (as detailed in Figure 3 for peptide 49-55). The results of this analysis are summarized in Figure 4. As shown, β enolase is involved in cross-reactivity too, although to a lesser extent. Conformational epitopes Peptides 49-55, 59-65 and 60-66 of one molecule of γ enolase identify a conformational epitope with peptide 188-194 of the other subunit of the homodimer γ-γ (see Figure 5, left). Peptides 49-55, 59-65 and 60-66 define a conformational epitope on the surface of β enolase, too (not shown). Peptides 52-58, 62-68 and 63-69 from one molecule of Borrelia burgdorferi enolase generate a conformational epitope with peptide 186-192 from another molecule, when the two subunits combine to form the glycolytic dimeric enzyme (Figure 5, right). If we assume the presence of serum reactivity to both γ enolase and α enolase (which is not the case of our patient) we find, with the same methods described above, a conformational B cell epitope shared by all the three human enolases and Borrelia enolase (Figure 6). It partially overlaps with the previously discussed conformational epitope. 4 Discussion There is currently debate on whether Lyme patients with long-term symptoms are suffering from persistent infection and/or from some sort of immune-mediated disorder. These symptoms most commonly include fatigue, cognitive deficits and musculoskeletal pain (3). Persistent cognitive abnormalities have been associated to hypometabolism in several brain regions (25, 26). In this study, we hypothesize that chronic symptoms are caused by an autoantibody - triggered by Borrelia infection through a mechanism of molecular mimicry - which interferes with glycolysis in neuronal tissues. The patient presented in this study had a complex picture of chronic symptoms, which matches the clinical picture of ME/CFS (23). He had both positive PCR for borrelial DNA and serum antibodies against two antigens of B. burgdorferi (p41 and p19, Table 1), and responded only partially to antibiotic treatment. Taken higher level and frequency of anti-neural antibodies in patients with PTLDS (10, 11), we further investigated for some possible autoimmune mechanism. Reactivity of patient’s serum against a wide range of neuronal antigens was tested, resulting in positivity to anti-γ enolase Abs (Table 2). This anti-neural autoimmunity could be due to several factors, but the observation that enolase is highly conserved from archaebacteria to mammals (27) strongly supports for an infectious agent as a possible source of the autoantibody, through a cross reactive immune response between microbial and human enolase. Borrelia enolase (BEN) shares almost 50% of sequence identity with human enolase and more than 130 similar positions (on a total of 433 residues) (Figure 1). Furthermore, BEN has recently been observed on the outer membrane of the spirochete (28-30), and has been demonstrated to be immunogenic, both in mice and in humans (30, 31). Therefore, we hypothesize that surface-exposed BEN could elicit antibodies that cross-react with human γ enolase, due to the similarity between amino acid sequences (molecular mimicry). To further support our hypothesis, an anti-human α enolase antibody has been shown to recognize purified recombinant BEN (28). Our computational analyses allowed us to find a conformational B cell epitope shared by human γ and Borrelia enolase (and to a lesser extent by β enolase), but not by α enolase (Figure 5). B cell epitopes, as it is well known, need surface exposure (32). Most of them are conformational and have linear peptides of maximum 4-7 residues (33). These characteristics are fulfilled by our conformational epitope. Human enolases are functionally active as five dimers (α-α, α-β, β-β, α-γ, γ-γ) (12). As the γ-γ dimer is specific for neuronal and neuroendocrine tissues (12), we hypothesize the central nervous system (CNS) as a possible target for this autoantibody. Autoantibodies against enolase have already been reported in several autoimmune diseases, while being relatively rare in healthy controls (15, 34, 35), in neurologic diseases (36), and in autism spectrum disorder (37). The involvement of enolase autoantibodies on our patient’s symptoms is supported by the detection of anti-enolase autoimmunity in brain disorders such as PANDAS (15, 38), obsessive-compulsive disorder (OCD) (35, 39), encephalitis lethargica (EL) (34), multiple sclerosis (MS) (40), and Hashimoto’s encephalopathy (HE) (41-43). In particular, HE resembles some features of Lyme encephalopathy (25). To support our hypothesis on an infectious involvement in anti-enolase antibodies production, other authors reported on a causal link between PANDAS and Streptococcus pyogenes (15, 36, 38). A similar association was proposed for OCD in adults (35), and for EL (34). Cross-reactivity between streptococcal surface enolase and human α enolase was experimentally demonstrated in sera of patients with acute rheumatic fever reacting with both molecules (44). Although anti-enolase autoimmunity seems to be diffused in a wide range of autoimmune diseases, different diseases appear to be linked to auto-Ab against specific isoforms: while PANDAS seems to be mainly associated with anti-γ enolase autoimmunity (15), Behçet’s disease (45), systemic lupus erythematosus, and systemic sclerosis (46), are linked to antibodies against α enolase, but not against γ. Furthermore, diseases in which an Ab against the same isoenzyme of enolase is present, 5 have often been linked to different epitopes. For instance, sera from patients with cancer-associated retinopathy bind residues 56-63 of α enolase (47), while peptide 207-238 of α enolase seems to be a specific autoepitope in endometriosis (48), and in HE patients, most of sera reacts against the amino terminal of α enolase (residues 1-134) (41, 43). These data support the hypothesis that the auto-Ab to the conformational epitope of γ enolase here discussed may be specific for PTLDS. In our hypothesis, the autoantibody against the conformational epitope of Borrelia enolase binds γ enolase exposed on the surface of neurons and it is then internalized (Figure 7, III). Once in cytoplasm, the antibody inhibits the catalytic activity of γ enolase in the ninth step of glycolysis, depleting the intracellular level of ATP (Figure 7, IV). We propose alteration of brain distal glycolysis by anti-γ enolase antibodies to explain changes observed in Lyme encephalopathy, where lower cerebral blood flow and metabolic rate in many areas of the brain have been reported, and suggested as primarily metabolic driven (25, 26). As γ enolase is expressed on the surface of neurons and neuroendocrine cells (49), the proposed internalization of anti-enolase autoantibodies is theoretically possible, and it has in fact been described in vitro, in at least one study (50). In the same paper, inhibition of enolase enzymatic activity by the internalized autoantibody, and a consequent reduction in cellular ATP, was reported, in agreement with our hypothesis. To further support the possible pathogenicity of anti-γ enolase antibody, the putative conformational epitope here proposed is relatively close to the active site cavity of the dimeric enzyme (51). Pathogenic effects of autoantibodies against surface-exposed epitopes can also be linked to activation of complement cascades and Fc receptors (52). This leads to the hypothesis that anti-γ enolase antibodies could induce damage in neuronal tissues, through the activity of the complement system and of leukocytes such as natural killer cells and macrophages (Figure 7, II). We acknowledge as limitations in our study that the analysis presented based on peptides alignments miss all cross-reactive conformational epitopes due exclusively to protein folding. Furthermore, we recognize that it is possible that anti-γ enolase antibody found in our patient may be due to cross-reactivity with antigens from other organisms. In previous studies (10, 11), heightened production of IgGs against brain proteins has been detected in patients exposed to B. burgdorferi and with persistent symptoms, but a common autoantigen was not found. That observation could support the hypothesis that anti-enolase autoimmunity could develop only in a subset of PTLDS patients, perhaps in those who fulfil diagnostic criteria for ME/CFS, who represent about 13% of cases (53). Conclusion In this study, the presence of antibodies against γ enolase was found in a patient with a complex picture of neurological symptoms and with exposure to B. burgdorferi. We hypothesize that the infection by B. burgdorferi elicits an antibody to Borrelia enolase that cross-reacts - via molecular mimicry - with human γ enolase. A possible cross-reactive conformational epitope has been found and a possible mechanism for the causal role of this antibody in the genesis of cognitive manifestations has been suggested. We acknowledge that our hypothesis needs to be validated in case studies investigating on the presence of anti-γ enolase antibodies in LD patients with persistent symptoms. The co-presence of anti-BEN antibodies should also be investigated to support the causal role of BEN for anti-enolase autoimmunity via molecular mimicry in PTLDS. Acknowledgments We are grateful to Margarita Bitetti and Giada Da Ros for their support in this study. References 6 [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] G. Trevisan and M. Cinco, Lyme disease. A general survey, Int J Dermatol 29 (1990), pp. 18. A. Mygland, U. Ljostad, V. Fingerle, et al., EFNS guidelines on the diagnosis and management of European Lyme neuroborreliosis, Eur J Neurol 17 (2010), pp. 8-16, e11-14. J.N. Aucott, L.A. Crowder and K.B. Kortte, Development of a foundation for a case definition of post-treatment Lyme disease syndrome, Int J Infect Dis 17 (2013), pp. e443449. I.o. Medicine, Critical Needs and Gaps in Understanding Prevention, Amelioration, and Resolution of Lyme and Other Tick-Borne Diseases: The Short-Term and Long-Term Outcomes: Workshop Report, The National Academies Press, Washington, DC (2011). E.E. Drouin, L. Glickstein, W.W. Kwok, G.T. Nepom and A.C. Steere, Human homologues of a Borrelia T cell epitope associated with antibiotic-refractory Lyme arthritis, Mol Immunol 45 (2008), pp. 180-189. E.E. Drouin, R.J. Seward, K. Strle, et al., A novel human autoantigen, endothelial cell growth factor, is a target of T and B cell responses in patients with Lyme disease, Arthritis Rheum 65 (2013), pp. 186-196. L.H. Sigal, S. Williams, B. Soltys and R. Gupta, H9724, a monoclonal antibody to Borrelia burgdorferi's flagellin, binds to heat shock protein 60 (HSP60) within live neuroblastoma cells: a potential role for HSP60 in peptide hormone signaling and in an autoimmune pathogenesis of the neuropathy of Lyme disease, Cell Mol Neurobiol 21 (2001), pp. 477495. J.C. Garcia-Monco, R.J. Seidman and J.L. Benach, Experimental immunization with Borrelia burgdorferi induces development of antibodies to gangliosides, Infect Immun 63 (1995), pp. 4130-4137. A. Alaedini and N. Latov, Antibodies against OspA epitopes of Borrelia burgdorferi crossreact with neural tissue, J Neuroimmunol 159 (2005), pp. 192-195. A. Chandra, G.P. Wormser, M.S. Klempner, et al., Anti-neural antibody reactivity in patients with a history of Lyme borreliosis and persistent symptoms, Brain Behav Immun 24 (2010), pp. 1018-1024. E. Jacek, B.A. Fallon, A. Chandra, M.K. Crow, G.P. Wormser and A. Alaedini, Increased IFNalpha activity and differential antibody response in patients with a history of Lyme disease and persistent cognitive deficits, J Neuroimmunol 255 (2013), pp. 85-91. V. Pancholi, Multifunctional alpha-enolase: its role in diseases, Cell Mol Life Sci 58 (2001), pp. 902-920. P. Pauluzzi, S. Bonin, M.A. Gonzalez Inchaurraga, G. Stanta and G. Trevisan, Detection of spirochaetal DNA simultaneously in skin biopsies, peripheral blood and urine from patients with erythema migrans, Acta dermato-venereologica 84 (2004), pp. 106-110. A.J. Church, F. Cardoso, R.C. Dale, A.J. Lees, E.J. Thompson and G. Giovannoni, Antibasal ganglia antibodies in acute and persistent Sydenham's chorea, Neurology 59 (2002), pp. 227-231. R.C. Dale, P.M. Candler, A.J. Church, R. Wait, J.M. Pocock and G. Giovannoni, Neuronal surface glycolytic enzymes are autoantigen targets in post-streptococcal autoimmune CNS disease, J Neuroimmunol 172 (2006), pp. 187-197. S.B. Needleman and C.D. Wunsch, A general method applicable to the search for similarities in the amino acid sequence of two proteins, J Mol Biol 48 (1970), pp. 443-453. S.F. Altschul, Amino acid substitution matrices from an information theoretic perspective, J Mol Biol 219 (1991), pp. 555-565. S. Karlin and S.F. Altschul, Methods for assessing the statistical significance of molecular sequence features by using general scoring schemes, Proc Natl Acad Sci U S A 87 (1990), pp. 2264-2268. 7 [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32] [33] [34] [35] [36] [37] [38] S.M. McAleese, B. Dunbar, J.E. Fothergill, L.J. Hinks and I.N. Day, Complete amino acid sequence of the neurone-specific gamma isozyme of enolase (NSE) from human brain and comparison with the non-neuronal alpha form (NNE), Eur J Biochem 178 (1988), pp. 413417. P.G. Leonard, N. Satani, D. Maxwell, et al., SF2312 is a natural phosphonate inhibitor of enolase, Nat Chem Biol 12 (2016), pp. 1053-1058. J. Echave, S.J. Spielman and C.O. Wilke, Causes of evolutionary rate variation among protein sites, Nat Rev Genet 17 (2016), pp. 109-121. M. Vollmar, Krysztofinska, E., Chaikuad, A., Krojer, T., Cocking, R., Von Delft, F., Bountra, C., Arrowsmith, C.H., Weigelt, J., Edwards, A., Yue, W.W., Oppermann, U., Crystal Structure of Human Beta Enolase Enob, RCSB PDB (To Be Published). M. Institute of, Beyond Myalgic Encephalomyelitis/Chronic Fatigue Syndrome: Redefining an Illness %@ 978-0-309-31689-7, The National Academies Press, Washington, DC (2015). G. Trevisan, P. Cattonar, C. Nobile, V. Perkan and G. Stinco, Dermatological manifestations of Lyme borrelosis, Acta Dermatovenerologica Alpina, Pannonica et Adriatica 5 (1996), pp. 101-107. B.A. Fallon, R.B. Lipkin, K.M. Corbera, et al., Regional cerebral blood flow and metabolic rate in persistent Lyme encephalopathy, Arch Gen Psychiatry 66 (2009), pp. 554-563. A. Newberg, A. Hassan and A. Alavi, Cerebral metabolic changes associated with Lyme disease, Nucl Med Commun 23 (2002), pp. 773-777. M. Piast, I. Kustrzeba-Wojcicka, M. Matusiewicz and T. Banas, Molecular evolution of enolase, Acta Biochim Pol 52 (2005), pp. 507-513. A.M. Floden, J.A. Watt and C.A. Brissette, Borrelia burgdorferi enolase is a surfaceexposed plasminogen binding protein, PLoS One 6 (2011), p. e27502. S.V. Nogueira, A.A. Smith, J.H. Qin and U. Pal, A surface enolase participates in Borrelia burgdorferi-plasminogen interaction and contributes to pathogen survival within feeding ticks, Infect Immun 80 (2012), pp. 82-90. A. Toledo, J.L. Coleman, C.J. Kuhlow, J.T. Crowley and J.L. Benach, The enolase of Borrelia burgdorferi is a plasminogen receptor released in outer membrane vesicles, Infect Immun 80 (2012), pp. 359-368. A.G. Barbour, A. Jasinskas, M.A. Kayala, et al., A genome-wide proteome array reveals a limited set of immunogens in natural infections of humans and white-footed mice with Borrelia burgdorferi, Infect Immun 76 (2008), pp. 3374-3389. J.M. Thornton, M.S. Edwards, W.R. Taylor and D.J. Barlow, Location of 'continuous' antigenic determinants in the protruding regions of proteins, EMBO J 5 (1986), pp. 409-413. P. Haste Andersen, M. Nielsen and O. Lund, Prediction of residues in discontinuous B-cell epitopes using protein 3D structures, Protein Sci 15 (2006), pp. 2558-2567. R.C. Dale, A.J. Church, R.A. Surtees, et al., Encephalitis lethargica syndrome: 20 new cases and evidence of basal ganglia autoimmunity, Brain 127 (2004), pp. 21-33. T.R. Nicholson, S. Ferdinando, R.B. Krishnaiah, et al., Prevalence of anti-basal ganglia antibodies in adult obsessive-compulsive disorder: cross-sectional study, Br J Psychiatry 200 (2012), pp. 381-386. A.J. Church, R.C. Dale and G. Giovannoni, Anti-basal ganglia antibodies: a possible diagnostic utility in idiopathic movement disorders?, Arch Dis Child 89 (2004), pp. 611614. S. Kalra, P.D. Burbelo, A. Bayat, et al., No Evidence of Antibodies against GAD65 and Other Specific Antigens in Children with Autism, BBA Clin 4 (2015), pp. 81-84. H. Nicolini, Y. Lopez, A.D. Genis-Mendoza, et al., Detection of anti-streptococcal, antienolase, and anti-neural antibodies in subjects with early-onset psychiatric disorders, Actas Esp Psiquiatr 43 (2015), pp. 35-41. 8 [39] [40] [41] [42] [43] [44] [45] [46] [47] [48] [49] [50] [51] [52] [53] R.C. Dale, I. Heyman, G. Giovannoni and A.W. Church, Incidence of anti-brain antibodies in children with obsessive-compulsive disorder, Br J Psychiatry 187 (2005), pp. 314-319. F. Forooghian, G. Adamus, M. Sproule, C. Westall and P. O'Connor, Enolase autoantibodies and retinal function in multiple sclerosis patients, Graefes Arch Clin Exp Ophthalmol 245 (2007), pp. 1077-1084. A. Fujii, M. Yoneda, T. Ito, et al., Autoantibodies against the amino terminal of alphaenolase are a useful diagnostic marker of Hashimoto's encephalopathy, J Neuroimmunol 162 (2005), pp. 130-136. H. Ochi, I. Horiuchi, N. Araki, et al., Proteomic analysis of human brain identifies alphaenolase as a novel autoantigen in Hashimoto's encephalopathy, FEBS Lett 528 (2002), pp. 197-202. M. Yoneda, A. Fujii, A. Ito, H. Yokoyama, H. Nakagawa and M. Kuriyama, High prevalence of serum autoantibodies against the amino terminal of alpha-enolase in Hashimoto's encephalopathy, J Neuroimmunol 185 (2007), pp. 195-200. P.A. Fontan, V. Pancholi, M.M. Nociari and V.A. Fischetti, Antibodies to streptococcal surface enolase react with human alpha-enolase: implications in poststreptococcal sequelae, J Infect Dis 182 (2000), pp. 1712-1721. K.H. Lee, H.S. Chung, H.S. Kim, et al., Human alpha-enolase from endothelial cells as a target antigen of anti-endothelial cell antibody in Behcet's disease, Arthritis Rheum 48 (2003), pp. 2025-2035. B. Terrier, N. Degand, P. Guilpain, A. Servettaz, L. Guillevin and L. Mouthon, Alphaenolase: a target of antibodies in infectious and autoimmune diseases, Autoimmun Rev 6 (2007), pp. 176-182. G. Adamus, D. Amundson, G.M. Seigel and M. Machnicki, Anti-enolase-alpha autoantibodies in cancer-associated retinopathy: epitope mapping and cytotoxicity on retinal cells, J Autoimmun 11 (1998), pp. 671-677. M. Walter, H. Berg, F.A. Leidenberger, K.W. Schweppe and W. Northemann, Autoreactive epitopes within the human alpha-enolase and their recognition by sera from patients with endometriosis, J Autoimmun 8 (1995), pp. 931-945. A. Hafner, N. Obermajer and J. Kos, gamma-1-syntrophin mediates trafficking of gammaenolase towards the plasma membrane and enhances its neurotrophic activity, Neurosignals 18 (2010), pp. 246-258. A. Magrys, T. Anekonda, G. Ren and G. Adamus, The role of anti-alpha-enolase autoantibodies in pathogenicity of autoimmune-mediated retinopathy, J Clin Immunol 27 (2007), pp. 181-192. L. Lebioda and B. Stec, Mapping of isozymic differences in enolase, Int J Biol Macromol 13 (1991), pp. 97-100. B. Diamond, P.T. Huerta, P. Mina-Osorio, C. Kowal and B.T. Volpe, Losing your nerves? Maybe it's the antibodies, Nat Rev Immunol 9 (2009), pp. 449-456. D.I. Bujak, A. Weinstein and R.L. Dornbush, Clinical and neurocognitive features of the post Lyme syndrome, J Rheumatol 23 (1996), pp. 1392-1397. 9 LEGEND TO THE FIGURES Figure 1. Identity and similar positions shared by Human and Borrelia Enolases. Henα1: human α enolase 1 (P06733-1); Henβ1: human β enolase 1 (P13929-1); Henγ1: human γ enolase 1 (P091041); Baen: Borrelia afzelii (strain PKo) enolase (Q0SNH5-1); Bben: Borrelia burgdorferi ss (strain ZS7) enolase (B7J1R2-1); Bgen: Borrelia garinii (strain PBr) enolase (B7XT07). Identities and similar positions have been calculated by Clustal Omega, with default settings. Figure 2. Peptides of human γ enolase not shared with human α enolase. In the first column 7 amino acids peptides of γ enolase which do not cross-react with α enolase (score below 50) are reported. For each epitope of the first column, the best local alignment with human α enolase 1 (P06733-1), human β enolase 1 (P13929-1), human γ enolase 1 (P09104-1), Borrelia afzelii (strain PKo) enolase (Q0SNH5-1), Borrelia burgdorferi ss (strain ZS7) enolase (B7J1R2-1), Borrelia garinii (strain PBr) enolase (B7XT07) has been searched. Those local alignments with crossreactivity are green colored, while those which are not cross-reactive are in orange. Borderline reactivity is in white. For each alignment, p value have been included. Figure 3. Study of cross-reactivity for epitope 49-55 of γ enolase. Figure 4. Cross-reactivity among enolases. Cross-reactions between Borrelia enolases and β enolase are marked in red, cross-reactions between Borrelia enolases and γ enolase are marked in blue. Cross-reactions between γ and β enolase are marked in black. Figure 5. The conformational epitope shared by human γ enolase and Borrelia enolase, but not by α enolase. On the left: peptides 49-55, 59-65, 60-66 and 188-194 of γ enolase form a conformational epitope (in yellow) when two molecules of γ enolase are bound. On the right: peptides 52-58, 6268, 63-69 and 186-192 of Borrelia burgdorferi enolase form a conformational epitope (in yellow) when the two subunits form the homodimeric enzyme. Figure 6. The conformational epitope shared by human enolases and Borrelia enolase. Left: peptides 37-43, 47-53, and 48-54 of γ enolase form a conformational epitope (in yellow). Center: peptides 37-43, 47-53, and 48-54 of β enolase form a conformational epitope (in yellow). Right: peptides 40-46, 50-56, and 51-57 of B. burgdorferi enolase form a conformational epitope (in yellow). Peptides 37-43, 47-53, and 48-54 identify a similar conformation epitope on α enolase (not shown). Figure 7. The proposed pathogenic role of anti-γ enolase antibody in PTLDS. I: Enolase from Bb binds its specific B cell receptor (BCR), then the antigen is internalized, cut into a T cell epitope and presented by major histocompatibility complex II (MHC II) to a Th cell with a specific T cell receptor (TCR). This leads to B cell activation and consequent production of anti-enolase antibodies. II: The interaction between anti-enolase Ab and surface exposed human enolase activate complement cascade and leukocytes (e.g. macrophages), through the Fc region, with possible induction of apoptosis. III: Internalization of enolase-Ab complex. IV: Anti-enolase Ab interferes with the ninth step of glycolysis. 10 11 12 13 14 15 16 17 Table 1. Infectious diseases workup in this patient (peripheral blood, unless otherwise indicated). Category Herpesviruses Other viruses Tick borne infections Pathogen Test Herpes simplex virus 1, 2 IgM, IgG Epstein-Barr virus IgM, IgG, PCR Cytomegalovirus IgM, IgG, PCR HHV-6 IgM, IgG, PCR Positive IgG HHV-7 PCR Negative Parvovirus B19 IgM, IgG, PCR Enterovirus IgM, IgG West Nile virus IgM, IgG Hepatitis C Abs Hepatitis B Anaplasma phagocytophilum Babesia Abs PCR TBE virus IgM, IgG Rickettsia spp. Aspergillus fumigatus Weil-Felix, PCR IgM, IgG PCR Ab by ELISA Bartonella henselae IgM, IgG, PCR Brucella spp. IgM, IgG Campylobacter jejuni IGA, IgG Candida albicans Mannan antigen Chlamydia pneumoniae IgM, IgA, IgG Chlamydia trachomatis IgM, IgA, IgG Coxiella burnetii IgM, IgG Mycoplasma spp IgM, IgG, PCR Streptococcus pyogenes ASO, throat swab culture Toxoplasma gondii IgM, IgG Treponema pallidum T.P.H.A. Helicobacter pylori stool sample culture Borrelia burgdorferi sl Other pathogens Result Negative Negative Positive (immunization) IgM, IgG 18 Negative Positive IgG (p41, p19) Positive Negative Positive Table 2. Serum autoantibodies workup in this patient. Category Autoantibody Test Results Anti-IgG Fc (Rheumatoid factor, RF) Anti-cyclic citrullinated peptide (CCP) General autoimmunity Autoimmune inflammatory myopathies and scleroderma panel Anti-nuclear (ANA) IFI on Hep2 Anti-double stranded DNA (ds-DNA) EIA Anti-extractable nuclear antigen (ENA) EIA Anti-smooth muscle antigen (ASMA) IFI Anti-neutrophil cytoplasmic (ANCA) IFI Anti-myeloperoxidase (MPO) EIA Anti-proteinase 3 (PR3) EIA Anti-tissue transglutaminase (t-TG) IgA-EIA Anti-endomysial (EMA) IgA-IFI Anti-thyroglobulin CLIA Anti-thyroid peroxidase (TPO) CLIA Anti-cardiolipin ELISA Anti-liver-kidney microsomial (LKM) IFI Anti-Mi-2, Ku, PM-Scl100, PM-Scl75, Jo-1, SRP, PL-7, PL-12, EJ, OJ, Ro-52, Scl-70, CENP A, CENP B, RP11, RP155, Fibrillarin, NOR90, Th/To, PDGFR EUROLINE Anti-Ach receptor RIA Negative Negative Anti-ganglionic Ach receptor (α3-α7) Neuronal surface-exposed antigens Neuronal intracellular antigens Other neuronal antigens Anti-NMDA receptor transfection Anti-AMPA 1,2 receptor transfection Anti-GABA B receptor transfection Anti-Myelin associated glycoprotein (MAG) IgM-ELISA Anti-GD1a, b; GM1, 2, 3; GQ1b, GT1b transfection Anti-VGKC (LGI 1, CASPR 2) transfection Anti-amphiphysin Rec. protein Anti-Hu, Yo, Ri, CV2, Ma2/Ta, GAD Rec. protein Negative Anti-basal ganglia H. brain IB Positive Anti-α enolase Rec. protein IB Negative Anti-γ enolase Rec. protein IB Positive Anti-pyruvate kinase Rec. protein IB Negative 19 Negative 20
1/--страниц