Clin Exp Immunol 2000; 120:420±423
Inhibition of aromatic l-amino acid decarboxylase activity by human autoantibodies È MPE³, A. AAKVAAG*, T. RYGH*, T. FLATMARK² & E. S. HUSEBYE, A. S. BéE, F. RORSMAN³, O. KA J. HAAVIK² Division of Endocrinology, Institute of Medicine, *Department of Biochemical Endocrinology and ²Department of Biochemistry and Molecular Biology, University of Bergen, Bergen, Norway, and ³Department of Medical Sciences, Uppsala University, Uppsala, Sweden
(Accepted for publication 2 March 2000)
SUMMARY A full-length rat cDNA clone encoding aromatic l-amino acid decarboxylase (AADC) (E.C. 4.1.1.28) was used for in vitro transcription and translation. The enzyme had catalytic activity (0´2 pmol serotonin/m l lysate per min), and was stimulated 2´5-fold by the addition of excess pyridoxal phosphate. On size exclusion chromatography, AADC eluted as a single activity peak with an apparent mol. wt of 93 kD. This activity peak was immunoprecipitated by sera from patients with autoimmune polyendocrine syndrome type I (APS I) containing autoantibodies against AADC. Serum and purified IgG from these patients inhibited the enzyme activity (non-competitively) by 10±80%, while sera from APS I patients without autoantibodies and controls did not. This finding confirms and extends previous observations that APS I patients have inhibitory antibodies against key enzymes involved in neurotransmitter biosynthesis. Keywords aromatic l-amino acid decarboxylase serotonin dopamine autoantigen
autoimmune polyendocrine syndrome type I
INTRODUCTION
PATIENTS AND METHODS
Aromatic l-amino acid decarboxylase (AADC) decarboxylates aromatic amino acids such as 5-hydroxytryptophan and 3,4dihydroxyphenylalanine in a pyridoxal phosphate-dependent manner, as part of the biosynthetic pathway of catecholamine and indolamine neurotransmitters, respectively [1]. Besides being present in nervous tissues, AADC is found in large amounts in neuroendocrine cells, liver and kidney [2]. We have previously cloned AADC from a rat insulinoma cDNA library by immunoscreening with sera from patients with autoimmune polyendocrine syndrome type I (APS I) [3], and shown that the presence of autoantibodies against AADC is correlated with the presence of autoimmune hepatitis and vitiligo [4, 5]. The aim of the present investigation was to study the catalytic properties of AADC synthesized by a coupled in vitro transcription and translation system, and to examine whether sera from patients with autoantibodies against AADC also affect the enzyme activity.
Patients Sera from 12 Norwegian and nine Swedish patients with APS I were investigated. Sera from 11 healthy Norwegian blood donors were used as controls. All APS I patients fulfilled the clinical criteria for the diagnosis, having at least two of the three following components: hypoparathyroidism, adrenocortical insufficiency and chronic mucocutaneous candidiasis [6,7]. The clinical characteristics of the patients are presented in Table 1. None of the patients used medication containing inhibitors of AADC activity.
Correspondence: Eystein Husebye MD, PhD, Division of Endocrinology, Institute of Medicine, Haukeland University Hospital, N-5021 Bergen, Norway. E-mail:
[email protected]
420
Synthesis of AADC by in vitro transcription and translation The full-length rat cDNA clone encoding AADC [3] was used for in vitro transcription and translation (ITT) using the TNT T3 coupled reticulocyte lysate system (Promega, Madison, WI). Typically, about 5% of the radioactivity was regularly incorporated into the protein. 35S-methionine-labelled products were used for the size exclusion chromatography experiments and measurement of autoantibodies against AADC. Assay of enzyme activity was performed with unlabelled AADC. Size exclusion chromatography Samples of the ITT product of AADC were analysed by size q 2000 Blackwell Science
Inhibition of aromatic AADC activity by human autoantibodies Table 1. Clinical features of Norwegian (nos 1±12) and Swedish (nos 13± 21) patients with autoimmune polyendocrine syndrome type I (APS I)
1 1 1 1 1 1 1
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
1 1
1 1
1 1
1
1
1
1 1
1 1 1 1
1 1 1
7
V0 (a)
14
18
22 Vt
6 5 4 3 2
1
1 0
1 1
10
A280
1 1 1 1 1 1 1 1 1
8
1
1
1
1 1 1 1 1 1 1
AI, Adrenal insufficiency; HP, hypoparathyroidism; Can, chronic mucocutaneous candidiasis; DM, diabetes mellitus; CAH, chronic autoimmune hepatitis; Alo, alopecia; Vit, vitiligo; Ena, enamel hypoplasia; Mal, malabsorbtion; Gon, gonadal failure.
exclusion chromatography using a Pharmacia (Stockholm, Sweden) Superdex 200 HR 30/10 column and a BioRad (Bedford, MA) BioLogic HR chromatography system. The mobile phase contained 0´1 m NaHEPES, 0´25 mm EDTA and 0´2 m NaCl, pH 7´50, and was pumped at a flow rate of 0´50 ml/min. Blue dextran and acetone were used to determine the void volume (V0) and the total liquid volume of the column (Vt), respectively. The column was calibrated using proteins with mol. wt taken from the literature as compiled by Uversky and Corbett & Roche [8, 9]. The following proteins were used: thyroglobulin, apoferritin, catalase, phenylalanine hydroxylase, yeast alcohol dehydrogenase, bovine serum albumin, haemoglobin, ovalbumin, RNase 1 and cytochrome c. Measurement of AADC activity AADC was incubated in a total volume of 50 m l at 308C for 40 min in 160 mm HEPES pH 8´0 containing 0´1 mm pyridoxal phosphate, and 0´5 mm 5-hydroxytryptophan as the substrate. The reaction was started by adding 2´5±5 m l of the ITT product. The reaction was stopped by an equal volume of ice-cold ethanol containing glacial acetic acid to pH 4´1. After precipitation of protein for 30 min at 48C, centrifugation at 10 500 g for 5 min, samples of the supernatant were analysed by high performance liquid chromatography (HPLC) on a Whatman Partisphere SCX (4´6 110 mm) column and fluorescence detector as previously described [10]. The mobile phase contained 50 mm Na-acetate buffer pH 4´2. Since the total ITT product was used as the enzyme source it was not possible to express the activity in moles per mg of protein, but only as moles serotonin/m l lysate per min, or as
(b) Immunoprecipitation (relative values)
1957 1958 1948 1990 1986 1981 1990 1967 1962 1976 1980 1986 1975 1986 1977 1971 1973 1978 1979 1985 1984
35S
F F M F F F M F M F M M F M F M M M F F M
AADC activity (relative values )
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
Elution volume (ml)
(ct/min 10 –5 )
No. Sex Year of birth AI HP Can Gon DM CAH Alo Vit Ena Mal
421
0·5
0
0
20
40
60
Fraction number Fig. 1. Size exclusion chromatography of aromatic l-amino acid decarboxylase (AADC) synthesized by in vitro transcription and translation. (a) 35S-methionine (W); A280 (X). (b) AADC activity (A); 35Slabelled AADC immunoprecipitated by a serum containing autoantibodies against AADC (patient 10) (P). V0, Void volume; Vt, total volume.
relative specific activity (inhibition studies). The reticulocyte lysate itself did not contain measurable AADC activity. Assay of antibodies against AADC by immunoprecipitation Antibodies against AADC were assayed by a method based on the in vitro transcribed and translated 35S-methionine-labelled AADC [4]. After an overnight incubation with sera, immune complexes were recovered using protein A Sepharose (Pharmacia) and microtitre plates with filter bottoms (MABV N12; Millipore, Hercules, CA) as described by Husebye et al. [4]. Other methods IgG was purified using protein A-Sepharose (Pharmacia) according to Ey et al. [11]. Protein concentration was measured according to Bradford [12]. RESULTS Characterization of AADC synthesized by in vitro transcription and translation AADC synthesized by ITT was catalytically active with an enzyme activity of about 0´2 pmol serotonin/m l lysate per min and a pH optimum of pH 8´0. As the enzyme activity was stimulated about 2´5-fold by the addition of an excess amount of pyridoxal phosphate, this coenzyme was included in routine measurements. On size exclusion chromatography of AADC synthesized in the presence of 35S-methionine, two partially separated peaks of
q 2000 Blackwell Science Ltd, Clinical and Experimental Immunology, 120:420±423
422
E. S. Husebye et al: 140
25
20 1/v (arbitrary units) –1
Enzyme activity (%)
120 100 80 60 40 20 0 APS I AADC Ab+
APS I AADC Ab–
15
10
5
0
Blood donors
–15
–10
–5
0
5
1/[5-hydroxytryptophan] Fig. 2. Inhibition of aromatic l-amino acid decarboxylase (AADC) activity by sera from patients with autoimmune polyendocrine syndrome type I (APS I) with (AADC Ab1) and without (AADC Ab2) autoantibodies against AADC.
protein-bound radioactivity appeared, the larger corresponding to an apparent mol. wt of about 93 kD and the smaller with a mol. wt of about 40 kD. All enzyme activity appeared in the first peak which corresponded perfectly with the immunological reactivity (Fig. 1). The effect of patient sera on enzymatic activity Sera from 21 patients with APS I were tested for their ability to inhibit AADC activity. Autoantibodies against AADC were detected in sera from 13 of 21 patients. Sera from these patients inhibited the enzyme activity at a 1:50 dilution from 11% to 79%, with a mean of 55 ^ 23% (mean ^ s.d.). Sera from eight of the 21 APS I patients without autoantibodies against AADC, as well as from 11 control sera, inhibited AADC activity by 10 ^ 5% and 6 ^ 18%, respectively (Fig. 2). The inhibitory response was reproducible within experiments and between experiments using different batches of AADC. The inhibitory response of serum from patient 10 in two different experiments was 56 ^ 3´9% (n 4) and 53 ^ 5´8% (n 3), respectively. The corresponding results for patient 16 were 53 ^ 6´5% (n 3) and 41 ^ 8´1% (n 3), respectively. The inhibition was highly potent with some of the sera. In one case, a 1:10 240 dilution was necessary to maintain enzyme activity at control levels (^ 5%), while a dilution of 1:327 680 was necessary to reduce the autoantibody index to the range observed in controls. The inhibitory activity could also be demonstrated by using a purified IgG fraction from the patients. The inhibitory activity of 1 m g/m l IgG prepared from the serum of patient 4 was 36 ^ 5´1%, while the corresponding inhibition of serum was 52 ^ 3´8% (n 3) and 37 ^ 3´8% (n 3) at 1:10 and 1:50 dilutions of serum, respectively. The mode of inhibition was studied by testing the inhibitory activity of different sera at varying substrate concentrations. Figure 3 shows an experiment with serum from patient 4 revealing non-competitive inhibition. Sera from other patients revealed the same type of inhibition. The degree of inhibition (in percent) of four APS I sera with autoantibodies against AADC decreased when pyridoxal phosphate was omitted from the reaction mixture (data not shown).
10
15
20
(mM)–1
Fig. 3. Inhibition of aromatic l-amino acid decarboxylase (AADC) activity by serum from a patients with autoimmune polyendocrine syndrome type I (APS I) (patient no. 4). The serum was used at 1:50 dilution and incubations were performed as described in Patients and Methods using varying concentrations of 5-hydroxytryptophan. Serum from APS I patient (W), control without serum (X). The lines were calculated by linear regression analysis.
DISCUSSION AADC decarboxylates the aromatic amino acids 5-hydroxytryptophan and 3,4-dihydroxyphenylalanine in a pyridoxal phosphatedependent manner, as part of the biosynthetic pathway of catecholamine and indolamine neurotransmitters [1]. It has an apparent mol. wt on SDS±PAGE of 51 kD and calculated subunit molecular mass of 54 kD [3], and the dimer has been reported to be the catalytically active form of the enzyme [13±15]. We here show for the first time that the AADC synthesized by a coupled in vitro transcription and translation system is catalytically active using 5-hydroxytryptophan as the substrate. The majority of the protein-bound radioactivity eluted with an apparent mol. wt of about 93 kD, while a smaller fraction eluted at a mol. wt of about 40 kD. Enzyme activity as well as immunological activity (immunoprecipitation) was confined to the larger peak (Fig. 1), which appears to represent the dimer. Although the mol. wt of 93 kD is lower than the theoretical molecular mass of the dimer (108 kD), the difference is within the range of variation reported in the literature [13±15]. The peak with an apparent mol. wt of 40 kD probably represents a non-specific product of the in vitro transcription-translation reaction [16], since it had no AADC enzyme activity, and was not immunoprecipitated by specific autoantibodies against AADC (Fig. 1). Autoantibodies found in sera and purified IgG fractions from APS I patients with reactivity against AADC were found to inhibit the enzyme activity to a varying degree. None or only a slight inhibition was observed with sera from patients without autoantibodies (Fig. 2 and Results). Amino acids as competitive substrates and possibly other substances in the sera could explain this slight inhibition of AADC activity. The variation in the potency of inhibition in autoantibody-positive sera may reflect binding to various epitopes on the enzyme, some of which are more important for enzyme activity than others. The inhibition was found to be non-competitive (Fig. 3) and could not be reversed by the addition of an excess of pyridoxal phosphate (see
q 2000 Blackwell Science Ltd, Clinical and Experimental Immunology, 120:420±423
Inhibition of aromatic AADC activity by human autoantibodies Results). This finding indicates that the autoantibodies inhibit activity by binding to other sites than that for binding of substrate and cofactor. The ability to inhibit AADC activity was retained even at high dilutions of serum. In one case dilution to 1:10 280 was required in order to avoid inhibition. For this patient, still higher dilutions (1:327 680) were necessary to avoid immunoprecipitation of AADC, which probably reflects that the immunoprecipitation assay is more sensitive and measures all autoantibodies, whether they modulate enzyme activity or not. Organ-specific autoimmune diseases are characterized by the presence of autoantibodies directed against enzymes of key importance for the organs involved. In autoimmune adrenocortical insufficiency (Addison's disease) the patients display autoantibodies against 21-hydroxylase, one of the enzymes participating in cortisol and aldosterone biosynthesis, which is only expressed in the adrenal cortex in humans [17]. APS I patients who often have both adrenocortical insufficiency and gonadal failure also display autoantibodies against two other steroidogenic enzymes, i.e. side-chain cleavage enzyme and 17a -hydroxylase [18, 19]. These two enzymes are expressed both in the adrenal cortex and in the gonads. Recently, we have shown that APS I patients have autoantibodies against glutamic acid decarboxylase [20], AADC [3,4], tryptophan hydroxylase [7] and tyrosine hydroxylase [21], enzymes that are involved in the biosynthesis of GABA, serotonin and catecholamines. The activities of several of these enzymes are also inhibited by these human autoantibodies [7, 19], but there is no convincing evidence in the literature that the autoantibodies themselves inhibit enzyme activity in vivo [22]. In conclusion, we have shown that AADC synthesized by in vitro transcription and translation is catalytically active. Autoantibodies against AADC present in sera from patients with APS I are able to potently inhibit the activity of AADC in a noncompetitive manner. Why patients with autoimmune diseases produce autoantibodies against AADC and other neurotransmitter enzymes remains elusive. One hypothesis is that T cells reactive against intracellular organ-specific enzymes are more likely to have escaped negative selection in the thymus [23], another is that the conformation change in the enzyme induced by substrate binding may create new epitopes against which the immune system can react. ACKNOWLEDGMENTS This study was supported by grants from the Norwegian Diabetes Association, The Astri and Edvard Risùen Fund, The Aagot Giertsen Fund, Novartis, Bristol-Myers-Squibb, The Research Council of Norway and The Swedish Medical Research Council. Ms Wenke Trovik is thanked for technical assistance.
REFERENCES 1 Zhu MY, Juorio A. Aromatic l-amino acid decarboxylase: biological characterization and functional role. Gen Pharmacol 1995; 26:681± 96. 2 Rahman MK, Nagatsu T, Kato T. Aromatic l-amino acid decarboxylase activity in central and peripheral tissues and serum of rats with LDOPA and L-5-hydroxytryptophan as substrates. Biochem Pharmacol 1981; 30:645±9. 3 Rorsman F, Husebye ES, Winqvist O, BjoÈrk E, Karlsson FA, KaÈmpe O. Aromatic l-amino acid decarboxylase, a pyridoxal phosphatedependent enzyme, is a novel b-cell autoantigen. Proc Natl Acad Sci USA 1995; 92:8626±9.
423
4 Husebye ES, Gebre-Medhin G, Tuomi T, Perheentupa J, LandinOlsson M, Gustafsson J, Rorsman F, KaÈmpe O. Autoantibodies against aromatic l-amino acid decarboxylase in autoimmune polyendocrine syndrome type I. J Clin Endocrinol Metab 1997; 82:147±50. 5 Gebre-Medhin G, Husebye ES, Gustafsson J, Winqvist O, Goksùyr A, Rorsman F, KaÈmpe O. Cytochrome P450IA2 and aromatic l-amino acid decarboxylase are hepatic autoantigens in autoimmune polyendocrine syndrome type I. FEBS Letters 1997; 412:439±45. 6 Ahonen P, MyllaÈrniemi S, SipilaÈ I, Perheentupa J. Clinical variations or autoimmune polyendocrinopathy-candidiasis-ectodermal dystrophy (APECED) in a series of 68 patients. N Eng J Med 1990; 322:1829±36. 7 Ekwall O, Hedstrand H, Grimelius L et al. Identification of tryptophan hydroxylase as an intestinal autoantigen. Lancet 1998; 25:279±83. 8 Uversky VN. Use of fast protein size-exclusion liquid chromatography to study the unfolding of proteins which denature through the molten globule. Biochemistry 1993; 32:13288±98. 9 Corbett RJ, Roche RS. Use of high-speed size-exclusion chromatography for the study of protein folding and stability. Biochemistry 1984; 23:1888±94. 10 Flatmark T, Jacobsen SW, Haavik J. Fluorometric detection of tryptophan, 5-hydroxytryptophan, and 5-hydroxytryptamine (serotonin) in high-performance liquid chromatography. Anal Biochem 1980; 107:71±74. 11 Ey PL, Prowse SJ, Jenkin CR. Isolation of pure IgG1, IgG2a and IgG2b immunoglobulins from mouse serum using protein A-sepharose. Immunochemistry 1978; 15:429±36. 12 Bradford MM. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 1976; 72:248±54. 13 Dominici P, Tancini B, Barra D, Voltattorni CB. Purification and characterization of rat-liver 3,4-dihydroxyphenylalanine decarboxylase. Eur J Biochem 1987; 169:209±13. 14 Shirota K, Fujisawa H. Purification and characterization of aromatic lamino acid decarboxylase from rat kidney and monoclonal antibody to the enzyme. J Neurochem 1988; 51:426±34. 15 Mappouras DG, Stiakakis J, Fragoulis EG. Purification and characterization of l-dopa decarboxylase from human kidney. Mol Cell Biochem 1990; 94:147±56. 16 Jackson RJ. Potassium salts influence the fidelity of mRNA translation initiation in rabbit reticulocyte lysates: unique features of encephalomyocarditis virus RNA translation. Biochim Biophys Acta 1991; 1088:345±58. 17 Winqvist O, Karlsson FA, KaÈmpe O. 21-hydroxylase, a major autoantigen in idiopathic Addison's disease. Lancet 1992; 339:1559± 62. 18 Krohn K, Uibo R, Aavik E, Peterson P, Savilahti K. Identification by molecular cloning of an autoantigen associated with Addison's disease as steroid 17 alpha-hydroxylase. Lancet 1992; 339:770±3. 19 Winqvist O, Gustafsson J, Rorsman F, Karlsson FA, KaÈmpe O. Two different cytochrome P450 enzymes are the adrenal antigens in autoimmune polyendocrine syndrome type I and Addison's disease. J Clin Invest 1993; 92:2377±85. 20 BjoÈrk E, Velloso LA, KaÈmpe O, Karlsson FA. GAD autoantibodies in IDDM, stiff-man syndrome, and autoimmune polyendocrine syndrome type I recognize different epitopes. Diabetes 1994; 43:161±5. 21 Hedstrand HH, Ekwall O, Haavik J et al. Identification of tyrosine hydroxylase as an autoantigen in autoimmune polyendocrine syndrome type I. Biochem Biophys Res Commun 2000; 267:456±61. 22 Boscaro M, Betterle C, Volpato M, Fallo F, Furmaniak J, Rees Smith B, Sonino N. Hormonal responses during various phases of autoimmune adrenal failure: no evidence for 21-hydroxylase enzyme activity inhibition in vivo. J Clin Endocrinol Metabol 1996; 81:2801± 4. 23 Winqvist O, SoÈderbergh A, KaÈmpe O. The autoimmune basis of adrenocortical destruction in Addison's disease. Mol Med Today 1996; 2:282±9.
q 2000 Blackwell Science Ltd, Clinical and Experimental Immunology, 120:420±423
