Primary biliary cirrhosis following lactobacillus vaccination for recurrent vaginitis

Journal of Hepatology 49 (2008) 466–473 www.elsevier.com/locate/jhep

Case report

Primary biliary cirrhosis following lactobacillus vaccination for recurrent vaginitisq Dimitrios Bogdanos1, , Thomas Pusl2, , Christian Rust2, Diego Vergani1, Ulrich Beuers2,3,* 1

Institute of Liver Studies, King’s College, London School of Medicine at King’s College Hospital, London, United Kingdom 2 Department of Medicine II-Grosshadern, University of Munich, Germany 3 Department of Gastroenterology and Hepatology, AMC, University of Amsterdam, G4-213, Meibergdreef 9, NL-1105 AZ Amsterdam, The Netherlands

Background/Aims: Antimitochondrial antibodies directed against the E2 subunit of the pyruvate dehydrogenase complex, PDC-E2, and other mitochondrial 2-oxoacid dehydrogenases (AMA-M2) are the hallmark for diagnosis of primary biliary cirrhosis (PBC). AMA-M2 formation as an early step in the pathogenesis of PBC has recently been assumed to be triggered by bacterial mimics of the E2 subunit and certain reactant xenobiotics. We report a case of symptomatic PBC diagnosed after sequential immunization with a lactobacillus vaccine for recurrent vaginitis over years. Methods: Serum AMA-M2 specificity of the patient was evaluated by indirect immunofluorescence, immunoblotting and ELISA. Serum antibody responses against pyruvate dehydrogenase complex-E2 subunit (PDC-E2212–226), the major PBCspecific mitochondrial autoepitope, and microbial mimics revealed cross-reactivity with beta-galactosidase of Lactobacillus delbrueckii (LACDE BGAL266–280) which shows a high local homology with that of Lactobacillus species administered via the vaccine. The relative affinity of antibody reactivity to LACDE BGAL266–280 was significantly higher than that against human PDC-E2212–226. Conclusions: We conclude that lactobacillus vaccination therapy may be another culprit for the development of PBC in genetically susceptible women. Ó 2008 Published by Elsevier B.V. on behalf of the European Association for the Study of the Liver. Keywords: PBC; Molecular mimicry; Lactobacillus; AMA; UDCA; Budesonide

1. Introduction Primary biliary cirrhosis (PBC) is a chronic cholestatic liver disease characterized by destruction of small intrahepatic bile ducts and subsequent development of liver fibrosis and cirrhosis predominantly in middle-aged Received 23 September 2007; received in revised form 6 April 2008; accepted 12 May 2008; available online 25 June 2008 Associate Editor: M. Trauner q The authors declare that they do not have anything to disclose regarding funding from industries or conflict of interest with respect to this manuscript. * Corresponding author. Tel.: +31 20 566 2422; fax: +31 20 691 7033. E-mail address: [email protected] (U. Beuers).   These authors contributed equally to the manuscript.

women [1]. Fatigue and pruritus are the most frequent symptoms which may affect up to 80% of patients [2,3]. PBC is regarded as an organ-specific autoimmune disease based on the presence of lymphoid infiltration in the portal tracts and of high titers of specific antibodies directed against the E2 subunit of pyruvate dehydrogenase complex (PDC-E2) or related mitochondrial 2-oxoacid dehydrogenase complexes (AMA-M2) [4,5]. The induction of AMA-M2 formation may be multifactorial [6]. Modification of mitochondrial or bacterial proteins after exposure to reactant xenobiotics such as the cosmetic component, 2-nonynoic acid [7,8] or hazardous aromatic hydrocarbons such as benzene [9] as well as exposure to bacterial proteins with a high degree of sequence homology with the human mitochondrial 2oxoacid dehydrogenase complexes like those of

0168-8278/$34.00 Ó 2008 Published by Elsevier B.V. on behalf of the European Association for the Study of the Liver. doi:10.1016/j.jhep.2008.05.022

D. Bogdanos et al. / Journal of Hepatology 49 (2008) 466–473

Novosphingobium aromaticivorans [10] or Lactobacillus delbrueckii [11] might induce loss of tolerance to human mitochondrial proteins in genetically susceptible individuals. Aberrant expression of PDC-E2-like proteins possibly derived from phagocytosed apoptotic biliary epithelial cells [12] at the luminal surface of cholangiocytes [6] may contribute to an immune response leading to florid cholangitis and bile duct destruction in PBC [13]. Here, we report a case of primary biliary cirrhosis diagnosed after immunization with a lactobacillus vaccine for years. The laboratory work-up was aimed to investigate the fine specificity of antibody responses against pyruvate dehydrogenase complex-E2 subunit (PDC-E2212–226), the major PBC-specific mitochondrial autoepitope and its microbial mimics in this female patient vaccinated annually for 8 years with Lactobacillus antigens. 1.1. Case report A 39-year-old previously healthy woman was referred to our outpatient clinic in July 2005 with a 3-month history of fatigue, generalized severe pruritus and pruritusassociated sleep disturbance. She and her family had no history of liver disease or other serious disorders. She worked as a banker and reported no regular intake of drugs or herbs. She was a non-smoker and was neither passively exposed to smoke nor to chemical toxins. She had no children and had not used hormones in the past [14]. Her gynaecologist had administered intramuscular lactobacillus vaccine [GynatrenR, >7  109 Lactobacilli of eight strains, L. rhamnosus (3), L. vaginalis (3), L. fermentum (1), L. salivarius (1)] twice within several weeks in the first year and at one-year intervals for 8 years as an immunization therapy for prevention of recurrent vaginitis [15]; serum liver tests or serum samples of the patient from that time were not available. The patient’s physical examination was unremarkable. Laboratory investigations showed elevated levels of total serum bilirubin (1.3 mg/dL; normal (N) < 1.0 mg/ dL), alkaline phosphatase (363 IU/L; N < 135), c-glutamyltransferase (244 IU/L; N < 38), serum aspartate

467

aminotransferase (92 IU/L; N < 33) and serum alanine aminotransferase (158 IU/L; N < 35). Antimitochondrial antibody (AMA, 1:5120) and AMA-M2 were strongly positive, and serum IgG (17.2 g/L; N 7.0–16.0) and IgM (4.4 g/L; N 0.4–2.3) were elevated. Serological tests for hepatitis A, B and C and antinuclear (ANA), antismooth muscle (ASMA), and soluble liver antigen (SLA) antibodies were negative. A liver biopsy revealed portal and periportal lymphocellular infiltration with signs of bile duct injury, ductopenia, moderate interface hepatitis and portoseptal fibrosis, compatible with stage 2–3 PBC. The diagnosis of a PBC–AIH overlap syndrome was considered unlikely (AIH score < 10)[16]. Therapy with ursodeoxycholic acid at 1000 mg/day (15 mg/kg/d) was initiated, and budesonide at 2  3 mg/d was added after 9 months when serum liver tests had only partially improved. The patient reported a rapid improvement of pruritus and fatigue within weeks. After 9–24 months, the patient reported general wellbeing. Mild pruritus (<3/10) was observed only occasionally. Fatigue and serum liver tests had markedly improved (Table 1) whereas serum AMA titers remained unchanged (1:3840 after 24 months, AMA-M2 positive). The patient agreed to perform additional studies and her serum samples were sent for further testing to the laboratory of Liver Immunopathology, King’s College London.

2. Methods 2.1. Characterization of the patient’s AMA 2.1.1. Indirect immunofluorescence (IIFL) Two serum samples from the patient taken after more than 6 months of ursodeoxycholic acid treatment were tested for conventional autoantibodies by indirect immunofluorescence (IIFL) on 5 lm frozen sections of a composite substrate containing rat liver, kidney and stomach using as revealing reagent an antitotal human IgG-fluorescein isothiocyanate conjugate (Dako Ltd., High Wycombe, Bucks, UK) at a starting dilution of 1/40 [17]. Positive sera were titrated to extinction by

Table 1 Clinical symptoms and biochemical markers of the PBC patient during treatment with UDCA for 9 months and combined treatment with UDCA + budesonide thereafter

Pruritus (0–10) Fatigue (0–+++) Bilirubin (xN) Alkaline Phosphatase (xN) c-GT (xN) ALT (xN) AST (xN) N = upper limit of normal.

Before treatment

After 3 weeks (UDCA)

After 9 months (UDCA)

After 24 months (budesonide + UDCA)

8–9 +++ 1.2 6.3 15.4 5.5 3.9

3 + 1.3 2.7 6.4 4.5 2.8

0–3 (+) 1.0 2.7 5.9 2.7 1.8

0 (+) 0.7 1.9 6.6 1.9 1.3

468

D. Bogdanos et al. / Journal of Hepatology 49 (2008) 466–473

double dilution. The slides were examined by an independent observer (Dr. E.T. Davies, Department of Immunology, King’s College Hospital) using an Olympus BX40 (Olympus Optical, London, UK) fluorescence microscope. 2.2. Antimitochondrial antibody reactivity by immunoblotting Antimitochondrial antibody specificity was further evaluated by immunoblotting using strips pre-coated with electrophoretically separated human liver mitochondrial subfractions [18,19]. Strips were incubated for 30 min with individual serum samples at a dilution of 1/200 and after three 5-min washings, they were incubated for 1 h with an HRP-conjugated rabbit antitotal human IgG (Dako Ltd., High Wycombe, Bucks, UK), or HRP-conjugated murine antihuman IgG subclass-specific antisera (IgG1:clone JDC-1; IgG2:31-7-4; IgG3:HP6050; IgG4:HP6025, Southern Biotechnology Associates, Inc., Birmingham, AL, USA) diluted in 2% bovine serum albumin (BSA) in phosphate-buffered saline (PBS) as revealing reagents [11,18,19]. After washing, the blots were developed with diamino-benzidine at 0.25 mg/ml in PBS with 0.1% H2O2 as substrate. A reference strip from the same lot with immunofixed bands of known mitochondrial autoantigens was used for comparative interpretation of the results. 2.3. Autoantibody testing by ELISA AMA-M2 autoantibody specificity was also analysed by ELISA using as antigenic preparations purified M2 antigen from bovine heart mitochondria (Euroimmun). Antibody reactivity against the major mitochondrial autoantigens, namely the E2 subunits of the pyruvate dehydrogenase complex (PDC), the branched chain oxoacid dehydrogenase complex (BCOADC) and the oxoglutarate dehydrogenase complex (OGDC) has been further evaluated using a triple hybrid (-3) [20], expressing the immunodominant human PDC-E2, BCOADCE2 and OGDC-E2 regions (INOVA, San Diego, CA, USA). 2.4. Amino acid similarities The beta-galactosidase of Lactobacillus delbrueckii (LACDE BGAL266–280) was recently identified as a target of immunological cross-reactivity in patients with PBC due to its eight amino acids in a row (seven identities) similarity with the major human PDC-E2212–226 mitochondrial autoepitope [11]. The corresponding BGAL of Lactobacillus salivarius is highly homologous to LACDE BGAL266–280 (local homology 89%) and is contained in the vaccine.

2.5. Peptide synthesis Two 15-mer biotinylated peptides containing the key sequences of beta- galactosidase (lactase) of Lactobacillus delbrueckii, subsp. bulgaricus (BGAL266–280 LACDE) (RDSEGDLVAEKLGPI) and human PDC-E2212–226 (KLSEGDLLAEIETDK) and an irrelevant 15 aa control peptide (YVNQSLRPTPLEISV) [19,21,22] of comparable hydrophobicity were constructed using Fmoc chemistry (Mimotopes Ltd., Clayton, Australia) [23]. The identity of each purified peptide was confirmed by mass spectrometry. 2.6. ELISA Antibody binding to the peptides (5 lg/ml) was determined by ELISA (serum dilution 1/200) [19,21,22,24]. For detection of IgG class and IgG subclasses antibody reactivity, an HRP-conjugated rabbit antitotal human IgG, or antihuman IgG subclasses (see above) were used. Reaction for a given peptide was considered positive when optical density (OD)test/ODcontrol/peptide P 2, this cut-off representing mean + 5 SD of 135 readings using serum (tested in triplicate) from 15 healthy subjects against the mimicking peptides [19,21,22,24,25]. 2.7. Inhibition studies Competition ELISAs were performed to test for cross-reactivity to LACDE BGAL266–280 and human PDC-E2212–226, measuring residual anti-LACDE BGAL266–280 or anti-PDC-E2212–226 reactivity, after pre-incubation with LACDE BGAL266–280, PDCE2212–226, purified PDC antigen, control peptide and control antigen (CYP2D6, Euroimmun), respectively, as liquid phase competitors, as previously described [11,19,21,22,24–27]. 2.8. Determination of antipeptide antibody affinity The relative affinity of antipeptide antibody binding to LACDE BGAL266–280 and human PDC-E2212–226 was measured, as previously described [11,19,24,28] using as chaotropic agent increasing concentrations (1– 8 M) of urea. The relative affinity index (rAI), defined as the concentration of urea that elutes 50% of the antibody reactivity, for each of the two mimicking peptides was extrapolated. 2.9. Inhibition of PDC enzymatic activity The enzymatic inhibitory activity of the PBC serum sample reactive to LACDE BGAL266–280/human PDCE2212–226 was measured as previously described [11,29] before and after incubation of the serum with individual human PDC-E2212–226 or LACDE BGAL266–280 pep-

D. Bogdanos et al. / Journal of Hepatology 49 (2008) 466–473

tides. The source of the enzyme was purified porcine heart PDC (Sigma Aldrich, Dorset, UK). The enzymatic activity was measured spectrophotometrically at 405 nm as the formation of thio-NADH from thio-NAD (Sigma) in the presence of the substrate pyruvate, and the cofactors cocarboxylase (Sigma) and coenzyme A (Sigma). Absorbance was read immediately and then at minute intervals for 10 min. The rate of change in absorbance was calculated, and the percent inhibition was obtained using the standard reaction rate in wells to which PBS (pH 7.4) was added instead of serum. The PBC serum (1/10,000) was tested before and after pre-incubation with each peptide. Tests were done in triplicate and expressed with the following formula: (mean percentage inhibition of the sample after preincubation/mean percentage inhibition of the sample before pre-incubation)  100. Repeat experiments gave variations <10%. 2.10. Apoptosis studies 2.10.1. IgG purification IgG was purified from serum of the index patient (IP), three normal controls (C1–C3), three AMA positive PBC patients (PBC1–PBC3) and two patients with primary sclerosing cholangitis (PSC1–PSC2) using the Melon Gel IgG purification System (Pierce Biotechnology, Rockford, IL, USA). Briefly, 500 ll of melon gel IgG was added to the column, centrifuged (4000g) and washed twice with 300 ll of purification buffer. Fifty microliters of individual sera diluted 1/10 in melon gel purification buffer were added to the column, and incubated for 5 min at room temperature with end-over-end mixing. After removal of bottom cap from the column, loosen top cap and re-inserting spin column in the collection tube, followed by centrifugation for 1 min, the purified antibody was collected and stored at 20 °C. 2.11. H-69 cell culture The immortalized human intrahepatic biliary epithelial cell line H-69 was kindly provided by Dr. D. Jefferson, Boston. It was grown in DMEM/Ham’s F12 medium (3:1, pH 7.4, 37 °C, 5% CO2) containing 10% fetal calf serum, 100 U/mL penicillin, 100 lg/mL streptomycin (all from PAA; Pasching, Austria), 1.8  104 mol/L adenine, 5 lg/mL insulin, 5 lg/mL transferrin, 2  109 mol/L triiodthyronine, 1.1  106 mol/L hydrocortisone, 6 1.6  10 mol/L epidermal growth factor (all from Sigma Chemical; St. Louis, MO, USA), and 5.5  106 mol/L epinephrine (Jenapharm; Jena, Germany). 2.12. Caspase assays Apoptosis was evaluated by measuring activation of the effector caspases 3/7 in subconfluent H-69 cells incu-

469

bated with purified IgG (1 mg/ml) from both patients and controls for 48 h. Etoposide (30 lmol/L) was used as a positive control. Activity assays were performed using a caspase 3/7 assay kit (Promega; Madison, WI, USA). 2.13. Quantitative RT-PCR Total RNA was isolated using the NucleoSpin RNA II kit (Machery–Nagel, Du¨ren, Germany). Two micrograms of total RNA were transcribed into cDNA by incubation with 200–500 ng oligo-d(T)12–18 primer and 1 ll 10 mM dNTP-mix in a volume of 13 ll for 5 min at 65 °C. After cooling on ice, 4 ll of 5 First-Strand-buffer, 1 ll 0.1 M DTT, 1 ll RNase OUT and 1 ll Superscript III (Invitrogen, Karlsruhe, Germany) were added and incubated for 1 h at 50 °C. Enzymes and RNA were degraded by heating for 15 min at 70 °C and incubation with 1 ll RNase H (Invitrogen) for 20 min at 37 °C. For quantitative RTPCR, the PCR cycler Rotor Gene RG-3000 (Corbett Research, Mortlake, Australia) and the QuantiTect SYBR Green PCR-Kit (Qiagen, Hilden, Germany) were applied. Quantification was done using GAPDH as endogenous control. Primer sequences: GAPDH forward 50 -ACCACAGTCCATGCCATCAC-30 and reverse 50 -T CCACCACCCTGTTGCTGTA-30 ; human polymeric Ig receptor (pIgR) forward 50 -CCACCGTGGAGAT CAAGATT-30 and reverse 50 -CAGCCCGTGTTATTC CACTT-3; and human neonatal Fc receptor (FcRn) forward 50 -GGCTCCTGCTCTTTCTCCTT-30 and reverse 50 -TCTCCCAATACCAGGACACC-30 . 2.14. Statistical analysis Results are presented as means ± SD or percentages (%). Data were analysed using Student’s t-test, Mann– Whitney U test, chi-square (v2) test and the Fisher’s exact test as appropriate. A two-tailed p value < 0.05 was considered significant. Statistical analyses were performed using SPSS (SPSS Inc., Chicago, IL, USA) statistical package.

3. Results AMA at a titer of 1/640 were detected in the serum of this PBC patient after more than 6 months of medical treatment. The sample was also positive using the -3 (PDC-E2, BCOADC-E2 and OGDC-E2) triple hybrid (78 RU/ml, cut-off: 20 RU/ml) by ELISA. The serum immunofixed major mitochondrial antigens including the 74 kDa PDC-E2, the 51 kDa BCOADC-E2 and OGDC-E2 and to a lesser extent the 55 kDa PDC-E3 binding protein (E3BP) and the 45 kDa PDC-E1a by immunoblotting. IgG subclass distribution of AMA showed an IgG3 predominance for PDC-E2 (Fig. 1). Serum was reactive

D. Bogdanos et al. / Journal of Hepatology 49 (2008) 466–473

Relative antibody affinity (mol/L)

470 %

IgG subclass distribution

100 75

50 25 0 IgG1

IgG2

IgG3

3.1. Inhibition studies Competition ELISA experiments showed that human PDC-E2212–226/BGAL266–280 LACDE reactivity was cross-reactive. Preincubation with relevant and control peptide/antigen revealed that IgG3 subclass antibody binding to human PDC-E2212–226 was inhibited by 72% after preincubation with human PDC-E2212–226; by 78% after preincubation with purified PDC antigen; and by 73% after preincubation with LACDE BGAL266–280. Insignificant inhibition was observed after preincubation with irrelevant control peptide and antigen (Fig. 2). Antibody binding to LACDE BGAL266–280 PDC-E2 peptide

BGAL LACDE peptide

PDC antigen

Control peptide

Inhibition of IgG3 anti-PDC-E2 peptide reactivity

80

60

40

20

25

50

100

0

250

500

BGAL266-280 LACDE

Fig. 3. Relative antibody affinity to Lactobacillus LACDE BGAL266–280 was significantly higher than that against human PDC-E2212–226 in serum of a PBC patient with a history of Lactobacillus vaccination. Relative antibody affinity was estimated as the concentration (mol/L) of the chaotropic agent (urea) that elutes 50% of the peptide antibody reactivity. For details, see Sections 2 and 3. *p < 0.05, two-sided Student’s t-test.

was also significantly inhibited by 78%, 71%, and 68% after preincubation with LACDE BGAL266–280, PDCE2212–226, and PDC antigen, respectively. 3.2. Affinity of antibody reactivity The relative affinity of antibody reactivity to LACDE BGAL266–280 was significantly higher than that against human PDC-E2212–226 (Fig. 3). Inhibition of PDC enzymatic activity was 100% using the serum reactive to LACDE BGAL266–280/human PDC-E2212–226 (1/1000 dilution). After preincubation with LACDE BGAL266–280, this was reduced by 59%; after preincubation with human PDC-E2212–226 it was reduced by 34%. 3.3. Apoptosis studies In an immortalized human cholangiocyte cell line, H69, the human IgG neonatal Fc receptor (FcRn) and the human IgA polymeric Ig receptor (pIgR) were expressed as documented by quantitative RT-PCR (data not shown). Incubation of human cholangiocytes for 48 h with etoposide (positive control) induced a 5.5-fold activation of caspase 3/7 (Fig. 4) in comparison to control cells incubated with the carrier only. Neither purified IgG (1 mg/ml) of the index patient nor purified IgG of three PBC patients, two PSC patients and three healthy controls did affect the rate of apoptosis in human cholangiocytes when compared to control cells (Fig. 4). These data indicate that purified PBC IgG did not trigger apoptosis in a human cholangiocyte cell line.

Control antigen

5

3

PDC-E2212-226

with human PDC-E2212–226 (ODtest/peptide/ODcontrol/peptide = 3.2) and with BGAL266–280 LACDE (7.5). Antibody reactivity to human PDC-E2212–226 and BGAL266–280 LACDE belonged to the IgG3 subclass.

0

*

IgG4

Fig. 1. IgG subclass distribution of AMA in serum of a PBC patient with a history of Lactobacillus vaccination. Triplicate determination by Western blotting revealed an IgG3 predominance for PDC-E2. For details, see Sections 2 and 3.

%

6

µg/ml

Competitor concentration Fig. 2. Binding to human PDC-E2212–226 of IgG3 subclass antibody from serum of a PBC patient with a history of Lactobacillus vaccination was inhibited after preincubation with relevant antigens. Inhibition was determined by competition ELISA. For details, see Sections 2 and 3.

4. Discussion We present the case of a woman with an 8-year history of lactobacillus vaccination for recurrent vaginitis

D. Bogdanos et al. / Journal of Hepatology 49 (2008) 466–473

471

Caspase 3/7 activity (fold change)

8 7 6 5 4 3 2 1 e

2

os id

Et op

PS C

3

1 PS C

C PB

C

2

1 PB

C

3 PB

C

2 C

1 C

IP

N

eg .

co n

tr o

l

0

Fig. 4. Purified IgG of patients with PBC, PSC and healthy controls did not induce apoptosis in immortalized human cholangiocytes. H-69 cells were incubated with purified immunoglobulin (Ig)G (1 mg/ml) from both patients (index patient, IP; three AMA positive control PBC patients, PBC 1–3; two patients with primary sclerosing cholangitis, PSC 1–2) and controls (C 1–3) for 48 h. Etoposide (30  106 mol/L) was used as a positive control. Apoptosis was quantified by measuring caspase 3/7 activity. For details, see Sections 2 and 3.

who developed characteristic clinical, biochemical and histological features of primary biliary cirrhosis. In view of known immunological cross-reactivity between a lactobacillus sequence and the major autoepitope of antimitochondrial antibody [11] we characterized autoreactivity and possible cross-reactivity [11,18,19, 21,22,24–27,30] in this patient finding that her serum contains a high-titer AMA, which belongs mainly to IgG3 subclass and targets the immunodominant PDCE2212–226 autoepitope. This serum also reacts with the Lactobacillus delbrueckii subs. bulgaricus self-mimicking BGAL266–280 sequence. The antibody shows the same IgG subclass restriction and is indeed cross-reactive: binding to the lactobacillus mimic is inhibited by the lactobacillus and mitochondrial peptides but, importantly, also by the full-length PDC-E2 autoantigen implying that the cross-reactive microbial/self peptide recognition involves the naturally occurring autoantigen. The finding that the relative affinity of antibody binding to LACDE BGAL266–280 is greater than that against human PDC-E2212–226 (Fig. 3) suggests that antimicrobial reactivity may have preceded that to the self mimic [11]. Lastly, the ability of the cross-reactive antibody to inhibit PDC’s enzymatic activity emphasizes its biological significance [11,25]. Of the Lactobacilli strains contained in the vaccine, the BGAL sequences deposited are only few. However, that of L. salivarius is highly homologous (89% local aa similarity) to that of Lactobacillus delbrueckii subs. bulgaricus used as an antigenic target in the present experiments in view of its previous characterization in terms of immunological cross-reactivity with the mitochondrial autoepitope and its high specificity for primary biliary cirrhosis [11]. Incubation of human cholangiocytes with purified IgG from the index patient and healthy and disease controls revealed no detectable induction of apoptosis although expression of the IgG neonatal Fc receptor,

FcRn, was shown by quantitative RT-PCR. Thus, PBC IgG-AMA apparently did not cause apoptosis in contrast to findings reported previously by Matsumura et al. [31] who showed caspase activation by PBC IgA, but not control IgA in MDCK cells transfected with pIgR. The difference may be explained in various ways: (i) IgA-AMA may be more appropriate than IgG-AMA to induce apoptosis in human cholangiocytes; (ii) the FcRn expressed in H-69 cells was not functional; (iii) the experimental conditions were inappropriate to induce apoptosis mediated by IgG-AMA. The observation that also purified IgG from disease controls did not cause apoptosis under the conditions chosen, rather supports these assumptions. Thus, the pathophysiological relevance of IgG-AMA remains unclear. The patient and her physicians were unable to provide informations about serum liver tests before start of lactobacillus vaccination. In addition, no serum samples were available for AMA testing from the time before start of lactobacillus vaccination. This weakens the assumption that lactobacillus vaccination has caused AMA formation and development of PBC in the patient presented here considering that AMA formation may preceed development of biochemical and clinical signs of PBC by years [6]. Still, the greater affinity of antibody binding to LACDE BGAL266–280 than that against human PDC-E2212–226 remains intriguing. In our symptomatic patient, UDCA monotherapy has improved symptoms and serum liver tests (Table 1). A complete biochemical response had been unlikely considering that the patient presented with serum bilirubin >1 mg/dl and at least moderate periportal/periseptal lymphocyte interface hepatitis and portoseptal fibrosis, risk markers for a progressive course of PBC[32]. Evaluation of combined treatment with UDCA and budesonide has been suggested for this particular subgroup of patients for further evaluation within a controlled clinical trial[33]. Combined treatment with UDCA and

472

D. Bogdanos et al. / Journal of Hepatology 49 (2008) 466–473

budenoside at a moderate dose led to further improvement of biochemical parameters. Lack of significant changes in serum AMA titers made it unlikely that cessation of vaccination therapy during therapy has contributed to the beneficial effect of medical treatment. In conclusion, we report on a female patient in whom lactobacillus vaccination therapy for the prevention of recurrent vaginitis may have induced an immune response against the PDC-E2 autoantigen contributing to the loss of immunological tolerance and the induction of primary biliary cirrhosis [1,4,13]. Thus, lactobacillus vaccination therapy of recurrent bacterial vaginitis might be another culprit for development of PBC in genetically susceptible women. References [1] Kaplan MM, Gershwin ME. Primary biliary cirrhosis. N Engl J Med 2005;353:1261–1273. [2] Poupon RE, Chretien Y, Chazouilleres O, Poupon R, Chwalow J. Quality of life in patients with primary biliary cirrhosis. Hepatology 2004;40:489–494. [3] Talwalkar JA, Lindor KD. Primary biliary cirrhosis. Lancet 2003;362:53–61. [4] Nishio A, Keeffe EB, Gershwin ME. Immunopathogenesis of primary biliary cirrhosis. Semin Liver Dis 2002;22:291–302. [5] Shimoda S, Nakamura M, Ishibashi H, Kawano A, Kamihira T, Sakamoto N, et al. Molecular mimicry of mitochondrial and nuclear autoantigens in primary biliary cirrhosis. Gastroenterology 2003;124:1915–1925. [6] Gershwin ME, Mackay IR. The causes of primary biliary cirrhosis: convenient and inconvenient truths. Hepatology 2008;47:737–745. [7] Rieger R, Leung PS, Jeddeloh MR, Kurth MJ, Nantz MH, Lam KS, et al. Identification of 2-nonynoic acid, a cosmetic component, as a potential trigger of primary biliary cirrhosis. J Autoimmun 2006;27:7–16. [8] Amano K, Leung PS, Rieger R, Quan C, Wang X, Marik J, et al. Chemical xenobiotics and mitochondrial autoantigens in primary biliary cirrhosis: identification of antibodies against a common environmental, cosmetic, and food additive, 2-octynoic acid. J Immunol 2005;174:5874–5883. [9] Ala A, Stanca CM, Bu-Ghanim M, Ahmado I, Branch AD, Schiano TD, et al. Increased prevalence of primary biliary cirrhosis near superfund toxic waste sites. Hepatology 2006;43:525–531. [10] Selmi C, Balkwill DL, Invernizzi P, Ansari AA, Coppel RL, Podda M, et al. Patients with primary biliary cirrhosis react against a ubiquitous xenobiotic-metabolizing bacterium. Hepatology 2003;38:1250–1257. [11] Bogdanos DP, Baum H, Okamoto M, Montalto P, Sharma UC, Rigopoulou EI, et al. Primary biliary cirrhosis is characterized by IgG3 antibodies cross-reactive with the major mitochondrial autoepitope and its Lactobacillus mimic. Hepatology 2005;42:458–465. [12] Allina J, Hu B, Sullivan DM, Fiel MI, Thung SN, Bronk SF, et al. T cell targeting and phagocytosis of apoptotic biliary epithelial cells in primary biliary cirrhosis. J Autoimmun 2006;27:232–241. [13] Leung PS, Coppel RL, Gershwin ME. Etiology of primary biliary cirrhosis: the search for the culprit. Semin Liver Dis 2005;25:327–336. [14] Gershwin ME, Selmi C, Worman HJ, Gold EB, Watnik M, Utts J, et al. Risk factors and comorbidities in primary biliary cirrhosis: a controlled interview-based study of 1032 patients. Hepatology 2005;42:1194–1202.

[15] Siboulet A. Vaccination against nonspecific bacterial vaginosis. Double-blind study of Gynatren. Gynakol Rundsch 1991;31:153–160. [16] Beuers U, Rust C. Overlap syndromes. Semin Liver Dis 2005;25:311–320. [17] Vergani D, Alvarez F, Bianchi FB, Cancßado EL, Mackay IR, Manns MP, et al. Liver autoimmune serology: a consensus statement from the committee for autoimmune serology of the international autoimmune hepatitis group. J Hepatol 2004;41:677–683. [18] Bogdanos DP, Baum H, Gunsar F, Arioli D, Polymeros D, Ma Y, et al. Extensive homology between the major immunodominant mitochondrial antigen in primary biliary cirrhosis and Helicobacter pylori does not lead to immunological cross-reactivity. Scand J Gastroenterol 2004;39:981–987. [19] Bogdanos DP, Pares A, Baum H, Caballeria L, Rigopoulou EI, Ma Y, et al. Disease-specific cross-reactivity between mimicking peptides of heat shock protein of mycobacterium gordonae and dominant epitope of E2 subunit of pyruvate dehydrogenase is common in Spanish but not British patients with primary biliary cirrhosis. J Autoimmun 2004;22:353–362. [20] Moteki S, Leung PS, Coppel RL, Dickson ER, Kaplan MM, Munoz S, et al. Use of a designer triple expression hybrid clone for three different lipoyl domain for the detection of antimitochondrial autoantibodies. Hepatology 1996;24:97–103. [21] Bogdanos DP, Baum H, Sharma UC, Grasso A, Ma Y, Burroughs AK, et al. Antibodies against homologous microbial caseinolytic proteases P characterise primary biliary cirrhosis. J Hepatol 2002;36:14–21. [22] Bogdanos DP, Lenzi M, Okamoto M, Rigopoulou EI, Muratori P, Ma Y, et al. Multiple viral/self immunological cross-reactivity in liver kidney microsomal antibody positive hepatitis C virus infected patients is associated with the possession of HLA B51. Int J Immunopathol Pharmacol 2004;17:83–92. [23] Rodda SJ, Tribbick G. Antibody-defined epitope mapping using the multipin method of peptide synthesis. Methods 1996;9:473–481. [24] Bogdanos DP, Baum H, Grasso A, Okamoto M, Butler P, Ma Y, et al. Microbial mimics are major targets of crossreactivity with human pyruvate dehydrogenase in primary biliary cirrhosis. J Hepatol 2004;40:31–39. [25] Kerkar N, Choudhuri K, Ma Y, Mahmoud A, Bogdanos DP, Muratori L, et al. Cytochrome P4502D6(193–212): a new immunodominant epitope and target of virus/self cross-reactivity in liver kidney microsomal autoantibody type 1-positive liver disease. J Immunol 2003;170:1481–1489. [26] Gregorio GV, Choudhuri K, Ma Y, Pensati P, Iorio R, Grant P, et al. Mimicry between the hepatitis C virus polyprotein and antigenic targets of nuclear and smooth muscle antibodies in chronic hepatitis C virus infection. Clin Exp Immunol 2003;133:404–413. [27] Polymeros D, Bogdanos DP, Day R, Arioli D, Vergani D, Forbes A. Does cross-reactivity between mycobacterium avium paratuberculosis and human intestinal antigens characterize Crohn’s disease? Gastroenterology 2006;131:85–96. [28] Quinn J, Diamond AG, Palmer JM, Bassendine MF, James OF, Yeaman SJ. Lipoylated and unlipoylated domains of human PDC-E2 as autoantigens in primary biliary cirrhosis: significance of lipoate attachment. Hepatology 1993;18:1384–1391. [29] Jois J, Omagari K, Rowley MJ, Anderson J, Mackay IR. Enzyme inhibitory antibody to pyruvate dehydrogenase: diagnostic utility in primary biliary cirrhosis. Ann Clin Biochem 2000;37:67–73. [30] Rodda SJ, Tribbick G. Antibody-defined epitope mapping using the multipin method of peptide synthesis. Methods 1996;9:473–481.

D. Bogdanos et al. / Journal of Hepatology 49 (2008) 466–473 [31] Matsumura S, Van De Water J, Leung P, Odin JA, Yamamoto K, Gores GJ, et al. Caspase induction by IgA antimitochondrial antibody: IgA-mediated biliary injury in primary biliary cirrhosis. Hepatology 2004;39:1415–1422. [32] Corpechot C, Carrat F, Poupon R, Poupon RE. Primary biliary cirrhosis: incidence and predictive factors of cirrhosis develop-

473

ment in ursodiol-treated patients. Gastroenterology 2002;122:652–658. [33] Corpechot C, Carrat F, Bahr A, Chretien Y, Poupon RE, Poupon R. The effect of ursodeoxycholic acid therapy on the natural course of primary biliary cirrhosis. Gastroenterology 2005;128:297–303.