The

n e w e ng l a n d j o u r na l

of

m e dic i n e

brief report

Recurrence of Bile Salt Export Pump Deficiency after Liver Transplantation

Paloma Jara, M.D., Loreto Hierro, M.D., Pilar Martnez-Fernndez, Ph.D., Rita Alvarez-Doforno, Ph.D., Francisca Ynez, B.S., Mara C. Diaz, M.D., Carmen Camarena, M.D., Angela De la Vega, M.D., Esteban Frauca, M.D., Gema Muoz-Bartolo, M.D., Manuel Lpez-Santamara, M.D., Javier Larrauri, M.D., and Luis Alvarez, Ph.D.

SUM M A R Y
Severe bile salt export pump (BSEP) deficiency is a hereditary cholestatic condition that starts in infancy and leads to end-stage liver disease. Three children who underwent orthotopic liver transplantation for severe BSEP deficiency had post-transplantation episodes of cholestatic dysfunction that mimicked the original disease. Remission of all episodes was achieved by intensifying the immunosuppressive regimen. The phenotypic recurrence of the disease correlated with the presence of circulating high-titer antibodies against BSEP that inhibit transport by BSEP in vitro. When administered to rats, these antibodies targeted the bile canaliculi and impaired bile acid secretion. holestatic disorders are among the most severe liver diseases in infancy and childhood.1 For some patients, orthotopic liver transplantation is the only effective therapy, resulting in favorable outcomes and no recurrence of the original disease.2,3 Severe BSEP deficiency, also referred to as progressive familial intrahepatic cholestasis type 2, is one such disorder. It is caused by recessive mutations in ABCB11, the gene encoding BSEP.4,5 BSEP is expressed at the canalicular membrane of the hepatocytes and transports bile acids into the canalicular space, using ATP as an energy source.6 Hepatocyte canaliculi in most patients carrying ABCB11 mutations express little or no detectable BSEP.5 Children with severe BSEP deficiency typically have jaundice and pruritus within the first few months of life. Early-onset cholestasis progresses to hepatic fibrosis, cirrhosis, and end-stage liver disease.7 Affected children are also at increased risk for liver cancer.5,8 Biochemical and histopathological features of this disorder include elevated serum concentrations of bile acids, intrahepatic cholestasis, and often, giant-cell transformation. Serum values of -glutamyltransferase (GGT) activity are normal, despite the degree of conjugated hyperbilirubinemia.9-11 Liver disease caused by severe BSEP deficiency is usually resistant to medical treatment; therefore, for most patients, transplantation becomes necessary.7,11 We report on a rare phenotypic recurrence of BSEP deficiency that takes place after liver replacement. The recurrence correlates with the presence in serum of blocking antibodies against BSEP.

From the Pediatric Liver Service (P.J., L.H., M.C.D., C.C., A.D.V., E.F., G.M.-B.), Research Unit (P.M.-F., L.A.), Service of Immunology (R.A.-D., F.Y.), Department of Pathology (J.L.), and Department of Pediatric Surgery (M.L.-S.), La Paz University Hospital, Madrid. Address reprint requests to Dr. Alvarez at the Research Unit, La Paz University Hospital-FIBHULP, Paseo de La Castellana 261, 28046 Madrid, Spain, or at luisalvarez.hulp@ salud.madrid.org. Drs. Jara, Hierro, and Martnez-Fernndez contributed equally to this article. N Engl J Med 2009;361:1359-67.
Copyright 2009 Massachusetts Medical Society.

n engl j med 361;14

nejm.org

october 1, 2009

1359

The New England Journal of Medicine Downloaded from nejm.org on December 17, 2011. For personal use only. No other uses without permission. Copyright 2009 Massachusetts Medical Society. All rights reserved.

The

n e w e ng l a n d j o u r na l

of

m e dic i n e

Table 1. Characteristics of the Three Patients with Recurrent Severe Bile Salt Export Pump Deficiency after Liver Transplantation. Characteristic Sex Year of birth Age at transplantation (yr) ABCB11 mutation Nucleotide mutation Predicted amino acid mutation Homozygote, 907AG R303G Compound heterozygote, 1741CT and IVS12+1GT L581F and splice-site disruption Homozygote, IVS17+1TA Splice-site disruption Patient 1 M 1982 5.2 Patient 2 F 1991 3.7 Patient 3 F 1995 2.2

Me thods
We obtained approval for the study from a local ethics committee. Written informed consent was provided by the patients or, if the patients were under 18 years of age, their parents.
Sequence Analysis of ABCB11

We isolated genomic DNA from peripheral-blood leukocytes, using the PureGene DNA Isolation kit (Gentra Systems), and amplified all 28 exons of ABCB11 and flanking intronexon boundaries by performing polymerase-chain-reaction assays. (Primer sequences and assay conditions are available on request.) We sequenced both strands using the 1.1 Big Dye Terminator RRMix (Applied Biosystems).
Immunohistochemical Analysis

Immunohistochemical studies were carried out on paraffin-embedded explanted liver samples and needle-biopsy specimens obtained during posttransplantation cholestatic episodes. Detection of BSEP and multidrug resistanceassociated protein 2 (MRP2) was performed by means of the antigen-retrieval method, as previously described,12 with a mouse monoclonal anti-MRP2 antibody (clone M2III-6, Chemicon International) and a goat polyclonal anti-BSEP antibody (Santa Cruz Biotechnology) as primary antibodies.
Indirect Immunofluorescence Analysis

Serum samples from the patients were screened at dilutions ranging from 1:40 to 1:5120, with the use of commercially prepared sections from rat liver, kidney, and stomach (Euroimmun, Lbeck). Bound antibodies were detected with the use of fluorescein isothiocyanate (FITC)conjugated rabbit antihuman IgA, IgG, and IgM antibodies (DakoCytomation).
1360
n engl j med 361;14

Figure 1 (facing page). Characteristic Findings of Episodic Cholestasis after Liver Transplantation for Severe Bile Salt Export Pump (BSEP) Deficiency in Patient 2. Panel A shows immunohistochemical detection of BSEP and multidrug resistanceassociated protein 2 (MRP2) in a liver specimen from a child without cholestasis (control) and the explanted liver from Patient 2. Paraffinembedded sections were stained with primary antibodies at a 1:20 dilution. Sections were counterstained with hematoxylin. BSEP staining was also absent in Patients 1 and 3. Panel B shows the results of liver-function testing during the course of the first post-transplantation cholestatic episode in Patient 2. Serum aspartate aminotransferase (AST), alanine aminotransferase (ALT), bilirubin, and -glutamyl transpeptidase (GGT) levels were monitored from the time of onset of the cholestatic attack (day 0) to the time of recovery. GGT activity remained relatively constant despite the long-lasting period of jaundice. Panel C shows morphologic characteristics of the liver-biopsy specimen obtained from Patient 2 at 26 days after the onset of the first post-transplantation cholestatic attack. Multinucleated giant hepatocytes (arrows) are visible (hematoxylin and eosin). The inset shows a portal tract in the same specimen, with a moderately dense lymphocytic infiltrate but no evidence of damage to the bile duct or portal vessels. The left-hand image in Panel D shows the pattern of immunofluorescence staining of rat-liver sections incubated with serum (dilution, 1:640) obtained at the onset of the third episode of cholestatic dysfunction in Patient 2. No reaction was found in sections of rat kidney or stomach. Positive bile-canalicular staining was also detected with the use of serum samples from Patients 1 and 3. The right-hand image in Panel D shows the result of indirect immunofluorescence analysis of serum from a control (Patient 4) obtained during the second episode of acute cellular rejection (see the Supplementary Appendix for details). Panel E shows serum antibody titers during the fifth cholestatic attack in Patient 2. The intensity of the immunosuppressive regimen was increased on day 0. Resolution of pruritus was achieved on day 15. On complete recovery of normal liver function, the anticanalicular antibody titer was 6.25% of that at the peak of the cholestatic attack. (Titers are indicated within the plot.) For Panels B and E, to convert values for bilirubin to micromoles per liter, multiply by 17.1.

nejm.org

october 1, 2009

The New England Journal of Medicine Downloaded from nejm.org on December 17, 2011. For personal use only. No other uses without permission. Copyright 2009 Massachusetts Medical Society. All rights reserved.

Brief Report

A
Control
Staining for MRP2 Staining for BSEP Staining for MRP2

Patient 2
Staining for BSEP

B
1000

C
AST, ALT, and GGT (U/liter)
900 800 700 600 500 400 300 200 100 0
0 30 60 90 120 150 180 210 AST ALT GGT Total bilirubin Direct bilirubin

10 9 8 6 5 4 3 2 1 0 240

Days since Onset of Attack

D
Indirect Immunofluorescence Analysis
Patient 2 Control

E
650 600 550 500 450 400 350 300 250 200 150 100 50 0 1:5120

Bilirubin (mg/dl)

AST ALT GGT Anticanalicular antibody (titer)

Total bilirubin

6 5 4 3 2

AST, ALT, and GGT (U/liter)

1:1280 1:640 1:320

0 20 40 60 80 100

1 0 120

Days since Onset of Attack

Western-Blot Analysis

BSEP (at a 1:200 dilution). Immune complexes were 1st RETAKE AUTHOR: Jara (Alvarez) ICM 2nd Membrane vesicles (containing a total of 30 g of detected by means of horseradish peroxidase REG F FIGURE: 1 of 3 3rd protein) from Sf9 cells (derived from CASE Spodoptera conjugated goat antihuman immunoglobulins Revised Line 4-C frugiperda) expressing either human BSEP (Sigma- (1:16,000 dilution, Nordic Immunological LaboraEMail SIZE ARTIST: ts H/T H/T 36p6 Aldrich) or human MRP1 (BD Biosciences) were tories) or rabbit antigoat immunoglobulins (1:6000 Enon Combo fractionated by means of 10% sodium dodecyl sul- AUTHOR, PLEASE NOTE: dilution, DakoCytomation). Figure has fatepolyacrylamide-gel electrophoresis. The gels been redrawn and type has been reset. Please check carefully. were transferred onto polyvinylidene difluoride Evaluation of Functional Transport membranes and cut into strips, which JOB: 36114 were probed Uptake of 3H-labeled taurocholate (American RaISSUE: 10-01-09 with serum samples from patients (at a 1:100 di- diolabeled Chemicals) in membrane vesicles from lution) or with the goat antibody against human Sf9 cells expressing human BSEP was measured
n engl j med 361;14 nejm.org october 1, 2009

Total Bilirubin (mg/dl)

1361

The New England Journal of Medicine Downloaded from nejm.org on December 17, 2011. For personal use only. No other uses without permission. Copyright 2009 Massachusetts Medical Society. All rights reserved.

1362
Patient 2 Episode 1 3.5 EBV infection Unknown Withdrawal of cor- Unknown ticosteroids Unknown 5.2 8.1 12.1 13.0 2.1 Episode 2 Episode 3 Episode 4 Episode 5 Episode 1 Patient 3 Episode 2 4.0 12.0 Low cyclosporine Switch from tac(trough level, rolimus to cy<50 ng/ml) closporine for food allergy
The

Table 2. Clinical, Biochemical, and Histologic Features and Outcome of Episodes of Recurrent Cholestatic Graft Dysfunction in the Three Patients.*

Characteristic

Patient 1

Episode 1

Time after transplantation (yr)

Precipitating factors

Unknown

n e w e ng l a n d j o u r na l

n engl j med 361;14

Cyclosporine Cyclosporine (seImmunosuppressive Cyclosporine Methylprednisolone Methylprednisolone Tacrolimus (trough Tacrolimus (trough Tacrolimus (trough level, (trough level, rum level 2 hr therapy be(trough level, (0.43 mg/kg on (0.14 mg/kg on level, 10 ng/ml), level, 5.1 ng/ 5.8 ng/ml), <50 ng/ml), after injection, fore episode 63 ng/ml), azaalternate days), alternate days), azathioprine ml), methylmethylprednismethylprednis678 ng/ml), thioprine (1.1 cyclosporine tacrolimus (0.9 mg/kg/ prednisolone olone (0.09 olone (0.26 methylprednismg/kg/day), (trough level, (trough level, day) (0.15 mg/kg on mg/kg/day), mg/kg on alolone (0.26 methylprednis81 ng/ml) 11 ng/ml) alternate days), mycophenoternate days) mg/kg on alolone (0.1 mg/ mycophenolate late mofetil ternate days) kg/day) mofetil (1200 (1100 mg/ mg/m2/day) m2/day) Jaundice, pruritus Jaundice, pruritus Jaundice, pruritus Asthenia Pruritus Pruritus Jaundice, pruritus

nejm.org

Symptoms

Jaundice, pruritus

Markers of liver function 30.2 21.6 320 810 35 37 7 220 30 190 42 4.9 1.7 7.6 2.5 4.4 2.3 220 205 33 1.5 Not assessed 515 1415 38 At onset: multinu- At onset: multinucleated hepatocleated hepatocytes cytes No No No 0.9 0.2 150 360 16 Not assessed 2.5 1.7 110 150 7 9.4 7.0 280 300 21 At day 30: mild fi- At onset: multinubrosis, ductocleated hepatopenia cytes Not assessed No No

Bilirubin (mg/dl)

of

Total

october 1, 2009

Direct (conjugated)

m e dic i n e

AST (U/liter)

ALT (U/liter)

GGT (U/liter)

The New England Journal of Medicine Downloaded from nejm.org on December 17, 2011. For personal use only. No other uses without permission. Copyright 2009 Massachusetts Medical Society. All rights reserved. No No

Histologic features

At day 21: cholesta- At day 26: ductope- At onset: minimal sis, multinuclenia, multinuclecytolysis, norated hepatoated hepatomal ducts cytes cytes

Endotheliitis or bileduct damage from inflammation

Brief Report

* The assessment of antibodies and in vitro studies of bile acid transport were carried out with the use of serum samples obtained at the onset of cholestasis. EBV denotes EpsteinBarr virus. The liver-function data were measured at the time of the peak increase in bilirubin. Normal values are as follows: total bilirubin, less than 1.1 mg per deciliter (18.8 mol per liter); direct bilirubin, less than 0.3 mg per deciliter (5.1 mol per liter); aspartate aminotransferase (AST), less than 50 U per liter; alanine aminotransferase (ALT), less than 45 U per liter; and -glutamyl transpeptidase (GGT), less than 30 U per liter. To convert values for bilirubin to micromoles per liter, multiply by 17.1.

Switch to tacroli- Switch to tacroli- Corticosteroid bo- Maintained on tac- Tacrolimus (trough Tacrolimus (trough Corticosteroid bo- Switch to tacrolimus (trough mus, corticolevel, 9 ng/ml), luses (10 mg/ rolimus, cortilevel, 7 ng/ml), luses (10 mg/ mus, methyllevel, 15 ng/ steroid boluses corticosteroid kg/day for costeroids (1 methylpredniskg/day for 3 prednisolone ml), high-dose (10 mg/kg/day boluses (10 3 days) mg/kg/day), olone (1.3 mg/ days), switch to (0.44 mg/kg/ corticosteroids for 3 days), mymg/kg/day for mycophenolate kg/day for 15 tacrolimus plus day), myco(dose range, cophenolate 3 days), mycomofetil (820 days; then 0.17 azathioprine (1 phenolate 1.4 to 0.27 mg/ mofetil (750 phenolate mg/m2/day) mg/kg/day mg/kg/day) mofetil (580 2/day) kg/day), urmg/m mofetil (1200 thereafter), mg/m2/day) sodiol mg/m2/day) mycophenolate mofetil (1100 mg/m2/day)

in the presence and in the absence of 4 mM ATP, as previously described,13,14 with the use of a rapid filtration assay. For inhibition studies, membrane vesicles were preincubated at 37C for 30 minutes in the presence of serum samples from patients (dilution, 1:75) or 20 M cyclosporine.
In Vivo Studies

Normal at 4 mo

Normal at 3 mo

Disappearance of Normal at 4.5 mo jaundice at 4 mo

Four groups of three male rats (SpragueDawley rats, 4 weeks of age, weighing 60 to 65 g) were injected in the tail vein on two consecutive days with 0.2 ml of phosphate-buffered saline alone, phosphate-buffered saline containing 40 g of a rabbit polyclonal antibody against human BSEP (H-180, Santa Cruz Biotechnology), 10 mg of human IgG (Grifols), or filter-sterilized serum from Patient 2 at an antibody titer of 1:5210. Twentyfour hours after the last injection, we anesthetized the animals with isoflurane, collected specimens of their bile through a cannula placed in the common bile duct for 15 minutes, and then performed exsanguination through the ascending cava. We assayed total bile acids using the 3-hydroxysteroid dehydrogenase method15 and tested for differences among the four rat groups using Students t-test. Liver-tissue sections (thickness, 4 m) were incubated for 1 hour with secondary antibodies (AlexaFluor488-conjugated antirabbit IgG at a 1:100 dilution, or FITC-conjugated antihuman IgA, IgG, and IgM antibodies at a 1:20 dilution). The same sections were subsequently incubated with a mouse anti-MRP2 antibody (1:100 dilution) and antimouse AlexaFluor594-conjugated secondary antibodies (Molecular Probes). These procedures were approved by our institutional animal care and use committee.

Normal at 1 wk

Normal at 3 mo

Normal at 2 mo

Normal at 1 mo

R e sult s
Three children who had clinically diagnosed progressive familial intrahepatic cholestasis underwent liver transplantation involving a deceased donor. Several years later, they had episodes of graft dysfunction that mimicked the original disease. Mutational analysis of ABCB11, carried out in each of the children after transplantation (Table 1), and immunohistochemical studies of paraffin-embedded archival hepatectomy specimens (Fig. 1A) indicated that severe BSEP deficiency, a form of progressive familial intrahepatic cholestasis in which GGT activity is low,7,11 was the cause of the disease.
nejm.org october 1, 2009

Outcome of liverfunction testing

Treatment

n engl j med 361;14

1363

The New England Journal of Medicine Downloaded from nejm.org on December 17, 2011. For personal use only. No other uses without permission. Copyright 2009 Massachusetts Medical Society. All rights reserved.

The

n e w e ng l a n d j o u r na l

of

m e dic i n e

Episodes of pruritus or jaundice with pruritus were the first symptoms to occur after transplantation in each of the children. (See Table 2 for other clinical, biochemical, and histologic features of the patients disease.) The clinical course of each episode was characterized by elevated serum values of aminotransferases and bilirubin but normal GGT activity (Fig. 1B). Graft-biopsy specimens obtained during periods of such elevations did not show signs of acute rejection (e.g., inflammatory bile-duct damage or endothelialitis). Some of the biopsy specimens showed intrahepatic cholestasis and giant-cell transformation (Table 2 and Fig. 1C); on immunohistochemical analysis, the specimens had normal levels of BSEP expression. The onset of cholestasis was associated in some instances with a reduction of immunosuppressive therapy in response to infection with the Epstein Barr virus or food allergy. We could not ascertain the precipitating factor in four episodes of cholestasis. In all cases, an increase in immunosuppression, as detailed in Table 2, led to the resolution of symptoms. Serum samples obtained from the three children during the cholestatic attacks were tested for the presence of typical autoantibodies. None of the samples were found to be positive. However, all of them contained high titers of antibodies (range, 1:1280 to 1:5120) that specifically reacted with antigens in bile canaliculi, as assessed by means of indirect immunofluorescence (Fig. 1D). Serum specimens were available from Patients 2 and 3 when they were asymptomatic; in these samples, the antibody titer fluctuated between 1:40 and 1:160. The antibody titer was also monitored during the course of the fifth post-transplantation cholestatic episode in Patient 2. The titer gradually declined from 1:5120 to 1:320 concomitantly with the normalization of the patients clinical condition and liver chemical values (Fig. 1E). To test the hypothesis that these canalicular antibodies were directed against BSEP, serum samples from each of the three patients were analyzed by means of immunoblotting with membrane proteins from insect Sf9 cells expressing either human BSEP or human MRP1. A polyclonal antibody against human BSEP immunoreacted with a protein of approximately 140 kD in membrane lysates from BSEP-expressing Sf9 cells. All serum samples predominantly immunoreacted with a protein of the same molecular mass (Fig. 2A). This immunoreactive band was not visualized in control lysates from MRP1-expressing cells.
1364
n engl j med 361;14

To substantiate an association between cholestatic graft dysfunction and the presence of antiBSEP antibodies, we tested the serum samples from a control group of children (Patients 4, 5, and 6) who had undergone liver transplantation for severe BSEP deficiency and who did not have recurrence of the disease over a mean follow-up period of 5.5 years (see the Supplementary Appendix, available with the full text of this article at NEJM.org). Serum samples from two children who had undergone liver transplantation for extrahepatic biliary atresia and had been healthy for 5 years after transplantation were used as additional controls. None of the control samples were positive for anticanalicular antibodies (Fig. 1D) or anti-BSEP antibodies (Fig. 2A). We then determined whether circulating antiBSEP antibodies could interfere with BSEP function in grafts, by using an assay of bile acid transport with insideout membrane vesicles from BSEP-expressing Sf9 cells. The ATP-dependent vesicle uptake of 3H-labeled taurocholate (mean [SD], 27.51.7 pmol per milligram of protein per minute) was not affected by preincubation with serum samples from any of the controls but was completely inhibited by preincubation with cyclosporine, a competitive inhibitor of BSEP,13 and by preincubation with serum samples from each of Patients 1, 2, and 3 (Fig. 2B). We next tested the ability of these antibodies to reach the bile canaliculus and impair bile acid secretion in vivo. We intravenously injected rats with phosphate-buffered saline or with a rabbit antibody raised against human BSEP that crossreacts with rat Bsep. The rats were killed, and sections of their livers stained with fluorescent antirabbit secondary immunoglobulins. We found no specific staining of liver sections from rats injected with phosphate-buffered saline, whereas those from rats injected with the antibody against BSEP showed canalicular staining (Fig. 3A). Sequential incubation of the same liver sections with an antibody against the canalicular protein MRP2 further confirmed that the injected anti-BSEP antibody localized predominantly on the canalicular domains. Similarly, after injecting rats with serum from Patient 2 and then incubating liver sections with fluorescein-conjugated antihuman secondary antibodies, we observed staining of bile canaliculi, whereas there was no staining in sections from control animals injected with human IgG (Fig. 3A). Rats injected with anti-BSEP antibody or with serum from Patient 2 had biliary concentrations of
october 1, 2009

nejm.org

The New England Journal of Medicine Downloaded from nejm.org on December 17, 2011. For personal use only. No other uses without permission. Copyright 2009 Massachusetts Medical Society. All rights reserved.

Brief Report

Figure 2. Reactivity of Serum from the Patients to Human Bile Salt Export Pump (BSEP) and Blockade of Bile Acid Transport Function In Vitro. Panels A and B show the results of analyses of serum samples from children who had undergone liver transplantation for severe BSEP deficiency and who had disease recurrence after transplantation (Patients 1, 2, and 3) and from children without recurrence after transplantation (controls [Patients 4, 5, and 6]), as well as pooled serum samples from two children with extrahepatic biliary atresia (EHBA). Membrane vesicles (containing a total of 30 g of protein) from Sf9 cells expressing either human BSEP or human multidrug resistanceassociated protein 1 (MRP1), as a control, were fractionated through 10% sodium dodecyl sulfatepolyacrylamide-gel electrophoresis and subjected to Western-blot analysis (Panel A) with a goat antibody against human BSEP (dilution, 1:200) or with serum samples (dilution, 1:100) from patients and controls. Visualization was performed with the use of horseradish peroxidaseconjugated goat antihuman IgG or rabbit antigoat IgG. The arrowhead indicates the immunoreactive band obtained with use of the anti-BSEP antibody in membrane lysates from BSEP-expressing Sf9 cells. Panel B shows the results of analysis of ATPdependent 3H-labeled taurocholate uptake in membrane vesicles containing human BSEP in the presence of phosphate-buffered saline (PBS), cyclosporine (20 M), or serum samples from patients and controls (dilution, 1:75). The uptake of 3H-labeled taurocholate was measured by incubating samples for 10 minutes at 37C in the presence and absence of 4 mM ATP. Each data point represents the mean value of triplicate measurements. On average, the levels of cyclosporine in diluted serum samples were 1/20,000 of the levels in the control sample used to test for drug-induced inhibition of BSEP activity. T bars and I bars indicate standard deviations.

A
Sf9 Membrane Vesicles
BS EP BS E M P RP 1 BS E M P RP 1 BS EP M RP 1 BS E M P RP 1 BS EP M RP 1 BS E M P RP 1 BS E M P RP 1
1st

193 118 99

kD

C (P on at tr ie ol nt 4 Co ) (P n at tr ie ol nt 5 Co ) (P n at tr ie ol nt 6)
With ATP

tie nt 1

tie nt 2

tie nt 3

Pa

Pa

B
Without ATP 35

Taurocholate Uptake (pmol/mg protein/min)

30 25 20 15 10 5 0

3H-Labeled

Pa

Cy

clo

total bile acids that were 50.0 to 62.5% of the levels in the corresponding controls. Bile acid concentrations in serum from these animals were significantly greater than the values in serum from control rats (Fig. 3B).

Discussion
We describe late, post-transplantation phenotypic recurrence of severe BSEP deficiency in three children with an immune response against BSEP. The fact that anti-BSEP antibodies in one of these children bound the canalicular membrane of hepatocytes when injected into rats was not unexpected; it has been shown that IgG can be internalized by hepatocytes and even transcytosed from blood to bile.16,17 The modest reduction in biliary bile acid secretion seen in antibody-treated rats is consistent with the fact that the antibodies did not seem to
n engl j med 361;14

ICM be delivered to the FIGURE: liver. (We did not find whole 2 of 3 2nd REG F 3rd a uniform distribution of staining by these antiCASE Revised Line bodies amongEMail tissue sections.) Moreover,4-C con- SIZE in ARTIST: ts H/T H/T trast with humans, rodents have canalicular bile 22p3 Enon Combo acid transporters other than Bsep.18,19 NOTE: AUTHOR, PLEASE Our results supporthas been redrawn and phenotypic Figure the notion that type has been reset. Please check recurrence of severe BSEP deficiency carefully. from results an antibody-mediated blockade of BSEP in grafts. 10-01-09 JOB: 36114 ISSUE: This is consistent with findings derived from the sequential assessment of antibody titers during the course of one episode of cholestatic graft dysfunction, in which normalization of clinical conditions and the results of liver tests under an increasingly intense immunosuppressive regimen were accompanied by a gradual decrease in the antibody titer. Moreover, in the two children tested, the antibody titer in the interval between episodes was markedly lower than that during the cholestatic bouts. The affected children were not tested for antiBSEP antibodies before liver transplantation; all of them received transplants before ABCB11 was iden-

AUTHOR: Jara

nejm.org

october 1, 2009

The New England Journal of Medicine Downloaded from nejm.org on December 17, 2011. For personal use only. No other uses without permission. Copyright 2009 Massachusetts Medical Society. All rights reserved.

3 C (P on at tr ie ol nt Co 4) (P n at tr ie ol n C t5 (P on ) at tr ie ol nt 6) EH BA
RETAKE

PB

in

nt

nt

or

tie

tie

sp

Pa

Pa

Pa

tie

nt

EH

BA

1365

The

n e w e ng l a n d j o u r na l

of

m e dic i n e

A
Injection
PBS Anti-BSEP antibody Human IgG Serum from Patient 2

None

Primary Antibody

MRP2

B
3500 P=0.009 P=0.04 70 P=0.03 P=0.009

2500 2000 1500 1000 500 0

Serum Bile Acids (mol/liter)

S tiAn BSE tib P od y H um an Ig G Se ru m Pa fro tie m nt 2

Biliary Bile Acids (mol/liter)

3000

60 50 40 30 20 10 0

Figure 3. Deposition of AntiBile Salt Export Pump (BSEP) Antibodies or Serum from Patients on the Bile Canaliculi and Impairment of Biliary Secretion of Bile Acids. Jara 1st RETAKE AUTHOR: ICM 2nd Rats were injected intravenously on each FIGURE: 3 of 3 of 2 consecutive days with a rabbit anti-human BSEP antibody (three rats) REG F 3rd or with serum from Patient 2 (three rats). Phosphate-buffered saline (PBS)Revisedhuman IgG were administered as reand CASE spective controls (three rats per group). Bile, serum, and liver samples were collected 24 hours after the last injec4-C Line EMail SIZE ARTIST: first stained withH/T tion. Panel A shows liver-tissue sections that werets anti-rabbit or anti-human secondary antibodies H/T 33p9 Enon Combomultidrug resistanceassociated protein 2 (MRP2) (green) and were then incubated with a mouse antibody against followed by antimouse AlexaFluor594conjugated secondary antibodies (red). Liver specimens from antibody-treatAUTHOR, PLEASE NOTE: Figure products colocalized with been reset. ed rats show immunofluorescence-reaction has been redrawn and type hasthe canalicular marker MRP2. Representative Please check carefully. immunofluorescence images of liver sections from each group are also shown. Panel B shows the mean total bile acid concentrations in bile and serum specimens from the rats. T bars indicate standard deviations.
JOB: 36114 ISSUE: 10-01-09

tified as the gene causing this disorder. Anti-BSEP antibodies are unlikely to account for the disease in the native liver. In the three patients, the original disease can be accurately ascribed to a direct defect in BSEP, as determined by sequence analysis and immunohistochemical studies. The antibodies probably developed after liver transplantation, in response to foreign BSEP acquired with the graft. Epitopes present on wild-type BSEP
1366

introduced with the graft seem likely to be recognized as new antigenic determinants by the immune system of the recipient. None of the mutations present in Patients 1, 2, and 3 have been found in other children with severe BSEP deficiency.5 There are a few reports of a humoral response, after solid-organ transplantation, against the disease-causing gene product. Circulating antibodies

n engl j med 361;14

nejm.org

october 1, 2009

The New England Journal of Medicine Downloaded from nejm.org on December 17, 2011. For personal use only. No other uses without permission. Copyright 2009 Massachusetts Medical Society. All rights reserved.

PB S tiBS An E tib P od y H um an Ig G Se ru m Pa fro tie m nt 2 An

An

PB

Brief Report

against dystrophin were detected in a patient with Beckers muscular dystrophy (caused by mutations of the dystrophin gene); the patient underwent cardiac transplantation and then had graft failure.20 In a minority of patients who undergo kidney transplantation for Alports syndrome, a hereditary disease associated with mutations in one or several subunits of type IV collagen of the basement membranes, the presence of post-transplantation antibodies against the normal type IV collagen of the graft correlates with the development of glomerulonephritis.21 Our current findings constitute a rare phenomenon among liver diseases for which transplanReferences
1. Balistreri WF. Liver disease in infancy

tation is indicated in the pediatric population, because such diseases rarely recur after transplantation. We recommend that children who are undergoing liver transplantation for BSEP deficiency be checked for circulating anti-BSEP antibodies. If the patients are positive for the antibodies, special attention must be given to the management of immunosuppressive therapy.
Supported by a grant from the Fondo de Investigaciones Sanitarias (05/2373). No potential conflict of interest relevant to this article was reported. We thank Sara Andueza, Pablo Lorente, and Delia Aguado for excellent technical support and Dr. Francisco Hernndez for valuable assistance with rat surgery.

and childhood. In: Schiff ER, Sorrell MF, Maddrey WC, eds. Schiffs diseases of the liver. 8th ed. Philadelphia: LippincottRaven, 1999:1357-512. 2. Migliazza L, Lpez-Santamara M, Murcia J, et al. Long-term survival expectancy after liver transplantation in children. J Pediatr Surg 2000;35:5-8. 3. Tiao G, Alonso MH, Ryckman FC. Liver transplantation in children. In: Suchy F, Sokol RJ, Balistreri WF, eds. Liver disease in children. 3rd ed. New York: Cambridge University Press, 2007:975-93. 4. Strautnieks SS, Bull LN, Knisely AS, et al. A gene encoding a liver-specific ABC transporter is mutated in progressive familial intrahepatic cholestasis. Nat Genet 1998;20:233-8. 5. Strautnieks SS, Byrne JA, Pawlikowska L, et al. Severe bile salt export pump deficiency: 82 different ABCB11 mutations in 109 families. Gastroenterology 2008; 134:1203-14. 6. Stieger B, Meier Y, Meier PJ. The bile salt export pump. Pflugers Arch 2007; 453:611-20. 7. Thompson R, Strautnieks S. BSEP: function and role in progressive familial intrahepatic cholestasis. Semin Liver Dis 2001;21:545-50. 8. Knisely AS, Strautnieks SS, Meier Y, et al. Hepatocellular carcinoma in ten chil-

dren under five years of age with bile salt export pump deficiency. Hepatology 2006; 44:478-86. 9. Whitington PF, Freese DK, Alonso EM, Schwarzenberg SJ, Sharp HL. Clinical and biochemical findings in progressive familial intrahepatic cholestasis. J Pediatr Gastroenterol Nutr 1994;18:134-41. 10. Bull LN, Carlton VE, Stricker NL, et al. Genetic and morphological findings in progressive familial intrahepatic cholestasis (Byler disease [PFIC-1] and Byler syndrome): evidence for heterogeneity. Hepatology 1997;26:155-64. 11. Alissa FT, Jaffe R, Shneider BL. Update on progressive familial intrahepatic cholestasis. J Pediatr Gastroenterol Nutr 2008;46:241-52. 12. Alvarez L, Jara P, Snchez-Sabat E, et al. Reduced hepatic expression of farnesoid X receptor in hereditary cholestasis associated to mutation in ATP8B1. Hum Mol Genet 2004;13:2451-60. 13. Byrne JA, Strautnieks SS, Mieli-Vergani G, Higgins CF, Linton KJ, Thompson RJ. The human bile salt export pump: characterization of substrate specificity and identification of inhibitors. Gastroenterology 2002;123:1649-58. 14. No J, Stieger B, Meier PJ. Functional expression of the canalicular bile salt export pump of human liver. Gastroenterology 2002;123:1659-66.

15. Talalay P. Enzymic analysis of steroid

hormones. Methods Biochem Anal 1960;8: 119-43. 16. Telleman P, Junghans RP. The role of the Brambell receptor (FcRB) in liver: protection of endocytosed immunoglobulin G (IgG) from catabolism in hepatocytes rather than transport of IgG to bile. Immunology 2000;100:245-51. 17. Blumberg RS, Koss T, Story CM, et al. A major histocompatibility complex class I-related Fc receptor for IgG on rat hepatocytes. J Clin Invest 1995;95:2397-402. 18. Wang R, Salem M, Yousef IM, et al. Targeted inactivation of sister of P-glycoprotein gene (spgp) in mice results in nonprogressive but persistent intrahepatic cholestasis. Proc Natl Acad Sci U S A 2001;98:2011-6. 19. Lam P, Wang R, Ling V. Bile acid transport in sister of P-glycoprotein (ABCB11) knockout mice. Biochemistry 2005;44:12598-605. 20. Bittner RE, Shorny S, Streubel B, Hbner C, Voit T, Kress W. Serum antibodies to the deleted dystrophin sequence after cardiac transplantation in a patient with Beckers muscular dystrophy. N Engl J Med 1995;333:732-3. 21. Kashtan CE. Renal transplantation in patients with Alport syndrome. Pediatr Transplant 2006;10:651-7.
Copyright 2009 Massachusetts Medical Society.

n engl j med 361;14

nejm.org

october 1, 2009

1367

The New England Journal of Medicine Downloaded from nejm.org on December 17, 2011. For personal use only. No other uses without permission. Copyright 2009 Massachusetts Medical Society. All rights reserved.