- Email: [email protected]
After Intestinal Transplantation Kidney Function Is Impaired by Downregulation of Epithelial Ion Transporters in the Ileum
After Intestinal Transplantation Kidney Function Is Impaired by Downregulation of Epithelial Ion Transporters in the Ileum J. Reinera,b, C.-J. Hsieha, C. Straarupb, P. Bodammerb, H. Schäfflerb, F. Graeplera, D. Stükerc, T. Krattc, M. Linnebacherd, S. Nadalinc, M. Wittec,d, A. Königsrainerc, and G. Lamprechta,b,* a First Medical Department, University of Tübingen, Tübingen, bDivision of Gastroenterology and Endocrinology, University Medical Center Rostock, Rostock, cDepartment of General, Visceral and Transplantation Surgery, University of Tübingen, Tübingen, and d Department of General, Thoracic, Vascular and Transplantation Surgery, University Medical Center Rostock, Rostock, Germany
ABSTRACT Background. Intestinal transplantation is a treatment option for intestinal failure. Although nephrotoxic medication after transplantation is a major cause for posttransplant renal insufficiency, it remains unclear why kidney dysfunction is particularly frequent after intestinal transplantation. Methods. This study analyzed messenger RNA expression of NHE3, DRA, and CFTR in 404 biopsies obtained between day 2 and 1508 from the terminal ileum of 10 adult intestinal transplant recipients. Results. The time courses of immunosuppression and glomerular filtration rate were correlated. In the first posttransplant year, expression of NHE3 and DRA, which mediate NaCl absorption, was diminished to a greater degree than that of CFTR, which mediates chloride secretion. Reduced NHE3 and DRA expression was associated with high tacrolimus trough levels. Titration of tacrolimus to low levels by year 2 was paralleled by partially restored NHE3 and DRA expression. In cell culture experiments, similar effects of tacrolimus on transporter expression were detected. In patients, both reduced tacrolimus levels and recovery of NHE3 and DRA expression were associated with stabilization of renal function. Conclusions. Our data strongly suggest that tacrolimus impairs absorption of NaCl and water from the transplanted ileum, leading to volume depletion and impaired renal function. This may be reversible by reduction of tacrolimus to lower levels without increased rates of rejection or chronic graft failure.
I
NTESTINAL OR MULTIVISCERAL TRANSPLANTATION has become a treatment option for intestinal failure if home parenteral nutrition fails [1]. Usually, transplanted patients achieve oral autonomy with regard to maintaining proteineenergy requirements as well as volume and electrolyte homeostasis. Currently, standard immunosuppressive regimens after intestinal or multivisceral transplantation aim to achieve tacrolimus “monotherapy” plus low-dose prednisolone [2]. Protocol biopsies are obtained from the transplanted ileum to rule out rejection frequently early after transplantation and at more prolonged intervals later in the course. In the absence of rejection, the transplanted intestine has normal endoscopic appearance and shows normal mucosal histology [3]. Despite that, intestinal ª 2016 by Elsevier Inc. All rights reserved. 360 Park Avenue South, New York, NY 10010-1710
Transplantation Proceedings, 48, 499e506 (2016)
fluid and electrolyte losses can be high early and occasionally also late after transplantation. Together with tacrolimus and other drug-induced nephrotoxicity, these losses put the patient at risk for impaired kidney function [4]. In addition, even pretransplant kidney function may be impaired owing to cycled total parenteral nutrition and repeated exposure to nephrotoxic drugs, which may not be reflected by creatinine or compensated by continuous fluid supplementation [5,6].
*Address correspondence to G. Lamprecht, Division of Gastroenterology and Endocrinology, University Medical Center Rostock, Ernst-Heydemann-Str. 6, 18057 Rostock, Germany. E-mail: [email protected] 0041-1345/16 http://dx.doi.org/10.1016/j.transproceed.2015.12.068
499
500
Indeed, renal function is impaired more often after intestinal or multivisceral than after other nonrenal solid organ transplantations [7e9]. Thus, water and electrolyte absorption from the transplanted intestine are of utmost importance for the long-term prognosis of these patients. Despite that, expression of the ion transporters that mediate intestinal NaCl absorption, chloride secretion, and subsequent water transport has not been studied in the transplanted ileum. The intestinal epithelium is composed of a single layer of highly specialized columnar cells. Transepithelial movement of solutes can occur either via the transcellular pathway by traversing the apical and basal membrane or the paracellular pathway across intercellular junctions. The apical membrane of epithelial intestinal cells comprising the primary barrier to transcellular flux is composed of a vast diversity of membrane-associated proteins, including ion transporters and ion channels as the major effectors of transcellular ion transport. Two ion exchangers, NHE3 and DRA (also known as CLD, chloride anion exchanger) facilitating Na/H and Cl/HCO3 exchange, are responsible for electroneutral NaCl absorption in the human ileum and proximal colon [10]. Their quantitative importance has been documented in knockout mice, which display chronic diarrhea [11,12]. A genetic defect in DRA function also leads to the human disease congenital chloride diarrhea [13]. In addition, dysregulation of NHE3 [14,15] as well as DRA [16] has been associated with inflammatory processes in the intestinal mucosa. Furthermore, CFTR is expressed in the human ileum and more so in the colon [17]. CFTR mediates chloride secretion and (over)stimulation of CFTR by toxins leads to secretory diarrhea [18]. Under normal conditions, the activity of NHE3 and DRA on the one side and CFTR on the other side, seem to be balanced in the healthy ileum leading to net electroneutral NaCl absorption. One conceivable explanation for frequent renal dysfunction in intestinal transplant recipients is impaired intestinal absorption of NaCl and water with subsequent development of renal failure owing to poor volume status. To better understand the mechanisms of impaired renal function, nephrotoxic immunosuppression, and intestinal ion and water absorption, we have studied messenger RNA (mRNA) expression of NHE3, DRA, and CFTR in the transplanted ileum. Furthermore, we have correlated these data along the time course after transplantation with tacrolimus trough levels and prednisolone dose as well as with the calculated glomerular filtration rate (GFR). MATERIALS AND METHODS Between 2006 and 2012, 10 adult patients were transplanted at the University of Tübingen because of intestinal failure with a benign underlying etiology and because of failing home parenteral nutrition. One additional biopsy was taken from the transplanted terminal ileum whenever routine surveillance or clinically indicated biopsies were taken. Protocol biopsies were obtained 3 times per week in the first month, 1 to 2 times per week in the next 2 months, and once a month for the rest of the first posttransplant year similar to other programs [19,20]. Thereafter, routine biopsies were taken
REINER, HSIEH, STRAARUP ET AL once every 3 months. Of note, only about 10% of all biopsies were taken outside this schedule for clinical indications. In the same way, ileal biopsies were obtained from 7 healthy controls undergoing screening colonoscopy for colorectal cancer. The study was approved by the local institutional review board (Ethikkommission der Universität Tübingen 022/2011BO2) and all patients gave written informed consent. Ileal biopsies were immediately transferred into RNAlater (Ambion, Foster City, CA) and stored at 4 C for preservation until further use. RNA was isolated using the RNAqueous kit (Ambion). The integrity of the isolated RNA was confirmed by agarose gel electrophoresis and the concentration of isolated RNA was determined using a spectrophotometer (Nano Drop 2000; Peq Lab Biotechnologie GmbH, Erlangen, Germany). cDNA was synthesized from approximately 2 mg mRNA using RevertAid H Minus First Strand cDNA Synthesis Kit (ThermoScientific, Waltham, MA) according to the recommendations of the manufacturer. For quantitative reverse transcriptase polymerase chain reaction (RT-PCR), the following primers were selected using Primer-Express 2.0-Software (Applied Biosystems, Foster City, CA, USA): H3F3A: sense: 50 -TTCCAGAGCGCAGCTATCG-30 ; antisense: 50 -TCTTCAAAAAGGCCAACCAGAT-30 . DRA: sense: 50 -GTGTCCTTTCTTGATGTTTCTTCAGT-30 ; antisense: 50 -CGGTTAAGCTTCTCAATGAAGTCA-30 . CFTR: sense: 50 -GGAAAAGGCCAGCGTTGTC-30 ; antisense: 50 -CCAGGCGCTGTCTGTATCCT-30 . NHE3: sense: 50 -CTGTCCCTCTACGGCGTCTT-30 ; antisense: 50 -GCTGCCAAACAGGAGGAAGTC-30 . Extensive primer optimization using cDNA clones as templates proved that all primers facilitated optimal and identical PCR amplification conditions. Quantitative RT-PCR (qRT-PCR) was carried out on an ABI Prism 7000 Real Time PCR instrument (Applied Biosystems) using Sequence Detection System Software V. 1.2.3. (Applied Biosystems). Initial steps of 50 C of 2 minutes and 95 C for 10 minutes were followed by 40 cycles of amplification at 95 C for 15 seconds and 60 C for 60 seconds. qRT-PCR for each biopsy was performed in triplicates and the mean was calculated. Obtained Ct values were normalized to the amplification of H3F3A and DCt values from transplanted ilea were then compared with mean DCt values from healthy controls (DDCt) and are displayed as 2DDCt.
Collection of Clinical Data Data from transplanted patients were collected retrospectively from clinical charts and entered into a database according to the date of biopsy. The estimated GFR (eGFR) was calculated using the Modification of Diet in Renal Disease Study group’s equation [21].
Cell Culture Experiments HROC113 cells [22] were kept in DMEM/Ham-F12 medium supplemented with 5% fetal bovine serum and penicillin/streptomycin, and seeded into tissue culture treated 6-well plates (Corning, Corning, NY) between subculture number 49 and 57. Cells were grown to confluence and let differentiate for another 4 days. Cells were quickly rinsed with phosphate-buffered saline before undergoing Trizol RNA extraction. qPCR was carried out as described using ViiA 7 Seq Detection.
Statistical Analysis All real-time qPCR data were fed into a database and linked to clinical data that were recorded according to dates of biopsy. Queries were used to compare single parameters to mRNA levels
EPITHELIAL ION TRANSPORTERS
501
using MS Access, Excel and JMP 11.1.1 (SAS, Inc, Cary, NC). P values are listed when a statistical test was performed.
RESULTS Patient Cohort
All 10 patients included in this study were Caucasian adults of northern European ancestry with a benign underlying etiology of their intestinal failure who underwent intestinal or multivisceral transplantation at the transplantation unit Tübingen between 2006 and 2011. The indication for transplantation was failure or impeding failure of home parenteral nutrition [2,23]. Quality of life was not regarded as an indication for transplantation [2]. Demographics of the cohort as well as specific indication, type of transplant, details of the immunosuppression, and outcome are shown in Table 1. The demographic data resembled those from other larger programs. With regard to immunosuppression, all patients received induction with either ATG (200 mg), thymoglobulin (2x 75 mg), basiliximab (2x 20 mg) or alemtuzumab (20 mg). Furthermore, all patients had an initial immunosuppressive regime that contained tacrolimus (with trough Table 1. Characteristics of the Study Cohort Transplanted Between 2006 and 2012 Characteristics
Value
Patients Male/female Age at transplantation (y) HPN before transplant (mo) Follow-up (d), mean Underlying etiology of short bowel syndrome (Crohn’s/volvulus/ ischemia/other resection) Pretransplant renal failure NOD2 mutation (hetero-/homozygous) Indication for transplantation Loss of central venous access Recurrent sepsis Intestinal failure-associated liver disease Type of graft Isolated intestine Multivisceral Including right sided colon Immunosuppression Induction (alemtuzumab/antithymocyte globulin/basiliximab) Tacrolimus maintenance Prednisolone Early mycophenolate Intermittent mammalian target of rapamycin inhibitor Outcome Biopsy-proven rejection Graft loss Death
10 6/4 50.4 7.49* 129 (17e363)† 787 (47e1508)† 6/1/2/1
*Standard deviation. † Range.
4 2/1 3 2 5 6 4 4 2/3/3 10 10 6 5
1 2 1
levels of 10e20 ng/ml immediately after transplantation) and high dose prednisolone [24]. 6/10 patients received mycophenolate mofetil (Cellcept). Mammalian target of rapamycin inhibitors (rapamycin or everolimus) were only used intermittently in 5 patients in an attempt to decrease nephrotoxicity, but were discontinued because of lack of beneficial effect on kidney function or because of additional unwanted side effects. The time course of the immunosuppression in individual patients is depicted in Fig 1. As can be seen, at 1 year after transplantation, all patients achieved a comparably simple immunosuppression consisting of tacrolimus and low-dose prednisolone with 2 patients intermittently taking rapamycin. Target trough levels for tacrolimus were 15 to 20 ng/mL in the first 3 months, 12 to 15 ng/mL in the following 3 months, and <12 ng/mL thereafter. All patients achieved oral autonomy and all intravenous fluid supplementation was discontinued at the time of discharge. Patients were educated to keep their volume status balanced, which was evaluated at each outpatient visit. In no case was scheduled home parenteral fluid supplementation reinitiated after hospital discharge. Expression of Intestinal Ion Transport Proteins NHE3, DRA, and CFTR After Intestinal Transplantation
To address whether ion transport is altered in the transplanted ileal epithelium after intestinal transplantation, qRT-PCR was used to detect changes of cellular mRNA of the major epithelial ion transporters NHE3, DRA, and CFTR. Expression levels of mRNA in transplanted individuals were compared with a control group of 7 healthy individuals (mean age; 47.2 years; range, 17e77) undergoing screening ileocolonoscopy. DCt values from the control group were 1.18 0.33 for NHE3, 2.05 0.17 for DRA, and 4.36 0.72 for CFTR (mean standard deviation). The expression of intestinal ion transporters after transplantation is displayed in Fig 2. Fig 2, AeC shows the relative expression compared with healthy controls of NHE3, DRA, and CFTR in individual patients at individual time points. Early after transplantation, the expression of all 3 transporters varied substantially. In the individual patients, this phase lasted 4e12 months, reflecting their early clinical course. Thereafter, mRNA expression became more stable. The expression data are summarized in Fig 2D. During the first year after transplantation, expression of NHE3 (21.9%), DRA (43.0%), and CFTR (48.0%) was strongly decreased. During years 2 and 3 after transplantation, expression showed both less variation and NHE3 and DRA recovered to significantly higher expression levels compared with year 1 (NHE3, 58.9% vs 21.9% [P < .05]; DRA, 59.5% vs 43.0% [P < .05]); for CFTR, the recovery was not significant (53.8% vs 48.0%). Nevertheless, gene expression of all 3 ion transporters remained diminished compared with healthy controls. Because the data showed substantial variation, the question arose as to whether the observed reduction of expression levels represented independent downregulation of individual ion transporters or common changes of
502
REINER, HSIEH, STRAARUP ET AL
prednisolone [mg/d]
A
30 30 30 30 30 30 30 30 30 30
C
D
mycophenolate mofetil [mg/d]
B
mTOR-Inhibitor trough [ng/ml] tacrolimus trough levels [ng/ml]
0
200
400
600
800
1000
1200
1400
time [days] 20 20 20 20 20 20 20 20 20 20
transporter abundance in the epithelium. To differentiate between these two possibilities, regression analysis was performed (Fig 2E). Expression levels of all 3 ion transporters correlated strongly and significantly with each other. NHE3 and DRA correlated slightly stronger with each other (rS ¼ 0.7185; P < .0001) than with CFTR (NHE3 vs CFTR, rS ¼ 0.6699 [P < .0001]; DRA vs CFTR, rS ¼ 0.5143 [P < .0001]). The strong correlation suggests that expression of these 3 transporters was commonly affected; however, NHE3 and DRA, which govern salt and water absorption, were more vulnerable in year 1 compared with CFTR, which is involved in water secretion. Intestinal Ion Transport, Tacrolimus and Kidney Function
0
200
400
600
800
1000
1200
1400
1000
1200
1400
1000
1200
time [days] 10 10 10 10 10 10 10 10 10 10 0
0
200
400
600
800
time [days] 2 2 2 2 2 2 2 2 2 2 0
0
200
400
600
800
time [days]
1400
Fig 1. Immunosuppression over time. Dose or trough levels of immunosuppressive agents are shown. Patients are displayed in chronological order by date of transplantation. Graft loss occurred in patient 1 (chronic graft failure; graft enterectomy at day 1315) and in patient 5 (prolonged acute rejection and sepsis; graft enterectomy at day 119). Patient 10 died owing to graft-versus-host disease on day 47. (A) Long-term administration of prednisolone. Initially, a highdose methylprednisolone regimen together with a T-cell antibody was given (not shown here for reasons of clarity). (B) Tacrolimus trough levels. All patients received tacrolimus as basic immunosuppressant. Initially, trough levels of tacrolimus showed substantial variation but became more stable over time. (C) Mammalian target of rapamycin (mTOR) inhibitors were given intermittently to decrease nephrotoxicity. Dark grey, everolimus; light grey, sirolimus. (D) Mycophenolate mofetil was given initially in 6 patients and intermittently when indicated.
To address the mechanisms that lead to the high incidence of impaired kidney function after intestinal transplantation, the calculated eGFR was assessed as a marker for kidney function and the corresponding tacrolimus trough levels were recorded (Fig 3). Along the time axis after transplantation, these were together put into context with the reduced expression of ion transporters. As expected, the eGFR was somewhat reduced before transplantation in these severely ill patients. Despite high tacrolimus trough levels early after transplantation, eGFR was only moderately reduced between months 0 and 6. However, the majority of our patients had prolonged hospitalization and the eGFR calculated by the Modification of Diet in Renal Disease formula tends to overestimate the true GFR in these malnourished patients, especially when intravenous fluid supplementation is given in the hospital [4]. Tracer studies of kidney function, which analyze the true GFR more precisely, were not performed. After month 6 posttransplantation, that is, after discharge, the eGFR was significantly reduced. Because of the declining kidney function, tacrolimus levels were reduced. Interestingly, eGFR remained stable thereafter and did not decline further (Fig 3). Although tacrolimus levels were high, the eGFR decreased in association with strongly reduced NHE3 and DRA expression. CFTR expression on the other hand was reduced to a lesser degree (Figs 2 and 3). Remarkably, reduced tacrolimus levels correlated with partial recovery of NHE3 and DRA expression to a comparable level as the stably reduced CFTR expression. Thus, a toxic effect of tacrolimus on epithelial ion transport, predominantly affecting salt and water absorbing transporters, seemed to be possible as an explanation. To this end, tacrolimus toxicity was tested in a recently characterized cell line, HROC113, which expresses similar mRNA levels of NHE3, DRA, and even higher levels of CFTR compared with biopsies from human ileum (unpublished data). In vitro, tacrolimus treatment of confluent HROC113 cells decreased mRNA expression of both NHE3 and DRA in a dose-dependent manner (Fig 4). Although low tacrolimus concentrations (2 ng/mL) did not alter expression levels
EPITHELIAL ION TRANSPORTERS
503
Fig 2. Expression of epithelial ion transport proteins in the transplanted small bowel. (AeC) Time course of messenger RNA (mRNA) expression levels in individual patients. Different patients are denoted in different colors. Note logarithmic scale of the y axis. (D) Summary of mRNA expression of NHE3, DRA, and CFTR in transplanted ilea. Out of 404 biopsies, RNA was recovered successfully from 397 specimens and further analyzed. Data are normalized to the expression in healthy controls. The shaded area denotes the variation in healthy controls. Within the first year after transplantation, expression of NHE3 (21.9%), DRA (43.0%), and CFTR about 48% (D) was strongly decreased. After 1 year, expression showed both less variation and recovery to higher expression levels (NHE3, 58.9%; DRA, 59.5%; CFTR, 53.8 %; median standard error of the mean). (E) Regression analysis of ion transporter expression shows strong and highly significant (P < .0001) correlation between NHE3 and DRA as wells a less prominent but equally significant correlations between NHE3 or DRA with CFTR.
significantly, high concentrations, mimicking in vivo tissue levels after transplantation, repressed mRNA expression clearly. CFTR expression remained largely unaffected in these in vitro studies. DISCUSSION
Over the last decades, intestinal transplantation has been optimized with regard to surgical procedures as well as with regard to immunosuppression regimens and thus survival has improved significantly [2]. However, kidney function has
remained a major problem after solid organ transplantation in general and after intestinal transplantation in particular [4,7e9,25,26]. The reasons behind the fact that kidney function is more often impaired after intestinal transplantation than after transplantation of other organs has not been addressed directly. Long-standing volume depletion before transplantation, even despite optimized parenteral support, continued volume depletion after intestinal transplantation because of ongoing intestinal losses and finally intense immunosuppression are all likely to play a role [6,8,9,27,28].
504
REINER, HSIEH, STRAARUP ET AL
140
20
GFR [ml/min]
100
18 16 14 12
80
10 60
8 6
40
4
tacrolimus trough levels [ng/ml]
GFR tacrolimus
120
opposed to CFTR was also detected in vitro using a colonic cancer cell line. Our data strongly suggest that the net effect of differential inhibition of NHE3 and DRA on the one side and of CFTR on the other is reduced salt and water absorption. In patients, this translates into latent intestinal salt and fluid losses, latent volume depletion and subsequently impaired kidney function. The low expression levels of the transporters NHE3 and DRA early after intestinal or multivisceral transplantation were accompanied by high tacrolimus trough levels and a decline of renal function. The decline in renal function
A
20 2
-3 6 31
-3 0 25
-2 4 19
-1 8 13
12 7-
6
0
0-
Pr eTx
0
time [months] Fig 3. Glomerular filtration rate (GFR) and tacrolimus trough levels after intestinal transplantation. Mean calculated GFR (Modification of Diet in End-stage Renal Disease method) over time in red on the left axis. Mean tacrolimus trough levels in blue on the right axis. Pre-tx, pretransplant.
In this context, the general function and the specific (patho)physiologic role of the transplanted intestinal epithelium have not been studied extensively. One major function of the intestinal epithelium is transepithelial NaCl transport, which governs water absorption [10]. NHE3 and DRA are 2 functionally coupled ion exchangers that mediate electroneutral NaCl absorption [12,29e31]. Unlike these 2, CFTR is an anion channel that mediates electrogenic chloride secretion [18]. It is feasible to study these transporters because they are expressed in the ileum (and proximal colon), from which biopsies are routinely obtained for rejection surveillance. The expression of all 3 studied transporters varied greatly at low levels, particularly within the first months through 1 year after transplantation, reflecting the individual and sometimes difficult early posttransplant course. Subsequently, the expression levels became more stable at about 60% of the expression in healthy controls. Despite this variation, mRNA expression of the transporters correlated strongly with each other in the biopsies and thus at each time point (Fig 2E). This suggests that, after intestinal transplantation, not 1 individual transporter or 1 functionally coupled set of transporters is disturbed, but that transporter abundance in the epithelium is impaired in general. Expression of NHE3 and DRA was impaired to a greater degree than CFTR in the biopsies. The more pronounced effect of tacrolimus on NHE3 and DRA as
B
C
Fig 4. Effect of tacrolimus on expression of ion transporters in HROC113. Messenger RNA expression levels of NHE3 (A), DRA (B), and CFTR (C) were determined by real-time quantitative polymerase chain reaction analysis and compared with H3F3A. DDCt-values were based on controldDCtdvalues without tacrolimus. n ¼ 3 per condition.
EPITHELIAL ION TRANSPORTERS
became apparent only 6 to 12 months after transplantation, most likely because it was compensated by intravenous fluid administration as long as the patients were admitted to the hospital. Because of declining renal function, tacrolimus was titrated to lower trough levels after the first year post transplant. In some cases, we also tried to substitute tacrolimus with a mammalian target of rapamycin inhibitor to reduce nephrotoxicity [24,32] (Fig 1C). However, this was in general not successful. Nevertheless, tacrolimus dose reduction was associated with both a stabilization of renal function and a recovery of NHE3 and DRA to about the same level as CFTR allowing for a more balanced volume status. Beside tacrolimus, many other uncontrolled factors may negatively affect the expression of intestinal ion transporters. Thus, we addressed directly a potential inhibitory effect of tacrolimus on intestinal epithelial cells. Because other currently available cell lines do not mimic physiologic intestinal expression levels of NHE3, DRA, and CFTR, we used HROC113 cells, a recently characterized human cell line, which expresses these transporters at levels similar to healthy ileal epithelium. Intriguingly, incubation of these cells with tacrolimus at concentrations comparable with levels achieved in patients, resulted in a dose-dependent reduction of NHE3 and DRA mRNA, while CFTR expression remained largely unaffected. Thus, after intestinal transplantation, tacrolimus may not only exert its adverse renal effects by direct toxicity on the kidney, but also by interfering with intestinal NaCl transport resulting in insufficient hydration and latent volume depletion. It should be noted that tacrolimus also compromises nutrientcoupled ion transport [33] and intestinal barrier function [34], both of which may negatively affect volume status as well. Recently, Yang et al [35] showed that high levels of rapamycin reduce intestinal NHE3 expression in animal models and kidney transplant patients through a mechanism that involves increased autophagy. In addition, CFTR protein expression in the rat ileum was unaffected by rapamycin treatment. Thus, tacrolimus and rapamycin seem to impair intestinal NaCl absorption in a similar manner. The possibility of a common molecular mechanism, however, remains open, because it is not known whether tacrolimus also interferes with autophagy in the enterocyte [36,37]. Because they observed the effect of rapamycin on autophagy in renal transplant recipients, our findings may principally be applicable to the pathogenesis of impaired renal function in other solid organs transplantations as well, but they may be more pronounced after intestinal transplantation because the transplanted organ and the target organ are the same. To some extent, we were forced to reduce tacrolimus trough levels to preserve kidney function. It may be argued that this would lead to an higher incidence of episodes of acute cellular rejection or to chronic graft dysfunction and eventually graft failure [38]. But even despite comparably mild immunosuppression and relatively low tacrolimus levels, we did not see biopsy-proven acute cellular rejection
505
later than 6 months after transplantation. Also, chronic graft dysfunction was not clinically evident except for patient 1 (starting around day 1100). Thus, in the clinical context, careful reduction of tacrolimus reduced the degree of nephrotoxicity and was associated with increased expression of NHE3 and DRA without compromising graft tolerance or long-term graft function. Although qRT-PCR is a powerful quantitative approach to detect changes of cellular RNA, a limitation of this study is that mRNA levels cannot make a point about protein expression [39]. Furthermore, the cellular as well as the subcellular localization of NHE3 and DRA as well as CFTR may be altered by tacrolimus as well as by other uncontrolled factors after intestinal transplantation [40]. Taken together, our data suggest that relatively low tacrolimus levels are not only acceptable after intestinal transplantation, but may even be aimed for. In this small clinical cohort, graft tolerance and long-term organ function were not compromised by lower levels of immunosuppression. Under such conditions, the strong impairment of intestinal ion transporter expression is largely reversed and expression levels reach about 60% of healthy controls. This likely contributes to clinically apparent less latent volume depletion. Together with a lesser degree of direct tacrolimus inherent nephrotoxicity, this may facilitate stabilization of renal function.
REFERENCES [1] O’Keefe SJ, Buchman AL, Fishbein TM, et al. Short bowel syndrome and intestinal failure: consensus definitions and overview. Clin Gastroenterol Hepatol 2006;4:6e10. [2] Fishbein TM. Intestinal transplantation. N Engl J Med 2009;361:998e1008. [3] Sasaki T, Hasegawa T, Nakai H, Musiake S, et al. Zoom endoscopic evaluation of rejection in living-related small bowel transplantation. Transplantation 2002;73:560e4. [4] Herlenius G, Fagerlind M, Krantz M, et al. Chronic kidney diseaseea common and serious complication after intestinal transplantation. Transplantation 2008;86:108e13. [5] Buchman AL, Moukarzel A, Ament ME, et al. Serious renal impairment is associated with long-term parenteral nutrition. JPEN J Parenter Enteral Nutr 1993;17:438e44. [6] Boncompain-Gerard M, Robert D, Fouque D, et al. Renal function and urinary excretion of electrolytes in patients receiving cyclic parenteral nutrition. JPEN J Parenter Enteral Nutr 2000;24: 234e9. [7] Ojo AO, Held PJ, Port FK, et al. Chronic renal failure after transplantation of a nonrenal organ. N Engl J Med 2003;349: 931e40. [8] Pironi L, Lauro A, Soverini V, et al. Renal function in patients on long-term home parenteral nutrition and in intestinal transplant recipients. Nutrition 2014;30:1011e4. [9] Lauro A, Zanfi C, Dazzi A, et al. Effect of age on native kidney function after adult intestinal transplants on long-term follow-up. Transplant Proc 2014;46:2322e4. [10] Kato A, Romero MF. Regulation of electroneutral NaCl absorption by the small intestine. Annu Rev Physiol 2011;73: 261e81. [11] Schultheis PJ, Clarke LL, Meneton P, et al. Renal and intestinal absorptive defects in mice lacking the NHE3 Naþ/Hþ exchanger. Nat Genet 1998;19:282e5. [12] Walker NM, Simpson JE, Yen PF, et al. Down-regulated in adenoma Cl/HCO3 exchanger couples with Na/H exchanger 3 for
506 NaCl absorption in murine small intestine. Gastroenterology 2008;135:1645e53.e1643. [13] Hoglund P, Haila S, Socha J, et al. Mutations of the downregulated in adenoma (DRA) gene cause congenital chloride diarrhoea. Nat Genet 1996;14:316e9. [14] Rocha F, Musch MW, Lishanskiy L, et al. IFN-gamma downregulates expression of Na(þ)/H(þ) exchangers NHE2 and NHE3 in rat intestine and human Caco-2/bbe cells. Am J Physiol Cell Physiol 2001;280:C1224e32. [15] Clayburgh DR, Musch MW, Leitges M, et al. Coordinated epithelial NHE3 inhibition and barrier dysfunction are required for TNF-mediated diarrhea in vivo. J Clin Invest 2006;116:2682e94. [16] Yang H, Jiang W, Furth EE, Wen X, et al. Intestinal inflammation reduces expression of DRA, a transporter responsible for congenital chloride diarrhea. Am J Physiol 1998;275:G1445e53. [17] Strong TV, Boehm K, Collins FS. Localization of cystic fibrosis transmembrane conductance regulator mRNA in the human gastrointestinal tract by in situ hybridization. J Clin Invest 1994;93:347e54. [18] Barrett KE, Keely SJ. Chloride secretion by the intestinal epithelium: molecular basis and regulatory aspects. Annu Rev Physiol 2000;62:535e72. [19] Abu-Elmagd KM, Tzakis A, Todo S, et al. Monitoring and treatment of intestinal allograft rejection in humans. Transplant Proc 1993;25:1202e3. [20] Horslen SP. Optimal management of the post-intestinal transplant patient. Gastroenterology 2006;130(2 Suppl 1):S163e9. [21] Levey AS, Bosch JP, Lewis JB, et al. A more accurate method to estimate glomerular filtration rate from serum creatinine: a new prediction equation. Modification of Diet in Renal Disease Study Group. Ann Intern Med 1999;130:461e70. [22] Maletzki C, Stier S, Gruenert U, et al. Establishment, characterization and chemosensitivity of three mismatch repair deficient cell lines from sporadic and inherited colorectal carcinomas. PLoS One 2012;7:e52485. [23] Pironi L, Hebuterne X, Van Gossum A, et al. Candidates for intestinal transplantation: a multicenter survey in Europe. Am J Gastroenterol 2006;101:1633e43. [24] Fishbein TM, Florman S, Gondolesi G, et al. Intestinal transplantation before and after the introduction of sirolimus. Transplantation 2002;73:1538e42. [25] Boyer O, Noto C, De Serre NP, et al. Renal function and histology in children after small bowel transplantation. Pediatr Transplant 2013;17:65e72. [26] Ueno T, Kato T, Gaynor J, et al. Renal dysfunction following adult intestinal transplant under tacrolimus-based immunosuppression. Transplant Proc 2006;38:1762e4. [27] Lauverjat M, Hadj Aissa A, Vanhems P, et al. Chronic dehydration may impair renal function in patients with chronic
REINER, HSIEH, STRAARUP ET AL intestinal failure on long-term parenteral nutrition. Clin Nutr 2006;25:75e81. [28] Suzuki M, Mujtaba MA, Sharfuddin AA, et al. Risk factors for native kidney dysfunction in patients with abdominal multivisceral/small bowel transplantation. Clin Transplant 2012;26: E351e8. [29] Lissner S, Nold L, Hsieh CJ, et al. Activity and PI3-kinase dependent trafficking of the intestinal anion exchanger downregulated in adenoma depend on its PDZ interaction and on lipid rafts. Am J Physiol Gastrointest Liver Physiol 2010;299: G907e20. [30] Lamprecht G, Gaco V, Turner JR, et al. Regulation of the intestinal anion exchanger DRA (downregulated in adenoma). Ann N Y Acad Sci 2009;1165:261e6. [31] Malakooti J, Saksena S, Gill RK, et al. Transcriptional regulation of the intestinal luminal Na(þ) and Cl() transporters. Biochem J 2011;435:313e25. [32] Andres AM, Lopez Santamaria M, Ramos E, et al. The use of sirolimus as a rescue therapy in pediatric intestinal transplant recipients. Pediatr Transplant 2010;14:931e5. [33] Malinowski M, Martus P, Lock JF, et al. Systemic influence of immunosuppressive drugs on small and large bowel transport and barrier function. Transpl Int 2011;24:184e93. [34] Gabe SM, Bjarnason I, Tolou-Ghamari Z, et al. The effect of tacrolimus (FK506) on intestinal barrier function and cellular energy production in humans. Gastroenterology 1998;115:67e74. [35] Yang J, Zhao X, Patel A, et al. Rapamycin inhibition of mTOR reduces levels of the Naþ/Hþ exchanger 3 in intestines of mice and humans, leading to diarrhea. Gastroenterology 2015;149: 151e62. [36] Nakagaki T, Satoh K, Ishibashi D, et al. FK506 reduces abnormal prion protein through the activation of autolysosomal degradation and prolongs survival in prion-infected mice. Autophagy 2013;9:1386e94. [37] Karapetyan YE, Sferrazza GF, Zhou M, et al. Unique drug screening approach for prion diseases identifies tacrolimus and astemizole as antiprion agents. Proc Natl Acad Sci U S A 2013;110: 7044e9. [38] Abu-Elmagd KM, Costa G, Bond GJ, et al. Evolution of the immunosuppressive strategies for the intestinal and multivisceral recipients with special reference to allograft immunity and achievement of partial tolerance. Transpl Int 2009;22:96e109. [39] Gygi SP, Rochon Y, Franza BR, et al. Correlation between protein and mRNA abundance in yeast. Mol Cell Biol 1999;19: 1720e30. [40] Bertelsen LS, Eckmann L, Barrett KE. Prolonged interferon-gamma exposure decreases ion transport, NKCC1, and Naþ-Kþ-ATPase expression in human intestinal xenografts in vivo. Am J Physiol Gastrointest Liver Physiol 2004;286:G157e65.
ABSTRACT Background. Intestinal transplantation is a treatment option for intestinal failure. Although nephrotoxic medication after transplantation is a major cause for posttransplant renal insufficiency, it remains unclear why kidney dysfunction is particularly frequent after intestinal transplantation. Methods. This study analyzed messenger RNA expression of NHE3, DRA, and CFTR in 404 biopsies obtained between day 2 and 1508 from the terminal ileum of 10 adult intestinal transplant recipients. Results. The time courses of immunosuppression and glomerular filtration rate were correlated. In the first posttransplant year, expression of NHE3 and DRA, which mediate NaCl absorption, was diminished to a greater degree than that of CFTR, which mediates chloride secretion. Reduced NHE3 and DRA expression was associated with high tacrolimus trough levels. Titration of tacrolimus to low levels by year 2 was paralleled by partially restored NHE3 and DRA expression. In cell culture experiments, similar effects of tacrolimus on transporter expression were detected. In patients, both reduced tacrolimus levels and recovery of NHE3 and DRA expression were associated with stabilization of renal function. Conclusions. Our data strongly suggest that tacrolimus impairs absorption of NaCl and water from the transplanted ileum, leading to volume depletion and impaired renal function. This may be reversible by reduction of tacrolimus to lower levels without increased rates of rejection or chronic graft failure.
I
NTESTINAL OR MULTIVISCERAL TRANSPLANTATION has become a treatment option for intestinal failure if home parenteral nutrition fails [1]. Usually, transplanted patients achieve oral autonomy with regard to maintaining proteineenergy requirements as well as volume and electrolyte homeostasis. Currently, standard immunosuppressive regimens after intestinal or multivisceral transplantation aim to achieve tacrolimus “monotherapy” plus low-dose prednisolone [2]. Protocol biopsies are obtained from the transplanted ileum to rule out rejection frequently early after transplantation and at more prolonged intervals later in the course. In the absence of rejection, the transplanted intestine has normal endoscopic appearance and shows normal mucosal histology [3]. Despite that, intestinal ª 2016 by Elsevier Inc. All rights reserved. 360 Park Avenue South, New York, NY 10010-1710
Transplantation Proceedings, 48, 499e506 (2016)
fluid and electrolyte losses can be high early and occasionally also late after transplantation. Together with tacrolimus and other drug-induced nephrotoxicity, these losses put the patient at risk for impaired kidney function [4]. In addition, even pretransplant kidney function may be impaired owing to cycled total parenteral nutrition and repeated exposure to nephrotoxic drugs, which may not be reflected by creatinine or compensated by continuous fluid supplementation [5,6].
*Address correspondence to G. Lamprecht, Division of Gastroenterology and Endocrinology, University Medical Center Rostock, Ernst-Heydemann-Str. 6, 18057 Rostock, Germany. E-mail: [email protected] 0041-1345/16 http://dx.doi.org/10.1016/j.transproceed.2015.12.068
499
500
Indeed, renal function is impaired more often after intestinal or multivisceral than after other nonrenal solid organ transplantations [7e9]. Thus, water and electrolyte absorption from the transplanted intestine are of utmost importance for the long-term prognosis of these patients. Despite that, expression of the ion transporters that mediate intestinal NaCl absorption, chloride secretion, and subsequent water transport has not been studied in the transplanted ileum. The intestinal epithelium is composed of a single layer of highly specialized columnar cells. Transepithelial movement of solutes can occur either via the transcellular pathway by traversing the apical and basal membrane or the paracellular pathway across intercellular junctions. The apical membrane of epithelial intestinal cells comprising the primary barrier to transcellular flux is composed of a vast diversity of membrane-associated proteins, including ion transporters and ion channels as the major effectors of transcellular ion transport. Two ion exchangers, NHE3 and DRA (also known as CLD, chloride anion exchanger) facilitating Na/H and Cl/HCO3 exchange, are responsible for electroneutral NaCl absorption in the human ileum and proximal colon [10]. Their quantitative importance has been documented in knockout mice, which display chronic diarrhea [11,12]. A genetic defect in DRA function also leads to the human disease congenital chloride diarrhea [13]. In addition, dysregulation of NHE3 [14,15] as well as DRA [16] has been associated with inflammatory processes in the intestinal mucosa. Furthermore, CFTR is expressed in the human ileum and more so in the colon [17]. CFTR mediates chloride secretion and (over)stimulation of CFTR by toxins leads to secretory diarrhea [18]. Under normal conditions, the activity of NHE3 and DRA on the one side and CFTR on the other side, seem to be balanced in the healthy ileum leading to net electroneutral NaCl absorption. One conceivable explanation for frequent renal dysfunction in intestinal transplant recipients is impaired intestinal absorption of NaCl and water with subsequent development of renal failure owing to poor volume status. To better understand the mechanisms of impaired renal function, nephrotoxic immunosuppression, and intestinal ion and water absorption, we have studied messenger RNA (mRNA) expression of NHE3, DRA, and CFTR in the transplanted ileum. Furthermore, we have correlated these data along the time course after transplantation with tacrolimus trough levels and prednisolone dose as well as with the calculated glomerular filtration rate (GFR). MATERIALS AND METHODS Between 2006 and 2012, 10 adult patients were transplanted at the University of Tübingen because of intestinal failure with a benign underlying etiology and because of failing home parenteral nutrition. One additional biopsy was taken from the transplanted terminal ileum whenever routine surveillance or clinically indicated biopsies were taken. Protocol biopsies were obtained 3 times per week in the first month, 1 to 2 times per week in the next 2 months, and once a month for the rest of the first posttransplant year similar to other programs [19,20]. Thereafter, routine biopsies were taken
REINER, HSIEH, STRAARUP ET AL once every 3 months. Of note, only about 10% of all biopsies were taken outside this schedule for clinical indications. In the same way, ileal biopsies were obtained from 7 healthy controls undergoing screening colonoscopy for colorectal cancer. The study was approved by the local institutional review board (Ethikkommission der Universität Tübingen 022/2011BO2) and all patients gave written informed consent. Ileal biopsies were immediately transferred into RNAlater (Ambion, Foster City, CA) and stored at 4 C for preservation until further use. RNA was isolated using the RNAqueous kit (Ambion). The integrity of the isolated RNA was confirmed by agarose gel electrophoresis and the concentration of isolated RNA was determined using a spectrophotometer (Nano Drop 2000; Peq Lab Biotechnologie GmbH, Erlangen, Germany). cDNA was synthesized from approximately 2 mg mRNA using RevertAid H Minus First Strand cDNA Synthesis Kit (ThermoScientific, Waltham, MA) according to the recommendations of the manufacturer. For quantitative reverse transcriptase polymerase chain reaction (RT-PCR), the following primers were selected using Primer-Express 2.0-Software (Applied Biosystems, Foster City, CA, USA): H3F3A: sense: 50 -TTCCAGAGCGCAGCTATCG-30 ; antisense: 50 -TCTTCAAAAAGGCCAACCAGAT-30 . DRA: sense: 50 -GTGTCCTTTCTTGATGTTTCTTCAGT-30 ; antisense: 50 -CGGTTAAGCTTCTCAATGAAGTCA-30 . CFTR: sense: 50 -GGAAAAGGCCAGCGTTGTC-30 ; antisense: 50 -CCAGGCGCTGTCTGTATCCT-30 . NHE3: sense: 50 -CTGTCCCTCTACGGCGTCTT-30 ; antisense: 50 -GCTGCCAAACAGGAGGAAGTC-30 . Extensive primer optimization using cDNA clones as templates proved that all primers facilitated optimal and identical PCR amplification conditions. Quantitative RT-PCR (qRT-PCR) was carried out on an ABI Prism 7000 Real Time PCR instrument (Applied Biosystems) using Sequence Detection System Software V. 1.2.3. (Applied Biosystems). Initial steps of 50 C of 2 minutes and 95 C for 10 minutes were followed by 40 cycles of amplification at 95 C for 15 seconds and 60 C for 60 seconds. qRT-PCR for each biopsy was performed in triplicates and the mean was calculated. Obtained Ct values were normalized to the amplification of H3F3A and DCt values from transplanted ilea were then compared with mean DCt values from healthy controls (DDCt) and are displayed as 2DDCt.
Collection of Clinical Data Data from transplanted patients were collected retrospectively from clinical charts and entered into a database according to the date of biopsy. The estimated GFR (eGFR) was calculated using the Modification of Diet in Renal Disease Study group’s equation [21].
Cell Culture Experiments HROC113 cells [22] were kept in DMEM/Ham-F12 medium supplemented with 5% fetal bovine serum and penicillin/streptomycin, and seeded into tissue culture treated 6-well plates (Corning, Corning, NY) between subculture number 49 and 57. Cells were grown to confluence and let differentiate for another 4 days. Cells were quickly rinsed with phosphate-buffered saline before undergoing Trizol RNA extraction. qPCR was carried out as described using ViiA 7 Seq Detection.
Statistical Analysis All real-time qPCR data were fed into a database and linked to clinical data that were recorded according to dates of biopsy. Queries were used to compare single parameters to mRNA levels
EPITHELIAL ION TRANSPORTERS
501
using MS Access, Excel and JMP 11.1.1 (SAS, Inc, Cary, NC). P values are listed when a statistical test was performed.
RESULTS Patient Cohort
All 10 patients included in this study were Caucasian adults of northern European ancestry with a benign underlying etiology of their intestinal failure who underwent intestinal or multivisceral transplantation at the transplantation unit Tübingen between 2006 and 2011. The indication for transplantation was failure or impeding failure of home parenteral nutrition [2,23]. Quality of life was not regarded as an indication for transplantation [2]. Demographics of the cohort as well as specific indication, type of transplant, details of the immunosuppression, and outcome are shown in Table 1. The demographic data resembled those from other larger programs. With regard to immunosuppression, all patients received induction with either ATG (200 mg), thymoglobulin (2x 75 mg), basiliximab (2x 20 mg) or alemtuzumab (20 mg). Furthermore, all patients had an initial immunosuppressive regime that contained tacrolimus (with trough Table 1. Characteristics of the Study Cohort Transplanted Between 2006 and 2012 Characteristics
Value
Patients Male/female Age at transplantation (y) HPN before transplant (mo) Follow-up (d), mean Underlying etiology of short bowel syndrome (Crohn’s/volvulus/ ischemia/other resection) Pretransplant renal failure NOD2 mutation (hetero-/homozygous) Indication for transplantation Loss of central venous access Recurrent sepsis Intestinal failure-associated liver disease Type of graft Isolated intestine Multivisceral Including right sided colon Immunosuppression Induction (alemtuzumab/antithymocyte globulin/basiliximab) Tacrolimus maintenance Prednisolone Early mycophenolate Intermittent mammalian target of rapamycin inhibitor Outcome Biopsy-proven rejection Graft loss Death
10 6/4 50.4 7.49* 129 (17e363)† 787 (47e1508)† 6/1/2/1
*Standard deviation. † Range.
4 2/1 3 2 5 6 4 4 2/3/3 10 10 6 5
1 2 1
levels of 10e20 ng/ml immediately after transplantation) and high dose prednisolone [24]. 6/10 patients received mycophenolate mofetil (Cellcept). Mammalian target of rapamycin inhibitors (rapamycin or everolimus) were only used intermittently in 5 patients in an attempt to decrease nephrotoxicity, but were discontinued because of lack of beneficial effect on kidney function or because of additional unwanted side effects. The time course of the immunosuppression in individual patients is depicted in Fig 1. As can be seen, at 1 year after transplantation, all patients achieved a comparably simple immunosuppression consisting of tacrolimus and low-dose prednisolone with 2 patients intermittently taking rapamycin. Target trough levels for tacrolimus were 15 to 20 ng/mL in the first 3 months, 12 to 15 ng/mL in the following 3 months, and <12 ng/mL thereafter. All patients achieved oral autonomy and all intravenous fluid supplementation was discontinued at the time of discharge. Patients were educated to keep their volume status balanced, which was evaluated at each outpatient visit. In no case was scheduled home parenteral fluid supplementation reinitiated after hospital discharge. Expression of Intestinal Ion Transport Proteins NHE3, DRA, and CFTR After Intestinal Transplantation
To address whether ion transport is altered in the transplanted ileal epithelium after intestinal transplantation, qRT-PCR was used to detect changes of cellular mRNA of the major epithelial ion transporters NHE3, DRA, and CFTR. Expression levels of mRNA in transplanted individuals were compared with a control group of 7 healthy individuals (mean age; 47.2 years; range, 17e77) undergoing screening ileocolonoscopy. DCt values from the control group were 1.18 0.33 for NHE3, 2.05 0.17 for DRA, and 4.36 0.72 for CFTR (mean standard deviation). The expression of intestinal ion transporters after transplantation is displayed in Fig 2. Fig 2, AeC shows the relative expression compared with healthy controls of NHE3, DRA, and CFTR in individual patients at individual time points. Early after transplantation, the expression of all 3 transporters varied substantially. In the individual patients, this phase lasted 4e12 months, reflecting their early clinical course. Thereafter, mRNA expression became more stable. The expression data are summarized in Fig 2D. During the first year after transplantation, expression of NHE3 (21.9%), DRA (43.0%), and CFTR (48.0%) was strongly decreased. During years 2 and 3 after transplantation, expression showed both less variation and NHE3 and DRA recovered to significantly higher expression levels compared with year 1 (NHE3, 58.9% vs 21.9% [P < .05]; DRA, 59.5% vs 43.0% [P < .05]); for CFTR, the recovery was not significant (53.8% vs 48.0%). Nevertheless, gene expression of all 3 ion transporters remained diminished compared with healthy controls. Because the data showed substantial variation, the question arose as to whether the observed reduction of expression levels represented independent downregulation of individual ion transporters or common changes of
502
REINER, HSIEH, STRAARUP ET AL
prednisolone [mg/d]
A
30 30 30 30 30 30 30 30 30 30
C
D
mycophenolate mofetil [mg/d]
B
mTOR-Inhibitor trough [ng/ml] tacrolimus trough levels [ng/ml]
0
200
400
600
800
1000
1200
1400
time [days] 20 20 20 20 20 20 20 20 20 20
transporter abundance in the epithelium. To differentiate between these two possibilities, regression analysis was performed (Fig 2E). Expression levels of all 3 ion transporters correlated strongly and significantly with each other. NHE3 and DRA correlated slightly stronger with each other (rS ¼ 0.7185; P < .0001) than with CFTR (NHE3 vs CFTR, rS ¼ 0.6699 [P < .0001]; DRA vs CFTR, rS ¼ 0.5143 [P < .0001]). The strong correlation suggests that expression of these 3 transporters was commonly affected; however, NHE3 and DRA, which govern salt and water absorption, were more vulnerable in year 1 compared with CFTR, which is involved in water secretion. Intestinal Ion Transport, Tacrolimus and Kidney Function
0
200
400
600
800
1000
1200
1400
1000
1200
1400
1000
1200
time [days] 10 10 10 10 10 10 10 10 10 10 0
0
200
400
600
800
time [days] 2 2 2 2 2 2 2 2 2 2 0
0
200
400
600
800
time [days]
1400
Fig 1. Immunosuppression over time. Dose or trough levels of immunosuppressive agents are shown. Patients are displayed in chronological order by date of transplantation. Graft loss occurred in patient 1 (chronic graft failure; graft enterectomy at day 1315) and in patient 5 (prolonged acute rejection and sepsis; graft enterectomy at day 119). Patient 10 died owing to graft-versus-host disease on day 47. (A) Long-term administration of prednisolone. Initially, a highdose methylprednisolone regimen together with a T-cell antibody was given (not shown here for reasons of clarity). (B) Tacrolimus trough levels. All patients received tacrolimus as basic immunosuppressant. Initially, trough levels of tacrolimus showed substantial variation but became more stable over time. (C) Mammalian target of rapamycin (mTOR) inhibitors were given intermittently to decrease nephrotoxicity. Dark grey, everolimus; light grey, sirolimus. (D) Mycophenolate mofetil was given initially in 6 patients and intermittently when indicated.
To address the mechanisms that lead to the high incidence of impaired kidney function after intestinal transplantation, the calculated eGFR was assessed as a marker for kidney function and the corresponding tacrolimus trough levels were recorded (Fig 3). Along the time axis after transplantation, these were together put into context with the reduced expression of ion transporters. As expected, the eGFR was somewhat reduced before transplantation in these severely ill patients. Despite high tacrolimus trough levels early after transplantation, eGFR was only moderately reduced between months 0 and 6. However, the majority of our patients had prolonged hospitalization and the eGFR calculated by the Modification of Diet in Renal Disease formula tends to overestimate the true GFR in these malnourished patients, especially when intravenous fluid supplementation is given in the hospital [4]. Tracer studies of kidney function, which analyze the true GFR more precisely, were not performed. After month 6 posttransplantation, that is, after discharge, the eGFR was significantly reduced. Because of the declining kidney function, tacrolimus levels were reduced. Interestingly, eGFR remained stable thereafter and did not decline further (Fig 3). Although tacrolimus levels were high, the eGFR decreased in association with strongly reduced NHE3 and DRA expression. CFTR expression on the other hand was reduced to a lesser degree (Figs 2 and 3). Remarkably, reduced tacrolimus levels correlated with partial recovery of NHE3 and DRA expression to a comparable level as the stably reduced CFTR expression. Thus, a toxic effect of tacrolimus on epithelial ion transport, predominantly affecting salt and water absorbing transporters, seemed to be possible as an explanation. To this end, tacrolimus toxicity was tested in a recently characterized cell line, HROC113, which expresses similar mRNA levels of NHE3, DRA, and even higher levels of CFTR compared with biopsies from human ileum (unpublished data). In vitro, tacrolimus treatment of confluent HROC113 cells decreased mRNA expression of both NHE3 and DRA in a dose-dependent manner (Fig 4). Although low tacrolimus concentrations (2 ng/mL) did not alter expression levels
EPITHELIAL ION TRANSPORTERS
503
Fig 2. Expression of epithelial ion transport proteins in the transplanted small bowel. (AeC) Time course of messenger RNA (mRNA) expression levels in individual patients. Different patients are denoted in different colors. Note logarithmic scale of the y axis. (D) Summary of mRNA expression of NHE3, DRA, and CFTR in transplanted ilea. Out of 404 biopsies, RNA was recovered successfully from 397 specimens and further analyzed. Data are normalized to the expression in healthy controls. The shaded area denotes the variation in healthy controls. Within the first year after transplantation, expression of NHE3 (21.9%), DRA (43.0%), and CFTR about 48% (D) was strongly decreased. After 1 year, expression showed both less variation and recovery to higher expression levels (NHE3, 58.9%; DRA, 59.5%; CFTR, 53.8 %; median standard error of the mean). (E) Regression analysis of ion transporter expression shows strong and highly significant (P < .0001) correlation between NHE3 and DRA as wells a less prominent but equally significant correlations between NHE3 or DRA with CFTR.
significantly, high concentrations, mimicking in vivo tissue levels after transplantation, repressed mRNA expression clearly. CFTR expression remained largely unaffected in these in vitro studies. DISCUSSION
Over the last decades, intestinal transplantation has been optimized with regard to surgical procedures as well as with regard to immunosuppression regimens and thus survival has improved significantly [2]. However, kidney function has
remained a major problem after solid organ transplantation in general and after intestinal transplantation in particular [4,7e9,25,26]. The reasons behind the fact that kidney function is more often impaired after intestinal transplantation than after transplantation of other organs has not been addressed directly. Long-standing volume depletion before transplantation, even despite optimized parenteral support, continued volume depletion after intestinal transplantation because of ongoing intestinal losses and finally intense immunosuppression are all likely to play a role [6,8,9,27,28].
504
REINER, HSIEH, STRAARUP ET AL
140
20
GFR [ml/min]
100
18 16 14 12
80
10 60
8 6
40
4
tacrolimus trough levels [ng/ml]
GFR tacrolimus
120
opposed to CFTR was also detected in vitro using a colonic cancer cell line. Our data strongly suggest that the net effect of differential inhibition of NHE3 and DRA on the one side and of CFTR on the other is reduced salt and water absorption. In patients, this translates into latent intestinal salt and fluid losses, latent volume depletion and subsequently impaired kidney function. The low expression levels of the transporters NHE3 and DRA early after intestinal or multivisceral transplantation were accompanied by high tacrolimus trough levels and a decline of renal function. The decline in renal function
A
20 2
-3 6 31
-3 0 25
-2 4 19
-1 8 13
12 7-
6
0
0-
Pr eTx
0
time [months] Fig 3. Glomerular filtration rate (GFR) and tacrolimus trough levels after intestinal transplantation. Mean calculated GFR (Modification of Diet in End-stage Renal Disease method) over time in red on the left axis. Mean tacrolimus trough levels in blue on the right axis. Pre-tx, pretransplant.
In this context, the general function and the specific (patho)physiologic role of the transplanted intestinal epithelium have not been studied extensively. One major function of the intestinal epithelium is transepithelial NaCl transport, which governs water absorption [10]. NHE3 and DRA are 2 functionally coupled ion exchangers that mediate electroneutral NaCl absorption [12,29e31]. Unlike these 2, CFTR is an anion channel that mediates electrogenic chloride secretion [18]. It is feasible to study these transporters because they are expressed in the ileum (and proximal colon), from which biopsies are routinely obtained for rejection surveillance. The expression of all 3 studied transporters varied greatly at low levels, particularly within the first months through 1 year after transplantation, reflecting the individual and sometimes difficult early posttransplant course. Subsequently, the expression levels became more stable at about 60% of the expression in healthy controls. Despite this variation, mRNA expression of the transporters correlated strongly with each other in the biopsies and thus at each time point (Fig 2E). This suggests that, after intestinal transplantation, not 1 individual transporter or 1 functionally coupled set of transporters is disturbed, but that transporter abundance in the epithelium is impaired in general. Expression of NHE3 and DRA was impaired to a greater degree than CFTR in the biopsies. The more pronounced effect of tacrolimus on NHE3 and DRA as
B
C
Fig 4. Effect of tacrolimus on expression of ion transporters in HROC113. Messenger RNA expression levels of NHE3 (A), DRA (B), and CFTR (C) were determined by real-time quantitative polymerase chain reaction analysis and compared with H3F3A. DDCt-values were based on controldDCtdvalues without tacrolimus. n ¼ 3 per condition.
EPITHELIAL ION TRANSPORTERS
became apparent only 6 to 12 months after transplantation, most likely because it was compensated by intravenous fluid administration as long as the patients were admitted to the hospital. Because of declining renal function, tacrolimus was titrated to lower trough levels after the first year post transplant. In some cases, we also tried to substitute tacrolimus with a mammalian target of rapamycin inhibitor to reduce nephrotoxicity [24,32] (Fig 1C). However, this was in general not successful. Nevertheless, tacrolimus dose reduction was associated with both a stabilization of renal function and a recovery of NHE3 and DRA to about the same level as CFTR allowing for a more balanced volume status. Beside tacrolimus, many other uncontrolled factors may negatively affect the expression of intestinal ion transporters. Thus, we addressed directly a potential inhibitory effect of tacrolimus on intestinal epithelial cells. Because other currently available cell lines do not mimic physiologic intestinal expression levels of NHE3, DRA, and CFTR, we used HROC113 cells, a recently characterized human cell line, which expresses these transporters at levels similar to healthy ileal epithelium. Intriguingly, incubation of these cells with tacrolimus at concentrations comparable with levels achieved in patients, resulted in a dose-dependent reduction of NHE3 and DRA mRNA, while CFTR expression remained largely unaffected. Thus, after intestinal transplantation, tacrolimus may not only exert its adverse renal effects by direct toxicity on the kidney, but also by interfering with intestinal NaCl transport resulting in insufficient hydration and latent volume depletion. It should be noted that tacrolimus also compromises nutrientcoupled ion transport [33] and intestinal barrier function [34], both of which may negatively affect volume status as well. Recently, Yang et al [35] showed that high levels of rapamycin reduce intestinal NHE3 expression in animal models and kidney transplant patients through a mechanism that involves increased autophagy. In addition, CFTR protein expression in the rat ileum was unaffected by rapamycin treatment. Thus, tacrolimus and rapamycin seem to impair intestinal NaCl absorption in a similar manner. The possibility of a common molecular mechanism, however, remains open, because it is not known whether tacrolimus also interferes with autophagy in the enterocyte [36,37]. Because they observed the effect of rapamycin on autophagy in renal transplant recipients, our findings may principally be applicable to the pathogenesis of impaired renal function in other solid organs transplantations as well, but they may be more pronounced after intestinal transplantation because the transplanted organ and the target organ are the same. To some extent, we were forced to reduce tacrolimus trough levels to preserve kidney function. It may be argued that this would lead to an higher incidence of episodes of acute cellular rejection or to chronic graft dysfunction and eventually graft failure [38]. But even despite comparably mild immunosuppression and relatively low tacrolimus levels, we did not see biopsy-proven acute cellular rejection
505
later than 6 months after transplantation. Also, chronic graft dysfunction was not clinically evident except for patient 1 (starting around day 1100). Thus, in the clinical context, careful reduction of tacrolimus reduced the degree of nephrotoxicity and was associated with increased expression of NHE3 and DRA without compromising graft tolerance or long-term graft function. Although qRT-PCR is a powerful quantitative approach to detect changes of cellular RNA, a limitation of this study is that mRNA levels cannot make a point about protein expression [39]. Furthermore, the cellular as well as the subcellular localization of NHE3 and DRA as well as CFTR may be altered by tacrolimus as well as by other uncontrolled factors after intestinal transplantation [40]. Taken together, our data suggest that relatively low tacrolimus levels are not only acceptable after intestinal transplantation, but may even be aimed for. In this small clinical cohort, graft tolerance and long-term organ function were not compromised by lower levels of immunosuppression. Under such conditions, the strong impairment of intestinal ion transporter expression is largely reversed and expression levels reach about 60% of healthy controls. This likely contributes to clinically apparent less latent volume depletion. Together with a lesser degree of direct tacrolimus inherent nephrotoxicity, this may facilitate stabilization of renal function.
REFERENCES [1] O’Keefe SJ, Buchman AL, Fishbein TM, et al. Short bowel syndrome and intestinal failure: consensus definitions and overview. Clin Gastroenterol Hepatol 2006;4:6e10. [2] Fishbein TM. Intestinal transplantation. N Engl J Med 2009;361:998e1008. [3] Sasaki T, Hasegawa T, Nakai H, Musiake S, et al. Zoom endoscopic evaluation of rejection in living-related small bowel transplantation. Transplantation 2002;73:560e4. [4] Herlenius G, Fagerlind M, Krantz M, et al. Chronic kidney diseaseea common and serious complication after intestinal transplantation. Transplantation 2008;86:108e13. [5] Buchman AL, Moukarzel A, Ament ME, et al. Serious renal impairment is associated with long-term parenteral nutrition. JPEN J Parenter Enteral Nutr 1993;17:438e44. [6] Boncompain-Gerard M, Robert D, Fouque D, et al. Renal function and urinary excretion of electrolytes in patients receiving cyclic parenteral nutrition. JPEN J Parenter Enteral Nutr 2000;24: 234e9. [7] Ojo AO, Held PJ, Port FK, et al. Chronic renal failure after transplantation of a nonrenal organ. N Engl J Med 2003;349: 931e40. [8] Pironi L, Lauro A, Soverini V, et al. Renal function in patients on long-term home parenteral nutrition and in intestinal transplant recipients. Nutrition 2014;30:1011e4. [9] Lauro A, Zanfi C, Dazzi A, et al. Effect of age on native kidney function after adult intestinal transplants on long-term follow-up. Transplant Proc 2014;46:2322e4. [10] Kato A, Romero MF. Regulation of electroneutral NaCl absorption by the small intestine. Annu Rev Physiol 2011;73: 261e81. [11] Schultheis PJ, Clarke LL, Meneton P, et al. Renal and intestinal absorptive defects in mice lacking the NHE3 Naþ/Hþ exchanger. Nat Genet 1998;19:282e5. [12] Walker NM, Simpson JE, Yen PF, et al. Down-regulated in adenoma Cl/HCO3 exchanger couples with Na/H exchanger 3 for
506 NaCl absorption in murine small intestine. Gastroenterology 2008;135:1645e53.e1643. [13] Hoglund P, Haila S, Socha J, et al. Mutations of the downregulated in adenoma (DRA) gene cause congenital chloride diarrhoea. Nat Genet 1996;14:316e9. [14] Rocha F, Musch MW, Lishanskiy L, et al. IFN-gamma downregulates expression of Na(þ)/H(þ) exchangers NHE2 and NHE3 in rat intestine and human Caco-2/bbe cells. Am J Physiol Cell Physiol 2001;280:C1224e32. [15] Clayburgh DR, Musch MW, Leitges M, et al. Coordinated epithelial NHE3 inhibition and barrier dysfunction are required for TNF-mediated diarrhea in vivo. J Clin Invest 2006;116:2682e94. [16] Yang H, Jiang W, Furth EE, Wen X, et al. Intestinal inflammation reduces expression of DRA, a transporter responsible for congenital chloride diarrhea. Am J Physiol 1998;275:G1445e53. [17] Strong TV, Boehm K, Collins FS. Localization of cystic fibrosis transmembrane conductance regulator mRNA in the human gastrointestinal tract by in situ hybridization. J Clin Invest 1994;93:347e54. [18] Barrett KE, Keely SJ. Chloride secretion by the intestinal epithelium: molecular basis and regulatory aspects. Annu Rev Physiol 2000;62:535e72. [19] Abu-Elmagd KM, Tzakis A, Todo S, et al. Monitoring and treatment of intestinal allograft rejection in humans. Transplant Proc 1993;25:1202e3. [20] Horslen SP. Optimal management of the post-intestinal transplant patient. Gastroenterology 2006;130(2 Suppl 1):S163e9. [21] Levey AS, Bosch JP, Lewis JB, et al. A more accurate method to estimate glomerular filtration rate from serum creatinine: a new prediction equation. Modification of Diet in Renal Disease Study Group. Ann Intern Med 1999;130:461e70. [22] Maletzki C, Stier S, Gruenert U, et al. Establishment, characterization and chemosensitivity of three mismatch repair deficient cell lines from sporadic and inherited colorectal carcinomas. PLoS One 2012;7:e52485. [23] Pironi L, Hebuterne X, Van Gossum A, et al. Candidates for intestinal transplantation: a multicenter survey in Europe. Am J Gastroenterol 2006;101:1633e43. [24] Fishbein TM, Florman S, Gondolesi G, et al. Intestinal transplantation before and after the introduction of sirolimus. Transplantation 2002;73:1538e42. [25] Boyer O, Noto C, De Serre NP, et al. Renal function and histology in children after small bowel transplantation. Pediatr Transplant 2013;17:65e72. [26] Ueno T, Kato T, Gaynor J, et al. Renal dysfunction following adult intestinal transplant under tacrolimus-based immunosuppression. Transplant Proc 2006;38:1762e4. [27] Lauverjat M, Hadj Aissa A, Vanhems P, et al. Chronic dehydration may impair renal function in patients with chronic
REINER, HSIEH, STRAARUP ET AL intestinal failure on long-term parenteral nutrition. Clin Nutr 2006;25:75e81. [28] Suzuki M, Mujtaba MA, Sharfuddin AA, et al. Risk factors for native kidney dysfunction in patients with abdominal multivisceral/small bowel transplantation. Clin Transplant 2012;26: E351e8. [29] Lissner S, Nold L, Hsieh CJ, et al. Activity and PI3-kinase dependent trafficking of the intestinal anion exchanger downregulated in adenoma depend on its PDZ interaction and on lipid rafts. Am J Physiol Gastrointest Liver Physiol 2010;299: G907e20. [30] Lamprecht G, Gaco V, Turner JR, et al. Regulation of the intestinal anion exchanger DRA (downregulated in adenoma). Ann N Y Acad Sci 2009;1165:261e6. [31] Malakooti J, Saksena S, Gill RK, et al. Transcriptional regulation of the intestinal luminal Na(þ) and Cl() transporters. Biochem J 2011;435:313e25. [32] Andres AM, Lopez Santamaria M, Ramos E, et al. The use of sirolimus as a rescue therapy in pediatric intestinal transplant recipients. Pediatr Transplant 2010;14:931e5. [33] Malinowski M, Martus P, Lock JF, et al. Systemic influence of immunosuppressive drugs on small and large bowel transport and barrier function. Transpl Int 2011;24:184e93. [34] Gabe SM, Bjarnason I, Tolou-Ghamari Z, et al. The effect of tacrolimus (FK506) on intestinal barrier function and cellular energy production in humans. Gastroenterology 1998;115:67e74. [35] Yang J, Zhao X, Patel A, et al. Rapamycin inhibition of mTOR reduces levels of the Naþ/Hþ exchanger 3 in intestines of mice and humans, leading to diarrhea. Gastroenterology 2015;149: 151e62. [36] Nakagaki T, Satoh K, Ishibashi D, et al. FK506 reduces abnormal prion protein through the activation of autolysosomal degradation and prolongs survival in prion-infected mice. Autophagy 2013;9:1386e94. [37] Karapetyan YE, Sferrazza GF, Zhou M, et al. Unique drug screening approach for prion diseases identifies tacrolimus and astemizole as antiprion agents. Proc Natl Acad Sci U S A 2013;110: 7044e9. [38] Abu-Elmagd KM, Costa G, Bond GJ, et al. Evolution of the immunosuppressive strategies for the intestinal and multivisceral recipients with special reference to allograft immunity and achievement of partial tolerance. Transpl Int 2009;22:96e109. [39] Gygi SP, Rochon Y, Franza BR, et al. Correlation between protein and mRNA abundance in yeast. Mol Cell Biol 1999;19: 1720e30. [40] Bertelsen LS, Eckmann L, Barrett KE. Prolonged interferon-gamma exposure decreases ion transport, NKCC1, and Naþ-Kþ-ATPase expression in human intestinal xenografts in vivo. Am J Physiol Gastrointest Liver Physiol 2004;286:G157e65.