SEARCH

SEARCH BY CITATION

Keywords:

  • Acute organ rejection;
  • enteroendocrine cell;
  • intestinal stem cell;
  • Neuro D;
  • Neurogenin-3;
  • NEUROG3;
  • NOD2;
  • small bowel transplantation;
  • PYY;
  • GLP-1

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Conflicts of Interest Statement
  9. References

Acute small intestinal allograft rejection presents clinically as an abrupt increase in ileal fluid output in the absence of extensive inflammation. We questioned whether acute intestinal rejection might be accompanied by a disturbance of normal intestinal stem cell differentiation. We examined the intestinal epithelial secretory cell lineage among patients experiencing early rejection before and during rejection as well as following corrective therapy. Lineage-specific progenitors were identified by their expression of stage-specific transcription factors. Progenitors of the enteroendocrine cell (EEC) expressing neurogenin-3 (NEUROG3) were found to be disproportionately reduced in numbers, along with their more mature EEC derivatives expressing neuro D; the enteric hormone PYY was the most profoundly depleted of all the EEC products evaluated. No change in the numbers of goblet or Paneth cells was observed. Steroid treatment resulted in resolution of clinical symptoms, restoration of normal patterns of EEC differentiation and recovery of normal levels of enteric hormones. Acute intestinal rejection is associated with a loss of certain subtypes of EEC, most profoundly, those expressing PYY. Deficiency of the mature EECs appears to occur as a consequence of a mechanism that depletes NEUROG3 EEC progenitors. Our study highlights the dynamics of the EEC lineage during acute intestinal rejection.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Conflicts of Interest Statement
  9. References

Intestinal transplantation is increasingly undertaken as a replacement therapy for gut failure (1,2). Graft failure secondary to rejection occurs in about 30–40% of transplants within the first 3 years postoperatively. The first episode of rejection occurs within the first 3 months following transplantation in over 80% of allograft recipients (1,3). Intestinal allograft rejection is generally graded as ‘mild’, ‘moderate’ and ‘severe’, based on intestinal pathology and responsiveness to therapeutic intervention (1,3,4). Mild rejection of the small intestine is first recognized clinically by an increase in intestinal fluid output through the ileal stoma. Gross endoscopic examination of the graft generally reveals a normally appearing mucosa. Among the earliest microscopic abnormalities observed is an increase in apoptotic bodies within the crypts in the absence of extensive inflammation (1,3,4). In the majority of these patients, a brief course of augmented immunosuppression corrects the clinical symptoms and restores normal intestinal histology. This clinical–pathological state, including its responsiveness to a brief course of immunosuppression, is recognized as ‘mild acute cellular rejection’ (MACR).

The small intestine provides a unique system in which to evaluate the consequences of acute rejection on tissue renewal; it is the only transplanted solid organ in which the stages of cellular proliferation and differentiation are well characterized with specific transcription factors marking stages of commitment to different lineages (5–7). Furthermore, in our institution the small intestinal allograft is routinely prospectively biopsied endoscopically, as well as during and after episodes of rejection. Thus, it is possible to compare tissue specimens from an individual allograft obtained before any evidence of disease with those obtained during rejection and following resolution. We report here that in acute rejection, coincident with the onset of diarrhea, a deficiency of mature EEC and their corresponding hormones can be observed, with perturbation of the EEC lineage traceable to a depletion of NEUROG3 progenitor cells. Steroid treatment results in resolution of clinical symptoms, restoration of the normal representation of mature EECs and recovery of normal levels of enteric hormones. Our study identifies the EEC progenitor lineage as an immunological target of acute intestinal rejection.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Conflicts of Interest Statement
  9. References

Study population

Small bowel transplantation was performed under the direction of a single surgeon (T.F.). The basic transplantation protocol has been described in detail elsewhere (1). Endoscopic mucosal pinch biopsies were taken at transplant-protocol specified time points and additionally whenever clinically indicated. Multiple grasp biopsies were obtained from ileal allograft mucosa during endoscopy of the ileal stoma generally between 10 and 30 cm from the ileocolonic anastomosis. Some biopsy samples were fixed, paraffin embedded and stained with hematoxylin and eosin. Biopsies were then examined by one of two blinded pathologists and scored using standard rejection grading criteria. Additional biopsy material was used for crypt isolation and storage in RNALater (Ambion, Austin, TX) according to the manufacturer's instructions. Patients consented to additional biopsy material collected simultaneously for research purposes. The study was approved by the IRB of Georgetown University.

Tissues were collected prospectively from every patient transplanted at our institution from November 2003, the date when the intestinal transplant program began, through November 2007, comprising 34 consecutive patients. For this study, we selected individuals experiencing MACR that met the following criteria:

  • 1
    pathological evidence of early grade MACR (4), with the primary pathological criterion being the presence of at least six intracryptal apoptotic bodies visualized per 10 crypt cross-sections, in the setting of normal cellular infiltration of the lamina propria;
  • 2
    clinical signs consistent with MACR (i.e. increased stomal effluent volume in the absence of any confounding pathologic states such as graft infection);
  • 3
    the first occurrence of MACR within the first 90 days following transplantation;
  • 4
    NOD2 wild-type genotype for recipient and donor (8);
  • 5
    confirmation of the diagnosis of MACR by resolution of symptoms with standard steroid treatment, i.e. methylprednisolone (1 g i.v.) administered as a bolus once daily for 2 days.

We excluded patients with pathological- or laboratory-based evidence of a concomitant systemic or gastrointestinal infection. Viral infections were diagnosed by routine clinically indicated serum quantitative real-time PCR for Epstein-Barr virus, cytomegalovirus, and noroviruses, serology for adenoviruses, as well as microscopic evaluation of protocol-driven mucosal biopsies.

Five out of 34 patients met these criteria and their clinical characteristics are presented in Table 1. Patients were receiving an ad lib diet low in simple sugars and fats. Food intake was stopped at the time of rejection. As noted in Table 1, average daily ileal stomal output roughly doubled in each case, comparing the output of 5 consecutive days 1 week before to the 5 days beginning from the first biopsy demonstrating rejection. Patients were treated for 2 days with a single daily bolus of methylprednisolone (1 g, i.v.). Symptoms resolved in each patient within 5 days of completion of therapy.

Table 1.  Clinical characteristics of patients experiencing mild acute cellular rejection
PatientAge (years)Primary pathologyPostTx time (days) to RjDuration of Rj (days)Symptoms at RjStoma output1 (mL/day) (mean ± SE)
Before RjDuring Rj
  1. Rj = rejection; SB = small bowel; SBS = short bowel syndrome; SMV = superior mesenteric vein; Tx = transplant.

  2. 1Output was measured 5 days prior to rejection and during 5 days of rejection.

  3. *p < 0.05; **p < 0.01.

141SBS myopathic pseudo-obstruction375Increased stoma output965 ± 1171975 ± 172**
246SBS idiopathic pseudo-obstruction614Increased stoma output, fever910 ± 1292124 ± 236*
318SBS, meconium Ileus485Increased stoma output924 ± 1372047 ± 170**
450SBS, SMV thrombosis536Increased stoma output, stomal edema852 ± 96 1930 ± 208**
527SBS idiopathic pseudo-obstruction295Increased stoma output, abdominal distention742 ± 63 1862 ± 145**

Imunohistochemistry of ileal biopsies were performed as described previously (8)

Goblet cells were visualized with antibodies raised against Mucin 2 (Muc-2), Paneth cells with human defensin 5 (HD-5) antibodies. EECs were immunostained against synaptophysin, a universal EEC marker (9) (SYP, 1:50 dilution, Abcam Inc., Ab6245, Cambridge, MA), HD-5 (1:80, Alpha Diagnostic Int., HDEFA5I-A, San Antonio, TX) and Muc-2 (1:100, Abcam Inc., Ab7848).

Immunohistochemical analysis of isolated ileal crypts

Crypts were isolated from the ileal biopsies by our modification of a described nonenzymatic method (10). Immunohistochemical analyses of isolated crypts were performed as described (11) as modified here. Briefly, the crypts were affixed to a cover glass, which had been precoated with Cell Tak (BD Bioscience, Bedford, MA). After that crypts were treated with 70% ethanol for 30 min and then with 0.2% Triton X-100 in Hanks' balanced salt solution for 30 min at 20°C. Samples were blocked with a serum blocking solution (Reagent 1A, ZYMED® LAB-SA System, 95-9943B, ZYMED Laboratories, San Francisco, CA) for 1 h at 4°C. A washing buffer (0.5% normal goat serum : 1% BSA : 0.02% Triton X-100 in TBS) was used for the subsequent dilutions and washes. The samples were incubated for 12 h at 4°C with primary antibodies to Msi-1 (NB 100-1759, Novus Biological, Littleton, CO; 1:50 dilution), Hath-1 (PA1-17101, ABR-Affinity Bioreagents, Golden, CO; 1:80 dilution), NEUROG3 (PA1-17099, ABR-Affinity Bioreagents; 1:30 dilution) and NeuroD (16508, Abcam Inc.; 1:50 dilution). After washing, slide-mounted crypts were incubated for 20 min with a biotinylated secondary antibody (Reagent 1B, ZYMED® LAB-SA System, 95-9943B, ZYMED Laboratories, San Francisco, CA). Slides were rinsed with PBS for 2 min three times, incubated with an Enzyme Conjugate (Reagent 2, ZYMED® LAB-SA System, 95-9943B, ZYMED Laboratories) and detected using a DAB Substrate Buffer and DAB Chromogen (Liquid DAB Substrate Chromogen System, K3466, DakoCytomation North America, Carpinteria, CA). The crypts were counterstained with hematoxylin and the number of immunostained cells per crypt was counted.

Total RNA isolation and real-time PCR were conducted as described previously (8)

We measured the concentration of the following mRNAs: synaptophysin (SYP, NCBI RefSeq: NM 0031792), mucin 2 (Muc 2, NCBI RefSeq: NM_002457.2), motilin (MLN, NCBI RefSeq: NM 002418.2), somatostatin (SST, NCBI RefSeq: NM_001048.3), peptide YY (PYY, NCBI RefSeq: NM_004160.3), gastric inhibitory polypeptide (GIP, NCBI RefSeq: NM 004123.2), glucagon-like peptide 1 (GLP-1, NCBI RefSeq: NM_002054.2), vasoactive intestinal peptide (VIP, NCBI RefSeq: NM 194435.1), 5-hydroxy-tryptamine (serotonin) transporter (5-HTT, NCBI RefSeq: NM 001045.2), human defensin 5 (HD5, NCBI RefSeq: NM_021010.1), human defensin 6 (HD6, NCBI RefSeq: NM_001926.2), lysozyme (Lsz, NCBI RefSeq: NM_021797.2). TaqMan probes and primers (Hs00300531_m1, Hs03005094_m1, Hs00159150_m1, Hs00356144_m1, Hs00373890_g1, Hs00175030_m1, Hs00174967_m1, Hs00175021_m1, Hs00169010_m1, Hs00360716_m1, Hs00427001_m1, Hs00253976_m1), respectively, were purchased from Applied Biosystems (Foster City, CA). Target mRNA levels are expressed relative to an internal glyceraldehyde-3-phosphate dehydrogenase (HAPDH) control (Assay ID: Hs99999905_m1, Applied Biosystems, Foster City, CA).

Isolated total RNA (1 μg) was reverse-transcribed into cDNA in a 20 μL reaction mix using Superscript III reverse transcriptase and random hexamers, according to the manufacturer's instructions (Invitrogen, Carlsbad, CA). For cDNA amplification, 10 min incubation at 95°C was done to activate AmpliTaqGold DNA polymerase; this was followed by 40 cycles of 15 s at 95°C and 1 min at 60°C for each cycle.

To measure cDNA levels, the threshold cycle at which fluorescence was first detected above baseline was utilized, and a standard curve drawn between starting nucleic acid concentrations and the threshold cycle. Target mRNA levels are expressed as arbitrary units relative to glyceraldehyde-3-phosphate dehydrogenase (GAPDH) mRNA recovered from that tissue sample (Assay ID: Hs99999905_m1, Applied Biosystems). Relative expression was defined as a relative value in comparison with the arbitrary expression value 1 assigned to the normalized amounts of mRNA recovered from patients’ tissues prior to the onset of acute rejection. qRT PCR was performed with the 7500 Fast Real-Time PCR machine (Applied Biosystems). More information regarding the qRT PCR protocol utilized can be obtained from https://products.appliedbiosystems.com/ab/en/US/adirect/custom_taqman_assay_link_placeholder.html.

SDS-PAGE and Western blot analyses were conducted as described previously (8)

We used primary antibodies reactive to Motilin (MTL, 1:80 dilution, Abcam Inc., ab4580), somatostatin (SST, 1:50, RayBiotech Inc., AM-13-0007, Norcross, GA), gastric inhibitory polypeptide (GIP, 1:100, RayBiotech Inc., AM-08-0038, Norcross, GA), vasoactive intestinal peptide (VIP, 1:40, Santa Cruz Biotechnology, 25347, Santa Cruz, CA), serotonin (5-HydroxyTryptamine) transporter (5-HTT, 1:200, Abcam Inc, 36127), glucagon-like peptide 1 (GLP-1, 1:80, Abcam Inc., 23472), peptide YY (PYY, 1:20, Santa Cruz Biotechnology, 47318).

To evaluate the Notch1 activation by assessing the level of Notch intracellular domain (NICD) in nuclear protein (NP) extracts, ileal crypts were isolated as described above and cryptal cells were lysed by using the Nuclear Extract kit (version C4) from Active Motif (Carlsbad, CA) following the manufacturer's recommendations for preparation of NP.

Cell counting procedures on tissue samples and crypts

The numbers of globlet, Paneth and EECs were determined by examination of immunostained whole tissue specimens with well-oriented crypt-villus units (CVU). A scorable CVU was defined as a sagittal section of the crypt-villus axis and an unbroken epithelial layer extending to the villus tip. Each cell was identified as a positive if a dark brown granular reaction was produced by the method studied. Analysis was performed by an observer who was unaware of the clinical condition of the patients. The measurements were done in at least 10 CVU per section, with 5 sections analyzed per biopsy. For each patient, about three biopsies were recovered and studied from each sampling period (before, during, and after the episode of rejection). The numbers of specific intestinal cells/CVU were tallied, and the mean number of positive cells/CVU was determined.

Numbers of specific progenitor cells were determined by examining isolated intact crypts from each biopsy collected both before and during rejection and were immunostained for the specific cell type. One to two biopsies were obtained for each patient before and during rejection, and each biopsy yielded about 20 ‘scorable’ crypts. A scorable crypt was defined as a complete U-shaped ileal gland: the number of labeled cells in the crypt was determined by direct count of all labeled cells in crypt column.

Statistical analysis

Differences between groups were compared using either a two-tailed paired Student t-test or where noted, a one-way ANOVA. Values of <0.05 were considered to be statistically significant. The abundance of hormone proteins from Western blotting measurements was expressed in densitometry units, normalized against values obtained for control samples. mRNA data from TaqMan quantitative real-time PCR analyses were normalized to GAPDH mRNA expression.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Conflicts of Interest Statement
  9. References

Enteroendocrine cells are selectively depleted in the ileal epithelium during acute cellular rejection of the allograft

We explored whether the epithelial cellular composition of the engrafted ileum was affected during MACR. We used immunohistochemical markers to detect each of the three secretory cell types in tissue biopsies. We compared ileal biopsies collected per protocol from allografts before (obtained within 7 days before recovery of the ‘index’ biopsy confirming the diagnosis of MACR), during the period of active rejection as well as after the process was brought under control (the first biopsy with normal histology obtained following treatment of rejection).

Goblet cells were visualized by immunostaining for mucin (Muc2) (12). No difference in the numbers of Muc2-expressing cells within the terminal ileum was seen in the samples taken before (Figure 1B) and during (Figure 1C) rejection (24.2 vs. 26.5 cells per CVU, respectively) (Figure 1D). Similarly, no significant difference in Muc2 mRNA content in ileal mucosa before and during rejection could be detected (Figure 1E).

image

Figure 1. Immunohistochemical analysis of ileal tissue with antibody directed against mucin 2 (Muc2) for detection of goblet cells. Ileal biopsies were taken before and during rejection. (A) Characteristic histology of MACR (H&E). (B) Muc2+ cells, before rejection. (C) Muc2+ cells, during rejection. Muc2-positive cells are shown in brown. Sections were counterstained with hematoxylin. Representative images are shown. Bars: 100 μm. (D) Quantification of ileal Muc2-positive cells. Ileal biopsies were taken from five patients before (C) and during rejection (R). Five sections from the same biopsy were counted for Muc2-positive cells. An average number of positive cells per crypt-villus unit (here defined as a villus and one adjacent crypt) was calculated. The y-axis represents the Muc2-positive cell counts (mean ± SE). *p < 0.001. (E) Expression of Muc2 mRNA in ileal biopsies from five patients before (C) and during rejection (R). Total RNA was isolated from ileal tissue and the amount of Muc2 mRNA was determined by TaqMan qRT PCR and normalized to the amount of GAPDH mRNA recovered from that tissue sample. The normalized amount of Muc2 mRNA in ileal samples before rejection was assigned a value of 1.0. *p < 0.001.

Download figure to PowerPoint

Paneth cell numbers were not significantly different before and during rejection (data not shown). Quantitative measurement of the mRNA abundance of several antimicrobial products of the Paneth cell, including HD5, HD6 and lysozyme, also did not differ significantly (data not shown).

Mature EECs were visualized by immunostaining for synaptophysin (9), and decreased significantly during MACR (Figure 2B) compared with tissue obtained prior to rejection (Figure 2A) (2.1 EEC per CVU prior to rejection as compared to 1 EEC per CVU (p < 0.001) (Figure 2C)). To further validate this apparent decline in EECs, synaptophysin mRNA was quantitatively measured using TaqMan real-time PCR on a second sample taken at the time of the initial biopsy, both before and during rejecting states. Compared to nonrejecting control tissue, mRNA concentration during rejection was decreased by 51% (p < 0.001) (Figure 2E).

image

Figure 2. Immunohistochemical detection of enteroendocrine cells (EEC) in ileal tissue. Synaptophysin (SYP) immunostaining was performed on adjacent sections of the corresponding biopsies shown in Figure 1. General details as in Figure 1. *p < 0.001. Bars: 100 μm.

Download figure to PowerPoint

Analysis of enteroendocrine ontogeny identifies a stage in the stem cell lineage perturbed during mild acute cellular rejection

Mature EECs are known to derive from an early crypt stem cell progenitor through a series of cellular precursors, each identifiable by specific transcription factors. To determine at which stage in the life history of the EEC the apparent decrease in mature cell number arose, we examined the expression of particular progenitor markers in isolated, intact crypts, using immunohistochemistry. Musashi 1 (Msi 1) is a transcription factor expressed by an early progenitor of the enterocyte lineage and was initially believed to mark the true stem cell of the crypt (13). The number of Msi1 positive cells was not significantly decreased during rejection, nor a significant difference in Msi1 mRNA concentration was observed (Figure 3B). This observation suggests that MACR is not associated with a disturbance in the rate of renewal of very early cellular derivatives of the intestinal epithelial stem cell.

image

Figure 3. Effects of rejection on specific gut epithelial progenitor cell populations in isolated ileal crypts. Representative images of isolated ileal crypts and total counts of positive cells for specific markers studied are shown. (A) Western blot analysis of Notch intracellular domain (NICD) in nuclear protein extracts from ileal cryptal cells before (C) and during (R) rejection; representative blot images and densitometry measurements are shown. Whole mount immunostained ileal crypts: (B) Musashi; (C) Hath-1; (D) NEUROG3; (E) Neuro D. For each progenitor marker, approximately 20 immunostained crypts were examined from each biopsy obtained from each of the five patients both before and after rejection. An average # of positive cells/crypt were tallied for each patient's biopsy. The average # of positive cells/crypt for each of the five patients from biopsies obtained before rejection were compared with values obtained from specimens after rejection; statistical significance of these differences was determined by a one-way ANOVA using the SPSS software (SPSS Inc., Chicago, IL). ***p < 0.01.

Download figure to PowerPoint

The Notch signaling pathway appears to play a role in the regulation of the entire secretory enterocyte lineage through control of a repressor, called Hes-1. Hes-1, in turn, controls the expression of transcription factors that are required for the differentiation of the secretory enterocyte lineage. Furthermore, pharmacologically modulating the activity of the Notch pathway in adult mice and humans can profoundly alter the cellular composition of intestinal enterocytes (14–16). To assess the level of activity of the Notch pathway, we measured the intranuclear content of NICD (17) and the activated intracellular component of Notch, in crypts isolated from tissue samples obtained before and during rejection. As anticipated from the normal intestinal representation of Paneth and goblet cells, no apparent difference between levels of intranuclear NICD detected during rejection and prior could be distinguished (p > 0.05) (Figure 3A). This observation demonstrates that the decreased density of EECs arises in a setting where Notch signaling within the secretory lineage has not been noticeably perturbed.

Hath-1 is expressed by the progenitor of the entire secretory lineage (7). The number of Hath-1 + cells remained nearly unchanged during MACR (p > 0.05) (Figure 3C), again suggesting that during rejection commitment to the complete secretory lineage had not been disturbed.

Hath-1 + progenitors are believed to normally differentiate into two cellular derivatives, one that subsequently develops into Paneth and goblet cells and the other that yields the EEC population (7). Since we had observed no significant perturbation in the numbers or phenotypes of either Paneth or goblet cells, we measured the number of cells expressing NEUROG3, a marker of the early EEC progenitor. While 3.3 NEUROG3 + cells per CVU were observed in tissue biopsies prerejection, 1.3 NEUROG3 + cells per CVU were detected during rejection (p < 0.01) (Figure 3D). This result suggested that a significant decrease in the numbers of progenitor cells committed to the EEC population occurred during rejection.

Early EEC progenitors subsequently differentiate into specific subtypes of intestinal EECs, each characterized by the specific pattern of neurotransmitters they express. NeuroD is a transcription factor expressed by cells that derive from NEUROG3+ precursors. Using a specific antibody to NeuroD, we observed a marked depression in NeuroD positive cells comparing crypts isolated before and during rejection (p < 0.01) (Figure 3E). These data suggested that in the setting of MACR, an arrest in the EEC lineage occurs at a stage responsible for the formation of NEUROG3 + EEC progenitors, and is perpetuated in a marked deficiency of NeuroD positive EECs.

As many as 10 different EECs have been described in man, each differing with respect to location within the intestine and the specific cell products they manufacture. To assess whether the deficiency of EECs associated with rejection affected specific types of EECs, we quantitatively measured the amounts of both mRNA and peptide for several of the major gut hormones that derive from the intestinal epithelium. Vasoactive intestinal peptide, somatostatin and motilin were significantly decreased during rejection compared to control (p < 0.05) (Figure 4, upper). A greater relative reduction in mRNA abundance was observed for gastric inhibitory peptide and the 5-hydroxytryptamine transporter (p < 0.01). Of the EEC gene products measured, glucagon-like peptide and peptide YY mRNAs (products of the L-type EEC) were reduced by the greatest magnitude during rejection in comparison to prerejection concentrations (greater than 10-fold, p < 0.001). Semiquantitative analysis of the individual peptides isolated from ileal tissues before and during rejection confirmed the pattern observed in the mRNA analysis (Figure 4, lower).

image

Figure 4. Enteroendocrine cell gene products. (A) mRNA of specific hormones. mRNA was quantitated as described in Figure 1. SST—somatostatin; MTL—motilin; GIP—gastric inhibitory peptide; GLP1—glucagon-like peptide 1; VIP—vasoactive intestinal peptide; PYY—peptide YY; 5-HTT—5-hydroxy-tryptamine transporter. Bars represent the mean and SEM. *p < 0.05; **p < 0.01; ***p < 0.001. (B) Western blot analysis of protein samples extracted from ileal tissue before (C) and during (R) rejection. Representative blot images and densitometry measurements are shown. Immunoblots were prepared as described in the Methods section. SST—somatostatin; MTL—motilin; GIP—gastric inhibitory peptide; GLP1—glucagon-like peptide 1; VIP—vasoactive intestinal peptide; PYY—peptide YY; 5-HTT—5-hydroxy-tryptamine transporter. Data are shown as mean ± SE (n = 5). A P-value is given as determined by Students's t-test. *p < 0.05; **p < 0.01; ***p < 0.001.

Download figure to PowerPoint

Enteroendocrine cell populations within the ileal epithelium recover following immunosuppressive control of mild acute cellular rejection

We examined tissue samples obtained within 1 month following standard immunosuppressive treatment of MACR. At this time ileal fluid output had returned to prerejection values, the clinical symptoms associated with rejection were gone and ad lib oral feeding had resumed. The numbers of EECs per CVU unit had returned to normal (data not shown). The relative amounts of the enteroendocrine gene products examined in ileal biopsy specimens, including the mRNAs of GLP-1 and PYY, were statistically indistinguishable from prerejection samples (Figure 5).

image

Figure 5. Enteroendocrine gene products, pre- and postrejection. mRNA was isolated from tissue specimens obtained before rejection (as in Figure 4) and about 1 month following remission, and quantitated as described in Figure 1 (C, prior to rejection; R, remission). p > 0.05 for all pairs.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Conflicts of Interest Statement
  9. References

In this report we demonstrate that MACR is associated with the selective loss of certain EEC subtypes, sparing Paneth and goblet cells. In our study we have followed the change in epithelial composition in individual allografts, comparing cellular densities prior to the onset of rejection, during the process and after resolution. We show that reduced numbers of mature EECs can be traced to a reduction in the numbers of the NEUROG3+ EEC progenitor, establishing the cellular stage in the EEC lineage, which first appears to be perturbed during rejection.

As yet we do not know whether the reduced number of the NEUROG3+ EEC precursor during MACR is due either to an increased rate of cellular destruction or to a reduced rate of renewal. The pathological hallmark of MACR is the presence of increased numbers of apoptotic bodies visible within intestinal crypts. In the rat acute small intestinal allogeneic allograft, rejection is associated with epithelial apoptosis, a consequence of a host versus graft immune assault (18). The most straightforward hypothesis to explain EEC depletion in human MACR would be that certain EEC precursors of the donor allograft are disproportionately targeted by the recipient. This hypothesis is supported by the recovery of the normal representation of EECs following a brief course of immunosuppression. The precise mechanism involved in EEC depletion remains unknown.

The report of Wang and coworkers identifying the association of malabsorptive diarrhea with the genetically based loss of EECs established a role for the EEC in the regulation of fluid dynamics within the human small intestine (19); the mechanism linking diarrhea to EEC depletion in the three reported patients, however, was not established (20). In our study the L cell appears to be the most profoundly depleted of the EEC subtypes with more than a 90% inhibition of its principal hormone product, PYY, noted. PYY has been shown, in rats, to inhibit water and electrolyte secretion from crypt epithelial cells through binding to a specific G protein coupled receptor, recently shown to be expressed in humans; (21,22) PYY introduced into the rat jejunum in vivo inhibits fluid secretion, and the effect is blocked with a specific PYY receptor antagonist (21). Indeed, PYY peptidomimetic agonists are under development as antidiarrheal therapeutics (23). The PYY receptor appears to be most robustly expressed on epithelial cells within the crypts of the distal small intestine, with no evidence of significant receptor expression on villus epithelial cells or within the colon. The increase in ileal fluid output that occurs during the rejection period, although reduced by cessation of oral nutrient intake, continues in the absence of oral feeding; thus, MACR-associated diarrhea is not purely ‘malabsorptive’, as was the case in the three reported patients with the congenital absence of intestinal EECs (19).

In a previous report we described an aggressive form of acute small intestinal allograft rejection observed in recipients who have Crohn's disease-associated mutations in their NOD2 alleles (8). Acute rejection in these patients is generally not fully corrected by augmented immunosuppression; the likelihood of persistent rejection, and subsequent graft loss and death, is manyfold higher than observed in recipients with a NOD2 wild-type genotype. In these NOD2 mutant recipients, within several weeks of transplantation, and prior to any visible pathological change in the mucosa of the donor bowel, several antimicrobial peptides (AMPs) expressed by the Paneth cells of the donor bowel, such as HD5, decrease in abundance (8). We speculated that in this recipient population, NOD2-expressing cells of the recipient, such as the dendritic cell, normally designed to sense microbes in the lumen and transduce the information to the Paneth cell, failed to fulfill this function. One consequence is the inadequate production of certain AMPs required to prevent microbial access to the epithelium and perhaps to influence the species diversity of the luminal bacteria (24,25). A secondary inflammatory response is mounted leading to progressive destruction of mucosa. In this population of allograft recipient, the pathophysiology of rejection resembles the mechanisms proposed to explain the ileal form of Crohn's disease (25–27).

In this report we have studied the more commonly recognized form of mild rejection that can be distinguished in recipients with a wild-type NOD2 genotype. The epithelium is targeted by an as yet uncharacterized immunological mechanism that depletes the normal representation of EEC cells, a process that can be reversed by adequate immunosuppression.

In summary, we report here that MACR is associated with a reversible reduction in the numbers of specific subtypes of EECs, and their corresponding hormones. The extent to which other human inflammatory intestinal conditions are associated with loss of specific enterocyte precursor populations remains to be established.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Conflicts of Interest Statement
  9. References

We wish to acknowledge support for this research through an American Society of Transplant Surgeons-Wyeth faculty award, and a grant from the Eli and Edythe L. Broad Medical Research Foundation. We wish to thank Drs. Steve Evans and Lynt Johnson for generous departmental support and encouragement. We acknowledge the help of Yulia Rekhtman, M.D., Jacqueline Laurin, M.D., and Cheryl Little, M.D., for help in collecting several of the biopsy specimens, and Bhaskar Kallakury, M.D., for help in histopathological evaluation. We are deeply grateful to the pediatric and adult nursing staff of the Georgetown University Hospital and to Erin Fennelly, R.D., for their tireless support and dedication to our patients.

American Society of Transplant Surgeons-Wyeth faculty award, and a grant from the Eli and Edythe L. Broad Medical Research Foundation provided funding for this work.

Conflicts of Interest Statement

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Conflicts of Interest Statement
  9. References

The authors declare no conflicting interests.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Conflicts of Interest Statement
  9. References