Laser Capture Microdissection as a New Tool to Assess Graft-Infiltrating Lymphocytes Gene Profile in Islet Transplantation

Abstract

Innovative tolerogenic protocols in transplantation would take advantage of the development of new tools capable of evaluating the impact of these treatments on the immune system. These assays have potential for clinical application. Currently, many of these studies are based on the analysis of peripheral lymph nodes and blood-derived cells, where the percentage of alloantigen-specific cells can be low or even unpredictable. We combined a laser capture microdissection (LCM) technique with real-time PCR (RT-PCR) to evaluate gene profile of islet-infiltrating lymphocytes. Donor Lewis rats islets were transplanted under the kidney capsule in diabetic Brown Norway rats. Administration of anti-LFA1 mAb or anti-CD28 F(Ab)′ was able to prolong islet survival, while the combined treatment resulted in indefinite survival. The analysis of gene expression profile for IL-2, IFN-γ, and IL-10 production of graft-infiltrating cells revealed high IL-2, IFN-γ, and IL-10 in untreated rats; on the contrary, the combined treatment selectively abrogated IL-2- and IFN-γ-producing cells infiltrate. The comparison between cytokine profile in periphery (even during an allogenic extra stimulus) and in the graft revealed the dichotomy between graft and peripheral cytokine assessment. We thus propose that direct analysis of graft-infiltrating cells should be used whenever possible to evaluate the effects of a new immunomodulatory protocol.

Keywords

Islet transplantation Laser capture microdissection (LCM)Gene expression Graft-infiltrating lymphocytes Costimulation blockade

Introduction

The development of new immunomodulatory protocols remains a major goal for transplant immunology (6, 7, 10,15). Moreover, it is also necessary to characterize, in preclinical models with assays that have potential for clinical application, the impact of these immune-suppressive strategies on the immune system. Most of the techniques are based on the analysis of blood; however, probably due to the low number of alloantigen specific cells in peripheral blood, contrasting data have emerged (8). Patient lymph nodes may offer an alternative source of immune cells to be used in alloimmune monitoring. However, being lymph nodes distant from the graft, the number of graft antigen-experienced cells may be low (8). Moreover, it can be debated if the results obtained away from the graft will ultimately reflect the immunological events at the graft level, and thus if they give reliable and reproducible results.

To date, the most reliable technique to study immunological phenomena at the graft level in patients is the histological characterization of graft-infiltrating lymphocytes (5,12). Unfortunately, histological analysis is limited for technical reasons to highly expressed surface antigens and to a few transcription factors (like Foxp3) (17). In our work we used a new combined technique of laser capture microdissection (LCM) and real-time PCR (RT-PRC) to characterize the state of the immune system at graft level through gene expression profiling of graft-infiltrating lymphocytes (3,11). We also compared the results obtained with analysis of peripheral lymph node.

Engagement of costimulatory molecules on lymphocyte cell membranes is a crucial step in the priming of the immune response and in graft transplantation rejection; interfering with this step has been shown to positively prolong allogenic graft survival (16).

We evaluated the innovative tool of laser microdissection in a new tolerogenic treatment based on the blockade of the LFA1-ICAM1 and CD28-B7.1/2 costimulatory pathways to produce long-term survival in a rat model of allogenic islet transplantation.

Materials and Methods

Animals

Inbred male Brown Norway and Lewis rats were purchased from Charles River Laboratories (Wilmington, MA). Protocols were approved by the Institutional Animal Care and Use Committee of our Institution.

Islet Transplantation

Pancreatic islets from Lewis rats were isolated by collagenase digestion followed by density gradient separation and then hand-picking, as described previously (4). Brown Norway rats were rendered diabetic with 225 mg/kg of streptozotocin (STZ, Sigma-Aldrich, St. Louis, MO, administered IP). Six hundred islets were transplanted under the renal capsule of clearly diabetic rats (2 consecutive days of glycemia >400 mg/dl), and glycemia was monitoring trough capillary blood analysis twice per week. Rejection of islet allografts was defined by blood glucose levels of >250 mg/dl for 2 consecutive days.

Antibodies

Monoclonal antibodies used in this study were purified from low-serum culture supernatants obtained from the following hybridomas, currently available in our lab: JJ319 (anti-CD28, IgG1) and WT.1 (anti-LFA1, IgG2a).

Dendritic Cells (DCs) Assay

To stimulate peripheral lymph node antigen-specific cytokine production, we customized a test, modifying the technique of the local lymph node assay (9), based on the injection of donor DCs in the subcutaneous tissue of the recipient to challenge draining lymph nodes. Fully activated DCs were obtained according to standard protocols. Briefly, bone marrow cells were dispersed in complete medium (RPMI 10% FCS p/s), GM-CSF (20 ng/ml; Invitrogen, Carlsbad, CA), and IL-4 (5 ng/ml; Invitrogen); every other day 75% of the medium was replaced with fresh medium. At day 9 nonadherent cells were dispersed in fresh media with GM-CSF (10 ng/ml) and at day 11 nonadherent cells were collected and activated with LPS (Sigma-Aldrich) (0.5 ng/ml) for 2 days. DCs were then injected in the footpad and dorsal subcutaneous tissue 12 days after transplantation, and draining nodes (popliteal and axillaries) were collected 3 days after injection. In preliminary experiments, DCs were labeled with CMFDA (Invitrogen) following the manufacturer's instructions. CMFDA loaded DCs (30 × 10⁶) were injected in the dorsal subcutaneous tissue and 24 and 48 h later excisional biopsy of axillary and immunofluorescence analysis for CMFDA was performed (2).

RT-PCR

RNA extraction from node and infiltrating cells was performed through RNeasy Mini and Micro Kit (Quiagen, Valencia, CA), respectively. DNase optional digestion step was performed. cDNA was obtained through M-MLV Reverse Transcriptase, RNase H Minus, Point mutation (Promega, San Luis Obispo, CA). Random hexamers (Amersham Pharmacia Biotech, Pittsburgh, PA) and dNTPs (Invitrogen) were added according to protocol. RT-PCR was performed using an ABI Prism 7700 sequence detector system (Applied Biosystems, Foster City, CA). Predesigned primers and probes were also obtained from Applied Biosystem, with β-actin as the internal control. Data are expressed as fold increase compared to baseline (lymph node lymphocytes derived from a pool of untransplanted, untreated, not DC-injected animals).

Histological Evaluation and LCM

Graft-bearing kidneys were included in OCT and snap frozen. Sections were stained with H&E. Sections in which mononuclear cell infiltration were detected were then mounted on a membrane-coated slide and hematoxylin stained. Leica AS LMD for LCM was used (Leica, Bannockburn, IL), according to the manufacturer's instructions, to detach mononuclear cell infiltration zone (14). Microdissected tissue was treated for RNA extraction and expression was evaluated by RT-PCR.

Statistical Analysis

Kaplan-Meier analysis was performed for rejection-free survival determination, and differences were assessed with the Mantel-Cox log-rank test. Values of p < 0.05 were considered significant.

Results

Combined Blockade of CD28 and LFA1 Produces Indefinite Survival of Allogenic Transplanted Rats

We first tested the effect of the blockade of the LFA1/ICAM1 and the B7.1/2-CD28 pathways in an allogenic model of islet transplantation in rats (13). We used anti-LFA1 mAb and the F(Ab)′ portion of the anti-CD28 mAb, because the use of the entire anti-CD28 mAb can activate lymphocytes in vitro (data not shown). Islets from Lewis rats were transplanted under the kidney capsule of Brown Norway rats rendered diabetic with streptozotocin. Graft survival was monitored through capillary blood analysis for glycemia level. Treated animals received IV injection of 1 mg of anti-LFA1 mAb or anti-CD28 F(Ab)′ or both, every 12 h for 3 days, starting 12 h before transplantation. Untreated animals (controls) promptly rejected islet graft in a few days (MST: 9.5 days; n = 5) (Fig. 1A, B). Administration of anti-LFA1 mAb significantly prolonged graft survival (MST: 30 days, p = 0.017 vs. control) while the use of anti-CD28 F(Ab)′ antibody has minor effects (MST: 13, p = 0.025 vs. control) (Fig. 1A, B). Notably, when anti-LFA1 and anti-CD28 blocking antibody were used in combination, long-term survival was achieved in 5 out of 8 animals and in the 3 remaining substantial prolongation was assessed [MST: indefinite; p = 0.0001 vs. controls; p = 0.01 vs. anti-LFA1 mAb treated; p = 0.0002 vs. anti-CD28 F(Ab)′ treated] (Fig. 1A, B).

Figure 1.

Anti-CD28 combined with anti-LFA1 treatment produced indefinite survival of allogenic islet transplantation and modified peripheral cytokine profile. Lewis islets were transplanted in hyperglycemic Brown Norway rats. (A) Prolonged survival was seen with anti-LFA1 treatment while indefinite survival was seen in combination treatment [MST: control = 9.5, n = 5, anti-LFA1 mAb treated = 30 days, n = 5, p = 0.017 vs. contorl; anti-CD28 F(Ab)′ = 13 days, n = 5, p = 0.025 vs. control; anti-LFA1 mAb + anti-CD28 F(Ab)′ = indefinite, n = 8, p = 0.0001 vs. control]. (B) Individual glycemia values (mg/dl) are reported for control rats (black dashed line), anti-CD28 F(Ab)′ treatment (left), anti-LFA1 mAb treatment (center), and combined treatment (right). (C) Immunofluorescence analysis of axillary lymph node 48 h after injection of CMFDA-labeled donor DCs in dorsal subcutaneous tissue of the recipient showed DC migration at this level (original magnification 10x). (D, E, F) Gene expression profile for cytokine production at axillary and popliteal lymph node level 72 h after DC injection showed clear pattern associated with the treatment compared to controls [fold increase vs. baseline, controls: IL-2 = 1.2x; IFN-γ = 1.7x; IL-10 = 2.3x n = 4; anti-LFA1 mAb + anti-CD28 F(Ab)′ treated: IL-2 = 0.7x, IFN-γ = 6.2x, IL-10 = 13.9, n = 4; controls vs. treated IFN-γ and IL-10, p = 0.03]. Representative curves are plotted; top panels control, bottom panels treated; x-axis: cycle number (0 to 40); y-axis: ΔRn (10⁻³ to 10¹). *p < 0.05.

Analysis of Donor DCs Injected in the Draining Node Revealed a Different Cytokine Profile in Treated and Untreated Animals

We first assessed the cytokine gene expression profile of treated and untreated (control) animals in the peripheral lymph nodes (popliteal and axillary), in a site that has potential for clinical application. No difference was evident between transplanted untreated and transplanted treated rats compared to baseline (gene expression of peripheral lymph node of untransplanted untreated rats) (data not shown). We thus tried to stimulate cytokine production. Aiming to produce a local alloantigen-specific response, we challenged recipient rats with donor DCs as an in vivo assay to challenge direct allorecognition. Bone marrow-derived donor DCs (10 × 10⁶) were injected into the footpaths and the dorsal subcutaneous tissue of transplanted animals, untreated or treated with anti-LFA1 mAb and anti-CD28 F(Ab)′ 12 days after transplantation. Preliminary experiments showed as CMFDA-labeled DCs injected at this level can migrate to popliteal node and to the axillary lymph node stations (Fig. 1C). Seventy-two hours after DC injection, poplietal and axillary nodes were harvested and the cytokine gene profile (IL-2, IL-10, and IFN-γ) was evaluated. Untreated animals showed, like in not DC-injected rats, a cytokine expression level very similar to baseline (fold increase: IL-2 = 1.2x, IFN-γ = 1.7 x, IL-10 = 2.3x compared to baseline, n = 4), while treated animals had a different cytokine profile level characterized by high IL-10 and IFN-γ production (fold increase: IL-2 = 0.7x, IFN-γ = 6.2x, IL-10 = 13.9x compared to baseline, n = 4). Statistical difference was reached between treated and untreated rats for IL-10 (p = 0.03) and IFN-γ (p = 0.03) (Fig. 1D-F).

We investigated in a few animals (two untreated control and two treated) which population was responsible for the IL-10 production. CD25⁺ cells were purified from draining lymph nodes trough magnetic beads separation and gene expression profile of CD25⁺ cells was compared to CD25^- cells. CD25⁺ mRNA expression was higher in the CD25⁺ group and this paralleled with a higher FoxP3 expression (fold increase CD25⁺ vs. CD25~: CD25 6.5 ± 1.4; FoxP3: 11.4 ± 2.6; n = 4) with no differences between treated and untreated control rats (data not shown). We concluded that CD25⁺ cells were also highly enriched in FoxP3⁺ cells. In treated rats IL-10 production was relatively similar between CD25⁺ and CD25^- cells; on the contrary, in control rats IL-10 was produced mostly by CD25⁺ cells (IL-10 fold increase CD25⁺ vs. CD25~: treated rats 1.8x and 1.2x; control rats: 5.4x and 5.1x).

We then analyzed the effect of DCs injection on allograft survival. Bone marrow-derived donor DCs (10 × 10⁶) were injected in treated rats and graft survival was monitored. Unfortunately, the boosting of the immune response produced by the injection of DCs invariably caused allograft rejection within 3 weeks in treated rats (data not shown).

LCM Is a Useful Instrument to Assess Cytokine Production at Islet Graft Level

We then analyzed cytokine expression at islet graft level. Most of the analyses of graft infiltrating immune cells are based on histology; unfortunately, histology's capacity to detect cytokine production is low. We thus evaluated intragraft cytokine production through gene expression analysis associated to LCM. LCM is a relatively new technique that allows extracting RNA from a group of cells, or even single cells, of interest starting from histology specimens (11,14) and it has been used to analyze the graft parenchyma in the context of syngeneic islet transplantation (1). Twelve days after transplantation, two rats untreated and two treated with anti-LFA mAb and anti-CD28 F(Ab)′ were sacrificed and graft-bearing kidneys were harvested, included in OCT, and snap frozen. Graft-bearing kidneys were then dissected through a cryostat and stained with H&E to individuate graft area. At this point LCM of graft-infiltrating inflammatory cells was performed and gene expression profile for IL-10, IL-2, and IFN-γ was assessed in untreated (Fig. 2A) and treated rats (Fig. 2B). In untreated rats, high levels of the two Th1 cytokines (IFN-γ and IL-2) were found as well as of IL-10 (fold increase: IL-2 = 4.5x; IFN-γ = 10.7x; and IL-10 = 17.8x compared to baseline). In contrast, in treated rats, very few infiltrating cells were detectable, with almost nondetectable expression of IFN-γ and IL-2, but demonstrating high expression of IL-10 (fold increase: 38.6x compared to baseline) (Fig. 2C-E).

Figure 2.

LCM of islet allograft. Anti-LFA1 mAb and anti-CD28 F(Ab)′-treated and control grafts showed different cytokine gene expression. Graft-bearing kidney was harvested 12 days after transplantation. Graft was individuated by kidney sectioning. LCM of controls (A, original magnification 20x) and anti-LFA1 mAb and anti-CD28 F(Ab)′-treated rats (B, original magnification 20x) was performed and samples were analyzed for gene expression. (C, D, D) Controls showed high IL-2, IFN-γ, and IL-10 expression (fold increase: IL-2 = 4.5x; IFN-γ = 10.7X; and IL-10 = 17.8x vs. baseline, n = 2), while only high IL-10 expression was seen in treated grafts (fold increase: IL-2 and IFN-γ under detection level; IL-10 38.6x vs. baseline, n = 2). Representative curves are plotted, top panels control, bottom panels treated; x-axis: cycle number (0 to 40); y-axis: ΔRn (10⁻³ to 10¹).

Discussion

In this article we note the use of LCM as a new tool to assess immunomodulatory protocol in islet transplantation setting. We generated a new protocol based on the blockade of the LFA1-ICAM1 and CD28-B7.1/2 pathways to produce indefinite survival in a rat model of allogenic islet transplantation. We then characterized the state of graft-infiltrating lymphocytes by combining for the first time in allogenic islet transplantation the techniques of LCM to RT-PCR. We therefore evaluated cytokine production, at a graft level, overwhelming the limitations of histological technique. This procedure, once established, will be able to provide more information regarding the immune system (presence of Treg, Th1/Th2 profile) and it can be suitable also for clinical studies. Our data showed that in rejecting untreated rats, IL-10 mRNA expression within islet allografts parallels the presence of a population of cells with a clear Th1 profile (high IFN-γ and IL-2). Our treatment appeared to be capable of abrogating the Th1 infiltrate, preserving and maybe expanding IL-10-producing cells.

Our DCs assay shows potential for assessing the state of the immune response versus alloantigen; this is proven by the significant difference in cytokine production between treated and untreated rats. However, that analysis of peripheral lymph nodes did not represent the immunological phenomena at the graft level. The Th1 profile we observed at graft level could not be detected in antigen-stimulated peripheral lymph nodes. Again, the presence of a population of IL-10-producing cells, while evident at graft level in both treated and untreated rats, could be detected in the peripheral lymph nodes only in treated animals. And last, the population of high IFN-γ-producing cells seen at the lymph node level in treated rats was not actually present in graft infiltration. We conclude that, at least in our model, peripheral assessment of the immune system is only partially capturing graft-infiltrating cells. Moreover, while peripheral lymph nodes assessment without stimulation was inconclusive, the challenge with donor-derived DCs carried the risk of alloimmune response boosting.

For these reasons, we note that our technique of directly assessing the effect of an immunomodulatory protocol at graft level, if it can be confirmed by further studies and tested in other models, may be a major breakthrough in transplantation medicine, being more reliable than peripheral assessment of the immune response and potentially useful for clinical application. Moreover, the potential of LCM in the context of islet transplantation may go further than the analysis of cytokine production. If fully characterized, LCM can be proposed as a technique that can truly address the effects of an immune treatment on the immune system at a local level.

Footnotes

Acknowledgments

A. Vergani is recipient of an AST-JDRF fellowship grant. P. Fiorina is recipient of an AST-JDRF Faculty Grant. This work was supported in part by grants from AIRC and MIUR (PRIN projects) to R. Pardi.

References

Ahn

Y. B.

; Xu

; Marselli

; Toschi

; Sharma

; Bonner-Weir

; Sgroi

D. C.

; Weir

G. C.

Changes in gene expression in beta cells after islet isolation and transplantation using laser-capture microdissection. Diabetologia 50(2): 334–342; 2007.

Asquith

; Debacq

; Florins

; Gillet

; Sanchez-Alcaraz

; Mosley

; Willems

Quantifying lymphocyte kinetics in vivo using carboxyfluorescein diacetate succinimidyl ester (CFSE). Proc. Biol. Sci. 273(1590): 1165–1171; 2006.

Berg

; Wu

; Levay-Young

; Heuss

; Pan

; Kirchhof

; Sutherland

D. E.

; Hering

B. J.

; Guo

Comparison of tolerated and rejected islet grafts: A gene expression study. Cell Transplant. 13(6): 619–629; 2004.

Fiorina

; Jurewicz

; Tanaka

; Behazin

; Augello

; Vergani

; von Andrian

U. H.

; Smith

N. R.

; Sayegh

M. H.

; Abdi

Characterization of donor dendritic cells and enhancement of dendritic cell efflux with CC-chemokine ligand 21: A novel strategy to prolong islet allograft survival. Diabetes 56(4): 912–920; 2007.

Girlanda

; Kleiner

D. E.

; Duan

; Ford

E. A.

; Wright

E. C.

; Mannon

R. B.

; Kirk

A. D.

Monocyte infiltration and kidney allograft dysfunction during acute rejection. Am. J. Transplant. 8(3): 600–607; 2008.

Hatipoglu

; Benedetti

; Oberholzer

Islet transplantation: Current status and future directions. Curr. Diab. Rep. 5(4): 311–316; 2005.

Hering

B. J.

; Kandaswamy

; Ansite

J. D.

; Eckman

P. M.

; Nakano

; Sawada

; Matsumoto

; Ihm

S. H.

; Zhang

H. J.

; Parkey

; Hunter

D. W.

; Sutherland

D. E.

Single-donor, marginal-dose islet transplantation in patients with type 1 diabetes. JAMA 293(7): 830–835; 2005.

Hernandez-Fuentes

M. P.

; Warrens

A. N.

; Lechler

R. I.

Immunologic monitoring. Immunol. Rev. 196: 247–264; 2003.

Kimber

; Dearman

R. J.

; Basketter

D. A.

; Ryan

C. A.

; Gerberick

G. F.

The local lymph node assay: Past, present and future. Contact Dermatitis 47(6): 315–328; 2002.

10.

Lechler

R. I.

; Sykes

; Thomson

A. W.

; Turka

L. A.

Organ transplantation—how much of the promise has been realized?

Nat. Med. 11(6): 605–613; 2005.

11.

Marselli

; Thorne

; Ahn

Y. B.

; Omer

; Sgroi

D. C.

; Libermann

; Otu

H. H.

; Sharma

; Bonner-Weir

; Weir

G. C.

Gene expression of purified beta-cell tissue obtained from human pancreas with laser capture microdissection. J. Clin. Endocrinol. Metab. 93(3): 1046–1053; 2008.

12.

Moreso

; Seron

; O'Valle

; Ibernon

; Goma

; Hueso

; Cruzado

J. M.

; Bestard

; Duarte

; del Moral

R. G.

; Grinyo

J. M.

Immunephenotype of glomerular and interstitial infiltrating cells in protocol renal allograft biopsies and histological diagnosis. Am. J. Transplant. 7(12): 2739–2747; 2007.

13.

Nicolls

M. R.

; Coulombe

; Beilke

; Gelhaus

H. C.

; Gill

R. G.

CD4-dependent generation of dominant transplantation tolerance induced by simultaneous perturbation of CD154 and LFA-1 pathways. J. Immunol. 169(9): 4831–4839; 2002.

14.

Pinzani

; Orlando

; Pazzagli

Laser-assisted microdissection for real-time PCR sample preparation. Mol. Aspects Med. 27(2-3): 140–159; 2006.

15.

Ricordi

; Strom

T. B.

Clinical islet transplantation: Advances and immunological challenges. Nat. Rev. Immunol. 4(4): 259–268; 2004.

16.

Rothstein

D. M.

; Sayegh

M. H.

T-cell costimulatory pathways in allograft rejection and tolerance. Immunol. Rev. 196: 85–108; 2003.

17.

Truong

; Plester

J. C.

; Hancock

W. W.

; Merani

; Murphy

T. L.

; Murphy

K. M.

; Kaye

; Anderson

C. C.

; Shapiro

A. M.

Combined coinhibitory and costimulatory modulation with anti-BTLA and CTLA4Ig facilitates tolerance in murine islet allografts. Am. J. Transplant. 7(12): 2663–2674; 2007.