Am J Respir Crit Care Med Vol 166. pp 417–422, 2002
DOI: 10.1164/rccm.2102106
Internet address: www.atsjournals.org

Obliterative bronchiolitis (OB), or chronic allograft rejection, is a major
cause of morbidity and mortality after lung transplantation. The goal
of these experiments was to determine whether several important
growth factors were upregulated during OB in the mouse heterotopic
trachea model. Isografts (BALB/c into BALB/c) and allografts (BALB/c
into C57BL/6) were implanted in three sets of cyclosporine-treated an-
imals and were harvested from 2 to 10 weeks. Ribonucleic acid was iso-
lated using the cesium chloride-guanidine method and was reverse
transcribed and semiquantitated with the polymerase chain reaction
using specific primers for platelet-derived growth factor (PDGF)-A and
PDGF-B chains, fibroblast growth factor (FGF) isoforms 1 and 2, trans-
forming growth factor-



, tumor necrosis factor-



(TNF-



), edothelin-1,
(prepro) epidermal growth factor, insulin-like growth factor-1, and



-actin as a control. Transforming growth factor-



, TNF-



, endothelin-1,
and insulin-like growth factor-1 expression were increased 1.5-fold to
5.0-fold (p



0.04 for each) in the allografts compared with the
isografts at Weeks 2 through 6. Significantly increased expression of
FGF-1, FGF-2, and PDGF-B was noted in the allografts at 4 weeks (p



0.05 for each), which reversed at 6 and 10 weeks. No differences were
found with the PDGF-A chain. The isografts expressed more epidermal

growth factor than allografts (p



0.001). Treatment with a TNF-



–sol-
uble receptor (human TNFR:Fc) significantly reduced epithelial injury
(p



0.01) and lumenal obstruction (p



0.037) in this model. We con-
clude that increased expression of a large number of growth factors
occurs during OB in this model. Growth factor blockade (in particular
with regard to TNF-



) may be useful in ameliorating OB in this model.
Keywords:

obliterative bronchiolitis; chronic rejection; lung; growth
factors; mouse

Although lung transplantation has become a successful clini-
cal therapy for end-stage pulmonary disease as surgical tech-
niques and immunosuppression regimens have improved,
long-term survival of lung transplant recipients has been ad-
versely affected by obliterative bronchiolitis (OB) or chronic
graft rejection. In fact, OB is largely responsible for the ap-
proximately 30% lower 5-year graft survival rates (i.e., 40%
versus 56–76%) between lung (or heart–lung) and other solid-
organ transplants (1). OB is an inflammatory disorder that leads
to airway injury and fibrosis. It affects approximately 50% of
lung transplant patients and is the leading cause of late-trans-
plant deaths (2). Therapies for OB are largely ineffective be-
cause little is known about the underlying mechanisms of this
disease. For these reasons, OB has been considered the peren-
nial “thorn in the side of lung transplantation” (3).
In the past 7 years, a number of animal models of OB have
been developed to investigate the pathogenesis of this disor-
der and have, quite rapidly, expanded the knowledge base on
this problem (4–6). Hertz and colleagues first described the
histologic changes of OB in a heterotopic mouse model (7),
and subsequently demonstrated the efficacy of cyclosporine in
slowing the rate of disease progression (8). The present au-
thors and others have characterized the inflammatory cell re-
cruitment during OB in this model (9, 10). Large numbers of
CD4



cells, CD8



cells, and macrophages are present during
an early phase of “cellular” airway inflammation. Subse-
quently, T cell numbers decline, and macrophages and myofi-
broblasts predominate. The elaboration of cytokines from T
cells (both Th1 and Th2) and macrophages during OB sug-
gests the pleiotropic nature of the alloimmune response (11).
The pathogenesis of the fibropoliferative phase of OB has
generated considerable interest as well because antagonism of
important fibrotic pathways may prove beneficial in slowing
airway scarring and airflow obstruction. Individually, the plate-
let-derived growth factor (PDGF)-A and PDGF-B chains and
the



receptor, fibroblast growth factor (FGF)-2, and trans-
forming growth factor-



(TGF-



) have all been implicated in
the pathogenesis of human and animal model OB (12–15). In
the experiments described herein, we simultaneously studied
the expression of a large number of profibrotic cytokines, in-
cluding PDGF, A and B chains, FGF-1 and FGF-2, TGF-



1,
tumor necrosis factor-



(TNF-



), endothelin-1, and insulin-
like growth factor-1 (IGF-1) to test the hypothesis that these
growth factors are upregulated during the fibro-obliterative
process that characterizes airway fibrosis in OB. The time
course of study was chosen to encompass fully the progression
of OB in the mouse model from cellular (acute-type) rejection
with epithelial injury and destruction through the fibroprolifer-
ative phase marked by mature lumenal scarring. Second,
TNFR:Fc, the soluble TNF-



receptor, was administered to de-
termine whether it could slow chronic rejection in this model.

METHODS

Mice

Seventy-two BALB/c (H2-d) and 16 C57BL/6 (H-2b) (Charles River,
Raleigh, NC) were obtained from pathogen-free colonies and were
housed and used in accordance with the rules of the Institutional Ani-
mal Care and Use Committee.

Tracheal Transplantation and Immunosuppression

Allografts and isografts were obtained by transplanting BALB/c tra-
cheas into C57BL/6 and BALB/c mice, respectively. Transplantation
and immunosuppression (cyclosporine A; Sandoz Pharmaceuticals,
East Hanover, NJ; 25 mg/kg intraperitoneally 5 days/week) were per-
formed as previously described (9). Briefly, tracheas were harvested
from donor animals, stored in Dulbecco’s modified Eagle’s medium at
4



C for 30 minutes, implanted two per recipient into subcutaneous
pockets in the dorsum of the neck, and subsequently, harvested at 2, 4,
6, and 10 weeks. One isograft and one allograft from each time point
were used for hematoxylin and eosin staining (9).

(

Received in original form February 26, 2002; accepted in final form April 16, 2002

)
Funded in part by the National and North Carolina chapters of the American
Lung Association, the Cystic Fibrosis Foundation, and the National Heart, Lung,
and Blood Institute.
Correspondence and requests for reprints should be addressed to Robert Aris,
M.D., CB# 7020, 420 Burnett-Womack Building, The University of North Caro-
lina at Chapel Hill, Chapel Hill, NC 27599-7524. E-mail: aris@med.unc.edu

Growth Factor Upregulation during Obliterative
Bronchiolitis in the Mouse Model

Robert M. Aris, Sean Walsh, Worakij Chalermskulrat, Vasantha Hathwar, and Isabel P. Neuringer

Divisions of Pulmonary and Critical Care Medicine, Department of Medicine and the Cystic Fibrosis Research and Treatment Center,
University of North Carolina at Chapel Hill, Chapel Hill, North Carolina

418

AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE VOL 166 2002

RNA Isolation from Tracheal Grafts

Sixty BALB/c (48 donors and 12 recipients) and 12 C57BL/6 (all re-
cipients) mice were used for the reverse transcription-polymerase chain
reaction (RT-PCR). Trachea grafts (two per recipient) from six (three
C57BL/6 and three BALB/c) different animals were harvested at each
time point and immersed in liquid nitrogen. Total RNA was isolated
using the cesium chloride-guanidine method as previously described
(11). The purity (260/280 nm absorbance ratio



2) and yield of RNA
were determined spectrophotometrically. The integrity of RNA was
verified using Nusieve agarose gel electrophoresis. Genomic DNA
contamination was removed with RNAse-free DNAase (Promega,
Madison, WI) and subsequent ethanol precipitation.

Reverse Transcription and Polymerase Chain Reactions

RT-PCR was performed as previously described (11, 16, 17) with mi-
nor modifications. PCR was performed using 0.5



M of each target
(i.e., PDGF-A and PDGF-B chains, FGF-1 and FGF-2, TNF-



, TGF-



,
IGF-1, endothelin-1, and epidermal growth factor [EGF]) or a control
(



-actin) 3




and 5




primer pair (

see

Table 1) and Taq DNA polymerase
(Invitrogen Corp., Carlsbad, CA). Water was used for a negative con-
trol.
The tissues from each (i.e., 2-, 4-, 6-, and 10-week allograft/isograft
pair) set of mice were analyzed simultaneously for each growth factor
and



-actin mRNA using optimal cycle numbers within the linear
phase of amplification (IGF-1, TGF-



, FGF-1, and endothelin-1: 23–
25 cycles; PDGF-B, TNF-



, and EGF-1: 25–27 cycles; PDGF-A and
FGF-2: 27–29 cycles; actin: 21–23 cycles). The PCR products were sep-
arated by 3% agarose Tris acetate (TAE) gel electrophoresis, stained
for 15 minutes in ethidium bromide, digitally photographed under ul-
traviolet light to quantify the band intensity (ImageQuant software;
Molecular Dynamics, Sunnyvale, CA), and normalized to



-actin.

TNFR:Fc Treatment of Mouse Heterotopic Tracheal OB

Allografts and isografts were generated as previously described. Hu-
man TNFR:Fc (a kind gift from Jacques Peschon; Immunex Corp., Se-
attle, WA), which has a high binding efficiency for mouse TNF-



, was
administered at a dose of 100



g per mouse subcutaneously every
other day from Days 3–21 to isorecipients and allorecipients. Human
immunoglobulin G (100



g per mouse subcutaneously every other day,
Polygam R; Baxter Healthcare Corp., Glendale, CA) was adminis-
tered as a negative control. Trachea grafts from 15 different animals
(five TNFR:Fc-treated allografts, five TNFR:Fc-treated isografts, and
five immunoglobulin G-treated allografts) were harvested at 2, 3, 4,
and 6 weeks and were examined for the primary endpoint, graft occlu-
sion (using ImageQuant software), and a secondary endpoint, graft epi-
thelialization (morphometric analysis of the percentage of lumenal cir-
cumference covered by ciliated epithelium), by two blinded readers.

Statistical Analysis

A two-way analysis of variance was used to test the null hypothesis
that growth factor transcript levels were not different between al-
lografts and isografts over the 2- to 6-week course of mouse hetero-
topic trachea OB (18). The 10-week time point was excluded from the
analysis of variance because of the marked upregulation of the major-
ity of the growth factors in the isografts. Additionally, isograft/al-
lograft mRNA intensities were compared at each individual time
point with unpaired

t

tests. The TNFR:Fc experiment was analyzed
with a repeated-measures analysis of variance (SigmaStat; SPSS Inc.,
Chicago, IL). A two-sided



of less than 0.05 indicated significance.

RESULTS

Histology

Hematoxylin and eosin-stained frozen sections confirmed our
previous findings (6) of acute cellular (mononuclear) inflam-
mation in the allografts that peaked at the 2- to 4-week time
points and subsequently subsided, followed by progressive lu-
menal scarring, which began at 4 weeks and culminated at 10
weeks (Figure 1). Allograft epithelial injury in the form of tis-
sue shedding and basement membrane denudation was
present at 2 weeks, and epithelial destruction was complete by
4 weeks. Isograft morphology remained normal throughout
the study period.

RNA Isolation and Quantification

Mean total RNA yields from two tracheal grafts per sample
were 19.0



17.2



g (range 4.1–90.3



g) using the cesium-guani-
dine method but were much lower, usually immeasurably so,
using rapid RNA isolation kits, including the RNeasy Total RNA
System (Qiagen Inc., Valencia, CA) and RNAzol B Method
(Cinna Scientific Inc., Friendswood, TX). The isolated RNA
from each sample lacked digested, low molecular weight RNA
bands and displayed the presence of two distinct ribosomal
RNA bands after integrity gel electrophoresis (data not shown).

TABLE 1. PCR PRIMER SEQUENCES

cDNA of Interest Mouse Primer Sequences (a) Sense and (b) Antisense
Amplified cDNA
Sequence Length
FGF-2 (designed in-house)
Genbank #M30644
(a) 5




AAC TAC AAC TCC AAG CAG AAG AGA GA 3




(b) 5




TTA AGA TCA GCT CTT AGC AGA CAT 3




292 bp
FGF-1 (designed in-house)
Genbank #U67610
(a) 5




TGC GGG CGA AGT GTA TAT AAA G 3




(b) 5




GCA GAA ACA AGA TGG CTT TCT G 3




250 bp
PDGF-A
Stratagene, La Jolla, CA
(a) 5




GCC CCT GCC CAT TCG GAG GAA GA 3




(b) 5




GGC CAC CTT GAC GCT GCG GTG G 3




224 bp
PDGF-B (designed in-house)
from reference 44
(a) 5




CTG AGC TGG ACT TGA ACA TG 3




(b) 5




TTA AAC TTT CGG TGC TTG CC 3




508 bp
TGF-



Clonetech, Palo Alto, CA
(a) 5




TGG ACC GCA ACA ACG CCA TCT ATG AGA AAA CC 3




(b) 5




TGG AGC TGA AGC AAT AGT TGG TAT CCA GGG CT 3




525 bp
TNF-



Stratagene
(a) 5




ATG AGC ACA GAA AGC ATG ATC 3




(b) 5




TAC AGG CTT GTC ACT CGA ATT 3




276 bp
IGF-1
from reference 45
(a) 5




CTT CTG AGT CTT GGG CAT GTC AGT 3




(b) 5




TCG TCT TCA CAC CTC TTC TAC CTG 3




320 bp
ET-1 (designed in-house)
Genbank #U35233
(a) 5




TCA GAC ACG AAC ACT CCC TAA G 3




(b) 5




CAC AAC CGA GCA CAT TGA CTA C 3




392 bp
EGF-1 (designed in-house)
Genbank #J00380
(a) 5




AAG GAG AAG GGA TTC CTA TCT G 3




(b) 5




TAT TTA GCT GCC TTT CCA GGT C 3




328 bp



-actin
Clonetech
(a) 5




GTG GGC CGC TCT AGG CAC CAA 3




(b) 5




CTC TTT GAT GTC ACG CAC GAT TTC 3




540 bp

Definition of abbreviations

: EGF



epidermal growth factor; ET



endothelin; FGF



fibroblast growth factor; IGP



insulin-like growth
factor; PCR



polymerase chain reaction; PDGF



platelet-derived growth factor; TGF-

 

transforming growth factor-



; TNF-

 

tumor
necrosis factor.