Transcript Differential gene expression profiles of CAN

DIFFERENTIAL GENE
EXPRESSION PROFILES OF
CHRONIC ALLOGRAFT
NEPHROPATHY
Enver Akalin MD
Medical Director
Kidney and Pancreas Transplantation
Mount Sinai Medical School
New York, NY
CHRONIC ALLOGRAFT
NEPHROPATHY
• Leading cause of allograft failure after patient
death
• Despite decrease in acute rejection rates, protocol
biopsies still indicated ~ 50% CAN findings at 1
year protocol biopsies
» Bohmig et al. Transplant Int 2005; 18: 131
• Current Banff CAN score is based on extent of
fibrosis and does not provide information about
cause, progression, or the treatment of allograft
injury
Morphologic Features of Transplant Glomerulopathy
Chronic Allograft Nephropathy: nodular arteriolar
hyalinization (calcineurin inhibitor toxicity)
CHRONIC ALLOGRAFT
NEPHROPATHY
Immunologic Factors
Acute or subclinical rejection
HLA mismatch
Donor specific antibodies
Inflammation
and injury
Non-immunologic factors
Ischemia/reperfusion
Delayed graft function
Donor factors: Age
Calcineurin inhibitor toxicity
Viral nephritis (Polyoma, CMV)
Post-transplant HTN
Hyperlipidemia
Accelerated Senescence
Adhesion molecules, cytokines and growth factors
Interstitial fibrosis, tubular
atrophy, transplant glomerulopathy,
and fibrous intimal thickening of arteries
Chronic renal failure after transplantation of a
nonrenal organ (69,321 persons who received
transplantation in USA between 1990-2000)
N Engl J Med 2003; 349:931
The natural history of chronic allograft nephropathy
(Follow-up 119 kidney/pancreas transplant recipients
by protocol biopsies up to 10 years)
New Engl J Med 2003;349:2326
DONOR SPECIFIC ANTI-HLA
ANTIBODIES
• Anti-donor HLA antibodies are detected in 1260% of recipients after transplantation
• The sensitivity and specificity of detecting antiHLA antibody is increased by utilizing ELISA or
Flow Beads, as compared to CDC methods
• Kidney transplant recipients with donor specific
anti-HLA antibodies demonstrated 2-10 fold
higher incidence of acute rejection and 5-6 times
more likely developing chronic rejection
• 5 year graft survival was 34% in patients with
anti-HLA antibodies, compared to 76% without
antibodies.
» McKenna et al. Transplantation 2000; 69:319
ANTI-HLA ANTIBODIES
• Prospective analyses of 139 patients for 8 years by yearly
anti-HLA antibodies demonstrated that 29 patients
developed chronic rejection and 100% of these patients
had anti-HLA antibodies before rejection. 14 developed
antibodies de novo. In contrast, 27% of 110 stable patients
developed post-transplant anti-HLA antibodies.
» Lee et al. Transplantation 2002; 74:1192
• In a prospective, multicenter study 2,278 patients F/U by
anti-HLA antibodies for a year. 500 patients with anti-HLA
antibodies had 6.6% graft failure in a year compared to
3.3% of 1,987 patients without anti-HLA antibodies.
» Terasaki et al. Am J Transplant 2004; 4: 438
C4d
• The presence of C4d in allograft biopsy
samples is accepted as a finding of an acute
or chronic humoral alloimmune response
• 61% of biopsy samples with chronic
transplant glomerulopathy/arteriopathy had
positive C4d staining (12% of all CAN
biopsies), and 88% of these patients had
donor anti- HLA antibody
» Mauiyyedi et al. J Am Soc Nephrol 2000; 12:574
C4d
• 50% of CAN biopsies had + C4d staining and
these patients demonstrated lower graft survival (4
vs 8 years of half life)
» Lederer et al. Kidney Int 2001; 59: 334
• C4d deposits in PTC were detected in 73 of 213
late biopsies (34%) performed in 213 patients
more than 12 mo after Tx (median, 4.9 yr) because
of chronic allograft dysfunction. Endothelial C4d
deposition was found to be associated with
chronic transplant glomerulopathy and basement
membrane multilayering
» Regele et al. J Am Soc Nephrol 2002; 13: 2371
IMMUNE ACTIVATION IN CAN
We investigated the role of chemokines
(Mig, IP-10, and MCP-1) and chemokine
receptors (CXCR3 and CCR2) and ICOS in
chronic allograft nephropathy by
immunohistopathologic examination of
renal biopsy samples
» Akalin et al. Am J Transplant 2003; 3:1116
Leukocyte labeling within glomeruli of renal biopsies from
patients with CAN
Pathology Number ICOS
CXCR3 Mig
IP-10
CCR2
MCP-1
CAN +
TGP+
6/6
6/6
6/6
4/6
0/5
0/6
100%
100%
100%
67%
0%
0%
0/9
0/7
0/9
0/9
0/11
0/11
0%
0%
0%
0%
0%
0%
CAN+
TGP-
6
11
Chronic Allograft Nephropathy plus Transplant Glomerulopathy
CONCLUSIONS
• Increased expression of chemokines, their receptors,
and ICOS on leukocytes in biopsies showing CAN
and TGP vs. CAN alone suggest different
underlying pathologic mechanisms
• CXCR3-Mig/IP-10 pathway may play important
role in the pathogenesis of TGP
• Targeting chemokines or chemokine receptors
and/or immune intervention may have clinical
application in prevention and treatment of TGP
Diagnostic criteria's for chronic humoral
rejection
A. Allograft histopathology (3/4 required)
1.
2.
3.
4.
Arterial intimal fibrosis
Duplication of glomerular basement membrane
Laminated PTC basement membrane
Interstitial fibrosis and/or tubular atrophy
B. C4d positivity at PTC
C. Serologic evidence of donor anti-HLA
antibodies
Takemoto et al. Am J Transplant 2004; 4: 1033
GENECHIP TECHNOLOGY
• Has the power to measure the expression of
thousands of human genes simultaneously.
• Currently two types of methods,
oligonucleotide and cDNA microarrays,
have been successfully used to analyze the
gene expression patterns of different types
of cancers and inflammatory diseases for
diagnosis, and to predict prognosis and
response to therapy.
• Hypothesis-generating, rather than
hypothesis-driven
GENECHIP AND CHRONIC
ALLOGRAFT NEPHROPATHY
• Gene expression profiles of 17 protocol kidney biopsies at 6
months after transplantation were studied by oligonucleotide
arrays to predict the development of CAN at 12 months
• Patients were on RAD B251 study (Certican versus Cellcept
with Neoral and prednisone)
• 9 patients developed CAN at 12 months (All were on Certican)
• 10 genes might predict the development of chronic rejection (8
upregulated, 2 downregulated)
• Those genes were APRIL (acidic protein rich in leucins),
OBCML (opiate-binding protein-cell adhesion molecule like),
the tumor suppressor gene NPRL2, cytokeratin 15, homeobox
gene B7, prolactin receptor, and guanine nucleotide binding
protein gamma 7.
 Scherer et al. Transplantation 2003; 75: 1323
GENECHIP ANALYSIS OF
TRANSPLANT NEPHRECTOMIES
• Gene expression patterns of 13 transplant nephrectomy
samples with chronic rejection comparing to 16 normal
kidneys (histologically normal part of tumor nephrectomy
samples) and 12 PCKD nephrectomy samples were studied
by using cDNA arrays
• Hierarchical clustering analysis revealed two distinct
subsets of chronic rejection
• 22 genes were upregulated and 6 were downregulated in
CAN biopsies compared to PCKD nephrectomy samples
» Donauer et al. Transplantation 2003; 76: 539
GENECHIP AND KIDNEY
TRANSPLANTATION
• 67 allograft kidney biopsies from 50
pediatric patients were analyzed by cDNA
arrays
• While the gene expression patterns
suggested at least 3 possible distinct
subtypes of acute rejection, 19 CAN biopsy
samples were clustered together without
demonstrating any subset
» Sarwal et al. N Engl J Med 2003; 349: 125
GENECHIP IN CALCINEURIN
INHIBITOR AVOIDANCE STUDY
• Patients were randomized into 2 groups:
– CsA+MMF+Pred+Basiliximab
– Rapa+MMF+Pred+Basiliximab
• 61 patients were enrolled.
• 55 patients underwent renal function studies and
48 patients had protocol biopsies at 2 years after
transplantation
• 41 were analyzed by Affymetrix GeneChip
» Flechner et al. Am J Transplant 2004, 4:1776
GENECHIP IN CALCINEURIN
INHIBITOR AVOIDANCE STUDY
• Comparison of CsA/MMF/P group to
SRL/MMP/P group yielded 379 differentially
expressed genes and 97% of the genes are
upregulated in the CsA-treated patient biopsies
• Comparison of combined Banff 2 and 3 biopsies
to Banff 0 biopsies in revealed significant
upregulation of genes responsible for
immune/inflammation and fibrous/tissue
remodeling
GENECHIP ANALYSIS OF PROTOCOL
BIOPSIES AT 2 YEARS
Gene names
Fold change by GeneChip
GeneChip p-values2
Fold change by Q-PCR
Q-PCR p-values4
Matrix metalloproteinase 7 (47)
3.22
0.000001
15.01
0.026
TIMP1 (48)
2.25
0.000003
3.02
0.0855
CCL5 (RANTES) (49)
1.55
0.0004
5.51
0.035
VEGF (50)
1.57
0.0008
5.62
0.040
Collagen III (51)
2.52
0.003
24.31
0.010
MCP1/CCL2 (52)
1.36
0.003
2.82
0.043
Fibronectin (38)
1.70
0.005
19.66
0.005
TNF alpha (53)
1.58
0.008
189.09
0.026
Angiotensin II receptor (54)
1.12
>0.013
12.80
0.006
TGF beta (55)
1.28
>0.013
2.17
0.1625
PDGF (56)
1.06
>0.013
2.32
0.1775
STUDY AIMS
• To characterize gene expression patterns in
transplant kidneys with CAN by comparing
them to the expression patterns of transplant
kidneys with normal histopathology.
• To correlate gene expression profiles in
CAN with clinical features including
positive donor specific antibodies, C4d
staining, transplant glomerulopathy, and
nodular arteriolar hyalinization
METHODS
• One piece of allograft biopsy was snap frozen
in liquid nitrogen at the time of a clinically
indicated biopsy (rising creatinine and/or
proteinuria).
• Biopsies were evaluated according to the
Banff criteria, and were further stained for
C4d antibodies.
• Donor specific antibodies were studied by
Flow Beads (Luminex)
METHODS
• High-density oligonucleotide arrays (Affymetrix
HG-U133A contains 22,283 probe sets
representing about 16,000 genes per chip) were
used for gene expression analyses
• Hierarchical cluster analysis was chosen as a
cluster analysis technique
• Statistical modeling included Robust Multi Array
(RMA) to normalize the data and Significance
Analysis of Microarray (SAM) to identify the
most significant genes
Biotin - labeled
cRNA transcript
Cells
AAAA
Poly (A)
RNA
B
IVT
+
cDNA
Biotin-UTP
Biotin-CTP
B
B
B
B
B
B
Fragment
heat, Mg2+
Add
polyA
Controls
B
B
Hybridize
B
B
B
Wash & Stain
Scan
(8 minutes)
(16 hours)
(75 minutes)
Add Oligo B2 &
Staggered Spike
Controls
PATIENT CHARACTERISTICS
Age
D1
D2
D3
D4
D5
D6
D7
D8
D9
D10
D11
D12
D13
D14
D15
D16
Sex/Race
F/H
M/H
M/W
M/H
M/W
M/W
M/W
M/B
M/H
F/W
F/B
F/A
F/W
F/W
M/H
M/H
33
42
50
47
31
42
64
54
46
53
47
25
34
48
41
34
Tx Type
LR
LUR
CAD
LUR
LR
CAD
CAD
CAD
LR
CAD
CAD
CAD
CAD
CAD
LR
LR
Years
11
5
4
5
16
1
2
1
6
9
5
3
10
7
6
6
S Cr
3.1
2.5
2.7
2.9
3.6
2.1
2.3
1.8
2.3
3
3.7
4.1
1.0
3.5
3.8
3.5
UA
1.3 g
3+
negative
1+
1.5 g
negative
3+
negative
negative
negative
1.9 g
3+
4g
negative
1+
2+
1997 Banff
Grade 2
Grade 2-3
Grade 2
Grade 1-2
Grade 3
Grade 1
Grade 2-3
Grade 2
Grade 1
Grade 1-2
Grade 3
Grade 3
Grade 2
Grade 2
Grade 1
Grace 2
C1
C2
C3
C4
C5
C6
F/B
M/H
M/B
M/H
F/B
M/H
28
31
38
58
36
53
LR
LR
LUR
LR
LR
CAD
3 mos
13 mos
28 mos
7 wks
13 mos
3weeks
1.3
1.3
1.7
1.7
1.1
1.1
30
30
30
30
negative
negative
Grade 0
Grade 0
Grade 0
Grade 0
Grade 0
Grade 0
RESULTS
D1
D2
D3
D4
D5
D6
D7
D8
D9
D10
D11
D12
D13
D14
D15
D16
Donor Specific Abs
Neg
Neg
Neg
Neg
DR8
A2
Neg
Neg
Neg
Neg
Neg
Neg
Neg
Neg
Neg
ND
C4d
Neg
Neg
Neg
Neg
Pos
Neg
Neg
Neg
Neg
Neg
Neg
Neg
Neg
Neg
Neg
Neg
Nodular Arteriolar Hyalinization
Yes
Yes
No
Yes
Yes
No
Yes
No
No
Yes
Yes
No
No
No
No
Yes
GENECHIP RESULTS
• Among 22,283 probe sets, representing about
16,000 genes per chip, 37-59% of genes were
present in all samples
• Replicated samples demonstrated less than
0.03% gene expression difference with a high
degree of correlation reflecting by R2 of 0.983
and 0.989
RESULTS
• 455 probe sets representing 324 genes were differentially
expressed in CAN biopsies compared to controls. While 212
genes were upregulated a minimum of 1.5 fold, 112 genes were
downregulated in CAN samples.
• Biological functions of differentially expressed genes:
•
•
•
•
•
•
•
•
•
•
•
Cellular metabolism: 171 (53%)
Cellular communication: 105 (32%)
Cell growth/maintenance: 103 (32%)
Signal transduction: 79 (24%)
Cell adhesion: 39 (9%)
Immune response: 31 (9%)
Apoptosis: 12 (4%)
Blood coagulation: 12 (4%)
Humoral defense mechanism: 5 (2%)
Complement: 5 (2%)
Function unknown: 24 (7%)
RESULTS
• Selected upregulated genes related to immunology
– Transforming growth factor-β induced factor and
thrombospondin-1, which played role in TGF-
signaling pathway, and platelet derived growth factor-C
– The chemokine CXCL-6, and the adhesion molecules
vascular cell adhesion molecule, cell adhesion molecule
with homology to L1CAM, activated leukocyte cell
adhesion molecule and selectin P
– The complement-related upregulated genes were
complement component 4A and 4B, I factor, B factor
(properdin), and clusterin (complement lysis inhibitor)
– The other upregulated immune response related genes
were; MHC class II DR 4, and DP1, interferon-
inducible protein, interferon induced protein 44, IL-17B
receptor, natural killer cell transcript 4, CD24, and
CD59.
RESULTS
• Other selected genes:
– Genes related to fibrosis and extracellular
matrix deposition, such as; versican
(chondroitin sulfate proteoglycan 2), integrinbeta 3 and 6, matrix metalloproteinase 7, 9 and
10, laminin, fibronectin, tenascin, glypican 3
and 4, collagen type IV alpha 2, disintegrin, and
cadherin 2 and 6
– claudin 3, annexin A1, A3 and A4, connexin,
which are related to tight junction between cells
RESULTS
• Selected downregulated genes:
– Significant number of downregulated genes
were podocyte related, such as; podocin,
nephrin, Wilms tumor 1, podocalyxin, and
synaptopodin, or renal-specific genes; renin,
calbindin, adrenomedullin, stanniocalcin, and
kininogen
– vascular endothelial growth factor, epidermal
growth factor, and fibroblast growth factors 1
and 9, insulin-like growth factor binding
protein 3 and 5
RESULTS
• In one sample with both strong C4d staining and positive
DSA, more than 50% of the 100 genes with the highest
hybridization intensities in comparison analyses were
related to:
• immunoglobulin structures,
• complement (C3, C4a, C4b, C1q, complement factor H
and I),
• B cells (CD20 and CD48),
• T cell receptor, nuclear factor of activated T cells,
natural killer cell,
• cytokine receptors (Interleukin-2, 7, and 10 receptor),
• chemokines (Humig, MCP 1, RANTES), chemokine
receptors (CCR2 and CCR7)
VASCULAR ENDOTHELIAL
GROWTH FACTOR
• VEGF regulates angiogenesis, vascular
permeability, endothelial cell migration and
proliferation.
• VEGF plays role in the maintenance and
formation of podocytes and establishes glomerular
filtration barrier
• VEGF-A knockout mice (homozygous) die at
birth with hydrops and renal failure and podocyte
specific heterozygosity of VEGF-A resulted in
proteinuria and endotheliosis at 2.5 weeks of age
resembling preeclampsia
» Eremina et al. J Clin Invest 2003; 111: 707
VEGF FOR CYTOPROTECTION
• VEGF levels are decreased in preeklampsia
• VEGF is protective against cyclosporin
nephrotoxicity
» Caramelo et al. Nephrol Dial Transpl 2004; 19: 285
• VEGF and fibroblast factor-1 was
downregulated in biopsies with diabetic
nephropathy by genechip and
immunohistopathology
» Am J Kidney Dis 2004; 43: 636
VEGF FOR PROINFLAMMATION
• VEGF is upregulated in mice models of
acute and chronic rejection
• VEGF is upregulated in human heart and
kidney biopsies with chronic rejection
» Reinders et al. Transplantation 2003; 76: 224
» Pilmore et al. Transplantation 1999; 67: 929
• VEGF is upregulated in experimental
models of diabetic nephropathy
CONCLUSIONS
• A number of genes are differentially expressed in biopsies with
CAN compared to normal transplant kidney biopsies by
GeneChip technology.
• Some of the upregulated genes are related to fibrosis (MMP-7, 9,
and 10, laminin, fibronectin, and tenascin) or factors/cytokines,
which induce fibrosis (TGF-β, thrombospondin-1, and PDGF).
• Gene expression of VEGF, EGF, FGF-1 and 9 were
downregulated in CAN. Immunpathologic examination indicated
the lack of VEGF production by the glomerular cells in CAN.
• Hierarchical cluster analyses demonstrated that CAN samples did
not show distinct subsets, including biopsies with nodular
hyalinosis.
• Gene expression profile of one patient with positive C4d staining
and donor-specific anti-HLA antibodies indicated the role of
immune-mediated mechanisms.
ROUTINE C4d STAINING
• Since July 2003, all clinically indicated transplant kidney
biopsies were routinely studied by C4d staining and
patients with positive C4d staining were routinely studied
for donor-specific anti-HLA antibodies (DSAs) by Flow
Beads (Luminex)
• 166 biopsies from 128 patients demonstrated that 22
biopsies (13%) from 11 patients (9%) were C4d +
• 9 patients with C4d+ had early AHR and 2 had CAN
• All C4d+ patients had DSAs
• Only 2 out of 62 biopsies (3%) with CAN (22 grade I, 26
grade II, and 14 grade III) were C4d +
• There were 11 CAN biopsies with transplant
glomerulopathy, and 4 with secondary focal segmental
glomerulosclerosis, and all were C4d negative
ROUTINE C4d STAINING
• 54 biopsies in 43 patients showed CAN
• 20 biopsies (37%) exhibited transplant
glomerulopathy and none showed
peritubular C4d staining
• Only 2 biopsies were C4d +. One had DSAs
and the other had no detectable DSAs
» Aly et al. Transplant Int 2005; 18: 800
LIMITATIONS OF MICROARRAY
STUDIES IN CAN
• GeneChip technology has been used in CAN patients in a
limited number of studies involving limited number of
patients (< 20)
• Difficulty in evaluating vast quantities of data to reach a
meaningful conclusion in a limited number of
heterogeneous group of patients and immunosuppressive
medications for a very heterogeneous disease like CAN
• Tissue sampling differences, as biopsy samples contain a
mixture of different cell and tissue types, involving varying
proportions of muscle, capsule, cortex and medulla of the
kidney
• Limited amount of total RNA
• The data analysis of microarrays is still in its infancy and
there is no universally accepted method
FUTURE DIRECTIONS
• In order for microarray studies to define certain
subgroups of patients for diagnosis, prognosis, and
management of CAN, or find candidate genes,
novel pathways, and/or surrogate markers,
hundreds of biopsies should be performed in
multicenter collaborative studies with a
prospective manner starting microarray analysis of
transplant kidney biopsies before implantation and
follow-up protocol biopsies until CAN develops
• Microdissected samples should be used to avoid
sampling errors