Tertiary lymphoid organs in renal allografts can be associated with donor-specific tolerance rather...

Tertiary lymphoid organs in renal allografts can beassociated with donor-specific tolerance rather thanrejection

Kathryn Brown, Steven H. Sacks and Wilson Wong

MRC Centre for Transplantation, King’s College London School of Medicine at Guy’s, King’s

and St. Thomas’ Hospitals, London, UK

Tertiary lymphoid organs can form at sites of chronic inflammation. Their presence has

been mainly associated with tissue destruction. In transplantation, there is a dynamic

immune response as in chronic inflammation. Indeed, the presence of tertiary lymphoid

organs has been associated with chronic rejection. In addition to a destructive alloimmune

response, secondary lymphoid organs are also important in transplant tolerance. We

hypothesised that tertiary lymphoid organs may also form during transplantation toler-

ance as this process also requires an active local immune response. If so, their presence

may enhance tolerance, resulting in better graft function rather than chronic rejection.

Using a mouse kidney allograft model of tolerance, we demonstrate the formation of

tertiary lymphoid organs within tolerated allografts. Tertiary lymphoid organs are

supplied by high endothelial venules, and contain T and B cells, macrophages, DC, Foxp31

T cells, donor MHC class II1 cells and recipient cells presenting donor-derived allopeptides.

Formation of tertiary lymphoid organs and the presence of immune cells within them are

associated with superior graft function, suggesting that tertiary lymphoid organs act to

amplify the prevailing immune response, be it a tolerant and beneficial immune response

or the previously described destructive alloimmunity.

Key words: Tertiary lymphoid organs . Tolerance . Transplantation

Introduction

Secondary lymphoid organs are crucial for the adaptive immune

response. Antigen-presenting cells migrate from the site of

inflammation to secondary lymphoid organs that provide a niche

for effective encounters between naı̈ve T cells and their cognate

antigens. In chronic inflammation, however, where antigens are

not cleared, tertiary lymphoid organs (TLO) can develop

ectopically in the periphery (reviewed in [1]). These are

organised structures, resembling lymph nodes, and have been

extensively studied in the context of autoimmune diseases such as

rheumatoid arthritis; chronic infections such as hepatitis C and

malignancy [1].

After organ transplantation, there are continuous low-level,

dynamic interactions between donor alloantigens and the reci-

pient’s immune system, in a situation similar to chronic inflam-

mation. However, as in autoimmune diseases, the antigens

cannot be cleared. As a result, TLO have been observed after

transplantation, for example in cardiac allgorafts undergoing

chronic rejection [2]. TLO within skin allografts were able to

mount an effective alloresponse, leading to rejection and devel-

opment of a memory response [3].

Germinal centres have been found in chronically rejecting rat

aortic, and human heart and kidney grafts [4]. Reports to date

have, in the main, suggested that formation of TLO within

transplanted organs is associated with poorer graft outcome.

SHORT COMMUNICATION

Correspondence: Dr. Wilson Wonge-mail: [email protected]

& 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.eji-journal.eu

Eur. J. Immunol. 2011. 41: 89–96 DOI 10.1002/eji.201040759 Cellular immune response 89

However, the fate of a transplanted organ depends on the balance

between graft destructive alloreactive T cells and graft protective

regulatory T cells [5]. There are examples in humans where the

balance is tipped in favour of regulation, allowing immunosup-

pressive therapy to be reduced, or ceased entirely [6, 7]. In

animal models, this has been described in numerous

situations [8–11]. Tolerance can be viewed as a dynamic inter-

action between donor antigens and recipient T cells [12]

analogous to the situation seen in chronic inflammation. It

involves an active local immune response which may lead to the

development of TLO. Indeed, lymphoid aggregates have been

described in two long-term tolerant human kidney allografts

[13]. The small number of tolerant transplant recipients, and the

sampling error inherent in biopsy samples, makes further inves-

tigation difficult. This prompted us to use a mouse kidney

allograft model where the recipient is tolerant to donor antigens,

resulting in indefinite graft survival in some recipients

and prolonged survival of donor type challenge skin grafts [11],

to see if TLO can be associated with donor-specific unrespon-

siveness.

Results and discussion

Lymphocytic clusters are present within tolerantkidney allografts

We have used a previously described fully allogeneic DBA/2 to

C57BL/6 mouse kidney allograft model in which 40% of

allografts were accepted indefinitely with normal graft function

and histology, whereas a further 40% of grafts were not rejected

acutely but survive up to 105 days with slowly deteriorating graft

function as reflected by blood urea nitrogen (BUN) and

histological features of interstitial fibrosis and tubular atrophy

(IFTA). Since there are no other confounding factors such as

calcineurin inhibitors in this model, the cause of the IFTA is likely

to be chronic rejection [11]. C57BL/6 recipients of DBA/2

kidneys were sacrificed at various time points after transplanta-

tion for histological examination. Syngeneic C57BL/6 kidney

grafts had no cellular infiltrates at day 14. Minimal infiltrates

were seen at day 45 and beyond (Fig. 1A–D). In DBA/2 allografts,

only small amounts of mononuclear cellular infiltrates were seen

at day 8 post-transplant scattered throughout the cortex; these

did not appear to be organised into clusters. At later time points,

clusters of lymphocytic infiltrates could be seen in most sections

(Fig. 1E–H). Their appearance differs from that seen at earlier

time points by being densely packed and having well-defined

margins.

Lymphocytic clusters satisfy the criteria for TLO

The exact definition of TLO has not yet been agreed. However,

the presence of T, B and DC, and high endothelial venules (HEV)

has been suggested. To determine whether the clusters of

lymphocytic infiltrates seen here were TLO, we carried out

histological analysis to investigate whether they fulfil these

criteria.

Constituents of cellular infiltrates resemble that oflymph nodes

Conventional secondary lymphoid tissues contain a variety of

immune cells that are essential constituents for co-ordinated

and effective immune responses. DBA/2 kidney allografts

were therefore examined to identify the cell types present

within the infiltrates. T cells (CD41, CD81 and regu-

latory T cells), B cells, DC and macrophages were all found

in lymphocytic clusters in all tissue sections examined

(Fig. 1I–N).

Clusters of cellular infiltrate contain HEV

Tissue sections were stained for the presence of HEV, sections of

vascular endothelium specialised to allow extravasation of

lymphocytes and normally found only in lymphoid organs, using

the specific marker peripheral lymph node addressin (PNAd).

PNAd1 vessels were seen in kidney allografts from 14 days post-

transplantation, and increased to 0.2970.17 positive vessels per

medium power field (mpf (� 160)) at day 45 post-transplanta-

tion, compared with 0.0870.1 in syngeneic kidney grafts

(Figs. 1O and 2A). This rose to 0.4870.16 positive vessels per

mpf 480 days post-transplantation. At this time point, all but one

kidney allograft had demonstrable PNAd1 vessels. The density of

PNAd1 vessels in conventional lymph nodes is usually much

higher than we have observed here. The reason and functional

relevance of this is unclear. We did not detect PNAd1 vessels in

some of the TLO, which may be at least partially due to their low

densities. Alternatively, PNAd1 vessels may take longer than 80

days to develop.

Put together, these data suggest that the cellular infiltrates

observed in the DBA/2 allografts were TLO and they will be

referred to as such from this point onwards.

Lymphatic neogenesis in kidney allografts

Lymphatic vessels may act to facilitate the trafficking of immune

cells through TLO [14]. We therefore stained DBA/2 kidney

allografts for the presence of podoplanin, a marker of lymphatic

endothelial cells. A higher density of podoplanin1 vessels was

seen from day 45 after transplantation, with an average of

1.6870.74 positive vessels per mpf (compared with 0.5670.15

in syngeneic grafts at the same time point), and increased to

8.6171.94 in the 480 days post-transplantation group (Figs. 1P

and 2B). In all kidney allografts sacrificed 480 days post-

transplantation, at least 60% of all TLO contained a podoplanin1

vessel.

Eur. J. Immunol. 2011. 41: 89–96Kathryn Brown et al.90


C57BL/6 donor kidneys

A B C D125µm 125µm 125µm 125µm

DBA/2 donor kidneys

E F H125µm 125µm 125µm 125µm

day 8 day 14 day 45 >day 80

G

Days post transplant day day day >daytransplant

J KI 125µm 125µm 125µm

NML

3pxoF8DC4DC

125µm 125µm 125µm

P QO

86DCc11DC91DC

200µm80µm 80µm

TSR

SAPninalpodopdANP

200µm 80µm 80µm

PAS I-Ad Y-Ae

Figure 1. Cellular infiltrates are present within tolerant transplant allografts. Kidneys from syngeneic C57BL/6 or allogeneic DBA/2 micetransplanted into C57BL/6 recipients were harvested at days 8 (A, E), 14 (B, F), 45 (C, G) or 480 (D, H) after transplantation and stained with PAS toidentify area of cellular infiltrates. Sections of a kidney allograft harvested at 95 days post-transplantation were also stained to identify cell typeswithin infiltrates: CD41 (I), CD81 (I), Foxp31 (K), CD191 B cells (L), CD11c1 DC (M) and CD681 macrophages (N). Sections of kidney allograft werestained to identify PNAd (O) and podoplanin1 (P) vessels to distinguish HEV and lymphatic vessels, respectively. PAS-stained sections of long-term(480 days post-transplant) kidney allografts with good (BUN 5 17.1 mmol/L at harvest) (Q) and poor (BUN 5 45.4 mmol/L at harvest) (R) graftfunction, showing cellular infiltrates in tolerant (Q) but not in chronically rejecting (R) kidney. Donor antigens were identified by staining for I-Ad1

donor DC (S); and Y-Ae1 cells expressing recipient MHC class II presenting donor peptide (T). Original magnifications �160 (A–H, I–N); � 250 (O, P,S, T); � 100 (Q, R). PAS, periodic acid Schiff; PNAd, peripheral lymph node addressin and BUN, blood urea nitrogen. A representative of at least threeexperiments is shown.

Eur. J. Immunol. 2011. 41: 89–96 Cellular immune response 91


Large cross-sectional area of TLO is associated withsuperior graft function

TLO thus far have mostly been associated in humans and

experimental animal models with having a destructive

effect. However, in our model TLO appear to be more pro-

minent in allografts with good function. We therefore investi-

gated the correlation between various attributes of TLO

objectively to see if any features were associated with superior

graft function.

Allograft sections were examined under mpf and the area of

individual TLO measured and expressed as mean area per mpf

(Fig. 2C). A correlation was found between mean TLO area and

graft function as measured by BUN, confirming our initial

subjective observation (r2 5 0.6541, p 5 0.0024, Figs. 1Q and

2D). Almost no TLO were seen in grafts from recipients

with BUN of Z40mmol/L (Fig. 1R), whereas grafts with TLO

area of Z150mm2/mpf were always associated with BUN of

30 mmol/L or less. Therefore, unlike other models, here TLO

formation was associated with a positive rather than a negative

outcome.

To determine whether area of TLO also correspond to graft

function in the tolerant state, we performed sub-group analysis

on recipients that had survived 480 days. Although there was a

trend towards larger TLO areas and better graft function,

(r2 5 0.399, p 5 0.068), it was just below the limit for statistical

significance. One possible explanation for this is that the majority

of the recipients in this sub-group were tolerant and therefore all

had good graft function; many recipients with IFTA do not

survive this long and are therefore under represented.

HEV marker PNAd is associated with superior graftfunction

Although not all of the lymphocytic clusters contained PNAd1

vessels, the presence of PNAd1 vessels within TLO was associated

with better graft function (r2 5 0.521, p 5 0.0281), in those

allografts harvested day 45 post-transplantation onward

(Fig. 2E).

Podoplanin1 vessels are associated with poor graftfunction

Significant amounts of podoplanin1 vessels were observed only

80 or more days after transplantation (Fig. 2B). Therefore, they

were used for analysis here. In contrast to staining for HEV using

PNAd, a strong correlation was seen between the presence of

podoplanin expressing vessels and the poor graft function

(r2 5 0.759, p 5 0.0239) (Fig. 2F).

A higher percentage of TLO containing podoplanin1 vessels

was found to be associated with poor graft function, whereas on

the contrary, a higher percentage of TLO containing PNAd1

vessels was associated with good graft function. These results

suggest that the entry of lymphocytes into the TLO (through

HEV) is beneficial for the allograft, whereas the exit of immune

cells (suggested to be through the newly formed lymphatic

vessels [15]) is detrimental to the allograft. This adds to the

theory that TLO here are not harmful, and may actually be of

benefit, in this model of tolerance. The podoplanin results were in

contrast to the studies by Stuht et al., who found that the

presence of lymph vessels within cellular infiltrates was asso-

ciated with improved graft function [15]. This may be because

their patients had chronic immune damage in their kidneys,

whereas our mouse model is of tolerant kidney grafts, further

supporting the notion that TLO perpetuate and amplify the

prevailing immune response in the local environment, be it

rejection or tolerance.

We have not found any major differences between

the tolerogeneic TLO seen in this model of spontaneous

acceptance, and the TLO found by other groups in

chronic rejection models. However, these structures are likely to

differ in the activity of the cells within them, the balance of

regulatory to effector cells and the cytokines and chemokines

present.

High percentage of leukocytes within TLO is correlatedwith better graft function

When the average number of CD41 cells per high power field (hpf

(� 400)) present in kidney allografts from all time points was

correlated with graft function as measured by BUN, there was

somewhat surprisingly a trend towards lower numbers of CD41

cells within the graft being associated with poor graft function,

although this trend did not reach statistical significance (Fig. 3A).

However, those grafts with poor function had a significantly

higher percentage of CD41 cells outside TLO (r2 5 0.566,

p 5 0.0048) (Fig. 3B). This suggests that although there are

large numbers of CD41 cells within allografts with good function,

the majority of these cells are within TLO and are not adversely

affecting the function of the graft.

The same pattern was observed with CD81 (Fig. 3C and D),

Foxp31 (Fig. 3E and F), CD191 (Fig. 3G and H), CD11c1 (Fig. 3I

and J) and CD681 (Fig. 3K and L) cells. There was a trend

towards lower cell numbers in allografts with poorer function.

However, the grafts with worse function had a signi-

ficantly higher percentage of these cells outside the boundaries of

TLO.

Using the total number of infiltrating leukocytes as an indi-

cator of severity of the rejection process may be misleading if the

presence of TLO is not taken into account.

Presentation of donor antigens within TLO

In organ transplantation, intact donor antigens on donor-derived

passenger dendritic leukocytes are recognised by recipient T cells

via the direct antigen presentation pathway, whereas donor-



derived peptides are presented by recipient antigen-presenting

cells in the context of self-MHC molecules, through the indirect

antigen presentation pathway. Lymphoid organs provide a

physical location for antigen-specific T cells from both antigen

presentation pathways to meet their cognate ligands. We

investigated whether TLO seen here contained antigens for the

direct and indirect pathways.

Sections of kidney allografts were stained for the presence of

the intact donor MHC class II molecule I-Ad. I-Ad1 cells could

clearly be seen within TLO at all time points, revealing the

presence of donor antigen-presenting cells within the TLO

(Fig. 1S).

Sections were also stained with the monoclonal antibody

Y-Ae which recognises the allopeptide Ea52–68 in the

context of I-Ab. It has been previously shown to identify I-Ab1

C57BL/6 cells presenting the relevant peptide derived from I-Ed

MHC class II molecules [16]. Staining was not seen at the earliest

time point analysed (14 days), but was observed from day 45

post-transplantation onwards, and the numbers of Y-Ae1 cells

increased again 480 days post-transplantation (1.6370.64 and

1.25 15

A BninalpodoPdANP

BL/6 d14

BL/6d45

BA/2 d14

BA/2d45

>80

days

0.00

0.25

0.50

0.75

1.00

no

. of

+ve

vess

els

per

mp

f

7BL/6

d14

7BL/6

d45

DBA/2 d14

DBA/2d45

>80

days

0

5

10

no

. of

+ve

vess

els

per

mp

f

500

750

1000

1250

area

C57 C57 D D

DBA/2>8

C57B

C57B DB DB

DBA/2>8

C

D

20

30

40

50

60

70

C57BL/6

d14

C57BL/6

d45

C57BL/6

>80 d

ays

DBA/2 d14

DBA/2d45

DBA/2>8

0 day

s0

250

0 100 200 300 400 500 600 7000

10

TLO area (µm2) per mpf

blo

od

ure

a n

itro

gen

(m

mo

l/L)

E

50

60

F

50

60

% TLO containing PNAd+ vessels

0 10 20 300

10

20

30

40

blo

od

ure

a n

itro

gen

(m

mo

l/L)

% TLO containing podoplanin+ vessels

50 60 70 80 90 100 1100

10

20

30

40

blo

od

ure

a n

itro

gen

(m

mo

l/L)

0

1

2

3

4

5

6

no

. of

Y-A

e+ c

ells

per

hp

f

HG

0

10

20

30

40

50

60

blo

od

ure

a n

itro

gen

(m

mo

l/L)

DBA/2 d14

DBA/2 d45

DBA/2 >8

0 days 0 1 2 3 4 5 6 7

no. of Y-Ae+ cells per hpf

I

40

50

60

10

20

30

0 10 20 30 40 50 60 70 800

%Y-Ae+ cells outside TLOs

blo

od

ure

a n

itro

gen

(m

mo

l/L)

Figure 2. Correlation between graft function and PNAd, podoplanin, TLO area and Y-Ae1 cells. Kidneys from syngeneic C57BL/6 or allogeneic DBA/2mice transplanted into C57BL/6 recipients were harvested at day 14, 45 or 480 days after transplantation and stained for (A) PNAd to show thedevelopment of HEV or (B) podoplanin to show the development of lymphatic vessels. (C) Degree of TLO formation as measured by the average areaper mpf covered by cellular infiltrate. (D) Correlation between average area of infiltrate per mpf and graft function as measured by BUN. (E) Usingonly those samples with significant PNAd staining (day 45 post-transplantation onwards), the percentage of clusters within the kidney allograftcontaining PNAd1 vessel(s) was correlated with BUN. (F) Using only those samples with significant podoplanin staining (480 days post-transplantation group only), the percentage of clusters within the kidney allograft containing podoplanin1 vessel(s) was correlated with BUN.(G) Staining for Y-Ae shows the presence of recipient MHC class II molecules presenting donor peptide. (H) Correlation between total Y-Ae1 cellnumbers and graft function. (I) Correlation between percentage of Y-Ae1 cells outside TLO and graft function. In (E, F, H and I), linear regressionanalyses were used to check for association between the features of TLO and BUN. Non-linear quadratic regression was used for (D) and comparedagainst a linear model. The best fitting one (quadratic) is shown. TLO, tertiary lymphoid organs; PNAd, peripheral lymph node addressin and BUN,blood urea nitrogen. Statistical significance measured using a Mann–Whitney test. ns, non significant. Each data point (solid circle) represents anindividual kidney recipient.



2.2671 Y-Ae1 cells per hpf, respectively), confirming the

presentation of donor antigen by recipient antigen-presenting

cells (Figs. 1T and 2G). The majority of these cells were found

within TLO.

The presence of donor I-Ad1 cells and Y-Ae1 recipient

cells presenting donor antigen within TLO suggests that

TLO are actively participating in the local immune

response by acting as a site of both direct and indirect allor-

ecognition.

When numbers of Y-Ae1 cells were correlated with graft

function, a non-significant trend towards higher numbers of

Y-Ae1 cells in grafts with poor function was seen (Fig. 2H).

However, a stronger association was found between a high

percentage of Y-Ae1 cells outside TLO and a poor graft function

20

30

40

50

60ea

nitr

oge

n (m

mol

/L)

20

30

40

50

60

ea n

itro

gen

(mm

ol/L

)A Br2=0.566p=0.0048

r2=0.188p=0.160

0 25 50 75 100 125 1500

10

no. of CD4+ cells per hpf

blo

od

ure

0 25 50 75 1000

10

%CD4+ cells outside TLOs

bloo

d ur

e

40

50

60

oge

n (m

mol

/L)C

r2=0.118p=0.275

0 50 100 150 2000

10

20

30


blo

od

urea

nitr

o

50

60

mol

/L)

r

F

0 25 50 75 100 1250

10

20

30

40

%Foxp3+ cells outside TLOs

bloo

d ur

ea n

itro

gen

(mm r2=0.523

p=0.011940

50

60

gen

(mm

ol/L

)D

r2=0.348p=0.043

0 25 50 75 1000

10

20

30


bloo

d ur

ea n

itro

50

60

mol

/L)

r

0 25 50 75 1000

10

20

30

40

no. of Foxp3+ cells per hpf

blo

od

urea

nitr

oge

n (m

m

E

r2=0 263p=0.089

0 25 50 75 1000

10

20

30

40

50

60

blo

od

urea

nitr

oge

n (m

mol

/L)

0 25 50 75 100 1250

10

20

30

40

50

60

bloo

d ur

ea n

itro

gen

(mm

ol/L

)HGr2=0.526p=0.0076

r2=0.079p=0.378


20

30

40

50

60

urea

nitr

oge

n (m

mol

/L)

20

30

40

50

60

urea

nit

roge

n (m

mol

/L)

0 25 50 75 100 125


I

KJ L

r2=0.066p=0.0013

r2=0.055p=0.464

0 25 50 75 100 125 1500

10

no. of CD11c+ cells per hpf

blo

od

u

30 40 50 60 70 80 900

10

%CD11c+ cells outside TLOs

bloo

d u

40

50

60

og

en (

mm

ol/L

)

40

50

60

ogen

(m

mol

/L)

r2=0.554p=0.0055

r2=0.002p=0.886

0 5 10 15 20 250

10

20

30

score per hpf

blo

od

ure

a n

itro

0 25 50 75 100 1250

10

20

30

%CD68 staining outside TLOs

bloo

d ur

ea n

itro

Figure 3. Correlation of graft function with histological parameters. (A, C, E, G, I, K) Numbers of CD4, CD8, Foxp3, CD19 and CD11c1 cells per hpfand CD68 score, respectively, versus BUN, in DBA/2 kidney allografts transplanted into C57BL/6 recipients, harvested 14, 45 and 480 days post-transplantation. (B, D, F, H, J, L) Correlation between the percentage of cells outside of TLO and BUN. In the case of Foxp3, one sample was excludedfrom the analysis due to the paucity of Foxp31 cells (one in entire section) within the section, making statistics unreliable. In all cases, linearregression analyses were used to check for correlation between the numbers or percentages of cells, and BUN. TLO, tertiary lymphoid organs andBUN, blood urea nitrogen. Statistical significance was measured using Mann–Whitney test.



(r2 5 0.419, p 5 0.0315) (Fig. 2I). Hence, again those cells

outside TLO appear to be more damaging to the allograft. It has

been proposed that regulatory T cells act mainly through the

indirect pathway [17], and TLO may therefore promote this

interaction.

Concluding remarks

We have described a model of TLO in which they behave in the

opposite manner to that previously described, suggesting that

TLO, like secondary lymphoid tissues, merely serve to amplify the

prevailing local immune response and may not necessarily cause

tissue damage. Further understanding of their function and the

circumstances under which they form may lead to the develop-

ment of therapeutic strategies to deviate the immune response

towards the desired direction.

Materials and methods

Animals

Female donor DBA/2 (H-2d) and recipient C57BL/6 (H-2b) mice

(Harlan UK, Oxon, UK) were kept in specific pathogen-free

facilities and used in accordance with the Animals (Scientific

Procedures) Act 1986.

Organ transplantation

Renal transplantation was performed as described previously

[18]. The left native kidney was removed at the time of

transplantation. The remaining native kidney was removed 7

days later, leaving the donor graft life sustaining. BUN was

measured at regular intervals to monitor graft function, using

Infinity Urea (Thermo Fisher Scientific, Middletown, CT, USA)

according to the manufacturer’s instruction. Tissues were

harvested as described previously [11].

Histology

Periodic acid Schiff (PAS) staining was carried out on 2mm thick

paraffin sections as described previously [19].

The area of the cellular infiltrate was measured in at least

12 mmpf of blinded samples using LUCIA G software (Nikon,

Tokyo, Japan) and means were calculated.

Immunohistochemistry

Frozen tissue samples were stained as described previously [11],

with the primary antibodies anti-podoplanin (Abcam plc,

Cambridge, UK), anti-CD4 (clone H129.19, BD, NJ, USA), anti-

CD8 (clone 53-6.7, BD), anti-CD19 (clone 1D3, BD), anti-CD11c

(clone N418, BioLegend, San Diego, CA, USA), anti-CD68 (clone

FA-11, Serotec, Oxford, UK), anti-PNAd (clone MECA-79,

BioLegend), anti-Foxp3 (clone FJK-16s, eBioscience, San Diego,

CA, USA), anti-I-Ad (clone AMS-32.1, BD) and anti-Y-Ae (clone

eBioY-Ae, eBioscience).

Positive vessels were counted in at least 10 mpf of

blinded samples and the means were calculated. Positive cells

were counted in 20 random hpf of each blinded sample,

and the means were calculated. For macrophage staining, a 63-

point grid was overlaid onto a �400 magnification of

macrophage staining. The percentages of grid points positive for

CD68 staining were counted in 20 hpf, and the means were

calculated.

Histological data were correlated to allograft function using

regression analyses (linear or quadratic) with GraphPad Prism 4

(GraphPad Software, San Diego, CA, USA). Mann–Whitney test

was used to test for statistical differences between histological

scores from different experimental groups. In all cases, values are

expressed as mean7SEM.

Acknowledgements: This work was supported by a grant from

the Genzyme Renal Innovations Program. The authors thank

Dr. Irene Rebollo Mesa for statistical help.

Conflict of interest: The authors declare no financial or

commercial conflict of interest.

References

1 Aloisi, F. and Pujol-Borrell, R., Lymphoid neogenesis in chronic inflam-

matory diseases. Nat. Rev. Immunol. 2006. 6: 205–217.

2 Baddoura, F. K., Nasr, I. W., Wrobel, B., Li, Q., Ruddle, N. H. and Lakkis, F.

G., Lymphoid neogenesis in murine cardiac allografts undergoing chronic

rejection. Am. J. Transplant. 2005. 5: 510–516.

3 Nasr, I. W., Reel, M., Oberbarnscheidt, M. H., Mounzer, R. H., Baddoura,

F. K., Ruddle, N. H. and Lakkis, F. G., Tertiary lymphoid tissues generate

effector and memory T cells that lead to allograft rejection. Am.

J. Transplant. 2007. 7: 1071–1079.

4 Thaunat, O., Field, A. C., Dai, J., Louedec, L., Patey, N., Bloch, M. F.,

Mandet, C. et al., Lymphoid neogenesis in chronic rejection: evidence for

a local humoral alloimmune response. Proc. Natl. Acad. Sci. USA 2005. 102:

14723–14728.

5 Zheng, X. X., Sanchez-Fueyo, A., Sho, M., Domenig, C., Sayegh, M. H. and

Strom, T. B., Favorably tipping the balance between cytopathic and

regulatory T cells to create transplantation tolerance. Immunity 2003. 19:

503–514.

6 Uehling, D. T., Hussey, J. L., Weinstein, A. B., Wank, R. and Bach, F. H.,

Cessation of immunosuppression after renal transplantation. Surgery

1976. 79: 278–282.



7 Zoller, K. M., Cho, S. I., Cohen, J. J. and Harrington, J. T., Cessation of

immunosuppressive therapy after successful transplantation: a national

survey. Kidney Int. 1980. 18: 110–114.

8 Ildstad, S. T. and Sachs, D. H., Reconstitution with syngeneic plus

allogeneic or xenogeneic bone marrow leads to specific acceptance of

allografts or xenografts. Nature 1984. 307: 168–170.

9 Larsen, C. P., Elwood, E. T., Alexander, D. Z., Ritchie, S. C., Hendrix, R.,

Tucker-Burden, C., Cho, H. R. et al., Long-term acceptance of skin and

cardiac allografts after blocking CD40 and CD28 pathways. Nature 1996.

381: 434–438.

10 Bickerstaff, A. A., Wang, J. J., Pelletier, R. P. and Orosz, C. G., Murine renal

allografts: spontaneous acceptance is associated with regulated T cell-

mediated immunity. J. Immunol. 2001. 167: 4821–4827.

11 Brown, K., Moxham, V., Karegli, J., Phillips, R., Sacks, S. H. and Wong, W.,

Ultra-localization of Foxp31 T cells within renal allografts shows

infiltration of tubules mimicking rejection. Am. J. Pathol. 2007. 171:

1915–1922.

12 Waldmann, H., Adams, E., Fairchild, P. and Cobbold, S., Infectious

tolerance and the long-term acceptance of transplanted tissue. Immunol.

Rev. 2006. 212: 301–313.

13 Xu, Q., Lee, J., Jankowska-Gan, E., Schultz, J., Roenneburg, D. A., Haynes,

L. D., Kusaka, S. et al., Human CD41CD25low adaptive T regulatory cells

suppress delayed-type hypersensitivity during transplant tolerance. J.

Immunol. 2007. 178: 3983–3995.

14 Kerjaschki, D., Regele, H. M., Moosberger, I., Nagy-Bojarski, K.,

Watschinger, B., Soleiman, A., Birner, P. et al., Lymphatic neoangiogenesis

in human kidney transplants is associated with immunologically active

lymphocytic infiltrates. J. Am. Soc. Nephrol. 2004. 15: 603–612.

15 Stuht, S., Gwinner, W., Franz, I., Schwarz, A., Jonigk, D., Kreipe, H.,

Kerjaschki, D. et al., Lymphatic neoangiogenesis in human renal

allografts: results from sequential protocol biopsies. Am. J. Transplant.

2007. 7: 377–384.

16 Ochando, J. C., Krieger, N. R. and Bromberg, J. S., Direct versus indirect

allorecognition: visualization of dendritic cell distribution and interac-

tions during rejection and tolerization. Am. J. Transplant. 2006. 6:

2488–2496.

17 Yamada, A., Chandraker, A., Laufer, T. M., Gerth, A. J., Sayegh, M. H. and

Auchincloss, H., Jr., Recipient MHC class II expression is required to

achieve long-term survival of murine cardiac allografts after costimula-

tory blockade. J. Immunol. 2001. 167: 5522–5526.

18 Han, W. R., Murray-Segal, L. J. and Mottram, P. L., Modified technique for

kidney transplantation in mice. Microsurgery 1999. 19: 272–274.

19 McManus, J., Histological and histochemical uses of periodic acid. Stain

Technol. 1948. 23: 99–108.

Abbreviations: BUN: blood urea nitrogen � HEV: high endothelial venule

� hpf: high power field � IFTA: interstitial fibrosis and tubular atrophy �mpf: medium power field � PAS: periodic acid Schiff � PNAd: peripheral

lymph node addressin � TLO: tertiary lymphoid organs

Full correspondence: Dr. Wilson Wong, MRC Centre for Transplantation,

5th Floor, Tower Wing, Guy’s Hospital, London SE1 9RT, UK

Fax: 144-2071885660

e-mail: [email protected]

Received: 21/6/2010

Revised: 12/8/2010

Accepted: 27/9/2010

Accepted article online: 13/10/2010



Tertiary lymphoid organs in renal allografts can be associated with donor-specific tolerance rather...

Documents

Transcript of Tertiary lymphoid organs in renal allografts can be associated with donor-specific tolerance rather...