Histopathology 2005, 46, 515521. DOI: 10.1111/j.1365-2559.2005.2129.

Characterization of macrophage subpopulations in colon

cancer using tissue microarrays
D Sickert, D E Aust,1 S Langer,1 I Haupt,1 G B Baretton1 & P Dieter
Institute of Physiological Chemistry and 1Institute of Pathology, Medical Faculty Carl Gustav Carus, University of
Technology Dresden, Dresden, Germany
Date of submission 5 March 2004
Accepted for publication 7 June 2004

Sickert D, Aust D E, Langer S, Haupt I, Baretton G B & Dieter P

(2005) Histopathology 46, 515521

Characterization of macrophage subpopulations in colon cancer using tissue microarrays

Aims: To determine the pattern of macrophage infiltration in colon cancers and its correlation with
clinicopathological characteristics.
Methods and results: Colon cancers from 100 patients
were arrayed into a tissue microarray (TMA). Four
cores per tumour were taken: three from the invasion
front (IF) and one from the tumour surface (TS).
Macrophages were quantified by immunohistochemistry with antibodies to the PG-M1, KP-1, MRP8, MRP14
and MRP8 14 antigens. The number of macrophages
was significantly higher in the TS cores than in the IF
cores and both tumour sites showed a higher number

of macrophages than the normal mucosa. The number

of macrophages decreased in higher stage tumours.
The different tumour-associated macrophage (TAM)
subpopulations were positively correlated with each
other.
Conclusions: The increased number of macrophages in
cancers compared with normal colon mucosa indicates
that macrophages are attracted to the tumour site.
However, decreasing macrophages in higher stage
colon cancers suggest that this attraction decreases
with tumour progression.

Keywords: colon cancer, KP-1, MRP14, MRP8, MRP8 14, PG-M1, tissue microarray, tumour-associated
macrophages
Abbreviations: IF, invasion front; TAM, tumour-associated macrophage; TMA, tissue microarray; TNF, tumour
necrosis factor; TS, tumour surface; VEGF, vascular endothelial growth factor

Introduction
There is considerable controversy as to whether
tumour-associated macrophage (TAM) subpopulations
promote or inhibit tumour progression.1 It is well
known that macrophages orchestrate immune responses through the secretion of cytokines, chemokines
and proteolytic enzymes.2 A model for the functional
heterogeneity observed in macrophage populations at
inflammatory sites in vivo suggests that cytokine
production at the site of inflammation might be due
largely to newly recruited monocytes, while phagocytosis and tissue remodelling are the work of latestage macrophages.3 Moreover, macrophages can
D.S. and D.E.A. contributed equally to this work.
Address for correspondence: Dr Daniela E Aust, Institute of
Pathology, TU Dresden, Fetscherstr. 74, D-01307 Dresden, Germany.
e-mail: daniela.aust@pathologie.med.tu-dresden.de

2005 Blackwell Publishing Limited.

present processed foreign antigens to primed T lymphocytes, allowing the enhancement or inhibition of a
specific immune response.4
Although previous studies have shown that the
above-mentioned functions of macrophages probably
play a role in anti-tumour response in colon cancer,57
there is no good evidence for tumour regression due to
TAMs. On the contrary, recent studies have pointed out
that tumours are able to escape the effects of cytotoxic
macrophages through the secretion of anti-inflammatory cytokines.8 Siegert et al.9 detected a significant
reduction of oxygen radical production in macrophages
after incubation with cell culture supernatants derived
from colon cancer cell lines.
Additionally, TAMs in colon cancer have angiogenic
potential due to TAM-derived cytokines such as tumour
necrosis factor (TNF)-a and vascular endothelial
growth factor (VEGF).1013 Analyses of colon cancer

516

D Sickert et al.

xenografts indicate that macrophages assist in tumour

invasion through the expression of lytic enzymes (e.g.
matrix metalloproteinase 2) digesting the extracellular
matrix.14 Furthermore, Niemi et al.15 hypothesize that
apolipoprotein E expression by macrophages surrounding the colon tumour area may modulate epithelial
integrity and thus contribute to tumour growth.
Previous findings of Hauptmann et al.16 suggest that
different macrophage phenotypes are associated with
different regions of colorectal carcinoma and different
effects on tumour cells.
Macrophages are a heterogeneous population of cells
derived from blood-borne monocytes that migrate into
tissues, where they undergo differentiation dependent
on the microenvironment.
We used a panel of five antibodies to determine the
infiltration of human colon carcinomas and normal
colon mucosa by monocytes macrophages. Pan
macrophage markers, PG-M1 and KP-1, characterize
TAMs, with PG-M1 showing more restricted reactivity
to cells of the monocyte macrophage lineage than
KP-1. Monocytes macrophages can also be detected
by differentiation-associated antigens belonging to the
Ca2+-binding S100 protein family, namely MRP8,
MRP14 and MRP8 14. MRP8- and MRP14-protein
expression is restricted to distinct stages of monocytic
maturation17 and appears very early at the site of
inflammation.18 Mature tissue macrophages do not
express these proteins. S100-calgranulins (MRP8,
MRP14) and calprotectin (MRP8 14) are abundantly
expressed in myeloid cells (monocytes macrophages,
neutrophils) and have been associated with various
inflammatory diseases19 as well as carcinomas.8,2022
These cells can be divided into an active inflammatory
type (MRP14+, MRP8 14+) expressing proinflammatory cytokines such as TNF-a as well as oxygen radicals
and a chronic inflammatory type (MRP8+).8
We designed a tissue microarray (TMA) comprising
100 well-characterized colon cancers. Previous studies
have demonstrated that TAMs provide insight into
molecular mechanisms important in carcinogenesis.23,24 There was scepticism that, due to tumour
heterogeneity, tissue cores used for the microarray
might not be representative of the biological properties
of the entire tumour. Therefore, several groups have
compared 0.6-mm cores from tumours with the corresponding conventional large sections. These analyses
have demonstrated excellent agreement (84100%)
between conventional tissue sections and single cores.
Such results indicate that tumour heterogeneity does not
influence the predictive power of the TMA results.23,2527
To date, no data exist on the use of TMAs for the analysis
of infiltrating immune cells in cancer.

The aim of our study was to determine the morphological pattern of macrophage infiltration in colon
cancers and to correlate it with clinicopathological
characteristics.

Materials and methods

clinical materials
One hundred colon cancers were selected from the
pathology archives, reviewed, and two representative
paraffin blocks chosen for the TMA. On each block an
area from the tumour surface (TS) and one from the
invasion front (IF) were marked. In addition, normal
colonic mucosa was marked on resection margins
(n 39). All the cancers were staged according the
International Union Against Cancer (UICC) and graded
according to World Health Organization criteria.28,29
The clinicopathological characteristics associated with
these samples are listed in Table 1.
tissue microarrays
TMAs were constructed by randomly selecting one core
(diameter 0.6 mm, length
0.7 mm) from the TS and
three cores from the IF region. Tissue cylinders were
punched out of two different donor blocks and placed
into a recipient paraffin block with defined array
coordinates (1.0 mm distance between the cores) using
Table 1. Clinical characteristics
Male

n 59

Age*

68 12 years

UICC stage
I

n 15

II

n 38

III

n 39

IV

n8

Grade
1

n1

n 44

23

n 23

n 32

Mucinous cancers

n 20

*Mean age in years SD.

UICC, International Union Against Cancer.

2005 Blackwell Publishing Ltd, Histopathology, 46, 515521.

Macrophages in colon cancer

517

Figure 1. Photomicrographs of five successive

invasion front cores of colon cancer immunohistochemically stained for: a, 27E10
(MRP8 14); b, KP-1; c, MRP14; d, MRP8;
e, PG-M1.

a tissue microarrayer (Beecher Instruments, Silver

Spring, MD, USA).
immunohistochemistry
For immunohistochemistry, 4-lm sections of the
microarray block were cut and transferred to glass
slides using a paraffin-sectioning aid system (Instrumedics Inc., Hackensack, NJ, USA). After deparaffinization
and hydration, the slides were treated with 1% H2O2
for 15 min at room temperature to abolish endogenous
peroxidase activity. Standard indirect immunoperoxidase procedures were used for immunohistochemistry
(ABC KIT-Elite; Vector Laboratories, Burlingame, CA,
USA). A panel of five primary monoclonal antibodies
for macrophage subtypes, i.e. PG-M1 (CD68), KP-1
(CD68) (Dako Corp., Carpinteria, CA, USA) and MRP8,
MRP14 and MRP8 14 (27E10) (Dianova, Hamburg,
Germany) were applied overnight at 4C. Antigen
retrieval was carried out for PG-M1 and MRP14
(microwave pretreatment, 15 min, 600 W, dilution
1 : 100 and 1 : 400) and MRP8 14 [pronase pretreatment, 15 min, 5% (v v) in TBSbuffer pH 7,2,
37C, dilution 1 : 30). KP-1 and MRP8 antibodies were
used without antigen retrieval (dilution 1 : 400 and
1 : 1200). The reaction was visualized with diam
2005 Blackwell Publishing Ltd, Histopathology, 46, 515521.

inobenzidine, and slides were counterstained with

haematoxylin (Figure 1). The primary antibody was
omitted for the negative control. Internal positive
controls (spleen, granulation tissue, tonsil) were included in every TMA block.
The percentage of tumour cells in each tissue core
was recorded semiquantitatively in five categories
(0, no tumour cells; a, > 025%; b, > 2550%;
c, > 5075%; and d, > 75100% tumour cells per
core). Stained macrophages and neutrophils were
scored quantitatively (details are given in Table 2).
Lymphocytes were recorded semiquantitatively in four
categories: no lymphocytes, small number of lymphocytes, moderate number of lymphocytes and large
number of lymphocytes.
statistics
The score results for tumour cells and different macrophage subgroups were converted into interval means
(Table 2) using Microsoft Excel 2000 (9.0). All subsequent statistical anayses were done with these values.
Statistical significance of differences was determined by
paired t-test, t-test for unpaired samples or U-test. A
P-value of 0.05 was considered significant. Statistical analysis was performed using the Spearman rank

D Sickert et al.

13

13

420

12

415

9,5

2140

30,5

1630

23

4160

50,5

3145

38

61100

80,5

4660

53

101140

120,5

6199

80

correlation for comparison between pairs of several cell

groups. All statistical tests were done with SPSS 11.0.1
(LEAD Technologies, Haddonfield, NJ, USA).

30
20
10
0

70
60
IF

50

KP-1
PG-M1
MRP8
MRP14
MRP8/14

40
30
20
10
0

Results

a
b
c
Percentage of tumour cells

Figure 2. Comparison of number of macrophages between cores with

different tumour cell content. TS, Tumour surface; IF, invasion front.

MRP8 14, P 0.0001). However (except for KP-1+

macrophages) monocyte macrophage densities were
higher in TS than in IF cores: PG-M1+ (P 0.005),
MRP8+ (P 0.031), MRP14+ (P 0.091) and
MRP8 14+ (P 0.135) TAMs (Figure 3). All macrophage subpopulations investigated in our study were
statistically significantly correlated with each other.

60
Numbher of macrophages

Although there was a significant variation of tumour

cell content within the cores, the number of macrophages was independent of the percentage of tumour
cells within the cores (Figure 2). In TS cores macrophages increased slightly with increasing tumour cell
content until the tumour cell content reached approximately 50%; in cores with a higher tumour cell
content, the number of macrophages decreased
(KP-1, PG-M1, MRP8, not significant; PG-M1 (c versus
d, P 0.013). No significant correlation between
number of macrophages and tumour cell content was
found in IF cores with different tumour cell contents
with the exception of MRP8 14 (b versus c, P
0.001). Significant differences between the three
IF cores were found in the number of infiltrating
macrophages, ranging from 19% (PG-M1) to 52%
(MRP14) deviation in the number of macrophages
from core to core, indicating that there is significant
heterogeneity in immune cell infiltration. However,
within one case the deviation between the IF cores
remained constant.
Comparison between tumour tissues and normal
mucosa revealed a significantly higher density of all
macrophage subsets in carcinoma tissue than in
normal mucosa. The number of macrophages in TS
and IF cores from the same patient correlated positively with each other (KP-1, P 0.0001; PG-M1,
P 0.006; MRP8, P 0.001; MRP14, P 0.001;

a
b
c
Percentage of tumour cells

PG-M1

50
40
KP-1

30
20

MRP8

MRP14

MRP8/14

0.135
0.001

TS
KP-1
PG-M1
MRP8
MRP14
MRP8/14

40

0.091
0.000

50

0.031
0.001

Positive
IntervalPositive
IntervalScore macrophages means Score macrophages means

60

10
0
*

0.005
0.000

MRP8, MRP14, 27E10

0.601
0.003

PG-M1, KP-1

70
Number of macrophages

Table 2. Score indices of interval-means of positively stained

macrophages

Number of macrophages

518

Figure 3. Comparison between tumour surface (TS), invasion front

(IF) and normal mucosa (N) with regard to macrophage infiltration.
*Significances of mean value comparison. , TS; hatched, IF; h, N.

2005 Blackwell Publishing Ltd, Histopathology, 46, 515521.

Discussion
In this paper, we describe the morphological patterns of
macrophage infiltration in colon cancer using TMAs.
TMAs have been used successfully for immunohistochemical studies focused on carcinogenesis,23,24,30

2005 Blackwell Publishing Ltd, Histopathology, 46, 515521.

Number of macrophages

60

TS

50

1 13 11 8 13 11
2 37 34 28 37 35
3 38 37 28 38 35
4 8 8 6 7 8

40
30

1
2
3
4

20
10
0
PGM-1

MRP-8

MRP-14 MRP-8/14
UICC
KP-1
PG-M1
MRP8
MRP14
MRP8/14

KP-1
60 IF
Number of macrophages

We found a higher number of lymphocytes within

normal mucosa than in TS (P 0.0001) and IF cores
(P 0.0001). The number of lymphocytes in TS and
IF cores correlated positively with each other. The
number of lymphocytes decreased with increased
tumour cell percentage (not significant).
Positive correlations were demonstrated between
the number of lymphocytes and the number of
macrophage subsets: KP-1 (TS, P < 0.05), PG-M1
(TS, P < 0.01, IF, P < 0.01), MRP8 (TS, P < 0.01),
MRP14 (TS, P < 0.01, IF, P < 0.01) and MRP8 14
(trend to significance).
The densities of KP-1+ and PG-M1+ macrophages in
TS and IF cores correlated positively with advanced
tumour stage (not significant). Contrary to this result,
MRP8+, MRP14+ and MRP8 14+ monocytes macrophages correlated inversely with advanced T-stage
(in IF, MRP14 P < 0.05, MRP8 14 P < 0.01)
The comparison between the different UICC stages
revealed that KP-1+ and PG-M1+ macrophages
increased with larger tumour size, but decreased in
metastatic tumours. For KP1+ TAMs in TS cores, there
was a significant increase in UICC stage 2 tumours
compared with UICC stage 1 tumours (P 0.043), and
a significant decrease in UICC stage 4 tumours when
compared with UICC stage 3 tumours (P 0.013). In
IF cores, the number of KP1+ macrophages decreased
significantly in stage 4 tumours when compared with
UICC stage 3 tumours (P 0.015). In contrast, the
MRP8-, MRP14- and MRP8 14-expressing monocytes macrophages did not show any differences
between different tumour UICC stages (Figure 4).
Although there was no correlation between the
number of infiltrating macrophages and tumour grade,
the comparison between mucinous and non-mucinous
cancers showed a higher infiltration for all subsets of
TAMs in non-mucinous adenocarcinomas with the
exception of KP-1. However, this difference between
both cancer types reached statistical significance only
for MRP14+ TAMs (P 0.004). In addition, nonmucinous tumours showed a higher infiltration with
lymphocytes than mucinous cancers (TS, P 0.038;
IF, P 0.001).
The density of all macrophage subtypes was a little
higher within tumours without necrosis than in
tumours showing necrosis.

519

UICC
KP-1
PG-M1
MRP8
MRP14
MRP8/14

Macrophages in colon cancer

50

1 15 15 14 15 14
2 38 38 34 38 38
3 39 39 37 39 37
4 8 8 8 8 8

40
30

1
2
3
4

20
10
0
KP-1

PGM-1

MRP-8

MRP-14 MRP-8/14

Figure 4. Relationship of tumour-associated macrophage

subgroups with International Union Against Cancer (UICC) stage.
The inserted table gives the number of cases per UICC stage.
TS, Tumour surface; IF, invasion front.

showing that the tissue cores are highly representative

of the whole tumour.23,2527 These studies, however,
were not dependent on the tumour cell content of the
individual cores. When starting this study, we did not
know whether the macrophage infiltration in tumours
would be dependent on the tumour cell content of the
cores, thus limiting the representativity of the TMA.
However, our findings have shown that the infiltration
with macrophages is independent of tumour cell
percentages within the individual cores. Since macrophage infiltration is known to be heterogeneous within
an individual tumour, multiple cores were taken from
each tumour to allow for the expected heterogeneity.
By arraying several cores per case and using a large
number of cases, we ensured that our TMA results are
representative of the tumours studied.
We found increased numbers of the different TAM
subgroups in tumour tissue compared with normal
colonic mucosa, indicating that TAMs are attracted to
the tumour site. These results may be explained by the
finding of Suzuki et al.,31 who reported a positive
correlation between macrophage numbers along the

520

D Sickert et al.

invasion front of colon cancers and the expression of

cell adhesion molecules on tumour microvessels.
Hemmerlein et al.32 found similar effects in advanced
stage renal cancer.
In addition to the expression of cell adhesion
molecules in endothelial cells of tumour microvessels,
MRP8+ and MRP14+ monocytes macrophages play a
regulatory role in the transendothelial migration of
human leucocytes and a modulatory role in inflammatory responses.33 In our study, the number of
MRP8+, MRP14+ and MRP8 14+ macrophages was
closely correlated with the number of CD68+ macrophages. These data may be explained by MRP8+,
MRP14+ and MRP8 14+ monocytes macrophages
migrating continuously into the tumour area and
maturing there due to tumour-induced signals.
MRP8+, MRP14+ and MRP8 14+ monocytes macrophages have been shown to inhibit tumour cell
proliferation22 due to the release of cytotoxic agents
such as oxygen radicals.34,35 A previous study by
Siegert et al.9 showed a significant reduction in oxygen
radical production in macrophages after incubation
with cell culture supernatants derived from colon
cancer cell lines. This is in line with our findings
that the number of monocytes macrophages with the
acute and chronic inflammatory phenotype (MRP8+,
MRP14+, MRP8 14+ macrophages) decreased with
increasing tumour size, while the overall number of
CD68+ macrophages increased with increasing
tumour size. These tumour-primed macrophages
may have failed to mature normally into tissue
macrophages. This may result in altered functions,
e.g. decreased production of reactive oxygen species
and tumour cytotoxicity.36,37
We have observed that lymphocyte density was
much higher within normal mucosa than in tumour
tissue. With increased tumour cell percentage and
within mucinous tumours the density of lymphocytes
decreased. Saio et al.38 showed that highly proapoptotic TAMs induce lymphocyte apoptosis in murine
tumours. Additionally, MRP8 14 protein has cytotoxic
activity against lymphocytes and induces lymphocyte
apoptosis.39 These functional data are corroborated by
our morphological finding of decreasing lymphocyte
infiltration in combination with increasing monocyte macrophage densities.
In conclusion, this is the first study showing
that cytotoxic macrophage subpopulations (MRP8+,
MRP14+, MRP8 14+ macrophages) and lymphocytes
decrease with increasing tumour size, suggesting a
decreasing anti-tumour response in higher stage colon
cancers, which is in line with the results of previously
published functional studies. Further morphological

and functional studies are needed to elucidate the

complex relation between immune cells and tumours.

References
1. Konur A, Kreutz M, Knuchel R, Krause SW, Andreesen R. Threedimensional co-culture of human monocytes and macrophages
with tumor cells: analysis of macrophage differentiation and
activation. Int. J. Cancer 1996; 66; 645652.
2. Schweyer S, Hemmerlein B, Radzun HJ, Fayyazi A. Continuous
recruitment, co-expression of tumour necrosis factor-alpha and
matrix metalloproteinases, and apoptosis of macrophages in gout
tophi. Virchows Arch. 2000; 437; 534539.
3. Riches DW. Signalling heterogeneity as a contributing factor in
macrophage functional diversity. Semin. Cell Biol. 1995; 6; 377
384.
4. Baxevanis CN, Voutsas IF, Tsitsilonis OE, Gritzapis AD, Sotiriadou
R, Papamichail M. Tumor-specific CD4+ T lymphocytes from
cancer patients are required for optimal induction of cytotoxic T
cells against the autologous tumor. J. Immunol. 2000; 164;
39023912.
5. Ohtani H, Naito Y, Saito K, Nagura H. Expression of costimulatory molecules B7-1 and B7-2 by macrophages along invasive
margin of colon cancer: a possible antitumor immunity? Lab.
Invest. 1997; 77; 231241.
6. Nakayama Y, Nagashima N, Minagawa N et al. Relationships
between tumor-associated macrophages and clinicopathological
factors in patients with colorectal cancer. Anticancer Res. 2002;
22; 42914296.
7. Herbeuval JP, Lambert C, Sabido O et al. Macrophages from
cancer patients: analysis of TRAIL, TRAIL receptors, and colon
tumor cell apoptosis. J. Natl Cancer Inst. 2003; 95; 611621.
8. Hemmerlein B, Markus A, Wehner M, Kugler A, Zschunke F,
Radzum HJ. Expression of acute and late-stage inflammatory
antigens, c-fms, CSF-1, and human monocytic serine esterase 1,
in tumor-associated macrophages of renal cell carcinomas.
Cancer Immunol. Immunother. 2000; 49; 485492.
9. Siegert A, Denkert C, Leclere A, Hauptmann S. Suppression of the
reactive oxygen intermediate production of human macrophages
by colorectal adenocarcinoma cell lines. Immunology 1999; 98;
551556.
10. Zhu GH, Lenzi M, Schwartz EL. The Sp1 transcription factor
contributes to the tumor necrosis factor-induced expression of
the angiogenic factor thymidine phosphorylase in human colon
carcinoma cells. Oncogene 2002; 21; 84778485.
11. Etoh T, Shibuta K, Barnard GF, Kitano S, Mori M. Angiogenin
expression in human colorectal cancer: the role of focal
macrophage infiltration. Clin. Cancer Res. 2000; 6; 35453551.
12. Barbera-Guillem E, Nyhus JK, Wolford CC, Friece CR, Sampsel
JW. Vascular endothelial growth factor secretion by tumorinfiltrating macrophages essentially supports tumor angiogenesis, and IgG immune complexes potentiate the process. Cancer
Res. 2002; 62; 70427049.
13. Bamba H, Ota S, Kato A, Kawamoto C, Matsuzaki F. Effect of
prostaglandin E1 on vascular endothelial growth factor production by human macrophages and colon cancer cells. J. Exp. Clin.
Cancer Res. 2000; 19; 219223.
14. Aharinejad S, Abraham D, Paulus P et al. Colony-stimulating
factor-1 antisense treatment suppresses growth of human tumor
xenografts in mice. Cancer Res. 2002; 62; 53175324.
15. Niemi M, Hakkinen T, Karttunen TJ et al. Apolipoprotein E and
colon cancer. Expression in normal and malignant human

2005 Blackwell Publishing Ltd, Histopathology, 46, 515521.

Macrophages in colon cancer

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.
26.

27.

intestine and effect on cultured human colonic adenocarcinoma

cells. Eur. J. Intern. Med. 2002; 13; 3743.
Hauptmann S, Zwadlo-Klarwasser G, Hartung P, Klosterhalfen B,
Kirkpatrick CJ, Mittermayer C. Association of different macrophage phenotypes with infiltrating and non-infiltrating areas of
tumorhost interface in colorectal carcinoma. Pathol. Res. Pract.
1994; 190; 159167.
Roth J, Goebeler M, van den Bos C, Sorg C. Expression of calciumbinding proteins MRP8 and MRP14 is associated with distinct
monocytic differentiation pathways in HL-60 cells. Biochem.
Biophys. Res. Commun. 1993; 191; 565570.
Kerkhoff C, Klempt M, Sorg C. Novel insights into structure and
function of MRP8 (S100A8) and MRP14 (S100A9). Biochim.
Biophys. Acta 1998; 1448; 200211.
Nacken W, Roth J, Sorg C, Kerkhoff C. S100A9 S100A8:
Myeloid representatives of the S100 protein family as prominent
players in innate immunity. Microsc. Res. Tech. 2003; 60; 569
580.
Endress H, Freudenberg N, Fitzke E, Grahmann PR, Hasse J,
Dieter P. Infiltration of lung carcinomas with macrophages
of the 27E10-positive phenotype. Lung Cancer 1997; 18; 3546.
Komatsu K, Kobune-Fujiwara Y, Andoh A et al. Increased
expression of S100A6 at the invading fronts of the primary
lesion and liver metastasis in patients with colorectal adenocarcinoma. Br. J. Cancer 2000; 83; 769774.
Hauptmann S, Zwadlo-Klarwasser G, Jansen M, Klosterhalfen B,
Kirkpatrick CJ. Macrophages and multicellular tumor spheroids
in co-culture: a three-dimensional model to study tumorhost
interactions. Evidence for macrophage-mediated tumor cell proliferation and migration. Am. J. Pathol. 1993; 143; 14061415.
Kononen J, Bubendorf L, Kallioniemi A et al. Tissue microarrays
for high-throughput molecular profiling of tumor specimens.
Nat. Med. 1998; 4; 844847.
Schraml P, Kononen J, Bubendorf L et al. Tissue microarrays for
gene amplification surveys in many different tumor types.
Clin. Cancer Res. 1999; 5; 19661975.
Bubendorf L. High-throughput microarray technologies: from
genomics to clinics. Eur. Urol. 2001; 40; 231238.
Torhorst J, Bucher C, Kononen J et al. Tissue microarrays for
rapid linking of molecular changes to clinical endpoints. Am. J.
Pathol. 2001; 159; 22492256.
Rimm DL, Camp RL, Charette LA, Costa J, Olsen DA, Reiss M.
Tissue microarray: a new technology for amplification of tissue
resources. Cancer J. 2001; 7; 2431.

2005 Blackwell Publishing Ltd, Histopathology, 46, 515521.

521

28. Wittekind C, Meyer HJ, Bootz F. TNM Klassifikation maligner

Tumoren. Berlin: Springer Verlag, 2002.
29. Hamilton SR, Aaltonen LA. Tumours of the digestive system. Lyon:
IARC Press, 2000.
30. Tzankov A, Zimpfer A, Pehrs AC et al. Expression of B-cell
markers in classical Hodgkin lymphoma: a tissue microarray
analysis of 330 cases. Mod. Pathol. 2003; 16; 11411147.
31. Suzuki Y, Ohtani H, Mizoi T et al. Cell adhesion molecule
expression by vascular endothelial cells as an immune inflammatory reaction in human colon carcinoma. Jpn J. Cancer Res.
1995; 86; 585593.
32. Hemmerlein B, Scherbening J, Kugler A, Radzun HJ. Expression
of VCAM-1, ICAM-1, E- and P-selectin and tumour-associated
macrophages in renal cell carcinoma. Histopathology 2000; 37;
7883.
33. Kerkhoff C, Eue I, Sorg C. The regulatory role of MRP8 (S100A8)
and MRP14 (S100A9) in the transendothelial migration of
human leukocytes. Pathobiology 1999; 67; 230232.
34. Mahnke K, Bhardwaj R, Sorg C. Heterodimers of the calciumbinding proteins MRP8 and MRP14 are expressed on the surface
of human monocytes upon adherence to fibronectin and collagen. Relation to TNF-alpha, IL-6, and superoxide production.
J. Leukoc. Biol. 1995; 57; 6371.
35. Bhardwaj RS, Zotz C, Zwadlo-Klarwasser G et al. The calciumbinding proteins MRP8 and MRP14 form a membrane-associated
heterodimer in a subset of monocytes macrophages present in
acute but absent in chronic inflammatory lesions. Eur. J.
Immunol. 1992; 22; 18911897.
36. Siziopikou KP, Harris JE, Casey L, Nawas Y, Braun DP. Impaired
tumoricidal function of alveolar macrophages from patients
with non-small cell lung cancer. Cancer 1991; 68; 10351044.
37. Alleva DG, Burger CJ, Elgert KD. Tumor growth increases
Ia-macrophage synthesis of tumor necrosis factor-alpha and
prostaglandin E2: changes in macrophage suppressor activity.
J. Leukoc. Biol. 1993; 53; 550558.
38. Saio M, Radoja S, Marino M, Frey AB. Tumor-infiltrating
macrophages induce apoptosis in activated CD8(+) T cells by a
mechanism requiring cell contact and mediated by both the cellassociated form of TNF and nitric oxide. J. Immunol. 2001; 167;
55835593.
39. Kerkhoff C, Hofmann HA, Vormoor J et al. Binding of two nuclear
complexes to a novel regulatory element within the human
S100A9 promoter drives the S100A9 gene expression. J. Biol.
Chem. 2002; 277; 4187941887.