CRP Present within Colorectal Cancer Cells is Taken Up
from the Circulation and not Produced by the Tumour; Evidence of a Complex
Relationship Between Host and Tumour
Arfon G M T Powell,
Scott T C Shepherd, Clare Orange, John H Anderson, Ruth F McKee, DC McMillan,
Paul G Horgan, Joanne Edwards
Arfon G M T Powell, Scott T C Shepherd, Joanne Edwards, Unit of
Experimental Therapeutics, Institute of cancer science, McGregor Building,
Western Infirmary. University of Glasgow, Glasgow G11 6NR, the United Kingdom
Arfon G M T Powell, Scott T C Shepherd, John H Anderson, Ruth F McKee, DC McMillan, Paul G Horgan, Academic Section of Surgery, University of Glasgow,
Royal Infirmary, Glasgow G31 2ER, the United Kingdom
Clare Orange, Department of
Pathology, Institute of cancer science, McGregor Building, Western Infirmary. University of
Glasgow, Glasgow G11 6NR, the United Kingdom
Correspondence to: Arfon G M T
Powell, Unit of Experimental
Therapeutics, Institute of cancer science, College of Medical, Veterinary and
Life of Sciences- University of Glasgow, 2nd floor McGregor Building, Western
Infirmary Glasgow, Glasgow G11 6NR, the United Kingdom.
Email: powella16@cardiff.ac.uk
Telephone: +0141-330-7244 Fax:+ 0141-211-2248
Received: April 26,
2014
Revised: May 27, 2014
Accepted: May 31, 2014
Published online: July 18, 2014
ABSTRACT
AIM: Preoperative raised serum CRP (sCRP)
is now widely accepted as being an independent predictor of poor survival.
Although CRP is known to made by the liver, whether the tumour is a site of CRP
production remains unknown. The aim if this study was to determine, in
colorectal cancer, if the tumour or liver is responsible for CRP production and
what is the relationship between tumoural CRP and survival.
MATERIALS AND
METHODS: To establish if
the tumour produces CRP, reverse transcriptase PCR for CRP mRNA expression was
performed on 30 malignant colorectal specimens, 19 non-malignant colorectal
specimens from the resection margin and 30 healthy liver specimens from
patients with no radiological evidence of liver metastasis. Tumour protein
expression of CRP was assessed in a cohort of 147 parraffin embedded colorectal
tumours.
RESULTS: This study demonstrates that
colorectal tumours are not a site of CRP production. Tumours do however,
express CRP and high cytoplasmic CRP was associated with necrosis (P=0.028)
and high nuclear CRP was associated with increasing T-stage (P=0.022).
Tumoural CRP was not associated with disease free or cancer-specific survival.
CONCLUSION: This study suggests that despite
colorectal tumours expressing CRP, there is no evidence of CRP production at
this site. Understanding the mechanism whereby CRP enters the cancer cells may
offer a novel therapeutic target for patients presenting with a systemic
inflammatory response.
© 2014 ACT. All
rights reserved.
Key words: Colorectal cancer; Recurrence; Systemic inflammation; Survival; CRP
Powell AGMT,
Shepherd STC, Orange C, Anderson JH, McKee RF, McMillan DC, Horgan
PG, Edwards J. CRP Present within Colorectal Cancer Cells is Taken Up from the
Circulation and not Produced by the Tumour; Evidence of a Complex Relationship Between Host and Tumour. Journal of Tumor 2014; 2(7):
187-192 Available from: URL: http://www.ghrnet.org/index.php/JT/article/view/777
INTRODUCTION
Colorectal cancer is the second most common cause of cancer death in
Western Europe and North America. Each year in the UK, there are approximately
35,000 new cases and 16,000 deaths attributable to this disease[1].
Surgical resection remains the primary treatment option with adjuvant
chemotherapy given to those with regional lymph node metastasis. Unfortunately
despite improvements in surgical techniques and the introduction of adjuvant
chemotherapy, there are still a significant proportion of patients who develop
recurrence and die of their disease[2].
In non-cancer
states, serum CRP becomes elevated in response to a complex interaction between
hormones, inflammatory cytokines and glucocorticoids. In colorectal cancer, it
remains unknown if circulating CRP is produced by the tumour, liver or both[3].
Possible mechanisms for serum CRP elevation include tumour growth and specific
immune responses to tumour antigens[4-8]. Cancer cells can increase
the production of pro-inflammatory cytokines, which in turn stimulate the
production of CRP in the liver[9-10]. CRP production was initially
thought to be restricted to the liver; however, recent studies have
demonstrated expression of the acute phase protein in numerous extrahepatic
tissues including the healthy kidney[11-12]. CRP expression has also
been observed in malignant tissue including; squamous cell carcinoma of the
oesophagus, renal cell carcinoma and bone marrow biopsies from myeloma
patients, however, its expression in colorectal cancer remains unknown[13-15].
The role of CRP within the tumour cell remains unclear however, Yang and
colleagues observed, in myeloma cells, that CRP activates numerous signaling
pathways including the PI3K/Akt and Ras/Erk pathways through membrane bound Fc¦Ã
receptors. This resulted in increased cancer cell numbers via inhibition of
apoptosis and increased proliferation[15]. However, the role of
cellular CRP in predicting survival and its relationship with
clinicopathological factors remain unknown.
Therefore, the
aims of the present study were to establish if colorectal tumour was a site of
CRP production and what is the relationship between tumoural CRP,
clinicopathological factors and survival in patients undergoing potentially
curative resection for colorectal cancer.
METHODS
Patient
characteristics
Fresh frozen
tissue cohort: The cohort consisted of 30 malignant samples taken from patients with
histologically proven colorectal cancer and 19 non-malignant colorectal tissue
samples taken from the specimens¡¯ resection margin. Thirty healthy liver
samples were taken from the above patients who had no histological evidence of
metastatic disease at presentation. All patients were diagnosed with primary
operable colorectal adenocarcinoma between January 1991 and August 1996 in a
single surgical unit at Glasgow Royal Infirmary and underwent potentially
curative resection following CT staging.
Immunohistochemistry cohort: Patients with
histologically proven colorectal cancer who, on the basis of pre-operative
staging CT scans and findings at laparotomy, were considered to have undergone
potentially curative resection for colorectal cancer (Stages I¨CIII) between
1997 and 2005 in a single surgical unit were included in the study. This cohort
was identified from a prospectively maintained database and included elective
and emergency patients. Patients with conditions known to acutely or
chronically evoke an acute or chronic systemic inflammatory response were
excluded. These included those patients with (1) pre-operative
chemo-radiotherapy; (2) clinical evidence of infection or (3) active chronic
inflammatory disease such as active rheumatoid arthritis. Patients who died
within 30 days of surgery were excluded from the survival analysis. The tumours
were staged according to conventional TNM classification and additional
pathological characteristics; (4) the presence of vascular invasion, peritoneal
involvement (PtI), margin involvement (MI) and tumour perforation (TP) were
recorded from pathology reports issued at the time of resection. Routine
laboratory measurements of haemoglobin (Hb), white-cell counts (WBC),
neutrophils (Neut), lymphocytes (lymph), platelet counts (Plt), albumin (Alb)
and C-reactive protein (sCRP) concentrations prior to surgery were recorded.
Lymph node ratio (LNR), neutrophil lymphocyte ratio (NLR), platelet lymphocyte
ratio (PLR), Klintrup score, Peterson Index and Ki67 were already available and
described previously[16-17].
Adjuvant
therapy was given to patients at the discretion of the clinical oncologist or
after multidisciplinary team assessment. Patient¡¯s clinical and pathological
information was available to the oncologist making these decisions and the
treatment offered was based on treatment guidelines for colorectal cancer at
that time. This study was approved by the Research Ethics Committee, Glasgow Royal
Infirmary, Glasgow.
RESEARCH METHODS
Quantitative
Reverse Transcriptase Polymerase Chain Reaction
Tissue
processing: After resection of the primary tumour representative parts of
malignant colorectal tissue were identified by a pathologist, Snap frozen and
stored in liquid nitrogen.
RNA isolation: 50-75 mg of
tissue was homogenised in a mortar and pestle in liquid nitrogen. 1 mL of
TRIZOL® (Invitrogen, UK) was used to extract total RNA according to
manufacturer¡¯s protocol. RNA quantity and quality was assessed by UV
spectrometry (GeneQuant machine) and by examining rRNA bands after agarose gel
electrophoresis. Only samples, that expressed both 18S band and a stronger 28S
band, were included in this study.
cDNA synthesis: To guarantee no
other DNA was present, samples were incubated with RNAse-free DNAse treatment
and removal reagent kit (Applera, UK) and incubated for 30 minutes at 37¡æ. To ensure the same amount of cDNA being used for quantification of
mRNA, a starting concentration of 1,000 ng of tRNA was used for each sample.
Random hexamer primers (50 ng) were used for First Strand cDNA Synthesis using
SuperScript TM II RT (Invitrogen, UK) according to manufacturer¡¯s instructions.
Before using cDNA for PCR amplification, 2 units of RNase H (Applera, UK) were
added to samples and incubated for 20 minutes at 37¡æ. Quality of cDNA
was assessed by using a PCR control run with human ß-actin. Product bands were
assessed by examination of agarose gel electrophoresis. Only samples, which
showed equal product bands at 330 bp, were included in this study.
Quantification of CRP gene
expression: Real-time quantative PCR
was performed in an ABI Prism 7,900 Sequence Detection System (Applied
Biosystems, UK) using CRP TaqMan® Gene Expression Assay (HS
HS00265044-M1, Applied Biosystems, UK). For the TaqMan® Gene
Expression Assays the manufacturer¡¯s protocol with recommended 40 rounds of
amplification was used. Thermal cycler condition were 50¡æ for 2 min, 95¡æ for 10 min followed by 40¡Á95¡æ for 15 sec and 60¡æ for 1 min. Product melting curve
analysis and gel electrophoresis experiments were used to ensure that only one
product of the expected size was amplified.
On each PCR
plate negative controls were included. Quantitative values were obtained from
the threshold cycle (Ct value) at which the increase TaqMan® probe
fluorescent signal associated with an exponential increase of each individual
PCR product reaching a fixed threshold value. Each individual primer had a
fixed threshold Ct value. These fixed threshold values were used for every cDNA
sample.
To enable
comparison of different mRNA expression levels, their relation to the average
expression level of the housekeeping gene HPRT (hypoxanthine-guanine
phosphoribosyltransferase, Hs 4310890E, Applied Biosystems, UK) were evaluated.
Data were analysed using the Sequence Detection Software, which calculates the
threshold cycle (Ct) value. The expression of the target assay was normalised
by subtracting the Ct value of the house keeping gene from the Ct value of the
relevant target assay. The fold increase, relative to the control, was obtained
by using the formula 2-¦¤Ct, and then expressed as a percentage (¡Á100). All samples
were measured in triplicate.
Immunohistochemistry: Blocks from the
primary tumour were fixed in 10% buffered formalin and embedded in paraffin
wax. One representative block of tumour was selected for each patient. Sections
(4 um) were cut and mounted on slides coated with aminopropyltriethoxysilane.
Prior to use, the specificity of all antibodies were stringently test and have
been reported previously[18]. Sections were dewaxed and rehydrated
through a series of xylene and graded alcohols. Antigen retrieval was performed
by immersing slides in Tris/EDTA buffer (10 mM Trizma Base, 0.25 mM EDTA, pH
9.0) for 20 minutes at 96¡æ and then cooled for 20 minutes.
Endogenous peroxidise activity was blocked by incubation in 3% H2O2
for 5 minutes. After transfer to a humidified chamber, sections were blocked by
incubation with casein solution (Vector Laboratories, Peterborough, England) at
a 1:10 dilution with Tris-buffered saline (TBS) for 20 minutes and then primary
antibody overnight at 4¡æ. CRP (rabbit polyclonal, Abcam
plc. Cambridgeshire, UK) 1:100 Antibody antigen binding was detected using the
peroxidase-based Envision (Dako, Cambridgeshire, UK) technique using a
diaminobenzidine chromogenic substrate system. Sections were then counter
stained using Haematoxylin, dehydrated, and mounted in DPX. Appropriate
positive controls were included in each run. Negative controls were omission of
the primary antibody.
Scoring method: Tissue staining
intensity was scored blind by 2 independent observers using a weighted
histoscore method also known as the Hscore[19]. Histoscores were
calculated from the sum of (1¡Á% cells staining weakly positive) +
(2¡Á% cell staining
moderately positive) + (3¡Á% cells staining strongly positive)
with a maximum of 300. Histoscore analysis, therefore, provides a
semi-quantitative classification of staining intensity. The inter-class
correlation coefficient (ICCC) for each protein was calculated to confirm
consistency between observers and the mean of the two observers¡¯ scores were
used for analysis. The ICCC values were consistently above 0.8 for the weighted
histoscore (ICCC values ¡Ý0.7 were considered acceptable).
The observers were blinded to the clinical outcome of the patients.
Statistical
Analysis: Differences in protein or gene expression levels were
analysed using the Mann-Whitney U test or Kruskal-Wallis test when appropriate.
Where no standard thresholds existed, continuous variables of protein/gene
expression levels were categorised according to quartiles or dichotomised at
the median (high/low expression) as indicated in results tables.
Inter-relationships between the various clinical parameters were calculated
using Cochran-Mantel-Haenszel Chi square test., a P-value of <0.05
was considered to be significant.
Cancer specific survival rates were generated using the Kaplan-Meier
method. The log rank test was used to compare significant differences between
subgroups using univariate analysis. Multivariate stepwise Cox-regression
analysis was performed to identify factors that were independently associated
with disease specific death. A stepwise backward procedure was used to derive a
final model of the variables that had a significant independent relationship
with survival. To remove a variable from the model the corresponding P-value
had to be greater than 0.05.
RESULTS
CRP mRNA expression levels in
liver and colorectal tissue
Thirty colorectal cancer patients were studied, 23 patients with colon
cancer and 7 patients with rectal cancer. There was matched non-malignant
tissue available for analysis in 19 cases. Twenty two males and eight women,
the median age of patients was 74 years (IQR 61-79 years). Sixteen patients
were classified as node negative and 14 patients had node positive disease.
Expression of
CRP from the liver samples had a median value of 6600 (IQR 4185.25-23670.90).
mRNA was successfully extracted from all 30 tumour samples and 19 normal
mucosal tissue samples, however CRP expression was not detected in any of these
samples (Table 1).
Protein
expression and correlation with clinical parameters
Clinicopathological
Details: One hundred and forty seven patients who underwent elective surgery
with curative intent for colorectal cancer were studied. The majority were
under 75 years of age (61%), female (53%), were node negative (63%), with a low
Peterson Index (93%). 20% received post-operative adjuvant chemotherapy. Forty
two percent had right sided cancer, 30% had left sided cancer and 28% had
rectal cancer. Most patients had a CRP (63%) and albumin (71%) concentration in
the normal range. The minimum follow-up was 120 months; the mean follow-up of
the survivors was 153 months. During this period, 41 patients had a cancer
related death and 52 patients had a non-colorectal cancer death
Tumour CRP expression
levels: Each cellular location was independently assessed for CRP expression
levels in the tumour cell. Forty-seven percent of tumours exhibited nuclear
expression, 50% cytoplasmic, and 0% membrane (Figure 1). Tumours were
subdivided into those with high (upper tertile) or low (lower 2 tertiles)
expression. On univariate analysis, neither cytoplasmic nor nuclear CRP
expression was associated with disease free (P=0.241 and P=0.965
respectively) or cancer-specific survival (P=0.268 and P=0.703
respectively) (Table 2 and 3). Chi squared analysis demonstrated that
cytoplasmic CRP only correlated with tumour necrosis (P=0.028) and
nuclear CRP correlated with advanced T stage (P=0.022) (Table 4). Serum
CRP was not associated with tumoural CRP expression at either nuclear (P=0.328)
or cytoplasmic (P=0.653) localization.
DISCUSSION
This study demonstrates that despite expressing CRP, colorectal tumours
are not a site of CRP production and possibly take this up from the
circulation. Although CRP at different cellular locations was associated with
increasing T stage and tumour necrosis, there was no association with survival.
CRP expression
has been observed in malignant tissue before; however, its exact role remains
unclear. In myeloma cells, CRP appeared to activate numerous signaling pathways
resulting in increased proliferation and inhibition of apoptosis[15].
This is particularly interesting given our finding that nuclear CRP was
associated with increasing T stage. Furthermore, our findings support those of
Contu and colleagues who saw that CRP was associated with tumour progression
but not survival in rectal cancer patients[20]. It is possible that
tumoural CRP may drive colorectal cancer growth without affecting its potential
for metastasis.
This is the
first study to demonstrate that despite expressing CRP, the colorectal tumour
is not a site of production. This suggests that a complex interaction between
the host and the tumour exists with the tumour utilizing growth factors such as
cytokines from the circulation. It is possible therefore, that CRP within the
tumour acts as a growth factor and promoting malignant behavior. This is
supported by our observation that nuclear CRP was associated with higher T
stage. Grivennikov and colleagues observed that IL-6 via STAT3 had a profound
impact on increasing tumour cell proliferation and reducing apoptosis[21].
It is therefore plausible that, in a state of systemic inflammation, the tumour
may utilize the abundant proinflammatory cytokines present in the host
circulation to facilitate growth.
There is now a
growing body of evidence to suggest that activation of the systemic
inflammatory response is associated with poorer cancer-specific survival in
patients undergoing potentially curative resection for colorectal cancer[16-17].
Interestingly, this study supported by others demonstrates no association
between tumoural CRP and survival despite a relationship with cancer
progression[20]. The systemic inflammatory response represents a
complex biosystem of hormones, inflammatory cytokines and glucocorticoids.
Serum CRP which is closely associated with IL-6 may represent a surrogate
marker of systemic inflammatory activity whilst not, in itself representing an
effector protein and further delineation of the inflammatory response is
required[9].
In conclusion,
this study has demonstrated for the first time that despite colorectal tumours
expressing CRP, the cancer cells and its microenvironment are not responsible
for production. The results of this study suggest that a complex interaction
between the tumour and the host exists. Identifying the mechanism of this
interaction may offer a novel therapeutic target for high risk patients.
ACKNOWLEDGMENTS
The authors would like to thank Think Pink for their
financial support in purchasing the automated scoring machine (Slidepath) and
Glasgow Royal Infirmary Endowment Board for funding this project.Contributors:
JE conceived the study. All authors contributed to data acquisition, analyses
and interpretations. AP, SS and JE drafted the manuscript. All authors
critically revised the manuscript. All authors approved the final version. PH
and JE provided funding support. AP and SS contributed equally to this work.
Funding: Glasgow Royal Infirmary Endowment Board. Ethics approval: This study
was approved by the West of Scotland Research Ethics Committee.
CONFLICT OF
INTERESTS
There are no
conflicts of interest with regard to the present study.
REFERENCES
1 Cancer stats, 2009 CancerResearchUK.org
[homepage on the internet] London: Cancer Research UK; [Page last updated: 17
November 2008; cited April 2009]; Available from:
http://info.cancerresearchuk.org/cancerstats/types/bowel/incidence/
2 Campbell Roxburgh,
Alec McDonald, Jonathan Salmond, Karin Oien, John Anderson, Ruth McKee, Paul Horgan,
Donald McMillan. Adjuvant chemotherapy for resected
colon cancer: comparison of the prognostic value of tumour
and patient related factors. Int J
Colorectal Dis 2011; 26: 483-492.
3 Volanakis JE. Human C-reactive protein: expression, structure, and function.
Mol Immunol 2001; 38: 189-197
4 Alexandrakis
MG, Passam FH, Moschandrea
IA, Christophoridou, AV, Pappa,
Constantina A, Coulocheri
SA, Kyriakou DS. Levels of serum
cytokines and acute phase proteins in patients with essential and
cancer-related thrombocytosis. Am J Clin Oncol 2003; 26:
135-140.
5 Balkwill
F, Mantovani A. Inflammation and cancer: back to
Virchow? Lancet 2001; 357: 539-545
6 Blann AD, Byrne
GJ, Baildam AD. Increased soluble
intercellular adhesion molecule-1, breast cancer and the acute phase response.
Blood Coagul Fibrinolysis
2002; 13:165-168
7 Basso D, Fabris
C, Meani A et al. C reactive protein in pancreatic
cancer and chronic pancreatitis. Ann Clin Res
1988; 20: 414-416
8 O¡¯Hanlon DM, Lynch J, Cormican M, Given HF. The acute phase
response in breast carcinoma. Anticancer Res 2002; 22:1289-1293.
9 Chikao Miki, Naomi Konishi, Eiki Ojima, Tsuyoshi Hatada, Yasuhiro Inoue, Masato Kusunoki.
C-reactive protein as a prognostic variable that reflects
uncontrolled up-regulation of the IL-1-IL-6 network system in colorectal
carcinoma. Dig Dis Sci
2004; 49: 970-976
10 Wigmore
SJ, Fearon KC, Sangster K et al. Cytokine regulation
of constitutive production of interleukin-8 and -6 by human pancreatic cancer
cell lines and serum cytokine concentrations in patients with pancreatic
cancer. Int J Oncol
2002; 21 :881-886
11 Hurlimann J, Thorbecke GJ, Hochwald
GM. The liver as the site of C-reactive protein formation. J Exp Med 1996; 123:
365-378
12 Jabs WJ, Logering BA, Gerke P, Kreft B, Eva-Maria Wolber,
Matthias Heinrich Friedrich Klinger, Lutz Fricke, J¨¹rgen
Steinhoff. The kidney as a second site of human C-reactive
protein formation in vivo. Eur
J Immunol 2003; 33: 152-161
13 Jabs WJ, Busse M, Kruger S, Jocham D, SteinhoffJ, Doehn C. Expression of C-reactive protein by
renal cell carcinomas and unaffected surrounding renal tissue. Kidney Int 2005; 68: 2103-2110
14 Nozoe
T, Korenaga D, Futatsugi M,
Saekib H, Maeharab Y, Sugimachi K. Immunohistochemical
expression of C-reactive protein in squamous cell
carcinoma of the esophagus - significance as a tumor marker. Cancer Lett 2003; 192: 89-95
15 Yang J, Wezeman
M, Zhang X, Lin P, Wang M, Qian JF, Wan B, Kwak LW, Yu L, Yi Q. Human C-reactive protein binds
activating Fcgamma receptors and protects myeloma
tumor cells from apoptosis. Cancer Cell 2007; 12: 252-265
16 Canna
K, Hilmy M, McMillan DC et al. The
relationship between tumour proliferative activity,
the systemic inflammatory response and survival in patients undergoing curative
resection for colorectal cancer. Colorectal Dis
2008; 10: 663-667
17 Roxburgh
Campbell S D, Salmond Jonathan M MBChB,
Horgan Paul G, Oien Karin
A, McMillan Donald C. Comparison of the prognostic value of inflammation-based
pathologic and biochemical criteria in patients undergoing potentially curative
resection for colorectal cancer. Ann Surg
2009; 249: 788-793
18 Elsberger B, Lankston L, McMillan DC, Underwood MA,
Edwards J. Presence of tumoural C-reactive protein
correlates with progressive prostate cancer. Prostate Cancer Prostatic Dis 2011; 14: 122-128
19 Kirkegaard T, Edwards J, Tovey S, McGlynn
LM, Krishna SN, Mukherjee R, Tam L, Munro AF, Dunne
B, Bartlett JMS. Observer variation in immunohistochemical analysis of protein expression, time for a change?
Histopathology 2006; 48: 787-794
20 Contu PC,
Contu SS, Rosito MA, Moreira LF. Expression of C-Reactive
Protein in Rectal Cancer. Gastroenterology Research 2009; 2:
104-109
21 Grivennikov S,
Karin E, Terzic J, Mucida
D, Yu GY, Vallabhapurapu S, Scheller
J, Rose-John S, Cheroutre H, Eckmann
L, Karin M. IL-6 and Stat3 are required for survival of intestinal epithelial
cells and development of colitis-associated cancer. Cancer Cell 2009; 15:
103-113
Peer reviewers: Takeru Wakatsuki, The Cancer
Institute Hospital of Japanese Foundation for Cancer Research. Dep. for
Internal Medicine of Gastroenterology, 3-8-31 Ariake, Koto, Tokyo, 135-8550
Japan; Marco Caricato, Department of Surgery, University Campus Bio-Medico di
Roma , Via Álvaro del Portillo, 200, 00128 Roma Italy; Zenia Saridaki Zoras,
Medical Oncologist, Scientific Director of the Oncology Department
"Asklepios", Maxis Kritis 10, Heraklion Crete 713 03, Greece.
Refbacks
- There are currently no refbacks.