Background
Liver resection or ablation remains the only potentially curative option for patients with colorectal liver metastases to date. Factors determining the risk of post-operative complications which may occur in 40-50% of patients have been investigated extensively [
1-
3]. Generally, morbidity is related to the ability of the remnant liver mass to regenerate and thus fulfil its metabolic functions [
4]. Monocytes as central players of the immune system play a pivotal role in liver regeneration. While overall numbers of monocytes increase after surgery, their immune function is partly suppressed [
5,
6]. Of note, a two-fold increase in monocytes on the first post-operative day has been found predictive in terms of enhanced overall survival after resection of colorectal liver metastases [
7].
The complex regulation of monocytes involves an array of proteins including C-reactive protein (CRP), macrophage colony stimulating factor (M-CSF), transforming growth factor beta 1 (TGFβ1) and angiopoietin 2 (ANG-2) [
8-
10]. These mediators are known to direct the recruitment and differentiation of monocyte populations for functional responses in phagocytosis, antigen presentation, angiogenesis and cytokine release. Furthermore, monocytes that migrate into the tissue develop into macrophages and dendritic cells which shape the local immune response and tissue regeneration [
11-
13]. The importance of M-CSF and monocyte-derived macrophages in liver regeneration has been investigated in mice genetically lacking functional M-CSF [
11]. These mice showed a reduction of Kupffer cells in the liver by 60%. After partial hepatectomy the mice exhibited a significantly reduced proliferation of hepatocytes and a delayed hepatic regeneration. Treatment with recombinant M-CSF resulted in the recovery of Kupffer cell counts and liver regeneration.
An indication as to how M-CSF is induced upon liver damage came from the
in vitro observation that the acute phase protein CRP may stimulate endothelial cells and macrophages to secrete M-CSF [
14]. M-CSF is known to function as a chemoattractant for monocytes to inflamed or injured tissues [
15] and can further promote survival and differentiation of monocytes [
16,
17]. Thus, it was shown that cytokines such as M-CSF and TGFβ1 trigger a phenotypic shift of circulating human monocytes by inducing the expression of CD16, the low-affinity immunoglobulin G receptor involved in antibody-dependent cellular cytotoxicity [
8,
10].
By differential expression of their receptors CD16 and CD14, the co-receptor for lipopolysaccharide (LPS), human monocytes are classified as CD14++CD16- "classical monocytes", CD14++CD16+ "intermediate monocytes" and CD14+CD16++ "non-classical monocytes" [
18]. Functional and genetic studies have indicated a developmental relationship between these subsets with gradual changes in surface markers during maturation as well as distinct biological properties [
19,
20]. Classical monocytes which amount to 85% of circulating monocytes, show the highest phagocytosis potential and are potent producers of pro-inflammatory cytokines in response to the bacterial component LPS [
19,
20]. In contrast, non-classical monocytes, amounting to 10% of total monocytes, exhibit a "patrolling" (crawling) behaviour along vessel walls and react strongly against viruses [
21,
22]. Intermediate monocytes have been linked to antigen presentation as well as angiogenesis [
20]. Thus, factors involved in MHCII complex formation (HLA-DR and CD74) and angiogenic molecules such as TIE2, endoglin and VEGFR2 (vascular endothelial growth factor receptor 2) are expressed at highest levels in the intermediate subset and point to functions in tissue remodelling [
9,
20].
A rare subset of monocytes is known to express TIE2, the receptor for angiopoietins, and is therefore termed TIE2 expressing monocytes (TEMs) [
23]. With respect to the official classification of monocytes, TIE2 expression is predominantly but not exclusively detected on intermediate monocytes [
9,
20]. TEMs were found to migrate to angiogenic sites in tumours potentially involving chemotaxis in response to ANG-2 [
23,
24]. They may stimulate angiogenesis by the paracrine secretion of growth factors such as VEGF and basic fibroblast growth factor [
24].
While the majority of functional studies have been conducted in mouse models, substantial evidence has additionally been gathered in the human setting demonstrating that human intermediate monocytes and TEMs carry angiogenic markers and have angiogenic properties [
9,
20,
25]. Thus, endothelial cell sprouting and tube formation were more potently induced by human TEMs than by TIE2-negative monocytes [
25], and intermediate as opposed to classical and non-classical monocytes were described as the subset with distinct tube forming capability [
20].
In contrast to studies performed on cancer patients, little is known about the regulation and contribution of pro-angiogenic monocyte subsets in the context of clinical liver resection and regeneration. Since surgery triggers the release of CRP from the liver and as a consequence may raise the levels of circulating M-CSF, a shift in monocytes towards CD16 positive subsets and their functional properties is conceivable. Given the pro-angiogenic potential of CD16+ intermediate monocytes and TEMs, we hypothesized that they are induced by liver surgery and subsequently localize at the resection site to promote tissue regeneration. Conversely, failure of the immune system to recruit the reportedly pro-angiogenic TEMs and intermediate monocytes might result in adverse outcome after surgery.
To address this subject, we conducted a study on 38 patients undergoing resection of colorectal liver metastases. We examined postoperative changes in the monocyte profile of patient blood as well as in wound fluid of subhepatic drainages. Alterations in monocyte subsets were investigated for a potential correlation with circulating levels of monocyte regulating cytokines including CRP, M-CSF, TGFβ1 and ANG-2. Furthermore, monocyte populations and related cytokines were assessed for a potential association with clinical parameters of liver function and injury after hepatic resection.
Methods
Patient collective
We analysed a total of 38 patients with liver metastases from colorectal cancer undergoing hepatic resection at the Department of Surgery, Medical University of Vienna between 2007 and 2011. Prior to liver surgery all patients underwent chemotherapy with a standard combination, mostly including bevacizumab. The majority of patients had their primary tumour resected before diagnosis and treatment of metastases. Patients with pre-existing liver disease were excluded from the study. The patient characteristics, their pre-operative variables of liver function and type of hepatic resection are summarized in Table
1. Liver resections were classified in minor (≤3 segments) and major (>3 segments) hepatectomy. Comparably, a sex- and age-matched control collective of 32 healthy individuals was included in the study. Analyses of patient samples were approved by the Institutional Ethics Committee (#300/2006, #437/2006, #791/2010). All patients gave written informed consent.
Table 1
Patient demographics
Sex | |
Male | 24 (63%) |
Female | 14 (37%) |
Site of primary tumour | |
Colon | 23 (61%) |
Rectum | 15 (39%) |
Concomitant primary resection | 7 (18%) |
Type of hepatic resection | |
Major | 20 (53%) |
Minor | 18 (47%) |
Preoperative parameters of liver function |
Median (Range)
|
Bilirubin (mg/dl) | 0.64 (0.35-1.71) |
ALAT (U/l) | 22 (9-346) |
ASAT (U/l) | 28 (8-404) |
gGT (U/l) | 35 (13-266) |
PT (%) | 107 (45-135) |
Age (years) | 65 (42-80) |
Sample collection
Blood was drawn before surgery (pre-OP) prior to the patients' transfer to the operating room and on post-operative day (POD) 1 and 3. Furthermore, in a subset of 12 patients wound fluid was additionally collected on POD 1 and POD 3. Surgical drainage catheters were placed intraabdominally directly subhepatic, to collect subhepatic wound exudate (SHW) close to the site of liver regeneration.
Analysis of monocyte populations
Blood was drawn into EDTA (ethylenediamine tetraacetic acid) containing tubes and processed at room temperature. Surface expression of CD14, CD16 and TIE2 was measured applying direct immunofluorescence staining followed by a lyse-no-wash procedure. In brief, 100 μl of whole blood were incubated with the following antibodies at saturating concentrations for 20 minutes: CD14-FITC (Becton-Dickinson, San Jose, CA, USA), CD16-PC5 (Beckman Coulter, Fullerton, CA, USA) and TIE2-PE (R&D Systems, Inc., Minneapolis, MN, USA). To eliminate erythrocytes the Versa Lyse solution (Beckman Coulter) was added for 20 minutes. Wound exudate was aspirated from drainage bags and transferred into an EDTA containing tube to allow for a comparable mode of sample processing and immunostaining as described for blood samples.
Flow cytometry was immediately performed with an FC500 cytometer (Beckman Coulter). Fluorescence gating parameters were established using antibody isotype controls, and values above the 99% negative staining threshold were considered positive. A total of 300.000 leukocytes were measured. Analysis of flow cytometry data was performed with Kaluza software (Beckman Coulter) and the gating strategy is documented in the Additional file
1. Intermediate monocytes (CD14++CD16+) were identified by high-level expression of CD14 and were further distinguished from classical monocytes (CD14++CD16-) by their co-expression of CD16. Numbers of intermediate and classical monocytes were combined to yield the total CD14++ monocyte count. TEMs were identified by the concomitant expression of CD14 and TIE2 (CD14++TIE2+ cells). Data are given in % frequency of total CD14++ monocytes.
Analysis of soluble blood parameters
For plasma preparation, blood was drawn into pre-chilled tubes containing CTAD (citrate, theophylline, adenosine and dipyridamole) and was processed on ice within 30 minutes as we have previously described [
26]. Plasma samples were stored in aliquots at −80°C until further analysis. For measurement of ANG-2, M-CSF (R&D Systems, Inc., Minneapolis, MN, USA) and TGFβ1 (eBioscience, San Diego, CA, USA) commercially available ELISA kits were applied according to the manufacturers' instructions. As the levels of active, unbound TGFβ1 were below the detection limit in CTAD plasma, total TGFβ1 was measured after acid-activation of plasma samples according to the manufacturer’s protocol. Thus, TGFβ1 values are given as the concentration of total protein including both, the active, freely circulating and the inactive, latency-associated peptide bound form of TGFβ1. Serum CRP levels and parameters of liver function or injury were available from routine hospital evaluation.
Statistical analysis
Statistical calculations were based on non-parametric tests using SPSS software version 20 (SPSS Inc., Chicago, IL, USA). Due to the challenges in collecting samples in a clinical setting, the available number of measured parameters varied with time points. Please note that statistical tests for paired samples were conducted, i.e., differences between time points were calculated using the Wilcoxon Test. Correlations between the investigated parameters were determined by Spearman’s rank correlation coefficient. The reported p-values were results of two-sided tests. P-values <0.05 were considered statistically significant. As the study design was explorative, no corrections for multiple testing were performed.
Discussion
This study demonstrates that liver resection triggers a strong shift in monocyte subsets and monocyte regulating cytokines. Elevated numbers of the reportedly pro-angiogenic CD16+ intermediate population and decreased numbers of CD16- classical monocytes were detected in patient blood on POD 1 and 3. This rapid increase in circulation was followed by an accumulation of intermediate monocytes in subhepatic wound fluid by POD 3. Comparably, M-CSF, TGFβ1, CRP and ANG-2 increased in patient plasma after surgery; in particular M-CSF and CRP concentrations correlated strongly with the level of intermediate monocytes.
Surgical tissue damage leads to an acute phase response in the liver and hence an increase in CRP release into the blood [
28]. Since CRP is known to induce M-CSF expression by endothelial cells [
14] and M-CSF up-regulates CD16 expression on human monocytes [
8], the observed changes in monocyte phenotype are likely to be triggered as a result of the altered cytokine milieu after surgery. In this regard our findings are in line with a recently published study showing increased values and a correlation of intermediate monocytes, M-CSF and CRP in trauma patients [
29].
We would like to note that our study was focused on the analysis of the potentially pro-angiogenic population of CD16+ intermediate monocytes as opposed to the majority of CD16- classical monocytes. With respect to the third subset of CD16++ non-classical monocytes, our detection method was prone to "contamination" with overlapping natural killer cells in flow cytometric measurements [
30]. Hence, we generally omitted this subset from our analysis. However, to gain preliminary information whether non-classical monocytes were similarly induced by liver surgery, we devised a modified gating strategy for non-classical monocytes (Additional file
4A): A logarithmic leukocyte density plot was introduced to improve resolution of monocytes, lymphocytes and granulocytes based on light scatter properties. Setting a tight gate on the monocyte population enabled us to minimize "contamination" of non-classical monocytes by CD16++ natural killer cells or granulocytes. We found that in contrast to the CD14++CD16+ intermediate monocytes (Additional file
4C), the CD16 positive non-classical (CD14+CD16++) subset exhibited a significant decrease (Additional file
4D) on POD 1 (p < 0.001, N = 24) and a partial recovery by POD 3 (p < 0.001, N = 21). In wound fluid, the subset of non-classical monocytes was essentially absent (data not shown). Thus, the intermediate monocyte population was selectively induced upon liver surgery which may relate to its reported angiogenic capacity and a pertaining potential to promote liver regeneration. While this study has focused on patients with liver resection, a comparable rise in intermediate monocytes to support tissue regeneration might also be envisaged for patients undergoing major surgery other than liver resection. This has, however, not been addressed in the current study.
With respect to the proposed pro-angiogenic TEM subset, there were no substantial fluctuations in the circulating numbers of TEMs after surgery and no correlation with the post-operative release of ANG-2 in patient blood. However, we found an accumulation of TIE2 positive monocytes (CD14++TIE2+ cells) in subhepatic wound fluid which is likely to be conditioned by the regenerating liver. Comparably, our results are in line with preclinical studies which detected TIE2+ monocytes in angiogenic mouse tissue [
24]. Rather than the induction of TEMs in blood, the local upregulation of TIE2 expression on monocytes due to environmental factors of the injured tissue might be responsible for this observation. Given the role of TIE2 in angiogenesis, the elevated expression of TIE2 on monocytes would be expected to enhance their angiogenic properties. It has to be noted that this study was designed to unravel monocyte changes and correlations in clinical parameters but did not include functional tests. Hence, the attribution of intermediate monocytes and TEMs as pro-angiogenic subsets relies on the assessment of previous reports [
9,
20,
25].
We further addressed the question whether alterations in the monocyte profile after liver resection are associated with hallmarks of liver injury and regeneration. Therefore, we included liver function parameters from routine hospital evaluation in our analysis. We found positive correlations of intermediate monocyte levels on POD 3 with liver parameters on POD 3-4, whereas measurements from POD 1 did not show significant correlations. While the surge of intermediate monocytes on the first post-operative day is likely an immediate reaction common to all patients, a sustained increase on POD 3 may indeed reflect a delayed, ongoing liver regeneration and reduced organ function as indicated by the elevated levels of liver enzymes on POD 3-4. In line, the median blood levels of angiogenic monocytes on POD 3 tended to be higher in patients with major (17% intermediate subset within CD14++ monocytes) as opposed to minor (8%) surgery (p = 0.065, N = 21), and intermediate monocytes significantly correlated with the level of tissue destruction as evidenced by LDH release. These observations strengthen our conclusion that monocyte populations with proposed angiogenic properties are associated with the process of liver regeneration after surgery. With respect to clinical implications the selective induction of this monocyte subset might therefore be envisioned as a therapeutic approach which merits further investigation: Bearing in mind that postoperative monocyte counts are predictive for overall survival after resection of colorectal liver metastases [
7], it has previously been attempted to raise monocyte counts by the perioperative treatment with recombinant GM-CSF [
31]. While GM-CSF is known to boost the immunoreactive properties of monocytes [
31], it induces the classical rather than the intermediate (CD16+) phenotype [
32]. Thus, the short-term perioperative application of M-CSF as opposed to GM-CSF might be more effective in promoting liver regeneration after surgery by selectively enhancing the angiogenic monocyte subsets. Further support to this notion was given by an experimental model of partial hepatectomy in M-CSF deficient mice, where perioperative application of recombinant M-CSF was found to rescue liver regeneration [
11].
Authors' contributions
DS contributed essentially to the study design, acquisition of monocyte and cytokine data, conducted the statistical analysis and delivered the initial manuscript draft. PS provided clinical data and critically revised the manuscript. PZ, LA, TM, and EB assisted in monocyte measurements and contributed to the paper discussion. LP and BG contributed to the acquisition of patient samples and clinical data and critically revised the manuscript. TG was involved in the study design, acquisition of clinical data and paper revision. CB contributed to the experimental design, statistical analysis of data, and revised the paper draft. All authors read and approved the final manuscript.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made.
The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.
Competing interests
The authors declare that they have no competing interests.