CXCR4 – a possible serum marker for risk stratification of abdominal aortic aneurysms
Abstract
Summary:Background: Abdominal aortic aneurysm (AAA) rupture is still associated with a mortality rate of 80–90%. Imaging techniques or molecular fingerprinting for patient-specific risk stratification to identify pending rupture are still lacking. The chemokine (C-X-C motif) receptor (CXCR4) activation by CXCL12 ligand has been identified as a marker of inflammation and atherosclerosis, associated with AAA. Both are highly expressed in the aortic aneurysm wall. However, it is still unclear whether different expression levels of CXCR4 and CXCL12 can distinguish ruptured AAAs (rAAA) from intact AAAs (iAAA). Patients and methods: Abdominal aortic tissue samples (rAAA: n=29; iAAA: n=54) were excised during open aortic repair. Corresponding serum samples from these patients (n=9 from rAAAs; n=47 from iAAA) were drawn pre-surgery. Healthy aortic tissue samples (n=8) obtained from adult kidney donors during transplantation and serum samples from healthy adult volunteers were used as controls (n=5 each). Results: CXCR4 was mainly expressed in the media of the aneurysmatic tissue. Focal positive staining was also observed in areas of inflammatory infiltrates within the adventitia. In tissue lysates, no significant differences between iAAA, rAAA, and healthy controls were observed upon ELISA analysis. In serum samples, the level of CXCR4 was significantly increased in rAAA by 4-fold compared to healthy controls (p=0.011) and 3.0-fold for rAAA compared to iAAA (p<0.001). Furthermore a significant positive correlation between aortic diameter and serum CXCR4 concentration was found for both, iAAA and rAAA (p=0.042). Univariate logistic regression analysis showed that increased CXCR4 serum concentrations were associated with AAA rupture (OR: 4.28, 95% CI: 1.95–12.1, p=0.001). Conclusions: CXCR4 concentration was significantly increased in serum of rAAA patients and showed a significant correlation with an increased aortic diameter. The level of CXCR4 in serum was associated with a more than 4-fold risk increase for rAAA and thus could possibly serve as a biomarker in the future. However, further validation in larger studies is required.
Introduction
An abdominal aortic aneurysm (AAA) is a potentially life-threatening disease that mainly affects male patients over the age of 65. Rupture (rAAA) is still associated with a mortality rate of 80–90%, which can only be significantly reduced by elective surgery [1]. Apart from comorbidities and patient’s preference, the indication for surgical treatment is mainly determined by the diameter (from 5.0 to 5.5 cm), speed of growth in diameter and patients symptoms like abdominal or back pain of the intact non-ruptured AAA (iAAA) [1].
However, there are patients, in whom AAAs rupture at smaller diameters, whereas in others remaining intact above the diameter threshold throughout life (rupture rate of up to 1.61 per 100 person-years with a diameter of 3.0–5.5 cm) [2]. Reliable imaging and/or molecular fingerprints for more accurate, patient specific risk stratification are still lacking. For some time now, research has focused on better diagnostic procedures, such as specific serum biomarkers or imaging, aiming for quantification of an imminent rupture risk in iAAA, independent of the aortic diameter [3].
Inflammatory processes in the aortic wall are pathogenic hallmarks responsible for the progression and eventual subsequent rupture of AAA [4]. Activation of the chemokine (C-X-C motif) receptor 4 (CXCR4) by the chemokine (C-X-C motif) ligand 12 (CXCL12) has been identified as an important marker of inflammation and atherosclerosis [5, 6, 7, 8]. Both proteins play an important role in immunologic and cancerogenic processes (e.g. evaluation of tumor growth and metastasis) [9, 10]. Furthermore they also seem to be involved in the pathophysiology of AAA [5, 11, 12]. Here, the exact mechanistic involvement is still unclear. Previously, our group as well as others have shown that CXCR4 and CXCL12 are highly expressed in aortic wall tissue derived from iAAA [7, 11]. Ruptured AAAs are believed to be associated with a different inflammatory response than intact AAA [13, 14], which might be quantifiable by the expression of CXCR4.
Thus, we hypothesized that different expression levels of CXCR4 and CXCL12 in rAAAs compared to iAAAs might serve as a potential biomarker to identify patients at increased risk of rupture.
Patients and methods
Study population and sample collection
All AAA tissue samples used in this study (total: n=83; rAAA: n=29; iAAA: n=54) were part of the Munich Vascular Biobank and were collected from patients who underwent open aortic repair in the Department for Vascular and Endovascular Surgery (Klinikum rechts der Isar, Technical University of Munich, Germany) [15, 16]. Tissue specimens were excised from the anterior or lateral part of the aneurysm sac during vascular graft interposition, still leaving enough aneurysm tissue to cover the prosthesis as much as possible. Healthy aortic tissue samples (n=8) obtained from adult kidney donors during transplantation in the Department of General Surgery (Klinikum rechts der Isar, Technical University of Munich, Germany) served as controls.
Following resection, the tissue samples were sectioned for histologic and molecular analyses and either fixed in formalin or immediately frozen in liquid nitrogen. The formalin fixed specimens were embedded in paraffin (FFPE). The frozen samples were stored at −80 °C.
Corresponding serum samples (n=56 in total; n=9 from rAAAs; n=47 from iAAA) were collected from blood samples drawn in close time proximity to the surgery. Unfortunately, for 27 patients it was not able to withdraw blood samples for the biobank due to the emergency admission.
Serum samples from healthy adult volunteers were used as controls (n=5). For serum separation, the collected whole blood samples were centrifuged for 15 min, and the supernatant was aliquoted immediately. Serum specimens were stored at −80 °C.
Protein extraction
For protein extraction, tissue samples were sectioned into pieces of approximately 2–3 mm3 and frozen in liquid nitrogen after adding Tissue Extraction Reagent I (Thermo Fisher Scientific, Waltham, MA, USA). Tissue homogenization was performed using the Bio-Gen PRO200 Homogenizer and Multi-Gen 7XL Probes (PRO Scientific, Oxford, CT, USA). Protein lysate was centrifuged at 4 °C for 20 min at 10,000× g and the supernatant was used for further protein analysis. Concentration of total protein was measured using the Pierce BCA Protein Assay Kit (Thermo Scientific, Rockford, IL, USA) and the samples were diluted to 1 μg/μl with Tissue Extraction Reagent I.
Quantitative analysis at protein level (ELISA)
Enzyme-Linked Immunosorbent Assays (ELISA) were used for quantitative analyses on the protein level for both tissue and serum samples. CXCR4 and CXCL12 were measured using the ELISA Kit for Human CXCR4 (Cloud-Clone Corp., Katy, TX, USA) and the Quantikine ELISA for Human CXCL12 (R&D Systems, Minneapolis, MN, USA). In addition, CXCR7, a second receptor for CXCL12, was measured in tissue samples with the ELISA Kit for Human CXCR7 (abbexa, Cambridge, UK), but could not be detected (data not shown). Concentration of GAPDH at protein level (Human GAPDH ELISA Kit, LifeSpan Biosciences Inc., Seattle, WA, USA) was used for normalization in tissue samples.
Histological staining
For histological characterization, 2–3 μm sections of ruptured and elective AAA tissue samples were cut from the paraffin blocks and mounted on slides. Hematoxylin-eosin (HE) and elastica van Gieson (EvG) staining were performed for morphological evaluation. Selected representative slides from both groups were also stained immunohistochemically (IHC) to evaluate the abundance and cellular localization of CXCR4 in relation to other cellular markers (CD68 for macrophages, CD45 for leucocytes and SM-α actin for smooth muscle cells). Consecutive tissue sections (2–3 μm) were cut and mounted on 0.1% poly-L-lysine coated (Sigma-Aldrich, St. Louis, MO, USA) slides. The samples were subsequently dried overnight at 56 °C, de-paraffinized, rehydrated and boiled in 10nM citrate buffer (pH 6.0) for 7 min using a pressure cooker for antigen retrieval. Pre-treatment with 3% hydrogen peroxide (H2O2) was used to block endogenous peroxidase activity. The samples were then incubated for 1 hour with the following antibodies, appropriately diluted in a REALTM Antibody Diluent (Dako, Glostrup, Denmark): anti-CXCR4 (ab124824, Abcam Inc., Cambridge, UK; dilution 1:300), anti-CD68 for macrophages (M0814, clone KP1, Dako, Glostrup, Denmark; dilution 1:2000), anti-CD45 for leucocytes (M0701, clones 2B11 + PD7126, Dako, Glostrup, Denmark; dilution 1:500) and anti-SM-α actin for smooth muscle cells (SMC) (M0635, clone HHF35, Dako, Glostrup, Denmark; dilution 1:200). Afterwards, the slides were incubated with a biotinylated secondary antibody (25 min) and peroxidase conjugated streptavidin (25 min), followed by visualization with a DAB chromogen (3–5 min) (REALTM Detection System Peroxidase/DAB+, Dako, Glostrup, Denmark). Mayer’s hematoxylin was used for counterstaining. Human tonsil tissue samples served as positive controls and for primary antibody control samples were incubated with the diluent alone, followed by secondary antibody incubation and detection.
Ethics approval and consent to participate
The study was carried out in accordance with the local ethical committee of the Klinikum rechts der Isar (reference number 2799/10) and with the international Guidelines of the World Medical Association Declaration of Helsinki. Written informed consent was obtained from all patients taking part in the study. It should be noted, that rAAA patients underwent immediate surgical intervention upon hospital admission, without any delay. Prior to surgery, blood samples were withdrawn for standard laboratory analyses, and one temporarily stored for our biobank. During surgery, the sample of aortic tissue was removed, as part of the standard treatment, and temporarily stored. Following successful therapy, the patients were asked to sign the necessary informed consent. If patients agreed, blood samples and tissue specimens were proceeded to our vascular biobank and were used in this study. If patients rejected or passed away before he could give his consent, both the collected blood sample and tissue were immediately disposed.
Statistics
Outliers in the protein expression were determined using Mahalanobis distance, visually checked, and removed from all analyses. Based on the data distribution, groups were compared using Chi Square test (categorical variables), t-test (normally distributed continuous variables), or Mann-Whitney test (non-normally distributed continuous variables). Data were tested for normal distribution using the Kolmogorov-Smirnov one-sample test.
Linear regression analysis was used to test the correlation of protein expression or concentration and aortic diameter. Logistic regression was used to assess the odds ratio of aortic rupture for the protein expression. Across all regression analyses, the protein expression was standardized a priori.
All tests were two sided and a p-value below 0.05 was considered significant. Statistical analyses were performed using R Statistical Software (v4.1.1), graphics were generated using the ggplot2-library.
Results
Patient characteristics
As summarized in Table I, patients with rAAA were significantly older than patients with iAAA (median (IQR) 74 (69–74) vs. 70 (63–73) years, p=0.013) and had a significantly larger AAA maximum diameter as measured by pre-operative CT scans (median (IQR) 90 mm (70–94) vs. 56 mm (51–67), p<0.001). The most prevalent cardiovascular-related comorbidities were hypertension (80.7%), hyperlipidemia (43.4%), and smoking (55.4%). Significantly more rAAA patients suffered from chronic kidney disease (51.7% vs. 18.5%, p=0.003).
Upon admission, serum analyses of patients with rAAA showed significantly higher levels of creatinine (1.4 mg/dl vs. 1.0 mg/dl, p<0.001) and leucocytes (14.0 G/l vs. 8.11 G/l, p<0.001), as well as lower hemoglobin (10.9 g/dl vs. 13.8 g/dl, p<0.001) and hematocrit levels (32.1% vs. 40.2%, p<0.001).
Control patients included n=8 (5 men) aortic healthy patients with a median age of 66 (54–72) years.
Cellular localization of CXCR4 in aneurysm samples
Cellular localization of CXCR4 was evaluated in the context of inflammatory cells and vascular smooth muscle cells (VSMCs). In general, rAAA showed more advanced morphologic disease characteristics including increased inflammatory infiltration (increased focal cellularity, neovascularization, accumulation of CD68-positive cells), lipid core formation and advanced vessel wall degeneration (intraluminal thrombus formation, intima-media thickening, disorganized fiber structure, fragmentation of elastin fibers), especially in comparison to the normal aortic architecture in healthy non-aneurysmatic donor samples (Figure 1).
Protein level CXCR4/CXCL12 in tissue and serum
In tissue lysates, no significant differences between iAAA (n=54), rAAA (n=29) and healthy controls (n=8) (CXCR4: for controls (Ctrl) vs. iAAA p=0.86, for Ctrl vs. rAAA p=0.40; CXCL12: for Ctrl vs. iAAA, p=0.28, for Ctrl vs rAAA p=0.89) were observed upon ELISA analysis (Figure 2).
In serum samples, the level of CXCR4 was significantly increased by 4-fold in rAAA compared to healthy controls (p=0.011) and by 3.0-fold for rAAA compared to iAAA (p< 0.001). In contrast, the expression of CXCL12 did not show significant differences in serum between AAA and healthy individuals (p=0.82) as well as between iAAA and rAAA (p=0.26) (Figure 2).
Correlation of the aortic diameter with the concentration of CXCR4 and CXCL12 within tissue and serum samples
When correlating the aortic diameter of all AAAs (intact and ruptured), no agreement was found for CXCL12, both in tissue lysate and serum. The same was observed for CXCR4 in tissue lysate. However, there was a significant correlation of an increase in aortic diameter with the concentration of CXCR4 in serum (p=0.042) (Figure 3).
In contrast, when iAAA and rAAA were considered separately, no correlation was observed between aortic diameter and serum concentration for either CXCL12 or CXCR4 (Figure 4).
Logistic regression analysis to test for risk factors for aortic rupture
Univariate logistic regression analysis for aortic rupture showed no significantly increased risk association for CXCR4 and CXCL12 concentrations in tissue lysate. However, increased CXCR4 concentration in serum samples was identified as an independent risk factor associated with rAAA (OR: 4.28, 95% CI: 1.95–12.1, p=0.001). In this model, again CXCL12 in serum was not associated with any risk of rAAA (Figure 5).
Discussion
This study showed for the first time that CXCR4 is significantly increased in serum of patients with rAAA compared to iAAA and that there is a significant correlation between the aortic diameter and the serum CXCR4 concentration. These results could point to a specific role of CXCR4 serum levels in ruptured compared to intact AAA.
As previously shown, there is a significant overexpression of the receptor CXCR4 in iAAA compared to healthy subjects [7]. This observation was confirmed by this study. Hinterseher et al. were also able to show that CXCR4 is upregulated in aortic aneurysms [17]. In addition to AAAs, CXCR4 has also been shown to be upregulated in other aneurysms, such as thoracic aortic aneurysms and intracranial aneurysms, and has been proposed as a potential therapeutic target or biomarker [18].
Patients with rAAA showed high white blood cell counts as an indication of an increased systemic inflammatory response, which is consistent with the fact that rAAAs are often associated with increased inflammation and shock [19, 20, 21, 22]. In the current samples, immunohistochemistry also showed increased infiltration of inflammatory cells, lipid core formation, and vessel wall degeneration in rAAAs. In addition, it was shown, that CXCR4 is mainly localized in the vascular media, focally also in the adventitia in the area of inflammatory foci. Interestingly, expression was found not only in macrophages but also in vascular smooth muscle cells. These results are in line with others, describing that in contractile SMCs the expression of CXCR4 is generally low and is significantly increased in injured arteries as well as synthetic SMCs [23]. Thus, the expression of CXCR4 in AAAs may also reflect the phenotypic state of these cells. SMCs are the main source of the components of extracellular matrix, particular collagens, and thus significantly contribute to the stability of the aortic wall. The expression pattern of CXCR4 suggests a broad mechanism of action albeit underscoring a potentially important biological role in AAA pathogenesis.
It is already known, that during inflammation, CXCR4 plays an important role in modulating the immune response. The CXCR4-CXCL12 axis is also widely involved in the development of immune cells [11]. Additionally, the CXCR4-CXCL12 axis plays a crucial role in the homing of stem and progenitor cells in the bone marrow and controls their mobilization into peripheral blood as well as tissues in homeostatic conditions, but also after tissue injury or stress [6]. CXCR4 and its ligand CXCL12 appear to play a key role in many immunological areas such as organogenesis, vascularization, and embryogenesis [24, 25, 26]. CXCR4 has been shown to be highly expressed in infiltrating cells, especially T and B lymphocytes, isolated from human AAAs [7, 11]. A possible mechanism that has been suggested for its immunological role is, that CXCR4, expressed on infiltrating lymphocytes, and CXCL12, expressed on stromal cells, are involved in the recruitment of lymphocytes within the arterial wall in AAA [11]. Interestingly, similar to rAAA, increased serum concentrations of CXCR4 were also detected in patients with sepsis [27]. One might therefore theorize that CXCR4 could possibly be counted among the acute phase proteins.
Of interest, inhibition of CXCR4 with a specific inhibitor AMD3100 leads to decreased inflammation in mouse and rat models of atherosclerosis. It decreases infiltration of adventitial macrophages, prevents aortic wall destruction, and inhibits AAA formation as well as stabilizes the existing AAA [12]. Here, a correlation of the concentration of CXCR4 in serum with aortic diameters was found, which was also demonstrated for human serum in the current study.
For the receptor ligand of CXCR4, CXCL12, on the other hand, no significant differences at protein level could be detected either in serum or in tissue. This could be due to the short half-life of CXCL12, which is only 26 min, especially in the blood circulation [28]. These results are thus in line with the findings of Merkelbach et al. in human carotid atherosclerotic plaques and implicate CXCL12 as less suitably in the context of circulating biomarkers [8].
Implications for future studies
This study cannot provide mechanistic insights of CXCR4 serum levels for the increased risk of rupture. As a sole biomarker CXCR4 may not be suitable, but it could still be used in explicit clinical applications, such as for instance CXCR4-labeled PET-CTs [29, 30] for the risk evaluation of AAA or in combination with other serum biomarkers in panel analysis. Quantitative analysis of histologic samples for CXCR4 positive areas, suggestive for a circulatory release of the receptor into the blood, comparing i.e. asymptomatic and symptomatic AAAs would be useful in the future. However, such effects need to be analyzed under the assumption of a broad heterogeneity of histologic appearance of AAA samples and might only be applicable to some subtypes of the disease [31].
Limitations
The study has a number of limitations, including its exploratory, single center, and retrospective nature on a small number of patients. Therefore, it is not possible to generalize the associations found and statistically significant results need to be considered with care. However, as these patients are rare, it is doubtful that enough patients could ever be recruited for larger scale studies.
Another limitation is that the control group used is small and not age-matched to the AAA study group.
It is also to mention, that CXCR4 plays a role in other biological processes such as e.g. immunologic and cancerogenic processes as already mentioned earlier [9, 10]. Cancer is known to occur more often in the elderly and in the current study cohort a significant difference between age in rAAA and iAAA could be observed. However, data on whether more patients in the rAAA group had occult cancer than in the iAAA group are not available.
Furthermore, CXCL12/CXCR4 axis plays a dual regulatory role in the initiation and development of diabetic kidney disease as well as chronic allogeneic nephropathy after kidney transplantation [32]. Thus, we cannot exclude also the role of kidney disorder in the increase of CXCR4 in rAAA patients.
Conclusions
CXCR4 is particularly localized in the tunica media of the aortic wall associated with SMCs and focally also in the adventitia in the area of inflammatory foci in both intact and ruptured AAA samples. The serum concentration in patients with rAAA compared to iAAA is significantly increased and associated with a more than 4-fold risk for rAAA. Additionally, serum CXCR4 concentrations correlate significantly with an increasing aortic diameter of AAA, suggesting its biomarker potential. However, further larger and particularly mechanistic studies are necessary to clarify its future value as a possible marker in AAA rupture risk prediction.
References
1 European Society for Vascular Surgery (ESVS) 2019 clinical practice guidelines on the management of abdominal aorto-iliac artery aneurysms. Eur J Vasc Endovasc Surg. 2019;57(1):8–93.
2 . Rupture rates of small abdominal aortic aneurysms: a systematic review of the literature. Eur J Vasc Endovasc Surg. 2011;41(1):2–10.
3 . Prospective and therapeutic screening value of non-coding RNA as biomarkers in cardiovascular disease. Ann Transl Med. 2016;4(12):236.
4 Abdominal aortic aneurysms. Nat Rev Dis Primers. 2018;4(1):34.
5 CXCR4 blockade induces atherosclerosis by affecting neutrophil function. J Mol Cell Cardiol. 2014;74:44–52.
6 . The CXCL12/CXCR4 chemokine ligand/receptor axis in cardiovascular disease. Front Physiol. 2014;5:212.
7 CXCR4: a potential marker for inflammatory activity in abdominal aortic aneurysm wall. Eur J Vasc Endovasc Surg. 2015;50(6):745–53.
8 Expression and cellular localization of CXCR4 and CXCL12 in human carotid atherosclerotic plaques. Thromb Haemost. 2018;118(01):195–206.
9 . CXCR4: a key receptor in the crosstalk between tumor cells and their microenvironment. Blood. 2006;107(5):1761–7.
10 . CXCL12 (SDF-1)/CXCR4 pathway in cancer. Clin Cancer Res. 2010;16(11):2927–31.
11 . Chemokine receptor expression on infiltrating lymphocytes from abdominal aortic aneurysms: Role of CXCR4–CXCL12 in lymphoid recruitment. Atherosclerosis. 2008;200(2):264–70.
12 . Chemokine (C-X-C Motif) Receptor 4 blockade by AMD3100 inhibits experimental abdominal aortic aneurysm expansion through anti-inflammatory effects. Arterioscler Thromb Vasc Biol. 2014;34(8):1747–55.
13 Matrix metalloproteinase-8 and -9 are increased at the site of abdominal aortic aneurysm rupture. Circulation. 2006;113:438–45.
14 Whole genome-expression profiling reveals a role for immune and inflammatory response in abdominal aortic aneurysm rupture. Eur J Vasc Endovasc Surg. 2009;37:305–10.
15 Biobanking: objectives, requirements, and future challenges – experiences from the Munich Vascular Biobank. J Clin Med. 2019;8(2):251.
16 Die Vaskuläre Biobank – hands on. Gefäßchirurgie. 2018;23:185–8.
17 Novel pathways in the pathobiology of human abdominal aortic aneurysms. Pathobiology. 2013;80(1):1–10.
18 . Bioinformatics analysis of common key genes and pathways of intracranial, abdominal, and thoracic aneurysms. BMC Cardiovasc Disord. 2021;21(1):14.
19 . Inflammation and cellular immune responses in abdominal aortic aneurysms. Arterioscler Thromb Vasc Biol. 2006;26(5):987–94.
20 . Cytokine pattern in aneurysmal and occlusive disease of the aorta. J Surg Res. 2001;101(2):152–6.
21 . Current concepts in the pathogenesis of abdominal aortic aneurysm. J Vasc Surg. 2003;38(3):584–8.
22 . Pathophysiology and epidemiology of abdominal aortic aneurysms. Nat Rev Cardiol. 2011;8(2):92–102.
23 Local CXCR4 upregulation in the injured arterial wall contributes to intimal hyperplasia. Stem Cells. 2016;34(11):2744–57.
24 The CXC chemokine SDF-1 is the ligand for LESTR/fusin and prevents infection by T-cell-line-adapted HIV-1. Nature. 1996;382(6594):833–5.
25 . A highly efficacious lymphocyte chemoattractant, stromal cell-derived factor 1 (SDF-1). J Exp Med. 1996;184(3):1101–9.
26 . Chemokines and their receptors in lymphocyte traffic and HIV infection. Adv Immunol. 2000;74:127 –80.
27 Flow cytometric analysis of peripheral blood dendritic cells in patients with severe sepsis. Cytometry B Clin Cytom. 2011;80(1):14–21.
28 . Stability of human stromal-derived factor-1α (CXCL12α) after blood sampling. Ann Clin Lab Sci. 2010;40(3):257–60.
29 . PET imaging of CXCR4 receptors in cancer by a new optimized ligand. ChemMedChem. 2011;6(10):1789–91.
30 [68Ga]Pentixafor-PET/CT for imaging of chemokine receptor CXCR4 expression in multiple myeloma - Comparison to [18F]FDG and laboratory values. Theranostics. 2017;7(1):205–12.
31 Aneurysm miRNA signature differs, depending on disease localization and morphology. Int J Mol Sci. 2016;17(1):81.
32 . The role of CXCL12 in kidney diseases: a friend or foe? Kidney Dis. 2021;7(3):176–85.
