Edited by: Kuzhuvelil B. Harikumar, Rajiv Gandhi Centre for Biotechnology, India
Reviewed by: Margot Zoeller, Heidelberg University, Germany; Yun Dai, Virginia Commonwealth University, United States
This article was submitted to Cancer Molecular Targets and Therapeutics, a section of the journal Frontiers in Oncology
†These authors have contributed equally to this work
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
Recurrence and metastasis are the primary causes of mortality in patients with colorectal cancer (CRC), and therefore effective tools to reduce morbidity and mortality of CRC patients are necessary. LOX-1, the ox-LDL receptor, is strongly involved in inflammation, obesity, and atherosclerosis, and several studies have assessed its role in the carcinogenesis process linking ROS, metabolic disorders and cancer. We have already demonstrated
Colorectal cancer (CRC) is one of the most prevalent and widely studied cancers in the world, conferring significant morbidity, mortality, and cost to the public health system (
LOX-1 has been first identified as a major receptor of oxidized-LDL (ox-LDL) and it is mainly expressed in endothelial cells and vascular-rich organs, playing a crucial role in cardiovascular diseases (
Most relevantly, a positive correlation has been reported among LOX-1 expression and the carcinogenesis process (
Epidemiological studies reveal that those individuals showing high levels of circulating ox-LDL, in addition to atherosclerotic plaques, are more susceptible to CRC (
The role played by LOX-1 in tumorigenesis also takes place through angiogenesis, which represents an early to mid-stage event in many human cancers. In particular LOX-1 upregulation increases VEGF-A165 expression (
In addition,
The known relationship between cancer and volatile compounds (VOCs) has motivated the study of tumor progression through the changes of VOC profile. The volatile portion of metabolism products, sometimes called volatilome, is supposed to reflect the metabolic changes due to a plethora of diverse factors, such as inflammation, necrosis, cancer degeneration, or alteration of microbiota. The relationship between volatilome composition and tumors has been observed in many experiments, both
In the last few years, VOC profile analysis for colorectal cancer diagnosis has been investigated, leading to promising results (
Recently, our research has shed light on the role of the LOX-1 receptor as a novel biomarker and molecular target to improve current therapeutic strategies of CRC (
In this paper, we explore
3.5 × 106 of DLD-1 cells (ATCC: CCL-221TM) in which LOX-1 has been knocked down by RNAi (LOX-1RNAi) and a scramble (scrambleRNAi), diluted in 150 μl of PBS, have been transplanted subcutaneously into the right flank of CD-1 male nude mice aged 5 weeks (
All animals injected subcutaneously and intravenously were sacrificed after 4 and 11 weeks, respectively, and tumor masses weighted. Further information is in
Morphometric evaluation of total tumoral area (mm2), necrosis area (mm2), percentage of necrosis, and residual tumoral area (mm2) has been performed on images captured by ACT-1 software (Nikon, Japan) connected to a Nikon microscope (EclipseE600), and analyzed by Scion Image software (Scion Corporation, Frederick, MA). Blinded evaluation of capillary density (vessels per mm2) at 40× magnification was performed on 10 independent fields/animals by three observers in 20 serial sections, with intervariability <5%. Necrotic and hemorrhagic areas have been excluded from the vessel count.
At autopsy, the left and right inguinal lymph nodes, axillary lymph nodes, liver, lung, and spleen were removed. All the serial sections (20 sections/organ) of 5 μm stained with H&E have been used for histopathological evaluation of the presence of metastasis and micro-metastasis and evaluated histologically by light microscopy. Any number of tumor cells present in the lymph node or in organ was considered as evidence of metastatic spread.
Serial four micron sections of xenograft tumors have been immunostained for Ki67 (MoAb, Dako, Denmark), VEGF-A165 (A-20, Santa Cruz biotechnology, U.S.), Acetyl Histone H4 (Upstate, NY), α-SMA (Dako, Denmark), CD31 (ab124432, Abcam), β-catenin (sc-7963, Santa Cruz Biotechnology, Inc.), and HIF-1α (MA1-16504, Affinity Bioreagents) following the streptavidin-biotin method, as previously described (
Mice were kept for 30 min without food and then moved in the experiment room and placed in a polypropylene box (KIS T box XXS- ABM Italia S.p.A.). The box was large enough to allow for normal physical activity. The lid of the box has been endowed with an inlet suitable for the insertion of either the Solid Phase MicroExtraction (SPME) sampler or the gas sensor array sampling tube. The total air released by the mice inside the measurement cage was collected on a SPME and then analyzed with the Gas Chromatography Mass Spectroscopy (GC/MS). The SPME fiber was a 50/30 μmDivinylbenzene/Carboxen/PDMS (SUPELCO, Bellefonte, PA, USA). In all measurements, the fiber was kept in the sampling box for 1 h, and analyzed over 3 h after their collection.
To standardize the headspace formation, the VOCs collection always began 15 min after the mouse entered the cage. The temperature and humidity of the experiment room was kept constant during the whole experiment.
The box floor was coated, before each measurement, with filter paper sheets (Biosigma srl) to collect solid and liquid dejection.
The GC/MS is a Shimadzu GCMS-QP2010 (Kyoto, Japan) equipped with a capillary column EQUITY-5 poly (5% diphenyl/95% dimethyl siloxane) (SUPELCO, Bellefonte, PA, USA). The size of the column is 30 m length × 0.25 mm I.D. × 0.25 μm thickness. The VOCs were desorbed from the SPME in splitless injection mode at 250°C for 3 min in the GC injection port. Compounds have been putatively identified using both NIST 127 and NIST 147 libraries. The identity of the compounds found in more than 80% of samples were confirmed comparing the mass spectra with those of pure standard obtained from Sigma Aldrich (
The gas sensor array was an ensemble of twelve quartz microbalances (QMB). These sensors detected the mass change (Δm) in absorbing layer deposited onto the surface of the quartz. The sensor signal is the change of the frequency (Δf) of the electric output signal of an oscillator circuit driven by the quartz. In the regime of small perturbations, Δm and Δf are linearly proportional. The adopted QMBs have a fundamental frequency of 20 MHz, corresponding to a mass resolution of the order of a few nanograms. The sensing materials are solid-state layers of porphyrinoids (porphyrins and corroles) (
In a previous study, after quantifying LOX-1 expression among several CRC cell lines, we focused our attention on two cell lines overexpressing LOX-1, HCT8 (non-metastatic), and DLD-1 (metastatic), in which we stably down modulated LOX-1 expression by RNAi (
Here in this study we decided to use DLD-1 cells, a colon cancer metastatic cell line, stably knocked down for LOX-1 (LOX-1RNAi DLD-1) as a model of colon cancer insurgence and spreading in nude mice. Before starting, the downregulation of LOX-1 mRNA expression was again confirmed by RT-qPCR (
Mice have been subcutaneously injected with LOX-1RNAi DLD-1 and scrambleRNAi DLD-1 cells in the right flank, as described in Materials and Methods. In parallel, a group of animals were injected with a saline solution. Tumor volume was monitored once a week for 4 weeks and measured by a caliper. Mice body weight was also measured once a week. For both, no significant differences were reported among the two groups of mice (
Even if the tumor weight and volume did not differ significatively between two mice groups (
Morphometric analysis of tumor area in LOX-1RNAi and scrambleRNAi mice evaluated on H.E. sections of the xenograft tumors.
10 | 58.89 ± 10.3 | 25.79 ± 4.6 | 42.58 ± 2.40 | 34.10 ± 6.1 | |
10 | 81.35 ± 17.24 | 27.06 ± 6.8 | 32.84 ± 2.35 | 54.29 ± 10.9 |
Blood vessels maturation is a crucial factor for colorectal cancer growth. Therefore, with the aim of assessing the role played by LOX-1 in neo-angiogenesis, histopathological and immunohistochemical analyses were performed for α-SMA, CD31 and VEGF-A165 expression on the tumoral mass and surrounding tissues of mice subcutaneously injected. The number of vessels has been identified by evaluating α-SMA expression in intratumoral and peritumoral tissues of scrambleRNAi and LOX-1RNAi xenograft tumors (
Representative images of the immunohistochemical analysis of α-SMA, CD31, and VEGF-A165 expression in xenograft tumoral tissues of mice injected subcutaneously. Neo vessels evidenced by α-SMA immunohistochemistry in scrambleRNAi
High number of neo vessels were present in scrambleRNAi xenograft tumors, and the aspect was sometimes discontinuous, especially in small caliber vessels. On the contrary, a very few number of vessels with a small caliber were evidenced in LOX-1RNAi xenograft tumors (
Finally, we have analyzed the expression of VEGF-A165 by immunohistochemistry performed on the same xenograft tumor tissues. The immunostaining of VEGF-A165 was observed in the cytoplasm of tumor cells, as well as in endothelial and in some stromal cells in scrambleRNAi DLD-1 xenograft tissues (
In order to understand the role of LOX-1 in promoting tumor growth and angiogenesis, we examined the expression of β-catenin and HIF-1α in the same tissues.
The immunohistochemical analysis of β-catenin is represented in
Representative images of the immunohistochemical analysis of β-catenin and HIF-1α expression. β-catenin expression in scrambleRNAi
In order to define the role of LOX-1 in proliferation and gene transcription modulation, we have performed IHC for Ki67 protein, a proliferative nuclear marker and histone H4 acetylation on xenograft tumor tissues.
The immunohistochemistry analysis for Ki67 (
Representative images of the immunohistochemical analysis of Ki 67 and acetyl histone H4 expression. Ki 67 expression in scrambleRNAi
The total volatile compounds released by mice were collected and analyzed with the Gas chromatograph/Mass spectrometer (GC/MS).
In total, 79 different VOCs were found in the collected samples (
The comparison of the abundances of VOCs in the different samples was limited to the recurrent compounds; these are long chain aldehydes and methylated hydrocarbons. It is important to note that the selection of the recurrent VOCs was performed avoiding compounds found exclusively in one group of animals.
For a second time, DLD-1 cells, as well as saline solution, was injected intravenously via the tail vein in a nude mice cohort (
Representative images of murine organs evaluated on H&E sections. Metastasis foci have been evidenced in liver, lung and lymph nodes in mice endovenously injected with scrambleRNAi DLD-1 cells. Metastatic repetitions with mucinous aspect were found in all liver of scrambleRNAi mice (*). Scale bar 20 μm. Quantitative analysis of metastasis number
Histopathological observations on tumor tissues evidenced the presence of multiple metastasis foci in lung and liver of all the mice injected with scrambleRNAi DLD-1 cells, and in three mice out of eight metastasis were evidenced also in lymph nodes (
Morphometric analysis of metastatic spread of LOX-1RNAi and scrambleRNAi DLD-1 cells in murine organs evaluated on H&E sections.
ScrambleRNAi | MO13240 | 12 | 2.4 | Lung, 2 lymph nodes | ||
ScrambleRNAi | MO13241 | 22 | 6.6 | Lung, liver | ||
ScrambleRNAi | MO13242 | 5 | 1.3 | Lung, liver | ||
ScrambleRNAi | MO13243 | 9 | 12 ± 3.2 | 1.8 | 3 ± 0.72 | 1 lymph node |
ScrambleRNAi | MO13244 | 8 | 1.5 | Lung, liver | ||
ScrambleRNAi | MO13245 | 7 | 1.8 | Lung, 1 lymph node | ||
ScrambleRNAi | MO13246 | 21 | 5.8 | Liver | ||
ScrambleRNAi | MO13247 | 13 | 2.7 | Lung, Liver | ||
LOX-1RNAi | MO13248 | 0 | 0 | |||
LOX-1RNAi | MO13249 | 0 | 0 | |||
LOX-1RNAi | MO13250 | 0 | 0 | |||
LOX-1RNAi | MO13251 | 0 | 0.37 ± 0.37 | 0 | 0.21 ± 0.21 | |
LOX-1RNAi | MO13252 | 3 | 1.7 | Lung, 2 lymph nodes | ||
LOX-1RNAi | MO13253 | 0 | 0 | |||
LOX-1RNAi | MO13254 | 0 | 0 | |||
LOX-1RNAi | MO13255 | 0 | 0 |
The total volatilome released by mice that underwent the intravenous injection was measured the day after the inoculation, and then after 4, 6, 9, and 11 weeks.
In total, 67 VOCs were found in the mice cage air (
Two of these compounds (octane-4-methyl and nonanal) are recurrent in both the experiments. These compounds are also released by saline injected mice. However, while in this group the abundances are uncorrelated (
Furthermore, while the abundance of nonanal is independent of both the injection method and injected cells, octane-4-methyl is produced in excess by scrambleRNAi mice, endovenously injected (
Plot of the abundance of nonanal vs. the abundance of octane-4-methyl found in mice volatilome. In each group the data collected along the whole experiments are plotted. While the abundance of nonanal is a specific of experimental condition, the abundance of octane-4-methyl is larger in mice endovenously injected with scrambleRNAi cells.
Gas Sensors were applied from the fourth week after the injection.
Principal component analysis of sensors data collected from endovenously injected mice.
A better appraisal of the differences between groups is obtained when PCA is calculated with the data collected in the same measurement session.
The results of the sensor array are in good agreement with the GC/MS data, but with some differences. In particular, at week 4 sensors fail to capture the differences between scrambleRNAi and LOX-1RNAi cell injected mice. This behavior suggests that the set of VOCs evidenced by GC/MS may not completely represent the volatile emissions from the mice. On the other hand, there might be other VOCs, which are not captured by the combination of the chosen SPME and GC/MS but are sensed by the gas sensors.
Invasion and metastasis are the deadly face of malignant tumors. Surgery represents the mainstay of treatment in early cases of CRC, but often patients are primarily diagnosed in an advanced stage of disease and sometimes also distant metastases are present. Five year survival rates are estimated to be between 85 and 90% for patients with localized cancer to colon or rectum. Survival decreases significantly for patients with distant metastasis, with 5 year survival of only 12.5%. Recurrence and metastasis are in fact the primary cause of mortality in patients with CRC. Neoadjuvant therapy is therefore needed. Recent therapeutic approaches that add epidermal growth factor receptor (EGFR) and VEGF targeted agents to standard chemotherapy have produced a prolonged overall survival (OS) of up to 30 months in patients with metastatic disease.
Within this context, a better understanding of the molecular mechanisms and biomarkers involved in early onset and in the metastatic process of CRC could play an important role to further improve the outcome of advanced CRC stage and metastatic CRC patients.
LOX-1, a marker for atherosclerosis (
In particular, LOX-1 is overexpressed in the presence of inflammatory stimulus by PCSK9, an enzyme involved in cholesterol homeostasis. In fact, in hypercholesterolemic states it is possible to reduce atherogenesis by silencing LOX-1 via PCSK inhibitor (
Also, our previous studies on human tissues have evidenced a strong up-regulation of LOX-1 in human colon cancer tissues (
In the present work we evaluate the contribution of LOX-1 in tumor development, in metastasis formation and organ colonization
Here we also report evidence of the pro oncogenic role of LOX-1 in supporting tumor growth, regulating the β-catenin expression. β-catenin is the most important regulator of proliferation, cell-cell adhesion, migration and epithelial-mesenchymal transition (EMT). The accumulation of β-catenin and the alteration of its degradation leads to colon cancer insurgence (
Furthermore, concomitantly to a reduced proliferative rate, knocking down LOX-1, we have also detected a modulation of histone H4 acetylation in LOX-1RNAi tumor tissues, suggesting that a metabolic factor such as LOX-1 could modulate DNA acetylation and thus drive the gene expression pattern in tumors. In fact, histone H4 acetylation is one of the major regulation mechanisms of DNA transcription and determines chromatin accessibility with effects on gene expression. Some of the anticancer therapies act as HDAC inhibitors, the enzyme responsible for DNA deacetylation; this highlights the fundamental role of acetylation in tumor progression.
The impact of these findings is further strengthened by the observations that we made in a cohort of nude mice in which LOX-1RNAi cells were intravenously injected. In these mice, we observed a significant inhibition of metastasis formation. Only one of these mice showed one metastasis in lung and two in lymph nodes. The small number of mice does not allow for a reliable statistical generalization of the results, however, the striking difference in metastasis development is in favor of the hypothesis of the role of LOX-1 in tumor progression. Importantly, all mice injected with scrambleRNAi cells showed metastasis in liver with aspects of mucinous cancers, the most aggressive colon cancer form. In fact, patients with mucinous colon cancer often have a poor prognosis and these tumors are prone to form metastasis, because mucin-associated carbohydrate plays a role in adhesive inter-structures that facilitate the survival of the metastatic clone (
The analysis of volatile compounds provides further evidence of the influence of LOX-1 in the progression of the tumor. Subcutaneously and endovenously injected mice have demonstrated to evolve differently, and this difference associated with the injection protocol has been found in the VOCs profile. The analysis was confined to the most recursive VOCs found in at least 80% of samples. In this way, episodical phenomena and random variations have been ruled out.
GC-MS analysis shows that mice injected with scrambleRNAi cells are characterized by the largest emission of volatile compounds. This behavior is particularly evident in endovenously injected mice while only a limited correlation has been found in mice treated with subcutaneous injection. For both the injection methods, all animals show the same kind of VOCs. Thus, the injection of DLD1 cells do not produce novel additional VOCs but a modulation of the abundance of the volatile metabolites that are usually emitted by mice.
The chemistry of the VOCs in the two experiments (i.e., subcutaneous and endovenous injection) is different and only two VOCs are common to both the experiments: octane-4-methyl and nonanal. The association of these compounds with cancer was found in previous studies. Octane-4-methyl was found at altered concentrations in the breath of patients affected by colorectal cancer (
In the intravenous experiment, GC-MS has been complemented by a gas sensor array that has been previously used for volatilome analysis in
At the current state of the art, the analysis of VOCs is mainly empirical, and the direct relationship between each VOC and tumor evolution mechanisms is still unclear. However, in this case the VOCs emission clearly correlates with the tumor evolution, providing a viable method for a simple, non-invasive alternative monitoring of tumor progression. Further studies are expected in this field to elucidate the intimate connection between VOCs and physiological processes.
Altogether, these results confirm the hypothesis that LOX-1 is a regulator of tumor progression, migration, invasion, metastasis formation, and tumor-related neo-angiogenesis, through the combination of specific molecular pathways. Further studies are necessary to deeper investigate the complex cancerogenesis mechanisms in which LOX-1 is actively involved.
All datasets generated for this study are included in the manuscript/
This study was carried out in accordance with the ethical standards (Declaration of Helsinki), in compliance with Tor Vergata animal care guidelines and following national and international directives to minimize animal suffering (Italian Legislative Decree 26/2014, Directive 2010/62/EU of the European Parliament and of the Council). The protocol was approved by the Italian Ministry of Health (protocol no. 954/2016-PR).
MM, RCa, CP, RCi, and SP conducted the experiments and wrote the manuscript. AC and RP prepared the electronic nose. EM analyzed data. AO, RM, and GN wrote the manuscript. CD and FS analyzed data, conceived the work, and wrote the manuscript.
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Authors thanks Graziano Bonelli for expert technical help.
The Supplementary Material for this article can be found online at:
volatile compound
colorectal cancer
oxidized-LDL
Lectin-like oxidized LDL receptor-1
vascular endothelial growth factor
gas chromatography mass spectrometer
principal component analysis
Hematoxylin and Eosin
solid phase micro-extraction
quartz microbalances
standard error mean.