Cancer Stem Cells in Moderately Differentiated Buccal Mucosal Squamous Cell Carcinoma Express Components of the Renin–Angiotensin System

Aim We have recently identified and characterized cancer stem cell (CSC) subpopulations within moderately differentiated buccal mucosal squamous cell carcinoma (MDBMSCC). We hypothesized that these CSCs express components of the renin–angiotensin system (RAS). Methods 3,3′-Diaminobenzidine (DAB) immunohistochemical (IHC) staining was performed on formalin-fixed paraffin-embedded MDBMSCC samples to investigate the expression of the components of the RAS: (pro)renin receptor (PRR), angiotensin converting enzyme (ACE), angiotensin II receptor 1 (ATIIR1), and angiotensin II receptor 2 (ATIIR2). NanoString mRNA gene expression analysis and Western Blotting (WB) were performed on snap-frozen MDBMSCC samples to confirm gene expression and translation of these transcripts, respectively. Double immunofluorescent (IF) IHC staining of these components of the RAS with the embryonic stem cell markers OCT4 or SALL4 was performed to demonstrate their localization in relation to the CSC subpopulations within MDBMSCC. Results DAB IHC staining demonstrated expression of PRR, ACE, ATIIR1, and ATIIR2 in MDBMSCC. IF IHC staining showed that PRR was expressed by the CSC subpopulations within the tumor nests, the peri-tumoral stroma, and the endothelium of the microvessels within the peri-tumoral stroma. ATIIR1 and ATIIR2 were localized to the CSC subpopulations within the tumor nests and the peri-tumoral stroma, while ACE was localized to the endothelium of the microvessels within the peri-tumoral stroma. WB and NanoString analyses confirmed protein expression and transcription activation of PRR, ACE, and ATIIR1, but not of ATIIR2, respectively. Conclusion Our novel findings of the presence and localization of PRR, ACE, ATIIR1, and potentially ATIIR2 to the CSC subpopulations within MDBMSCC suggest CSC as a therapeutic target by modulation of the RAS.

inTrODUcTiOn Oral cavity cancer is the sixth most common cancer globally (1,2) with more than 90% being squamous cell carcinoma (SCC) (2,3). Oral cavity SCC (OCSCC) arises from the squamous epithelium of the lips, oral tongue, floor of mouth, hard palate, buccal mucosa, maxillary and mandibular alveolus, and the retromolar trigone (1,4). Predisposing factors for BMSCC include tobacco use, alcohol abuse, and betel leaf chewing (5). BMSCC is most prevalent in South East Asia and Southern Asia due to the habitual use of betel quid and betel leaf chewing (2,4), more commonly in men (2). Current treatment for BMSCC involves surgery, often with postoperative radiotherapy (RT), and sometimes chemotherapy (ChT) (2). Despite advances in treatment, 5-year survival for BMSCC remains 50-58%, and the overall survival has only increased by 5% in the past 20 years (1,4,6). This poor prognosis is partly due to late presentation with advanced loco-regional disease and less commonly metastasis to the bone, brain, or liver (5).
Cancer stem cells (CSCs) have been demonstrated in many types of cancers and have been proposed to be the origin of these cancers, including BMSCC (7). CSCs are suggested to play a crucial role in carcinogenesis with their ability for self-renewal and differentiation into multiple lineages through symmetric and asymmetric division, giving rise to diverse cell populations (7).
Although the origin of CSCs remains unclear (8), they are distinguished from the majority of the cancer cell population by their expression of CSC markers (9). Overexpression of CSC markers has been associated with increased tumor size, local invasion, and metastasis (10,11). Increased expression of CSC markers has also been associated with worse prognosis (12)(13)(14), treatment resistance, and higher risk of loco-regional recurrence and distant metastasis following RT and ChT (15).
We have previously demonstrated the role of stem cells in the biology of infantile hemangioma (IH) putatively regulated by the RAS (23,24). This, coupled with recent publications, suggesting a role for the RAS in cancer growth (18), with components of the RAS: ACE, ATIIR1, and ATIIR2, being demonstrated in areas of cancer (25), led to the notion of RAS playing a role in tumor angiogenesis and tumor cell proliferation (25), both being determinants of tumor growth and metastasis. This suggests the RAS as a potential therapeutic target for cancer (18). Despite the proposed role of the RAS in carcinogenesis (18), there are currently no publications showing the presence of the RAS in BMSCC.
In this study, we investigated the expression of the components of the RAS: PRR, ACE, ATIIR1, and ATIIR2 within MDBMSCC using immunohistochemical (IHC) staining, Western Blotting (WB), and NanoString gene expression analysis. We also investigated the localization of these proteins in relation to the CSC subpopulations we have identified within this tumor (16).

Tissue samples
Moderately differentiated buccal mucosal squamous cell carcinoma (MDBMSCC) specimens from one female and five male patients, aged 38-80 years (mean 59.3 years), sourced from the Gillies McIndoe Research Institute's Tissue Bank, were used for this study, which was approved by the Central Regional Health and Disability Ethics Committee (ref. no. 12/CEN/74).

histochemical and immunohistochemical staining
Hematoxylin and eosin (H&E) staining was used to confirm the appropriate histological grade and to identify areas of MDBMSCC within the tissue sections by an anatomical pathologist (HDB). To determine co-expression of the proteins, immunofluorescent (IF) IHC staining was performed on two samples of MDBMSCC from the original cohort of six patients used for DAB IHC staining, utilizing a combination of Vectafluor Excel antimouse 488 (ready-to-use; cat# VEDK2488, Vector Laboratories, Burlingame, CA, USA) and Alexa Fluor anti-rabbit 594 (1:500; cat# A21207, Life Technologies, Carlsbad, CA, USA) to detect combinations that included OCT4 and SALL4 with PRR, ACE, ATIIR1, and ATIIR2. All IF IHC-stained slides were mounted in Vecta Shield Hardset mounting medium with 4′,6′-diamidino-2-phenylindole (Vector Laboratories).
Positive human control tissues used for primary antibodies were placenta for PRR; liver for ACE, ATIIR1, and ATIIR2; and seminoma for OCT4 and SALL4. A negative MDBMSCC control sample was also prepared by omitting the primary antibodies.
All antibodies were diluted with Bond primary antibody diluent (cat# AR9352, Leica), and all DAB and IF IHC staining was carried out on the Leica Bond Rx autostainer as previously described (26).

nanostring gene analysis
Five snap-frozen MDBMSCC samples from the original cohort of patients used for DAB IHC staining were utilized for isolation of total RNA for NanoString nCounter™ Gene Expression Assay (NanoString Technologies, Seattle, WA, USA). Extraction of RNA from tissues was performed using RNeasy Mini Kit (Qiagen, Hilden, Germany). RNA was quantitated by Qubit ® 2.0 Fluorometer (Life Technologies) and subjected to NanoString nCounter™ gene expression assay completed by New Zealand Genomics Ltd (Dunedin, NZ), according to the manufacturer's protocol. Probes for the genes encoding for PRR (ATP6AP2, NM_005765.2), ACE (NM_000789.2), ATIIR1 (AGTR1, NM_000685.3), ATIIR2 (AGTR2, NM_000686.3), and the housekeeping gene GUSB (NM_000181.1) were designed and synthesized by NanoString Technologies. Raw data were analyzed by nSolver™ software (NanoString Technologies) using standard settings. Results were normalized against the housekeeping gene and graphed using Excel (Microsoft Office 2013).

Western Blotting
Total protein was extracted from five MDBMSCC specimens by homogenization in ice-cold RIPA buffer (Sigma-Aldrich, St Louis, MA, USA) containing 10mM dithiothreitol (DTT) (Sigma-Aldrich) and 1× HALT protease and phosphatase inhibitor cocktail (Thermo Fisher Scientific). Solubilized proteins were precipitated for 1 h at −20°C (ProteoExtract ® Protein Precipitation Kit, Merck Millipore, Billerica, MA, USA) and then re-suspended overnight in 1× Laemmli sample buffer (Bio-Rad, Hercules, CA, USA) containing 10mM DTT. Equal amounts of protein extracts were heated in sample buffer and then separated by Bolt™ 4-12% Bis-Tris Plus gel (cat# NW04120, Invitrogen, Carlsbad, CA, USA) electrophoresis. Separated proteins were then transferred to PVDF membrane (cat# IB24001, Life Technologies), which were then blocked in TBS containing 0.1% Tween-20 and 2% skim-milk powder for 90 min at 4°C. The membranes were subsequently probed using the following primary antibodies: PRR  2000; cat# A21239, Invitrogen). All primary and secondary antibodies were diluted in TBS containing 0.1% Tween-20 and 2% skim-milk powder. Detection of the HRP-conjugated secondary antibodies was achieved using Clarity™ Western ECL substrate (Bio-Rad). All membranes were imaged using the ChemiDoc MP imaging system (Bio-Rad).
image analysis 3,3′-Diaminobenzidine IHC-and IF IHC-stained slides were viewed and imaged using an Olympus BX53 light microscope (Tokyo, Japan) and an Olympus FV1200 confocal laser-scanning microscope (Olympus), respectively. All IF IHC-stained images were processed with CellSens Dimension 1.11 software using the 2D deconvolution algorithm (Olympus).

resUlTs histochemical and 3,3′-Diaminobenzidine immunohistochemical staining
Hematoxylin and eosin staining confirmed the histological grade and the presence of MDBMSCC (n = 6) for each sample. DAB IHC staining showed cytoplasmic expression of PRR (Figure 1A, brown) localized predominantly to cells within the tumor nests ( Figure 1A, brown, short arrows). There was also faint immunoreactivity in the endothelium of the microvessels (Figure 1A, brown, arrowheads) and stronger cytoplasmic staining in cells within the peri-tumoral stroma (Figure 1A, brown, long arrows). ACE (Figure 1B, brown) was expressed on the endothelium of the microvessels, which were predominantly located around the periphery of the peri-tumoral stroma. ATIIR1 (Figure 1C, brown) was expressed predominantly in the cytoplasm and nuclei of cells within the tumor nests ( Figure 1C, brown, short arrows) with faint cytoplasmic staining in cells within the peri-tumoral stroma (Figure 1C, brown, long arrows). Faint cytoplasmic and stronger nuclear staining of ATIIR2 (Figure 1D, brown) was observed in cells within the tumor nests ( Figure 1D, brown, short arrows) and some cells within the peri-tumoral stroma (Figure 1D, brown, long arrows).
Positive controls for PRR, ACE, ATIIR1, and ATIIR2 demonstrated the expected staining patterns in human placenta (Image S1A in Supplementary Material, brown), liver (Image S1B in Supplementary Material, brown), liver (Image S1C in Supplementary Material, brown), and kidney (Image S1D in Supplementary Material, brown), respectively. The omission of the primary antibody in MDBMSCC samples provided an appropriate negative control (Image S1E in Supplementary Material, brown).

Western Blotting
Western Blot analysis confirmed the presence of PRR ( Figure 3A) and ATIIR1 (Figure 3B) within the extracts of all five MDBMSCC samples from the original cohort of patients used for DAB IHC staining, at their expected molecular weight of 39 and 43 kDa, respectively. ACE (Figure 3C) was detected in only one of the five MDMBSCC samples analyzed, at the expected molecular weight of 195 kDa. High image exposure was required to clearly visualize the ACE signal in the MDMBSCC sample, which implies that ACE was present at very low abundance. ATIIR2 was detected in all five MDBMSCC samples at approximately 51 kDa (Figure 3D), which is larger than the native form of the protein and may signify detection of a glycosylated isoform. A band at approximately 41 kDa was detected in one of the five total protein extracts ( Figure 3D, BMSCC_3) that may correspond to the native protein (27). Detection of β-actin confirmed approximately equal total protein loading across all five samples ( Figure 3E).

nanostring gene analysis
NanoString gene analysis of the components of the RAS: PRR, ACE, ATIIR1, and ATIIR2, was performed in five samples of MDBMSCC from the original cohort of six patients used for DAB IHC staining, normalized against the housekeeping gene, GUSB. Transcriptional profiling confirmed the presence of PRR and ACE mRNA in all five samples and ATIIR1 in one sample, but ATIIR2 mRNA was below the detectable level within all five samples (Figure 4).

DiscUssiOn
The novel finding of the expression pattern of PRR, ATIIR1, ATIIR2, and ACE on CSC subpopulations within MDBMSCC provides insights into the biology of this aggressive tumor. It is noteworthy that the CSC subpopulation within the peri-tumoral  Expression is depicted in relative units (RU) as a ratio to the GUSB housekeeper. Transcriptional profiling confirmed the presence of PRR and ACE mRNA in all five samples and ATIIR1 in one sample, but ATIIR2 mRNA was below the detectable level within all five samples. stroma does not express EMA. This CSC subpopulation may represent a "normal" stem cell population or cells that have undergone an epithelial to mesenchymal transition (EMT) (28), with the endothelial population possibly reflecting the phenomenon of vascular mimicry (29).
It is intriguing that although IHC staining demonstrated the expression of both ATIIR1 and ATIIR2, supported by WB detection of bands at the expected size, mRNA expression for ATIIR2 was not detected. This may be due to rapid degradation of mRNA of this protein.
Recent literature suggests that the RAS may play a role in cancer growth and metastasis, especially in promoting cell proliferation and angiogenesis (25,30), with a review highlighting a role for the RAS in other cancers, predominantly in cellular proliferation (28). We have demonstrated an interesting expression pattern, within MDBMSCC; the components of the RAS on the CSC subpopulations within the tumor nests, peritumoral stroma, and the endothelium of the microvessels within the peri-tumoral stroma.
The expression of the ESC marker SOX2 on the endothelium adjacent to the tumor nests is consistent with our recent publication demonstrating expression of the stem cell markers SALL4 and OCT4 (16). These findings and the observation of the expression of PRR and ACE on the endothelium is intriguing and may reflect a primitive phenotypic endothelium similar to that reported in colorectal cancer (29), although this remains a topic for future investigation.
While the analysis of normal buccal mucosa would strengthen the interpretation of our data, these novel findings of the localization of the components of the RAS to the CSC subpopulations within MDBMSCC mirror our similar finding within OTSCC (17,31). Although we have not demonstrated the functionality of these receptors in this report, it is exciting to speculate CSCs as a potential novel therapeutic target by modulation of the RAS using existing anti-hypertensive drugs, such as aliskiren, which targets renin; β-blockers, which block the production of (pro) renin and so decrease levels of renin; ACE inhibitors and ATII receptor blockers (32)(33)(34)(35).
The relatively small sample numbers and the lack of functional data for these receptors in this report are limitations of the study and demonstrate the need for further work to better understand the precise regulatory function of the RAS on the CSC subpopulations, which may lead to an effective treatment for BMSCC and other cancers. eThics aPPrOVal Central Regional Health and Disability Ethics Committee (ref. no. 12/CEN/74).