Aldehyde Dehydrogenase: An Off-Label Marker of Endothelial Activation in Oral Squamous Cell Carcinoma

Introduction: The vascular endothelial (VE) expression of aldehyde dehydrogenase (ALDH) 1/2 family in oral leukoplakia (OL) and oral squamous cell carcinoma (OSCC) cases has not been studied so far. The aim of this study was to illustrate the “off-label” endothelial expression of cancer stem cell (CSC) biomarker, ALDH1/2, adjacent to oral potentially malignant and malignant lesions in order to shed some light on the mechanisms leading to oral carcinogenesis. Materials and methods: The expression of CSC protein-biomarker ALDH1/2 was detected through immunohistochemistry (IHC) in 30 paraffin-embedded samples of OL and 21 samples of OSCC compared to five samples of normal oral mucosa. Statistical analysis was done using SPSS, Pearson Chi-square, and Fischer’s exact test. The significance level was set at 0.05 (p≤ 0.05). Results: In oral mucosal vessels, ALDH1/2 was not expressed. It was expressed significantly more in the vessels of OSCCs compared to the OLs (Fisher’s exact test, p-value= 0,001). However, when endothelial expression of ALDH1/2 in the vasculature of OLs was compared with that of the normal oral mucosa, no significant change was noticed (Fisher’s exact test, p-value=1.000). Discussion: The IHC VE expression of ALDH1/2 in OSCC vasculature but not in OL indicates a possible significantly stronger activation of endothelial cells during carcinogenesis, which could be an indicator of the role of inflammation in the development of field cancerization and of prognostic value for (vascular/lymphatic) metastasis.


Introduction
The term "oral potentially malignant disorder" (OPMD) is attributed to oral mucosal disorders/lesions that exhibit an increased risk for malignant transformation compared to healthy mucosa [1]. The most common OPMD is oral leukoplakia (OL) [2]. Malignant conversion of OL results in oral squamous cell carcinoma (OSCC), arising from cells of the stratified squamous epithelium. Its biological behavior and other clinical and microscopic/molecular parameters affect therapeutic procedures and prognosis [3].
Oral carcinogenesis, according to the cancer stem cell (CSC) theory, is caused by a particular population of CSCs. CSCs are at the top of the cancer cell hierarchy [4]. CSC biomarkers are used to detect subpopulations of cancer cells with "stemness" (characteristics of stem cells) and are subdivided into markers of embryonic stem cells and markers showing the presence of stem cell traits in certain cells (stemness). Aldehyde dehydrogenases (ALDH) are a group of enzymes that catalyze the oxidation of aldehydes and function as stemness markers [5]. There is expression of ALDH in vascular endothelial (VE) cells in the tumor blood vessels of in vivo mouse models of oral carcinoma while there is no expression in normal blood vessels [6]. Angiogenesis is crucial for the process of tumorigenesis and metastasis [7]. The most important angiogenic factors are the three peptide families of VE growth factor (VEGF), fibroblast growth factor (FGF), and platelet-derived growth factor (PDGF) [7].
The aim of this study is to illustrate the CSC biomarker ALDH1/2 expression in the VE cells adjacent to potentially malignant and malignant lesions.  For the IHC technique application, the incisions were mounted on slides. The material was processed using anti-ALDH1/2 antibody (sc-166362) at a dilution of 1:100 using the Dako EnVision FLEX+ Visualization Systems (Agilent Technologies, Inc., Santa Clara, California, United States). Specifically, the staining process included antigen recovery, the application of primary antibody, the application of the EnVision™ FLEX DAB+ Substrate Chromogen System (Agilent Technologies, Inc.), the chromogenic agent application (Dako Dab Envision) (Chromogen), and finally the application of hematoxylin.

Materials And Methods
The incision was then affixed to the mounting plate and coated to protect and preserve the preparation over time. The evaluation of IHC staining was performed by microscopically examining the incisions in order to observe and record the results. Two observers (DA and VZ) evaluated the staining. The process was blinded and during the evaluation, only a five-digit code, attributed to each tissue sample, was known to the observers. The vascular staining was evaluated either as positive or negative. The IHC staining was evaluated as positive when the cytoplasm and/or the membrane of at least one endothelial cell was depicted in brown in an area of the same size in all samples.

Results
Despite the IHC expression of ALDH1/2 in the normal epithelium (the pattern of positive epithelial expression includes the cytoplasmic and membranous staining of individual spindle cells and of the basal cell layer in contrast to the negative endothelial cells underlying the epithelium), the endothelial staining of ALDH was negative in the vessels of the oral mucosa ( Figure 1A). The endothelial staining was also negative in cases of mildly dysplastic and non-dysplastic OL. In contrast, in five moderately and severely dysplastic OL samples (out of 30 OL samples), ALDH1/2 was strongly expressed in the endothelial cells of vessels adjacent to the superficial dysplastic epithelium (the pattern of positive epithelial expression in moderately and severely dysplastic OL samples includes the cytoplasmic and membranous staining of two-thirds of the epithelium whereas the pattern of positive VE expression includes the cytoplasmic and membranous staining of the endothelial cells, underlying the OL lesion) (Figure 1 B-D). However, no statistically significant difference was noticed compared to the normal oral mucosa (Fisher's exact test, p-value= 1.000). Interestingly, this partial positivity in worse cases of OL was dramatically increased in endothelial cells of 13 out of 21 OSCC samples, irrespective of the grade of differentiation but with statistical significance compared to OL (Fisher's exact test, p-value= 0,001) (the pattern of positive epithelial expression in OSCC cases includes the cytoplasmic and membranous staining of more than twothirds of the epithelium whereas the pattern of positive VE expression includes the cytoplasmic and membranous staining of the endothelial cells adjacent to the cancerous foci and the overlying dysplastic epithelium) (Figure 1 E-F). The positivity was also statistically significantly higher in OSCC than in normal cases (Fisher's exact test, p-value=0,039) ( Table 2).  The aforementioned findings imply that the expression of ALDH is detected in higher levels in general in cancer. The positive staining of ALDH1/2 of the endothelial cells of the vessels adjacent to cancerous foci in OSCC cases was noticed. The demographical and clinical manifestation details of the patients from which the positively stained tissue samples were derived are given in Table 3.

Discussion
VE staining of ALDH has been reported only in tissues of kidney cancer (specifically renal cell carcinoma, a well-known angiogenic tumor) where double immunofluorescence staining of the frozen sections of human renal tumors and normal kidney tissues was performed using anti-ALDH antibody, and the ALDH staining was proven to be negative in normal blood vessels, but was strongly positive in tumor blood vessels [6].
Additionally, in in vivo mouse models of oral carcinoma, double immunofluorescence staining of oral carcinoma xenografts in mice was carried out using anti-ALDH antibody, and VE staining of ALDH has been reported in VE cells of tumor blood vessels within tumor foci [6]. However, ALDH was expressed in the tumor blood vessels of oral carcinoma xenografts, indicating that oral carcinoma contains ALDH high endothelial cells. ALDH was hardly expressed in normal blood vessels in vivo [6]. Additionally, the pattern of ALDH expression in tumor blood vessels was heterogeneous, which suggests that stem-like endothelial cells are present in tumor blood vessels in vivo [6].
Interestingly, endothelial cells positive for ALDH showed drug resistance to 5-FU in vitro and in vivo and manifested higher levels of aneuploidy [8]. Therefore, ALDH may be applied as a prognostic marker for drug resistance and genetic aberrations. High ALDH activity was found in a subset of human mesenchymal stromal cells with vascular regenerative potential [9], suggesting a preexisting cell capacity for a proangiogenic secretory role, which either initiates or mediates the angiogenesis taking place, close to dysplasia or cancer. Tumor endothelial cells (TECs) are generally different from their normal counterparts due to different gene expression [10,11] The chemokine receptor CXCR7 is upregulated in TECs and therefore constitutes a novel marker [12]. Lysyl oxidase is also upregulated in TECs and its knockdown inhibited cell migration, and thus may be applied as a biomarker for angiogenesis [13]. Prostacyclin receptor mediates reendothelialization and angiogenesis, constituting another novel biomarker for neoangiogenesis [14]. Biglycan is also upregulated in TECs and its knockdown inhibited cell migration, and thus may be applied as a biomarker for angiogenesis [15].
TECs may acquire cytogenetic abnormalities while in the tumor microenvironment and these cytogenetic alterations in tumor vessels of carcinoma may play a significant role in modifying tumor-stromal interactions [16,17]. Finally, TECs from high metastatic tumors have a more pro-angiogenic phenotype than those from low metastatic tumors [18]. In our study, we found (for the first time) a strong positivity of ALDH in OSCC tissues (mainly) and in the advanced stages of OL. Further studies in larger samples of patients with a wider variety of ALDH biomarkers may shed some light on the possible role of endothelial expression of ALDH in tumor neo-angiogenesis (blood or lymphatic) and possible relevant therapeutic implications. It may be the case, based on the findings of the aforementioned studies, that specific gene expression alters the TECs which in turn attract cancer cells on a locoregional level. The next step may be the epithelial to mesenchymal transition of cancer cells, enabling the migration through the adjacent vessels, and thus resulting in metastatic lymphadenopathy and distant metastasis. Additionally, the positive VE staining in severe OL may indicate that the microenvironment of dysplastic OL interacts with the underlying stroma.
Since FGF, especially FGF-2, plays a significant angiogenic role [7], fibroblasts may also mediate the interaction between dysplastic epithelial cells and endothelial cells. Inflammation may also influence angiogenesis through the function of transforming growth factor beta (TGF-β), interferons (IFNs), TNF-a, and interleukins [7]. TGF-β, IFNs, and TNF-a still play a controversial role but interleukins on the other hand have proven to be an important pro-angiogenic factor [7]. A reverse scenario may also be assumed where inflammation initiates the stroma alterations leading to VE alterations leading eventually to epithelial dysplasia and carcinogenesis. Finally, ALDH is implicated in initiating the process of field cancerization, which is a biological process in which large areas of cells at a tissue surface or within an organ are affected by carcinogenic alterations [19]. Also, ALDH1 activity is correlated with poor clinical prognosis and higher recurrence rates [20].
The limitations of the study include the lack of quantitative assessment, the lack of TNM (tumor, nodes, and metastases) classification and five-year survival rate of the OSCC cases, as well as the lack of information regarding the course of the disease in general, its therapeutic approach, and the malignant transformation rate of the OL cases involved.

Conclusions
The statistically significant high IHC VE expression of ALDH1/2 in OSCC than in OL and normal oral mucosa vasculature indicates a possible significantly stronger activation of endothelial cells during carcinogenesis that could be an indicator for the role of inflammation in the development of field cancerization and of prognostic value for (vascular/lymphatic) metastasis. Further studies in more patients and with additional information regarding TNM classification, the course of the disease, and the malignant transformation rate of OL may enhance the clinical importance of our findings.

Additional Information Disclosures
Human subjects: Consent was obtained or waived by all participants in this study. Ethics Committee of the Dental School of Aristotle University of Thessaloniki, Greece issued approval Nr 8/03.07.2019, dated July 3, 2019. Animal subjects: All authors have confirmed that this study did not involve animal subjects or tissue.

Conflicts of interest:
In compliance with the ICMJE uniform disclosure form, all authors declare the following: Payment/services info: All authors have declared that no financial support was received from any organization for the submitted work. Financial relationships: All authors have declared that they have no financial relationships at present or within the previous three years with any organizations that might have an interest in the submitted work. Other relationships: All authors have declared that there are no