Expression of Cathepsins B, D, and G in WHO Grade I Meningioma

Aim: We have recently demonstrated the presence of putative tumor stem cells (TSCs) in World Health Organization (WHO) grade I meningioma (MG) localized to the microvessels, which expresses components of the renin-angiotensin system (RAS). The RAS is known to be dysregulated and promotes tumorigenesis in many cancer types, including glioblastoma. Cathepsins B, D, and G are isoenzymes that catalyze the production of angiotensin peptides, hence providing bypass loops for the RAS. This study investigated the expression of cathepsins B, D, and G in WHO grade I MG in relation to the putative TSC population we have previously demonstrated. Methods: 3,3-Diaminobenzidine (DAB) immunohistochemical (IHC) staining with antibodies for cathepsins B, D, and G was performed on WHO grade I MG tissue samples from 10 patients. Three of the MG samples subjected to DAB IHC staining underwent immunofluorescence (IF) IHC staining to investigate co-expression of each of these cathepsins using combinations of smooth muscle actin (SMA) and embryonic stem cell marker OCT4. NanoString mRNA expression (n = 6) and Western blotting (WB; n = 5) analyses, and enzyme activity assays (EAAs; n = 3), were performed on snap-frozen WHO grade I MG tissue samples to confirm transcriptional activation, protein expression, and functional activity of these proteins, respectively. Results: DAB IHC staining demonstrated expression of cathepsins B, D, and G in all 10 MG samples. NanoString mRNA expression and WB analyses showed transcriptional activation and protein expression of all three cathepsins, although cathepsin G was expressed at low levels. EAAs demonstrated that cathepsin B and cathepsin D were functionally active. IF IHC staining illustrated localization of cathepsin B and cathepsin D to the endothelium and SMA+ pericyte layer of the microvessels, while cathepsin G was localized to cells scattered within the interstitium, away from the microvessels. Conclusion: Cathepsin B and cathepsin D, and to a lesser extent cathepsin G, are expressed in WHO grade I MG. Cathepsin B and cathepsin D are enzymatically active and are localized to the putative TSC population on the microvessels, whereas cathepsin G was localized to cells scattered within the interstitium, These results suggest the presence of bypass loops for the RAS, within WHO grade I MG.


INTRODUCTION
Meningioma (MG) is one of the most common primary neoplasms of the central nervous system with an age-adjusted incidence of 11.5 per 100,000 person-years (1)(2)(3). Risk factors for the development of MG include hereditary syndromes such as neurofibromatosis type 2, high-dose radiation, and being female (1,(4)(5)(6)(7). The standard of care for MG is gross total resection. However, due to tumor location and/or surgical morbidity, complete resection is not achievable in 50% of cases (8). Incomplete resection along with high recurrence rates of up to 32% during long-term follow-up (8-10) underscore a need for more optimal medical management of MG.
The prevailing concept proposes a neuroectodermal origin of MG, specifically arachnoid cap cells (11). However, there is accumulating evidence for the presence of tumor stem cells (TSCs) within MG (12). Tumorigenesis is proposed to be driven by a subpopulation of TSCs, which originate from acquired mutations in normal resident embryonic stem cells (ESC), progenitor cells or differentiated cells (13)(14)(15). Takahashi et al. (16) were the first to elucidate expression of ESC markers OCT4, SALL4, NANOG, SOX2, c-MYC, and KLF4, with the ability to induce and maintain pluripotency in addition to a deregulated capacity for renewal in induced pluripotent stem cells. We have recently demonstrated the presence of an ESClike population which expresses OCT4, NANOG, SOX2, KLF4, and c-MYC on both the endothelial and pericyte layers of the microvessels in World Health Organization (WHO) grade I MG (17). This ESC-like population expresses components of the renin-angiotensin system (RAS): pro(renin) receptor (PRR), angiotensin converting enzyme (ACE), angiotensin II receptor 1 (ATIIR1), and angiotensin II receptor 2 (ATIIR2) in the microvessels in this tumor (18).
While the RAS is classically associated with regulation of blood pressure and blood volume, it also plays a role in cancer by inducing angiogenesis, promoting cellular proliferation, and inhibiting apoptosis (19)(20)(21). The RAS is dysregulated in malignancy, promoting tumor growth in cancers such as nonsmall cell lung (22), ovarian (23), pancreatic (24), and prostate (25) cancer, as well as brain tumors such as glioblastoma (GB) (26).
The classical RAS cascade (Figure 1) is initiated by (pro) renin, physiologically secreted from the juxtaglomerular cells in the kidney, and binds to PRR. Renin then catalyzes the conversion of the precursor angiotensinogen (AGN), which is physiologically produced by the liver, to angiotensin I (ATI) (21,27). ATI is further cleaved by endothelial-bound ACE to form angiotensin II (ATII), which is in turn converted in to angiotensin III (ATIII) by aminopeptidase A. ATII and ATIII can produce vasoconstriction or vasodilatory effects depending on whether they bind to ATIIR1 or ATIIR2, respectively (19).
The RAS can be activated through the classical cascade mechanisms initiated by (pro)renin and PRR or via bypass loops by a group of lysosomal proteases called cathepsins (19). Cathepsin B is a cysteine lysosomal protease that catalyzes the conversion of pro (renin) into active renin without activating the PRR (28). Cathepsin D is an aspartic lysosomal protease that can directly convert AGN to ATI due to its significant homology to renin (29). Cathepsin G is a serine lysosomal protease with the capacity to generate ATII from ATI, or directly from AGN, without the action of renin or ACE (30). Therefore, any attempt at modulating RAS-induced tumorigenesis with traditional RAS modulators such as β-blockers, ACE inhibitors (ACEIs) or angiotensin receptor blockers (ARBs), may be circumvented by cathepsin-induced activation of the RAS.
The expression of cathepsin B by the stem cell niche on the microvessels in GB has been reported recently (31), with this staining pattern being a negative prognostic factor (32). This study was aimed at investigating the expression of cathepsins B, D, and G in WHO grade I MG, in relation to the putative TSC population we have previously identified (17).

Tissue Samples
WHO grade I MG tissue samples from one male and nine female patients, age 36-85 (mean, 61.2) years included in our previous studies (17,18), were sourced from the Gillies McIndoe Research Institute Tissue Bank for this study. Study approval was granted by the Central Regional Health and Disability Ethics Committee (Ref. no 15CEN28). Informed written consent was obtained from all patients.
To further characterize the expression of cathepsins B, D, and G, IF IHC staining was performed on three representative MG samples from the original cohort of 10 patients included in the DAB IHC staining. Dual IF IHC staining was performed for each of the cathepsins using identical primary antibodies and concentrations as for DAB IHC staining, in conjunction with SMA or OCT4. An appropriate fluorescent secondary antibody combination of Vectafluor Excel anti-rabbit 594 (readyto-use, cat# VEDK-1594, Vector Laboratories, Burlingame, CA, USA) and Alexa Fluor anti-mouse 488 (1:500, cat# A21202, Life Technologies) was used for detection.
DAB IHC-stained slides were counterstained with hematoxylin and mounted in Surgipath Micromount mounting medium (cat# 3801732, Leica), whereas IF IHC-stained slides FIGURE 1 | The classical renin-angiotensin system (RAS) and its bypass loops consisting of cathepsins B, D, and G. Classically (pro)renin, from the juxtaglomerular cells in the kidney, is activated by binding to the (pro)renin receptor. Renin coverts angiotensinogen (AGN), produced by the liver, to form angiotensin I (ATI), which is further cleaved to form angiotensin II (ATII), by angiotensin converting enzyme (ACE). Aminopeptidase A then converts ATII to angiotensin III; both of which activate either angiotensin II receptor 1 (ATIIR1) and/or angiotensin II receptor 2 (ATIIR2). Alternately, the RAS may be activated via its bypass loops. Cathepsin B and cathepsin D are renin-activating enzymes. Cathepsin D bears significant homology to renin and is therefore able to independently catalyze the conversion of AGN to form ATI. Cathepsin G may form ATII from either ATI or directly from AGN. Adapted with permission from Integrative Cancer Science and Therapeutics (19).
Positive controls for DAB IHC staining included human placenta for cathepsin B, human breast tissue for cathepsin D and human splenic lymphoma for cathepsin G. Negative controls were performed on sections of WHO grade I MG using a matched isotype control for rabbit primary antibodies (ready-to-use; cat# IR600, Dako), to determine the specificity of the antibody. Negative controls for IF IHC staining were performed using a section of MG tissue with the combined use of primary isotype mouse (ready-to-use; cat# IR750, Dako) and rabbit (ready-to-use; cat# IR600, Dako) antibodies.

Image Analysis
The DAB IHC-stained slides were viewed and images were captured, using an Olympus BX53 light microscope fitted with an Olympus DP21 digital camera and processed with CellSens 2.0 software (Olympus, Tokyo, Japan). The IF IHC-stained images were captured using the Olympus FV1200 biological confocal laser-scanning microscope and processed with CellSens Dimension 1.11 software using the 2D deconvolution algorithm (Olympus).

Enzymatic Activity Assays
Enzymatic activity of cathepsin B and cathepsin D were determined in snap-frozen WHO grade I MG tissue samples from three of the five samples used for WB analysis. Enzymatic activity assay (EAA) kits were utilized for cathepsin B (cat# ab65300; Abcam) and cathepsin D (cat# ab65302; Abcam).
All steps of the procedure were performed according to the manufacturer's protocol, and as recently described (33). Fluorescence was measured in a Nunc TM F96 MicroWell TM black polystyrene plate (cat# 136101, Thermo Fisher Scientific) using the Varioskan Flash plate reader (cat# MIB5250030, Thermo Fisher Scientific). Tonsil and denatured tonsil tissue lysates were used as appropriate positive and negative controls, respectively.
NanoString mRNA Expression Analysis mRNA expression of cathepsins B, D, and G was determined by NanoString mRNA analysis in six snap-frozen WHO grade I MG samples of the original cohort of 10 patients used for DAB IHC staining. Total RNA was extracted separately from ∼20 mg of each snap-frozen MG tissue and run through the NanoString nCounter TM Gene Expression Assay (NanoString Technologies, Seattle, WA, USA), as previously described (27,36). Probes for the genes encoding cathepsin B (NM_001908.2), cathepsin D (NM_001909.3) cathepsin G (NM_001911.2) and the housekeeping gene GUSB (NM_000181.1), were designed and synthesized by NanoString Technologies. Raw data were analyzed with nSolver TM software (NanoString Technologies) using standard settings and normalized against the housekeeping gene.

Statistical Analyses
Data obtained from WB and NanoString mRNA analyses were subjected to paired t-tests using SPSS version 24 software.

Histochemical and DAB IHC Staining
The diagnosis of WHO grade I MG was confirmed by H&E staining for all 10 tissue samples (Figure 2A). All DAB IHCstained sections exhibited ubiquitous, granular, cytoplasmic staining for cathepsin B (Figure 2B, brown) and cathepsin D (Figure 2C, brown). Cytoplasmic and nuclear staining for cathepsin G (Figure 2D, brown) showed scattered clusters of focally positive cells.
The expected positive staining patterns for cathepsin B (Supplementary Figure 1A, brown), cathepsin D (Supplementary Figure 1B, brown), and cathepsin G (Supplementary Figure 1C, brown) were demonstrated on human placenta tissue, human breast tissue, and human splenic lymphoma, respectively. Negative control tissues showed minimal staining for all three cathepsins on sections of WHO grade I MG tissue using a matched isotype control for both mouse and rabbit primary antibodies (Supplementary Figure 1D, brown).

IF IHC Staining
To determine whether each cathepsin was expressed by the putative TSC population on the microvessels that express the ESC markers we have previously reported in WHO grade I MG (17), we co-stained each cathepsin with the pericyte marker, SMA (37). Cathepsin B (Figure 3A, red) and cathepsin D (Figure 3B, red) were illustrated on both the endothelium as well as the outer pericyte layer. These were in contrast to cathepsin G (Figure 3C, red), which was expressed by stained cells away from the microvessels.
To confirm the expression of both cathepsin B (Figure 3D, red) and cathepsin D (Figure 3E, red) on the putative TSC population, we performed co-staining with the stem cell marker, OCT4 (Figures 3D,E, green), confirming co-expression of both markers in the microvessels. Negative isotype controls demonstrated minimal staining as expected ( Figure 2F).

Western Blotting
WB of the snap-frozen WHO grade I MG samples demonstrated the presence of bands at the expected molecular weight of 25 kDa for cathepsin B (Figure 4A, blue) and 28 kDa for cathepsin D (Figure 4B, blue) in all five MG samples. Cathepsin G was also detected at the expected molecular weight of 29 kDa (Figure 4C, blue), but at relatively low levels in all five samples. α-tubulin confirmed approximate equivalent protein loading for the MG samples examined (Figures 4A-C, green). Densitometry was performed to convert the WB data in to numerical values for quantitative analysis. Results for cathepsins B, D, and G (Figure 4D), relative to α-tubulin, confirmed significant differences in the relative abundance of cathepsin B (p < 0.002) and cathepsin D (p < 0.0001) compared to cathepsin G.

Enzymatic Activity Assays
In light of the WB analysis, detecting only cathepsin B and cathepsin D, we limited our study to determine the functional activity of these two cathepsins. EAAs demonstrated enzymatic activity for cathepsin B and cathepsin D (Figure 5).

NanoString mRNA Expression Analysis
NanoString mRNA expression analysis demonstrated the presence of mRNA transcripts for cathepsins B, D, and G, normalized against the housekeeping gene GUSB, in all six MG samples (Figure 6). Statistical analysis of the NanoString data, using paired t-tests, further confirmed that the mean levels of expression of cathepsin B (p < 0.004) and cathepsin D (p < 0.001) were significantly higher than that of cathepsin G, but only marginally different to each other (p < 0.019).

DISCUSSION
This study demonstrated the expression of cathepsins B, D, and G in WHO grade I MG. Cathepsin B and cathepsin D were enzymatically active and were demonstrated on the putative TSC population of the endothelial and pericyte layers of the microvessels (17), which expresses components of the RAS (18). NanoString mRNA expression and WB analyses demonstrated low levels of mRNA transcripts and protein expression of cathepsin G, respectively. DAB and IF IHC staining showed that this protein was expressed by cells scattered within the interstitium, away from the microvessels in the WHO grade I MG samples examined. It is interesting to speculate that the cathepsin G + cells may be of a mast cell phenotype, as previously reported (38); however, this topic requires further investigation.
The faint OCT4 IF IHC staining maybe due to the relative low abundance of this protein, and the use of dual DAB IHC staining could further accentuate the staining. Furthermore, the relative abundant staining for SMA, maybe more representative of arterioles or venules.
Elevated expression of cathepsin B and cathepsin D has been observed in many tumor types such as infantile hemangioma (38), oral tongue squamous cell carcinoma (33), MG (39), and gliomas (40,41) including GB (34). Interestingly, the cathepsin B expression pattern that we report here almost parallels that reported with GB cancer stem cells (31). Taken together, upregulation of cathepsin B within WHO grade 1 MG may indicate a growth phase in this tumor, although, this requires further investigation.
Research into the role of cathepsin B and cathepsin D in tumorigenesis, so far, has mainly focused on their independent influence on tumor invasiveness. Malignant transformation results in the overproduction of cathepsin B (42) and elevated cathepsin B levels correlate with increased invasiveness and recurrence of MG (32,39,43,44), astrocytoma and gliomas (39,45). As an intracellular lysosomal protease, cathepsin B has been found in the extracellular matrix (ECM) of colorectal cancer, localized to the invasive front of the tumor (46). The involvement of cathepsin B in tissue invasion has been attributed to its proteolytic activation of ECM components urokinase plasminogen activator, plasminogen and plasmin, which further activate metalloproteinases (MMPs), whilst destroying MMP inhibitors, thereby enabling ECM degradation (47). At the same time cathepsin B promotes autophagy and cannibalism allowing tumor cells to recycle nutrients and maintain a proliferative and infiltrative phenotype (42). Despite increasing evidence for its involvement in promoting invasive tumor grades, little is known about the role of cathepsin B in its RAS-related influence on tumorigenesis in non-invasive WHO grade I MG, which represent 80% of all tumors (48).
Evidence of a role for cathepsin D in brain tumors, particularly MG, has so far been inconclusive (49). Some studies have found elevated levels of cathepsin D in high-grade astrocytomas (45) and GB (34,50), while others have shown no association with invasive MG (51) or an inverse relationship between cathepsin D levels and MG tumor grade (52). Despite the ability of cathepsin G to activate the RAS via multiple bypass loops, there is a paucity of literature on its potential role in tumorigenesis. This study demonstrated low expression of cathepsin G, which was localized to cells scattered within the interstitium, away from the microvessels, in WHO grade I MG.
The RAS is known to be dysregulated in malignancy and promote tumorigenesis in many cancer types. The expression of cathepsins B, D, and G in WHO grade I MG suggests the presence of bypass loops for the RAS, which could potentially negate the effectiveness of traditional RAS modulators such as β-blockers, ACEIs and ARBs (30,53,54). We have previously demonstrated the putative presence of an ESC-like population localized to the microvessels within WHO grade I MG (12,17), that expresses components of the RAS (18). This study demonstrates that  Our results suggest that cathepsin B and cathepsin D may be novel therapeutic targets, in addition to the classical RAS, for WHO grade I MG. Limitations of the study include the relatively small sample size. Future studies including a larger sample size and functional in vivo experiments are needed to conclusively determine the role of these proteases within MG. In addition, more distinct cellular staining could be improved using chromogenic DAB IHC staining to localize the markers investigated.

AUTHOR CONTRIBUTIONS
TI and ST formulated the study hypothesis. TI, AW, and ST designed the study. AW, KW, and RJ recruited patients and obtained study samples. TI, HB, AW, RR, and ST interpreted the DAB IHC data. TI, RR, and ST interpreted the IF IHC data. BvS performed the WB analysis. BvS, TI, AW, and ST interpreted the WB data. TI, RR, and ST interpreted the NanoString mRNA analysis data. RM carried out statistical analysis. RR, BvS, AW, TI, and ST drafted the manuscript. All authors commented on and approved the manuscript.