Spatial tissue proteomics reveals distinct landscapes of heterogeneity in cutaneous papillomavirus‐induced keratinocyte carcinomas

Infection with certain cutaneous human papillomaviruses (HPV), in conjunction with chronic ultraviolet (UV) exposure, are the major cofactors of non‐melanoma skin cancer (NMSC), the most frequent cancer type worldwide. Cutaneous squamous cell carcinomas (SCCs) as well as tumors in general represent three‐dimensional entities determined by both temporal and spatial constraints. Whole tissue proteomics is a straightforward approach to understand tumorigenesis in better detail, but studies focusing on different progression states toward a dedifferentiated SCC phenotype on a spatial level are rare. Here, we applied an innovative proteomic workflow on formalin‐fixed, paraffin‐embedded (FFPE) epithelial tumors derived from the preclinical animal model Mastomys coucha. This rodent is naturally infected with its genuine cutaneous papillomavirus and closely mimics skin carcinogenesis in the context of cutaneous HPV infections in humans. We deciphered cellular networks by comparing diverse epithelial tissues with respect to their differentiation level and infection status. Our study reveals novel regulatory proteins and pathways associated with virus‐induced tumor initiation and progression of SCCs. This approach provides the basis to better comprehend the multistep process of skin carcinogenesis.


| INTRODUCTION
Mass spectrometry (MS)-based proteomic profiling of cell lines is well established to investigate protein or cell type-specific functions. The resulting data are analyzed by computational methods to identify the respective proteins and related quantitative information. 1 Spatial tissue proteomics complement genomic and transcriptomic methods to contextualize physiological and pathological processes 2-4 within a cellular network. However, despite methodological breakthroughs, most proteomic analyses focused on the bulk tumor, not considering the spatiotemporal heterogeneity within individual specimens or different dynamic stages. 5 Indeed, tumor diagnostics in general still involves histological analyses of tissue sections by combining conventional microscopy and immunostainings of marker proteins. 5 Hence, larger studies are hampered by limited possibilities for multiplexing on the same tissue section as well as by the availability of suitable antibodies. Due to their excellent preservation of tissue architecture over decades, formalin-fixed, paraffin-embedded (FFPE) samples represent wellsuited material to study intra and intertumoral heterogeneity. 5 Remarkably, despite the availability of MS-based proteomes from FFPE cancer tissue, 6 a combination with spatial resolution of tumors is rare, 7,8 mostly due to the low yield of material obtainable from a specific area of the specimen. This inherently limits both the number of identifiable proteins and also spatial resolution within the tissue. 5 While the stable crosslinking of proteins also represented challenges for sample preparation and MS analysis in the past, 9 recent technological advancements made FFPE tissues a highly valuable source for proteome analyses.
To our knowledge, such analyses have not yet been performed on non-melanoma skin cancer (NMSC), the most frequent type of cancer worldwide. NMSC represents a serious disease regarding its high incidence and particularly high morbidity in patients undergoing organ transplantation and systemic immunosuppression. 10 Besides chronic ultraviolet (UV) exposure, certain types of cutaneous human papillomaviruses (HPV) are essential cofactors in the development of NMSCs. 11,12 Papillomaviruses (PVs) are small nonenveloped doublestranded DNA viruses that infect epithelial keratinocytes. Due to the initial high load of transcriptionally active HPV DNA in premalignant precursor lesions, which is frequently lost in progressed cutaneous squamous cell carcinomas (SCC), epidemiological and functional studies have hypothesized that cutaneous PVs may favor skin carcinogenesis via a "hit-and-run" mechanism. 11,12 This multistep progression toward cutaneous SCCs is a concerted action between the initial presence of these viruses and chronic UV radiation, allowing the accumulation of mutated cells that finally makes viral presence dispensable. 11 Indeed, this scenario was first demonstrated in Mastomys coucha, an African multimammate mouse that is naturally infected with the cutaneous Mastomys natalensis papillomavirus (MnPV), the etiological agent inducing benign epithelial tumors such as papillomas. 13 This preclinical model mimics many aspects of cutaneous HPV infection in humans, 13 allowing to investigate the temporal and spatial spread of a cutaneous PV in its genuine host starting from primary infection to the final manifestation of a skin tumor. 14 During chronic UV irradiation, MnPV-infected animals develop considerably more cutaneous SCCs than virus-free counterparts. 14 Here, like in humans, two distinct types of carcinomas can be found: well-differentiated keratinized SCCs (KSCC) with high loads of transcriptionally active viral genomes, and poorly differentiated non-keratinizing SCCs (nKSCC) that lack viral DNA but frequently harbor p53 mutations together with signs of an epithelial-to-mesenchymal transition (EMT). 14 Since PV replication depends on differentiating squamous epithelium, a dedifferentiated tissue is apparently counteracting the permissive viral cycle, leading to the loss of episomal viral DNA in these tumors. Since this scenario could not yet be observed in other animal models used in this context 15,16 and this multistep process cannot be mimicked in cell culture, the underlying molecular details are not yet understood.
In this study, we applied a workflow to microdissect different regions of cutaneous SCCs in the Mastomys model, followed by ultrasensitive and rapid protein isolation using single-pot solid-phaseenhanced sample preparation (SP3) 17,18 and MS to decipher intraand intertumoral heterogeneity. For the first time, characteristic proteomic landscapes could be correlated to a particular tissue phenotype and viral infection status on a spatial level in vivo.

| Samples
Tissue samples from M. coucha were taken from our tissue collection obtained in previous studies, 14,[19][20][21] where skin biopsies and tumors were cut longitudinally with scalpels, fixed in 4% formalin, and embedded in paraffin (FFPE). M. coucha are naturally infected with the cutaneous MnPV, which like HPVs belongs to the family Papillomaviridae. The animals were chronically exposed to UV radiation (312 nm) to induce cutaneous SCCs with different histopathological properties in virus-positive skin, namely, welldifferentiated keratinizing SCCs (KSCC) and poorly differentiated non-keratinizing SCCs (nKSCC). 14

| Sample preparation for proteome profiling
Samples were thawed, vortexed, and subsequently transferred to adaptive focused acoustics (AFA)-TUBE TPX polymerase chain reaction (PCR) stripes (Covaris, Inc.) for AFA ultrasonication in a LE220plus Covaris device. Here, the peak incidence power (PIP) was set to 450, the duty factor (DF) to 50%, the cycles per burst (CPB) to 600, and the time to 300 s per TPX PCR stripe. Dithering of the AFA focus was applied with a 3 mm z-offset. Samples were centrifuged at 15,000g for 10 min before proceeding with a protein quantification assay (Pierce, ThermoFisherScientific). For each sample, 10 μg of extracted protein was transferred to a 96-well PCR plate for processing with autoSP3 on an Agilent Bravo liquid handling system. 17 In brief, the plate was prepared with each sample in a total volume of 12 μL 1% SDS, 100 mM ammonium bicarbonate (ABC). Proteins were automatically reduced and alkylated by the addition of 10 mM tris(2-carboxyethyl)phosphine (TCEP), 40

| MS
Liquid chromatography with tandem MS (LC-MS/MS) analyses were carried out on an Ultimate 3000 UPLC system (ThermoFisherScientific) directly connected to an Orbitrap Exploris 480 MS for a total of 120 min. Dried peptides were reconstituted before measurement and online desalted on a trapping cartridge (Acclaim PepMap300 C18, 5 µm, 300 Å wide pore; ThermoFisherScientific) for 3 min using 30 µL/min flow of 0.05% TFA in water. The analytical multistep gradient (300 nL/min) was performed using a nanoEase MZ Peptide analytical column (300 Å, 1.7 µm, 75 µm × 200 mm, Waters) using solvent A (0.1% formic acid in water) and solvent B (0.1% formic acid in ACN). For 102 min, the concentration of B was linearly ramped from 4% to 30%, followed by a quick ramp to 78%, after 2 min the concentration of B was lowered to 2% and a 10 min equilibration step appended. Eluting peptides were analyzed in the MS using datadepend acquisition (DDA) mode. A full scan at 120k resolution (380-1400 m/z, 300% AGC target, 45 ms maxIT) was followed by up to 2 s of MS/MS scans. Peptide features were isolated with a window of 1.4 m/z, fragmented using 26% NCE. Fragment spectra were recorded at 15k resolution (100% AGC target, 54 ms maxIT).
Unassigned and singly charged eluting features were excluded from fragmentation and dynamic exclusion was set to 35 s.

| Statistics
The following filtering, normalization, and imputation were per- condition-specific elements might lead to artificially increased distances in between the conditions. For missing values with no complete absence in one condition, the R package missForest was used for imputation. 26 The statistical analysis was performed with the R-package "limma." 27 The p-values were adjusted with the Benjamini-Hochberg method for the multiple testing. 28 For the principal component analysis (PCA) plots, the filtering, normalization, and imputation as described above were applied to the whole set of samples of the HE staining and E1^E4 staining set, respectively. The sample positions in the two-dimensional grid were calculated via the "plotMDS" function of the R-package "limma."

| Network and functional analyses
Regulatory networks and functional analyses were generated using ingenuity pathway analyses (IPA) software (Qiagen Inc.). A "core analysis" was conducted using all data sources available in IPA. All networks and node types were selected to be displayed. To specify results, only mouse species-related entries were used for analyses. A z-score of ±2.0 as a cut-off for a reliable prediction was automatically set by IPA. 29 Protein networks were visualized with STRING database (www.string-db.org). 30 Genes and proteins were named based on the standard nomenclature for describing mouse genes and proteins. 31

| Tissue stainings
FFPE tissue sections were deparaffinized as described above for subsequent stainings. HE staining was performed using a standard protocol. 32 For immunostainings heat-induced epitope retrieval (HIER) was achieved by boiling the sections in a steamer for 10 min in citrate buffer pH 6.0 or ethylenediaminetetraacetic acid (EDTA) buffer pH 9.0.
Sections were blocked with 5% goat serum and incubated overnight at  In preceding calibration tests, we realized that a minimum tissue volume of approximately 0.15 mm 3 is required to obtain satisfactory MS data. Since high SDS concentrations of the lysis buffer would interfere with MS measurements, lysates were purified via SP3 paramagnetic bead technology (outlined in Figure 1D) that facilitates ultrasensitive proteomic analyses of protein mixtures. 17

| Spatial MS analysis of tumor tissue reveals strong intra-and intertumoral heterogeneity
To assess the similarity between the proteomes of in total 25 samples (i.e. control epidermis, MnPV-induced benign tumors, well-and dedifferentiated areas of UV-induced SCCs), microdissection was based on HE stainings. The resulting MS data covered 3752 proteins, of which 866 proteins were shared between all samples. Intriguingly, subjecting these proteome data to a PCA (Figure 2A), the individual samples clustered in strikingly distinct areas. This clearly shows that the overall resemblance of the individual proteomes deriving from the same tissue type is principally higher than to those of other tissue origins. Spatial proximity revealed that proteomes of welldifferentiated tumor areas still preserved a higher similarity to morphologically normal epidermis, while proteomes obtained from MnPV-induced tumors and, most prominently, dedifferentiated areas of SCCs were more distinct ( Figure 2A). As depicted in the Venn diagram ( Figure 2B) based on the proteins shared between all samples, the high similarity between phenotypically normal epidermis and well-differentiated SCC areas is reflected by a low number ( = 97 proteins; see Figure 2B, red circle) of significantly (adj.p < 0.05) differentially expressed proteins. Conversely, dedifferentiated SCC areas ( = 240 proteins) and MnPV-induced tumors ( = 571 proteins) revealed substantially more differences from normal epidermis. Of note, these differentially abundant proteins were not equally altered, but each comparison revealed prominent outliers ( Figure 2C). Some of those, for instance, keratin 10 (K10; log2FC = -6.41, adj.p < 0.001) and K2 (log2FC = -9.71, adj.p < 0.001), marking postmitotic suprabasal and spinous keratinocytes, respectively, are exceedingly changed in dedifferentiated tissue when compared with normal epidermis ( Figure 2D). All significantly changed proteins are summarized in Supporting Information: STables 1-6.
Phenotypic alterations microscopically seen after HE staining ( Figure 1A), point to an EMT which is characterized by a switch of E-cadherin to N-cadherin expression. 33   SCC areas (log2FC = 1.56, adj.p = 0.064). Also linked with loss of E-cadherin during an EMT, an increase of VIM confers migratory abilities to cancer cells 34, 35 An EMT could lead to an increase of cancer stem cells (CSCs) 36 that are characterized by coexpression of major hyaluronan (HA) receptor CD44 and nerve growth factor (NGF) receptor CD271 (also known as p75 NTR ). 37 While the latter was not detected in any tissue type analyzed, CD44 was significantly   Figure 3A shows a selection of 30 diseases and bio functions ordered by relevance based on a z-score calculated by IPA. 29 In fact, cell death-associated processes belonging to the categories "Cancer, Organismal Injury and Abnormalities," such as necrosis and apoptosis were deactivated  (Figure 2A).
Interestingly, although peptides derived from NGF receptor CD271 were not detected by MS profiling, IPA analyses predicted an upregulation of NGF signaling in dedifferentiated SCC areas ( Figure 3B) that may favor the maintenance of stem-like properties and tumorigenicity. 40 Of potential interest, NGF signaling is linked to several enriched bio functions (e.g., metastasis and growth of tumor) ( Figure 3A). 41 Consistent with this assumption, phosphatase and tensin homolog (PTEN) PI3K/AKT signaling and mTOR pathway were -compared with morphologically normal epidermis-activated in dedifferentiated SCCs areas, thereby promoting tumor cell growth, metastasis and invasion to new healthy tissues. 42,43 In line with this, pathways related to EMT (e.g., ephrin receptor signaling) and metastasis (e.g., colorectal cancer metastasis signaling) were also predicted to be activated. Conversely, these pathways are deactivated in MnPV-induced tumors ( Figure 3B).

| Activity of upstream regulators depends on tissue origin
Next, upstream regulator analyses were performed with IPA to identify effector molecules that may explain the up-or downregulation of proteins in the MS datasets ( Figure 3C, see also Supporting Information: STables 21-27). Generally, well-known functional regulators associated with cancer (e.g., TP53) were obtained.
In our system, the predicted activity of the gene product of Nfe2l2 (NRF2, nuclear factor erythroid 2-related factor 2) was strongly increased in MnPV-induced benign tumors when compared with normal epidermis (z-score = 4.596). Intriguingly, NRF2 activation increases the expression of cytoprotective proteins, thereby contributing to DNA damage prevention. 44  IPA calculated z-scores to rank the relevance of differentially detected proteins and networks. A red square (positive z-score) indicates increased activity, a blue square (negative z-score) indicates decreased activity, a gray square indicates that no prediction could be made by IPA. (D) Protein networks based on Trp53-regulated proteins in well-differentiated (Trp53-regulated proteins: n = 24) or dedifferentiated tumor tissue (Trp53-regulated proteins: n = 66) when compared with phenotypically normal epidermis. Networks were created using a string database. 30 The upstream regulator Trp53 is marked bold. The connections represent protein-protein associations. regulator DEC1 stimulates PI3K/AKT/mTOR activation and causes a prosurvival and prometastatic phenotype in breast cancer. 45,46 In dedifferentiated SCC areas, its activation may explain increased mTOR signaling ( Figure 3B).
Previous studies have shown that MAP4K4 promotes pancreatic cancer 47 and acts as an upstream regulator of the ERK/MAPK pathway required for adenocarcinoma maintenance. 48 Consistently, the predicted activity of MAP4K4 in our comparisons ( Figure 3C) correlates with (de)activation of ERK/MAPK signaling in canonical pathway analyses ( Figure 3B).
With respect to the human TP53 gene (or its murine and Mastomys homolog Trp53), IPA predicted that its regulatory activity affects 24 differentially expressed proteins in well-differentiated but 66 proteins in dedifferentiated SCC areas when compared with phenotypically normal epidermis (Supporting Information: STables 22 and 23). This is in line with the previous finding that Trp53 mutations -potentially altering protein interactions-are more frequent in poorly differentiated nKSCCs than in well-differentiated KSCCs. 14 Consequently, reflecting tumor heterogeneity on a molecular level, the protein networks obviously differ when visualizing the connections of these hits ( Figure 3D) using STRING database. 30 Taken together, these analyses identified pathways and regulators involved in the multistep process of skin carcinogenesis and SCC dedifferentiation.

| Validation of proteomic data using tissue stainings
To verify the proteomic profiles, immunostainings against up/ downregulated proteins ( Figure 2D 49 and was found to be strongly upregulated in the infected epidermis (log2FC = 6.63, adj.p < 0.001) and

| Tissue stainings validate proteomic changes caused by MnPV infection
We also verified these proteome data using tissue stainings against several differentially expressed viral and cellular gene products, some of which were partially found among the top five up/downregulated candidates ( Figure 5E).   proteins based on HE staining and 2414 proteins based on E1^E4 staining could be achieved. This disproportion can be explained with a lower quality of sample material in the latter setting. However, despite being lower when compared with analyses of cell line lysates, the efficiency is comparable to other in vivo MS studies using FFPE tissue. 2,17,18,51 However, some expected high abundance epidermal proteins, for example, E-cadherin, were not detected by MS, which could be due to inherent limitations of this technology itself. 52 Notably, when compared in PCA, the extracted proteomes of morphologically normal epidermis, well-and poorly differentiated SCC areas as well as benign MnPV-induced tumors were remarkably distinct ( Figure 2A). Nevertheless, all individual samples of the same tissue type, regardless of whether they derived from different tumors and animals, clustered well within the same area. This suggests a high degree of heterogeneity between well-and poorly differentiated areas within an SCC, but also similar processes of malignization in different SCCs. Notably, consistent with histological observations ( Figure 1A) and according to the PCA (Figure 2A (Figures 2A and 3B).
Typically, tissue changes histologically characterized as dedifferentiation are associated with an EMT, allowing cancer cells to invade tissues, metastasize and evade immune surveillance without losing the pluripotency of epithelial stem cells. 34,56 In this process, the cell shape ( Figure 1A) changes due to a cytoskeletal rearrangement and loss of cell-cell adhesion, which is partially attributable to a decreased expression of different keratins. 35,57 Of those, more than 20 were significantly reduced when comparing dedifferentiated SCC areas to normal epidermis. Conversely, only nine were significantly reduced in well-differentiated SCC areas (Supporting Information: STables 1 and 2), which could also be confirmed by tissue stainings (Figure 4). Consistent with our data, a recent study identified two subtypes of human AKs and cSCCs by analyzing their methylome profiles. 58 Considering mainly the methylation patterns of keratin genes, two different epidermal differentiation stages could be clearly defined, namely lesions with a more differentiated keratinocyte profile or with similarity to epidermal stem cells. Furthermore, epigenetic analyses suggested an increased invasive and metastatic potential of tumors arising from undifferentiated keratinocytes. 59 In addition, tissue stainings also revealed strong E-cadherin expression in well-differentiated, but not in less differentiated SCC areas, which contrasted to VIM expression ( Figure 4). Both proteins are important EMT markers 34,35 and E-cadherin plays a pivotal role in cell-cell adhesion, while its loss alone can induce EMT and invasiveness. 33,60 Of note, the mutual regulation between Ecadherin and the ephrin receptor function is important for morphologic maintenance of epithelial cells while overexpressed ephrin receptors can promote migration and invasion by inhibiting Ecadherin function. 57 In line with these previous reports, IPA predicted ephrin receptor signaling activation in dedifferentiated SCC areas ( Figure 3B).
It is also of note that besides others, compared with normal epidermis, NGF signaling was predicted to be strongly activated in dedifferentiated SCC areas, contrasting the prediction for welldifferentiated SCC areas and benign MnPV-induced tumors ( Figure 3B). NGF acts via two cognate receptors, one of them being CD271. 61 Interestingly, the predicted NGF signaling activation is in line with the increased expression of CD271 in dedifferentiated SCC areas ( Figure 4). In normal epithelium, CD271 contributes to epidermal homeostasis and the "switch" between stem cells and their progeny. 62 The positive staining was in conjunction with increased levels of CD44, which mediates intercellular interactions with the extracellular matrix. 63 Co-expression of both proteins is considered a marker for CSCs, which are closely associated with EMT. 37,56,64,65 Dedifferentiated tumors usually have a bad prognosis, which was observed in humans, but also for non-keratinizing SCCs in our preclinical model. 14 Hence, the predicted upregulated MAP4K4 activity ( Figure 3C) is consistent with the findings that its elevated expression in different carcinoma types correlates with tumor size, metastasis, and decreased survival. [66][67][68] In the context of HPV16-positive cancers, PI3K/AKT signaling is activated by viral oncoproteins, 69,70 but it can also be induced by mutations. 71 Notably, IPA predicts both activated PI3K/AKT and mTOR signaling in dedifferentiated SCC areas ( Figure 3B). In cancer in general, genomic instability and mutations increase over time and in cutaneous SCCs, the background mutation rates associated with UV damage are higher compared with noncutaneous tumors. 72 One important gene, tumor suppressor p53, was often found to be inactivated, especially in dedifferentiated SCC areas. 14 Here, activation of the p53-dependent Bhlhe40 gene (encoding DEC1) ( Figure 3C) could inhibit DNA damage-induced cell death 73 and assure the growth of UV-damaged cells. Additionally, IPA predicted other tumor suppressors known from skin cancer to be inactivated, for example, PTEN ( Figure 3C) even in the HPV context, 74,75 which could contribute to EMT-based dedifferentiation in a multistep process 14 that could finally substitute the initial role of the virus role as a driver of skin carcinogenesis (Figure 7). Consistently, due to viral interference in various cellular pathways, 12 only a relatively small number of shared proteins (304 out of 2414) was found among all samples.
The comparison between MnPV-induced tumors and MnPVinfected epidermis revealed several upregulated Serine/arginine-rich splicing factors (SRSF) (Supporting Information: STables 28-30) that positively control cellular and viral splicing. 76 For example, the viral E2 transcription factor controls the expression of SRSF1 and SRSF3 that in turn regulate viral L1 protein expression in a differentiationdependent manner. 77,78 Transglutaminase 3 (TGM3), one of the top upregulated proteins, catalyzes the crosslinking of different structural envelope proteins, 79 for example, the strongly upregulated SPRR1A ( Figure 5E), during terminal differentiation. 80 Increased expression of TGM3 and SPRR1A in MnPV-induced tumors may account for their low level of aggressiveness similar to other cancer types. 81,82 Reducing COL1A1 levels in MnPV-infected tissue ( Figure 5E) may have the same purpose, as increased expression of COL1A1, as well as COL1A2, were previously associated with lower survival of ovarian cancer patients. 83 84 Whether this may promote UV-induced carcinogenesis remains to be investigated.
F I G U R E 7 Graphical summary of epithelial-to-mesenchymal transition (EMT) in cutaneous squamous cell carcinoma (SCC) development based on our data and references mentioned above. The permissive environment that allows the synthesis of progeny viruses is lost due to molecular changes accompanied during the multistep progression toward malignancy.
Notably, also HPV16-transformed cells downregulate CAV1 in a p53-dependent manner and its recovery suppresses HPV-mediated cell transformation. 85 It is also noteworthy that many proteins upregulated during viral infection belong to the family of 40 S ribosomal proteins ( Figure 5E, Supporting Information: STables 28 and 30). Besides potentially indicating ribosomal dysfunction, which is also linked to cancer progression, [86][87][88] some ribosomal proteins also have extra-ribosomal functions, for example, RPS27A, which is involved in posttranslational modifications, apoptosis inhibition, and proliferation of HPVimmortalized cells. 89,90 Consistently, upon viral infection, IPA predicted increased synthesis and metabolism of proteins in conjunction with elevated cell viability and reduced cell death ( Figure 5F). From a viral perspective, it is beneficial to increase the components required for protein synthesis, also including eukaryotic translation initiation factors of which some were upregulated in MnPV-induced tumors (Supporting Information: STable 30). Consequently, the most upregulated pathway seems to be the EIF2 pathway ( Figure 5F) that regulates both global and specific messenger RNA (mRNA) translation 91 and is subverted by HPV18 E6 to inhibit the expression of proapoptotic genes. 92 Interestingly, one of the kinases involved in this pathway also directly phosphorylates transcription factor NRF2 (the product of Nfe2l2) in response to environmental stress 91 and NRF2 was predicted to be activated in MnPV-induced tumors in both datasets ( Figures 3C and 5F). Its activation increases the expression of cytoprotective proteins, 44,93 thereby preventing DNA damage and virus integration into the host genome, while its deactivation could cause genome instability. 94 So far, cutaneous PVs are not known to integrate into the host genome and since NRF2 also inhibits the NLRP3 inflammasome, 95 its activation may be a strategy to maintain viral episomes.
IPA prediction of upstream regulators indicated a strongly reduced activity of RNA-binding protein LARP1 (La-related protein 1) ( Figure 5F) and RICTOR ( Figures 3C and 5F) upon viral infection.
Both are regulatory key components of the mTOR pathway that controls cell proliferation, migration, and tumor growth. 96,97 This suggests a deactivated mTOR pathway in benign MnPV-induced tumors, in contrast to its activation in dedifferentiated SCC areas ( Figure 3B) where it could promote tumor growth, invasion, and metastasis. 42,43 Although MnPV-induced benign lesions showed strong keratinization (Figures 1, 4, and 6), particularly K10 was reduced (Supporting Information: STables 3 and 30). K10 inhibits cell-cycle progression and thus may directly contribute to terminal keratinocyte differentiation. 98 Loss of suprabasal K10 causes increased proliferation of basal keratinocytes and accelerates their migration to the skin surface. 99 Since basal keratinocytes represent the reservoir of cutaneous PVs 100 and virion assembly occurs in the uppermost epithelial layers, K10 downregulation seems to be beneficial for the viral life cycle. This assumption is in line with an IPA-predicted reduced activity of ROCK1 ( Figure 5F), which promotes keratinocyte terminal differentiation. 101 Conversely, ROCK1 inhibition maintains stemness characteristics of primary hair follicle stem cells, 102 blocks senescence and differentiation, and induces indefinite proliferation of keratinocytes. 103,104 Terminal differentiation also involves the activity of CASP14. 49 As already shown for HPV8, disruption of this process causes an accumulation of unprocessed pro-caspase-14 levels, 105 which was also observed in the case of MnPV-infection ( Figures 3D, 5E, and 6). Interestingly, skin with higher CASP14 levels is also better protected from UVB radiation. 106 Consistent with the suppression of the CEBPA/microRNA-203 (miR-203) pathway in HPV8-induced lesions, a key pathway of epidermal differentiation and proliferation, 107 IPA also predicted an MnPV-dependent deactivation of CEBPA ( Figure 5F). Several differentially (de)activated upstream microRNA (miRNAs) were also predicted in our study, for example, activated miR-let-7a-5p (Supporting Information: S- Table 37-39), which can repress migration, invasion, and EMT. 108 Thus, a major aim of MnPV seems to be to maintain host cell proliferation keeping a certain degree of differentiation, thereby creating a favorable environment for virus replication and spread.
In summary, our comprehensive in vivo study identified a whole spectrum of interesting proteins involved in PV-induced skin carcinogenesis and progression of cutaneous SCCs. It revealed an opposite regulation of several molecular pathways in well-and dedifferentiated SCCs and increased stem cell characteristics in the latter, ultimately causing the diverging phenotypes of both tumor entities. Consequently, these data offer a basis to functionally dissect the hit-and-run mechanism of HPV-induced carcinogenesis. The more we understand proteomic networks leading to epithelial SCCs, the more we learn how to efficiently interfere by therapeutic means with one of the most frequent cancers worldwide.