The Pro le of Expression of SG2NAs Differs Between Cancer Types and it is Involved in the Reprogramming of Tumour Cell Proteome


 Striatin and SG2NA are scaffold proteins that from signalling complexes called STRIPAK. It has been associated with cancer and other diseases. Our earlier studies have shown that SG2NA forms a complex with the cancer associate protein DJ-1 and signalling kinase Akt, promoting cancer cell survival. In the present study, we used bioinformatics analyses to confirm the existence of two isoforms of human SG2NA i.e., 78 and 87 kDas. In addition, several smaller isoforms like 35 kDa were also seen in western blot analyses of human cell lysates. The expression of these isoforms varies between different human cancer cell lines. Also, the protein level does not corroborate with its transcript level, suggesting a complex regulation of its expression. In breast tumour tissues, the expression of the 35 and 78 kDa isoforms was higher as compared to the adjacent normal tissues, while the 87 kDa isoform was detected in the breast tumour tissues only. With the progression of stages of breast cancer, the expression of 78 kDa isoform decreased, while 87 kDa became undetectable. In coimmunoprecipitation assay, the profile of SG2NA interactome in breast tumor vis-à-vis adjacent normal breast tissues shows hundreds of common proteins, while some proteins specifically interacted in breast tumour tissue only. We conclude that SG2NA is involve in diverse cellular pathways and have roles in cellular reprogramming during tumorigenesis.


Introduction
Striatin is the prototype member of a small subfamily of WD-40 repeat superfamily of proteins [1]. The other two members are SG2NA and Zinedin (also called Striatin 3 and 4 respectively). Besides the signature WD-40 repeats in their C-terminus, they are also characterized by one each of caveolin-binding, coiled-coil, and calmodulin-binding domains that are involved in protein-protein interactions [2]. Striatininteracting phosphatase and kinase (STRIPAK) complexes are a group of multi-protein assemblies held together by the Striatin proteins as the scaffolds [3]. STRIPAK complexes are conserved in evolution, and they regulate diverse cellular processes viz., vesicular tra cking, Golgi assembly, Hippo signalling, autophagy, cell migration, cell cycle, cell differentiation, metabolism, and programmed cell death [3].
Dysfunctional STRIPAK assemblies have been associated with various developmental and degenerative diseases including cancer [4][5][6]. The protein phosphatase 2A (PP2A), a major regulator of the cellular phosphoproteome, is a dimer of the subunits A & C, that binds to the regulatory subunit B. There are four structurally distinct variants of the B subunit. Striatin & SG2NA are the organizers of the subunit B′′′ in which they assemble several kinases and other proteins [7,8].
Tumour progression and metastasis involves aberrant kinase signalling and cytoskeletal reorganization; and STRIPAKs have roles in both of these processes [4,6]. Hippo signalling regulates organ size and tissue homeostasis, and dysfunctional Hippo signalling has often been associated with cancer. Serinethreonine kinases MST1 & 4 are the constituents of STRIPAK and in association with MOB, it contributes to the Hippo signalling [9,10]. In prostate and hepatocellular carcinoma, increased expression of MST4 has been reported. It enhances the ERK signalling, and induces EMT [11]. MST3/4 are involved in the phosphorylation of Ezrin, Radixin, and Moesin (ERM proteins) that facilitate cell migration. Striatin-interacting protein STRNIP2 antagonizes the functions of MST3/4 [12]. STRIPAK complexes are also involved in cytoskeletal reorganization in cancer cells. Actin-binding protein cortactin facilitates cancer cell invasion. Striatin binds to cortactin-binding protein CTTNBP2, suggesting its role tumorigenesis [13].
SG2NA, the second member of the Striatin family was initially isolated as a novel auto-antigen in the sera of a lung and bladder cancer patient [14]. Mouse SG2NA has at least six variants generated by alternate splicing, intron retention and RNA editing [15]. Variants of SG2NA are conserved in evolution, have similar but distinctive structural characteristics and are functionally related [16,17]. Unlike Striatin that primarily expresses in the striatum region of the brain (and named accordingly), expression of SG2NA is more ubiquitous [15,18,19]. Our recent studies have suggested that SG2NAs have roles in certain cancer associated events like cytoskeletal reorganization, membrane sialylation, and modulation of the markers of EMT [20,21]. Variants of SG2NA associate with the anti-oxidant protein DJ-1 and protects neuronal and cancer cells from oxidative stress [19,22]. SG2NA also regulates ER homeostasis and the expression of 78 kDa SG2NA is modulated with the progression of cell cycle [23,24]. In the present study, we have examined whether the pro le of expression of SG2NA isoforms in cancer cell lines and biopsies can be related to the cancer type. Our study suggests that the expression of SG2NA is extensively modulated in several cancer cell lines and tissues; and it is involved in the reprogramming of the cellular proteome towards cancer phenotypes.
Materials And Methods 1. Cell culture Different human cancer cell lines were procured from NCCS, Pune, India. Cells were cultured as a monolayer in 10% FBS containing DMEM with antibiotic penicillin, amphotericin B and streptomycin in a humidi ed, 5% CO 2 containing incubator at 37°C.

RNA extraction and RT-PCR
Total RNA was extracted from different cancer cell lines using TRI reagent. First strand cDNA was synthesized from 1 µg of RNA with reverse transcriptase (Epicentre) and oligo (dT)18 primer (Fermentas) at 37°C for 2 h. PCR was done using isoform speci c primer pairs.

Page 4/29
Estimation of protein concentration was done by modi ed Bradford method. Equal quantities of protein sample from each cell lines/tissue (70 µg) were resolved on 10% SDS-PAGE. The separated proteins were then transferred on PVDF membrane in Towbin's buffer (25 mM Tris, 192 mM glycine and 20% methanol).
Membrane blocking was done for 2-3 h at room temperature in 5% BSA in 0.05% TBST buffer; then incubated overnight with primary antibody of SG2NA at 4°C. Immunodetection was done using HRPconjugated mouse secondary antibody.

Southern blotting
PCR product was resolved on 1.5% agarose gel in 1X TAE buffer and immobilized on a nylon membrane. Prehybridization followed by probing with labelled Sg2na cDNA probe (double stranded, random labelled, 10 6 cpm/ml) overnight at 55°C. Filters were nally washed twice and developed.

Stability studies on Sg2na transcript
Transcription was inhibited by the treatment with actinomycin D (10 µg/ml). Cells were then harvested at different time points and RNA was extracted using TRI reagent, converted to cDNA. Sg2na transcripts were assayed by real time PCR using 1× SYBR Green, Sg2na ORF speci c primers and 1 µl of cDNA. β-Actin transcript was used as an internal control.

Human Biospecimen
Human tissues used in the present study were collected from Acharya Harihar Regional Cancer Centre (AHRCC), Cuttack, from December 2018 to July 2019. An observational pilot study was done from the hospital data of patients with carcinoma of breast, colon and stomach. All patients were con rmed histopathologically and cases of primary tumors are included in this study. Altogether, 16 samples including 10, 3 and 3 patients of breast, colon and stomach carcinoma respectively were considered for the study. All the patients underwent surgery in AHRCC, Cuttack, followed by adjuvant treatment as indicated from postoperative histopathological features. The resected tissues were immediately collected and preserved immediately in liquid nitrogen for protein analysis. In this study respective clinical parameters like age, sex, clinical presentation, type of surgery, nal detailed histopathology, TNM Staging and Stage Grouping were retrieved from hospital records (Supplemental Table S4). The nal analysis of the study included comparison of the pro le of SG2NA variants between adjacent normal and tumor tissues in all the cancer specimens.

Co-immunoprecipitation
The Protein G agarose beads were washed with 1X PBS on a rotary shaker for 5 mins followed by centrifugation at 13000g for 30 sec at room temperature. The washed beads were then equilibrated with lter sterile IP lysis buffer (20 mM Tris HCl pH 8, 137 mM NaCl and 1% Triton X-100, 2mM EDTA). The swollen beads were then resuspended in a certain volume of lysis buffer. Breast tissue lysates (both tumor and control) were prepared by incubating in IP lysis buffer, supplemented with protein inhibitor cocktail and kept in ice for 1 hr followed by centrifuging at 12000g for 15 minutes. One milligram of each tissue lysate were added to the equilibrated beads and incubated for 2 hr at 4°C on rotatory shaker. Each lysate-beads mix was then centrifuged for 5 mins at 2000 rpm and supernatant were transferred into fresh tube. Each supernatant was added to the antibody crosslinked beads, mixed gently, followed by overnight incubation at 4°C on a rotatory shaker. Each supernatant was then removed by spinning at 13000 rpm for 30 sec at 4°C and the agarose pellet were resuspended in washing buffer (20 mM Tris HCl pH 7.4, 150 mM NaCl, 2 mM EDTA, 1% Triton X-100, 2mM EGTA,0.2mM sodium orthovanadate) for 5 mins followed by centrifugation. This step was repeated ve times with last wash in the same buffer without detergent. The bead bound immunocomplex were then resuspended in low pH Glycine buffer (0.1 M Glycine, pH 2) for 15 mins at 4°C. Each supernatant was then collected by centrifuging at 13000 g for 30 sec followed by neutralizing the protein mix by the adding 1.5 M Tris HCl (pH 8.8). This supernatant (pulled down proteins) was then collected for LC-MS analysis.

Sample Preparation for LC-MS/MS
25ul of sample was reduced with 5mM TCEP and further alkylated with 50mM iodoacetamide and then digested with Trypsin (1:50, Trypsin/lysate ratio) for 16 hr at 37 o C. Digests were cleaned using a C18 silica cartridge to remove the salt and dried using speed vac. The dried pellet was resuspended in buffer A (5% acetonitrile, 0.1 formic acid).

Mass Spectrometric Analysis of Peptide Mixtures
All the experiments were performed using EASY-nLC 1200 system (Thermo Fisher Scienti c) coupled to QExactive mass spectrometer (Thermo Fisher Scienti c) equipped with Nano electrospray ion source. One microgram of the peptide mixture was resolved using a 25 cm Pico Frit column (360µm outer diameter, 75µm inner diameter, 10µm tip) lled with 1.9 µm of C18-resin (Dr Maeisch, Germany). The peptides were loaded with buffer A (5% acetonitrile, 0.1% formic acid) and eluted with a 0-40% gradient of buffer B (95% acetonitrile, 0.1% formic acid) at a ow rate of 300 nl/min for 100 min. MS data was acquired using a data-dependent top10 method dynamically choosing the most abundant precursor ions from the survey scan.

LC-MS/MS Data Processing
Samples were processed and RAW les generated were analysed with Proteome Discoverer (v2.2) against the Human UniProt reference proteome database. For Sequest and Amanda search, the precursor and fragment mass tolerances were set at 10 ppm and 0.02 Da, respectively. The protease used to generate peptides, i.e., enzyme speci city was set for trypsin/P (cleavage at the C terminus of "K/R: unless followed by "P") along with maximum missed cleavages value of two. Carbamidomethyl on cysteine as xed modi cation and oxidation of methionine and N-terminal acetylation were considered as variable modi cations for database search. Both peptide spectrum match and protein false discovery rate were set to 0.01 FDR.

Bioinformatics analysis:
The protein list of interacting partners of SG2NA from co-immunoprecipitation in control and breast tumor tissue were analysed by BioVenn (biovenn.nl) comparison tool [25]. The protein-protein interaction network was analysed by STRING (Search Tool for the Retrieval of Interacting Genes/Proteins) tool.

Results
Identi cation of the human variants of SG2NA: We have reported earlier that in mouse, there are at least six isoforms of SG2NA (87, 82, 78, 52 and 38 and 35 kDas) arising out of alternative splicing, intron retention and transcript editing [15,18] Although SG2NA is highly conserved in evolution [17], its variants in human are not known. We analysed the EST and NCBI database (https://blast.ncbi.nlm.nih.gov/Blast.cgi) which predicted four transcript variants as shown in Fig. 1A. To validate the EST database, total RNA was isolated from HEK293T cells and RT-PCR was performed using primer pair anking the intron region (forward, exon 5 and reverse, exon 15). In RT-PCR we obtained 2 bands of length around 1000-1200 bp (Fig. 1B). To con rm the identity of these bands, southern blotting was performed using full length human cDNA for sg2na (www.origene.com) as the probe. As shown in Fig. 1C, two distinct bands corresponding to 78 and 87 kDa variants (derived from the size of the RT-PCR products) were seen in the autoradiogram.
Further analysis was performed using UniProtKB/Swiss-Prot, Gene cards, and NCBI databases (https://www.uniprot.org/, https://www.expasy.org/resources/uniprotkb-swiss-prot, https://www. genecards.org/, https://www.ncbi.nlm.nih.gov/gene). Only two isoforms viz., 87 and 78 kDa were found at the protein level. Finally, analyses using Ensemble (https://asia.ensembl.org/ index.html) and Gepia2 software (http://gepia2.cancer-pku.cn/#index), eight isoforms were mapped, out of which protein level expression were found for six isoforms (Table 1). One isoform i.e., STRN3-207 (332 aa) was found in the Uniport and Ensemble database but it was predicted to undergo nonsense mediated decay [28,29], hence it was not available in the EST database of NCBI. Three smaller isoforms i.e., STRN3-205, STRN3-206 and STRN3-208 appeared to be the fragments of the 78 and 87 kDa isoforms which might be formed by the proteolytic cleavage. Some other smaller isoforms (like the 35 and 52 kDa isoforms mouse) might be present in Human, but they could not be predicted by the bioinformatics analyses. As evident from all analyses, we inferred that the two isoforms viz; 78 and 87 kDa are expressed in Human tissues at the protein level. Expression pro le of SG2NA variants differ among cancer cell lines Earlier we have reported that in several cancer cell lines with increased levels of ROS; SG2NA recruits DJ-1 and phospho-Akt to the plasma membrane, promoting their survival and growth [19,22]. To further analyse the role of SG2NA in cancer, we checked its expression in a number of well-studied human cell lines ( Table 2). We rst estimated the transcript level of sg2na in those cell lines. As shown in Fig. 2A, the level of sg2na mRNA was low in HepG2, H1293, and HeLa cells, moderate in A549 cells, and high in DU145 and HEK293T cells. Western analyses of the extracts from those cells however showed a different pattern (Fig. 2B). In HepG2 cells, the expression of 78kDa SG2NA was quite high and that of 87 kDa was low but detectable. In HeLa, DU145 and Hep3B cells, the expression of both 87 and 78 kDas was very low but detectable. In HEK293T, H1299 and A549 no expression was detected. Therefore, there were major incompatibilities between the mRNA and protein levels In HepG2 cells, the transcript level was the lowest, but the protein level was the highest. Contrastingly, in HEK293 and A549 cells, while the transcript levels were high, there were hardly any expression of the protein. To address this issue, we measured the stability of sg2na transcript in HEK293T, A549 and HepG2 cells. Cells were treated with actinomycin D, an inhibitor of transcription; and harvested at different time points post treatment. Total RNA was isolated and sg2na mRNA level was estimated by qRT-PCR. As shown in Fig. 2C, in all the cell lines tested, sg2na transcript level decreased and reached to ~40-60% of the original level by 8 hours. However, it was more stable in A549 cells as the rate of decrease was < 25% in rst 4 hours. This corroborates with the observation that the level of sg2na mRNA was higher in A549 cells than in HepG2 cells. Such posttranscriptional decay thus showed a complex mode of regulation of SG2NA at both mRNA and protein levels in different cell lines.

HEK293T
Derived from human embryonic kidney 293 cells and contains the SV40 T-antigen.

A549
Generated through explant culture of lung carcinomatous tissue from a 58-year-old Caucasian male.

H1299
Established from the lymph node metastasis of the lung from a patient.

HepG2
Derived from a liver hepatocellular carcinoma of a 15-year-old Caucasian male.

Hep3b
Hepatoma cell line derived from the biopsies taken during lobectomies of an 8 year old black male.

DU145
Human prostate cancer cell line derived from a 69-year-old white male.

HeLa
First immortalized cell line to be developed from a 30-year-old black female suffering from aggressive cervical cancer.
Breast, gastric/stomach, and colon cancer tissues show distinctive expression pro le of SG2NAs Invasive and in ltrating ductal carcinoma are the most common form of breast cancer. We analysed four samples along with their normal counterparts (from the same patient collected from the adjoining region) by western blotting. As shown in Fig. 3, in all four samples, higher expression of 78 kDa and 35 kDa isoforms (not identi ed by bioinformatics analysis but it is predicted as it is present in mouse) of SG2NA were seen in tumour tissues as compared to normal counterparts. Noticeably, the 87 kDa isoform is expressed in tumour only. It is also found that with the increase in the stages of cancer, the expression of 78 kDa isoform decreases and that of 87 kDa isoform almost disappears in the 3rd stage. Much change was not seen in case of 35 kDa isoform. Details of the patients and the results is given in Table 3. Adenocarcinoma is the most common type of stomach cancer comprising about 90% of all cases. It originates in the mucosa cells present in the innermost layer of the stomach that produce the mucus. We collected three samples and the details are given in Table 4. In the cases of the stomach carcinoma, each patient showed a different pattern of expression of SG2NAs (Fig. 3). It might be due to the heterogeneity of stomach tissues and the associated subtypes of adenocarcinoma. We also analysed the pro le of SG2NA in three cases of colon cancer. However, unlike in the cases of breast and stomach, in the case of colon (both diseased and normal counter parts), the expression of SG2NA was quite low (Fig. 3) and no speci c pattern was evident.  The interactome of SG2NA in breast tumour and its normal counterparts partially overlap As described above, in the breast cancer tissues there were modulations in the expression of the variants of SG2NA, although no de nitive pattern was evident. Since SG2NAs are scaffold proteins, to probe into their potential functions in cancer, we looked into their interacting partners. Lysates were prepared from normal and breast tumour tissues from the same patient, coimmunoprecipitated with the SG2NA antibody, followed by LC-MS/MS. Total 215 and 291 proteins were immunoprecipitated from the normal and the tumour tissues respectively. Among those, 187 proteins were common between both, 104 were found only in tumour tissue and 28 were found only in the normal counterpart (details given in Supplemental Table S1 and Fig. 4). In the functional enrichment analysis for KEGG pathways, we found that the interacting partners of SG2NA present in both normal and tumour tissues are involved in diverse biological functions including metabolism; cell structure; protein synthesis and processing; neurodegenerative, cardiovascular and infectious diseases; cell signalling; etc. (Supplemental Table S2).
Similar analyses of the interacting partners of SG2NA found only in tumour tissues also showed their involvement in diverse pathways but those were not directly related to breast cancer (except oestrogen signalling, Supplemental Table S2). Upon comparison, we found that (i) Many interacting proteins common between normal and tumour tissues are involved in cancer related pathways viz., HIF-1 signalling, Tight/Gap junction, Pentose phosphate pathway, function of the ER etc., though they were not exclusive for the tumour sample (Supplemental Table S2). (ii) In the pathways that were common between the normal & tumour samples vis-à-vis the tumour only samples, certain proteins were found in the tumour samples only, but they were the isoforms of those found in the common pool (e.g., while ENO-1 and PGK-1 were present in the common pool; ENO-2/3 and PGK-2 were found in the tumour only pool; Supplemental Table S2). (iii) There were eighteen pathways that were exclusive for the tumour tissues but none except spliceosome, oestrogen signalling, and viral oncogenesis have been directly related to cancer (Supplemental Table S3). We then analysed the functions of 104 proteins that were found only in tumor samples and as shown in Table 5, most of them have been reported as the prognostic markers for various types of cancers. The signi cance of this analyses is discussed in the following section.

Discussion
With the advent of tools of genomics and proteomics, a surprising level of heterogeneity has been found in malignant tumours. It is now widely accepted that solid tumours often comprise of distinct subpopulations of transformed cells with unique physiology [30]. Nevertheless, certain characteristics viz., dedifferentiation, metabolic reprograming, attainment of unlimited proliferation potential, lack of response to the growth-inhibitory signals, evasion of apoptosis, tissue invasion etc.; make cancer cells highly distinguishable from their normal counterparts. Despite signi cant advancements in understanding the biology of cancer, due to such complexities; identi cation of novel therapeutic targets remains a challenge [31].
In recent years, Striatin has gained signi cant attention for its role in cancer in general and in hippo signalling in particular [4][5][6]. The role of its paralogue SG2NA in the context of cancer or any other diseases is rather obscure. Unlike Striatin, that has higher level of expression in the striatum region of the brain (and named accordingly), the expression of SG2NA is ubiquitous. It also has several splice variants with complex subcellular distributions [15,20]. Therefore, the interacting partners of SG2NA are likely to be highly diverse. In agreement, by co-immunoprecipitation and blue-native PAGE analyses, we have recently found that in rat midbrain extract SG2NA has ~ 200 interacting partners and three fourth of which also interact with the cancer associate protein DJ-1 (Manuscript under review in BBA Proteins and Proteomics). Proteins that interact with SG2NA are involved in a plethora of cellular pathways including metabolism, mitochondrial function, cell signalling etc. Since DJ-1 has been associated with the bladder, prostate, colorectal and gastric cancers etc. [32,33], it is anticipated that SG2NA and some of its associated proteins would also be involved in cancer. Such notion is further strengthened by our recent observation that overexpression of 35 kDa SG2NA and knocking down the 78 kDa in NIH3T3 cells induces certain markers of epithelial to mesenchymal transition [21]. To explore the possible role of SG2NAs in tumour progression, we rst tested its expression pro le in several well-established cancer cell lines. It appears from the results that rather than having an exclusive pro le of expression in those cells, it was highly variable among the cell lines originating from different types of tumours. Further, in several cell lines, there were major disconnect between the mRNA and protein levels, suggesting a dynamic regulation of its expression. Similarly, its expression pro le not only varied among three different cancer tissues we had tested, it also varied between the stages of breast cancers. Considering the large number of interacting partners of SG2NA (seen in mid-brain), it thus appears that rather than being involved in a few selective pathways, it might be dynamically engaged in more innate cellular networks. In agreement, coimmunoprecipitation analyses showed that while there are several hundred interacting partners of SG2NA common between both normal and tumour cells, there are also eighty-two proteins of diverse functions that interact with it only in tumour cells. Although these proteins are functionally diverse, we clustered them based on their functions viz., protein synthesis and ribosomal structure (21 proteins); chaperone & heat shock proteins (5 proteins); transporters (5 proteins); GTP binding tubulins (11 proteins); cell signalling (3 proteins); binding to DNA/RNA (5 proteins); glycolysis (7 proteins); complement pathway (5 proteins); structural proteins (5 proteins) etc. Since several of these proteins are part of multimeric assemblies like ribosome and microtubules, some might have been immunoprecipitated along with others which directly interact with SG2NA. Therefore, even if we accept that all those eighty-two proteins are not necessarily the unique interacting partners of SG2NA, most turned out to be the prognostic markers of various types of cancers. It thus strongly suggests that SG2NA is involved in modulating a plethora of cellular pathways, and while transiting from normal to cancer phenotype; it partially switches its interacting partners to facilitate the reprogramming of the cellular events, tuning it towards tumour progression. Though our inference is based on some preliminary studies, it nonetheless advocates nuanced role of SG2NA in normal cell function as well as tumorigenesis. Reverse). The ampli ed product was resolved on 1% agarose. Two bands obtained whose length corroborated with the 78 and 87 kDa isoforms. (C) The ampli ed products were con rmed by southern blotting using human SG2NA clone as probe.   proteins were found common between two datasets. Venn diagram showing SG2NA interacting proteins as identi ed by the Co-immunoprecipitation (details given in S1A).

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.