Determining KLF14 tertiary structure and diagnostic significance in brain cancer progression

Expression analysis of new protein targets may play a crucial role in the early detection and diagnosis of brain tumor progression. The study aimed to investigate the possible relation of KLF14, TPD52, miR-124, and PKCε in the development and progression of brain cancer and space occupying lesion (SOL) of the brain. One hundred human blood samples comprising varying diagnostic groups (SOL brain, grade I, II, III, IV) were analyzed by real-time quantitative PCR to determine the expression level of KLF14, TPD52, miR-124, and PKCε. TPD52 and PKCε were upregulated in brain cancer by 2.5- and 1.6-fold, respectively, whereas, KLF14 and miR-124 were downregulated in brain cancer. In metastatic and high-grade brain cancer, TPD52 and PKCε expression were up-regulated and KLF14 and miR-124 expression were down-regulated. Further, these genes were found to be differentially expressed in the blood of patients with SOL. Upregulation of TPD52 and PKCε, however, reduced expression of KLF14 and miR-124 in SOL of the brain as compared to healthy controls. Expression analysis of TPD52, KLF14, miR-124, and PKCε provided useful information on the differences existing between the normal brain and SOL, in addition to gliomas; thus, might prove to be useful having diagnostic or prognostic value.

www.nature.com/scientificreports/ sidered statistically significant when P-value was less than 0.05. ROC curve analysis was employed via Graphpad prism and Area under the curve (AUC) along with 95% confidence interval score was determined.
Construction of 3D structure for KLF14. 3D structure of KLF14 was determined by using an insilico approach. The amino acid sequence of KLF14 (gene ID: 136,259) was retrieved from NCBI. The sequence used was in FASTA format. The subcellular localization of KLF14 was predicted using DeepLoc-1.0 46 , Hum-mPLoc 3.0 47 and PSORT 48 . For multiple sequence alignment, all amino acid sequences of KLF family (1)(2)(3)(4)(5)(6)(7)(8)(9)(10)(11)(12)(13)(14)(15)(16) were collected from NCBI in FASTA format and aligned by using "ClustalW" 49 . The Aligned sequences were analysed to find out the conserved domains of all KLF members. Web-based tools TMHMM 2.0 50 , TmPred 2.0 51 and HHpred 52 were used to predict transmembrane domains of KLF14. Various tools such as Spider 2 53 , PSIPRED 54 , I-TASSER 55 and PSSPRED were employed to predict the secondary structure of KLF14. Phylogenetic analysis of the KLF proteins were performed using Mega X 56 . KLF4 structure was used as a template for homology modelling. The crystal structure of KLF4 (2wbu) was retrieved from Protein Data Bank RCSB 57 . Homology modelling server of Swiss-Model Workspace 58 was used to construct the 3D structure. Three dimensional structures were envisaged by using Chimera 59 .  www.nature.com/scientificreports/ Pathway construction. To construct a cellular pathway and establish a crosstalk between understudied genes, KEGG and String analyses were performed, and the pathway was built via DAVID software.

Ethics approval and consent to participate
The experimental protocol for the use of Human was approved (Ref: No: IRB-110) by the ethical committee of Combined Military Hospital and ASAB, NUST. Informed consent was taken from all participants of the study.

Results
Expression of TPD52, PKCε, KLF14 and miRNA-124 in blood of brain cancer patients. This study utilized 80 patient blood samples from different brain cancer subtypes, including astrocytoma, oligodendroglioma and glioblastoma and 100 blood samples from healthy control individuals. Gene expression analysis of the brain cancer samples by qRT-PCR revealed that there was 2.5-fold higher expression of TPD52 in blood of brain cancer patients as compared to healthy controls ( Fig. 1A, P = 0.0015). Elevated PKCε expression (1.5-fold) was also observed in cancer patients as compared to healthy controls ( Fig. 1B, P = 0.0046). KLF14 mRNA levels were 50% lower in cancer patients as compared to healthy controls ( Fig. 1C, P = 0.0025). A striking reduction in miR-124 level was observed in the blood of brain cancer patients with expression decreased to 0.002 of healthy controls ( Fig. 1D, P = 0.001). Diagnostic significance of the expression of studied genes in brain tumor was determined through receiver operating characteristic (ROC) curve analysis (Fig. 1E-H). The analysis revealed that the value for KLF14 and miR-124 in predicting risk of brain tumor is high with Area under the Curve (AUC) 0.86 and 0.84, respectively. The value for TPD52 and PKCε also depicted their significance in predicting the brain tumor risk (AUC 0.62 and 0.56, respectively).

Figure 1.
Expression and diagnostic significance of TPD52, PKCε, KLF14 and miRNA 124. Expression of (A) TPD52 (B) PKCε (C) KLF14 and (D) miR-124 in brain tumor patients compared to healthy controls. There is up-regulated expression of TPD52, and PKCε (P = 0.0015, P = 0.0046 respectively) and down-regulated expression of KLF14 and miRNA124 (P = 0.0025, P = 0.0010 respectively) in brain tumor patients compared to healthy controls. Representative data were presented as mean ± SEM of triplicate experiments and **refers to P value less than 0.05 and ***refers to P value less than or equal to 0.001. Statistical significance was measured by two-way ANOVA. ROC curve for TPD52, PKCε, KLF14 and miR-124 predicted high risk for brain tumor. Area under the ROC curve (AUC) for (E) TPD 52 was 0. 69 (Table 2). These genes showed an association with tumour grade (P < 0.0001). In the advanced grade group, there was higher expression of TPD52 and PKCε as compared to lower grade group ( Fig. 2A, B, respectively). KLF14 and miR-124 showed low expression in advanced grade groups as compared to lower grade groups (Fig. 2C, D respectively). Differential expression of these understudied genes in relation to primary and secondary metastatic status of brain cancer was also investigated ( Table 2). TPD52 and PKCε were observed to be highly expressed in the secondary metastatic group as compared to the primary metastatic group of brain cancer patients (P < 0.0001) ( Fig. 3A and Brespectively). Lower expression of KLF14 (Fig. 3C) and miR-124 ( Fig. 3D) was observed in the secondary metastatic group compared to the primary metastatic group (P = 0.0001, P < 0.0001 respectively).
TPD52, miRNA-124, PKCε and KLF14 were evaluated for altered expression levels separately in male and female patient groups ( Table 3). The expression of TPD52 and PKCε was higher while miR-124 and KLF14 was www.nature.com/scientificreports/ lower in both males and females in comparison to the control group (Fig. 4). However, the expression of TPD52 and PKCε were up-regulated in females in comparison to males (P = 0.0002 and P = 0.0003, respectively). KLF14 expression (P = 0.0027) was comparatively lower in males while miR-124 (P < 0.0001) expression was comparatively lower in females.
Expression of TPD52, PKCε, KLF14 and miRNA-124 in space occupying lesion (SOL) of the brain. Twenty blood samples from patients with space occupying lesions (SOL) of brain were collected and analysed for expression of TPD52, miR-124 PKCε and KLF14 (Table 3). These genes were found to be differentially expressed. There was elevated expression of TPD52 and PKCε compared to healthy control (p < 0.0001). The expression of TPD52 and PKCε was lower in SOL patients in comparison to patients with high grade cancer (Fig. 5A, B, respectively). KLF14 and miR-124 mRNA levels were decreased in SOL samples relative to healthy control (p < 0.0001), but to a lesser extent than the reductions in high grade cancer (Fig. 5C, D, respectively).
Expression of TPD52, PKCε, KLF14 and miRNA 124 in astrocytoma, glioblastomas, and oligodendrogliomas. Differential expression analysis of TPD52, PKCε, KLF14 and miRNA 124 in brain cancer subtypes: astrocytomas, glioblastomas, and oligodendrogliomas revealed that the expression of TPD52 and PKCε was up regulated in glioblastomas compared to astrocytomas and oligodendrogliomas ( Fig. 6A and B). However, TPD52 expression was lower comparative to glioblastomas, and oligodendrogliomas but significantly Figure 3. Expression of TPD52, PKCε, KLF14 and miRNA124 in primary and secondary groups of brain cancer. (A) Expression of TPD52 in metastatic and non-metastatic groups of brain cancer patients compared to healthy controls; (B) Expression of PKCε in metastatic and non-metastatic groups of brain cancer patients compared to healthy controls; (C) Expression of KLF14 in metastatic and non-metastatic groups of brain cancer patients compared to healthy controls; (D) Expression of miRNA 124 in metastatic and non-metastatic groups of brain cancer patients compared to healthy controls. There is an elevated expression of TPD52 and PKCε in metastatic group compared to non-metastatic group (P < 0.0001). KLF14 and miRNA124 are low in expression in metastatic group compared to non-metastatic group (P = 0.0005, P < 0.0001 respectively). Representative data were presented as mean ± SEM of triplicate experiments.). Statistical significance was measured by one-way ANOVA (****P < 0.0001 and ***P < 0.05). www.nature.com/scientificreports/ up-regulated relative to healthy control. In comparison to astrocytomas and glioblastomas, the expression of PKCε was lower in oligodendrogliomas but elevate comparative to healthy individual. The expression of KLF14 and miR-124 was downregulated when compared to control. However, KLF14 expression was higher in astrocytomas relative to glioblastomas, and oligodendrogliomas ( Fig. 6C) while expression of miR-124 was higher in astrocytomas and oligodendrogliomas compared to glioblastomas (Fig. 6D).
Identification of pathways instituting crosstalk between TPD52, KLF 14, PKCε, and miR-124. Several analysis tools were employed to determine potential pathways involving our genes of interest TPD52, PKCε, KLF14 and miR-124. Using KEGG and String, the output data indicated that these genes are linked with each other and are involved in the Akt pathway. Further pathway information obtained via DAVID software showed that PKCε is found upstream of the Ras/Raf pathway and regulates signal transduction from G-Protein-Coupled Receptor (GPCR) to KRas. A summary compilation of these analyses is shown in Fig. 7.

Multiple sequence alignment including KLF14.
To provide further understanding of KLF14, multiple sequence alignments were carried out among the members of the KLF family. Figure 8 shows the Clustal Omega sequence alignment of the fifteen Kruppel-like factors, which depicts the three conserved Zinc finger domains across the KLF family. The Zinc finger domain is the most important feature of Kruppel-like factors. It contains three classical Cys2-His2 zinc fingers and is observed to be conserved in all the known KLFs. Moreover, alignment of amino acid sequences of the zinc fingers reflects a high degree of sequence identities. These zinc finger domains interact with CAC CCC elements and GC-rich regions of DNA to initiate activation and repression of transcription 60 .
Phylogenetic analysis. Phylogenetic deduction is an important aspect in functionally analysing a gene family. Full-length amino acid sequences encoding KLF proteins were used to construct UPGMA tree ( 3D structure of KLF14. Structure of KLF14 (NCBI ID: NP_619638.2) (Fig. 10A) obtained by Swiss model was found to be 95.06% favoured by Ramachandran plot (Fig. 10B). The three-dimensional structure was constructed by using KLF4 (protein id: 2wbu) as a template retrieved from RCSB PDB that undertakes 61.45% sequence identity of KLF14 with KLF4. Three-dimensional (3D) structure of KLF14 (Fig. 10A) has three zincfinger domains near the C-terminus, all three are of classical C2H2 type. The dihedral angles (phi against psi) of amino acids that have particular role in secondary structure are visualized by Ramachandran plot (Fig. 10B).
MolProbity score was 1.23 and Ramachandran Outliers were 2.47%. However, some bad angles and bonds are formed as a result of derangements. In total, 1/719 bad bond and 19/965 bad angles were identified in the structure. In the upper right side, loops and helices are allowed, however, the upper left region favours beta sheets. The lower left region favours alpha helices, and left lower part disfavours the protein structure (Fig. 10B). Superimposition results (Fig. 10C) revealed 61.45% identity between the constructed structure of KLF14 (green) and the template KLF4 (red). Localization of KLF14. Intracellular localization was determined using DeepLoc 1.0 that predicted KLF14 cellular localization inside the nucleus. This prediction supports KLF14 function as transcriptional regulator (Fig. 11B).

Discussion
Brain cancer has been classified as the fatal cancer type that has relatively shorter survival rate Its increasing incidence has also become a source of concern 63,64 . Gene expression dysregulation plays an important role in the development of brain tumours. For example, elevated risk of sub ependymal giant cell astrocytoma is brought about due to a single gene disorder associated to the "p" arm of chromosome 9 56 . Therefore, it is necessary to investigate genes that may be involved in the progression of brain tumours in order to identify targets for (D) Expression of miRNA 124 in male and female patients compared to healthy male and female control, respectivelys. There is significant difference of PKCε, TPD52, miRNA124, and KLF14 between male and female patients group and healthy control (P = 0.0003, P = 0.0002, P < 0.0001, P = 0.0027 respectively). Representative data were presented as mean ± SEM of triplicate experiments. Statistical significance was measured by one-way ANOVA (****P < 0.0001, ***P < 0.01, **P < 0.05). www.nature.com/scientificreports/ treatment and diagnosis of the disease. The present study evaluated differential expression of four genes: PKCε, TPD52, KLF14 and miR-124 that are involved in modulation of Akt/PI3K and Ras/Raf/ERK1/2 signaling in different cancers. Previously, co-expression of PKCε, TPD52, and miR-124 with KLF3 was reported in breast cancer 36 . PKCε down-regulation at the transcriptomics level indicated its tumor-suppressive function in breast cancer. In the current study, elevated PKCε expression was found. Expression of miR-124 and TPD52 were shown to be decreased and increased, respectively, in blood of breast cancer patients, a result that is in accordance with the findings of the current study.
KLF14 has an important role in brain functions, including cell proliferation, apoptosis, senescence, angiogenesis, adhesion and migration 65 . To the best of our knowledge, dysregulated expression of KLF14 is not reported in brain cancer so far. This study reported that the expression of KLF14 in brain tumours is down-regulated as compared to healthy controls suggesting its expression is necessary for the normal functioning of brain. The expression of KLF14 is also suppressed in other human cancers, supporting its tumor suppressive role 13 . This study further focuses on the expression of KLF14 in relation with metastatic status and tumor grade. Downregulation of KLF14 was observed in metastatic group, suggesting KLF14 down-regulation might have role in cancer metastasis. KLF14 expression variation was also observed in different brain tumour grades. In lower grade of brain tumours (I + II), its expression is higher than the advanced grade tumours (III + IV), suggesting the continuous reduction of KLF14 expression in brain tumours may lead towards the high-grade cancer. The present study also demonstrated the novel finding of lower expression of KLF14 in space occupying lesion (SOL) of brain compared to healthy controls; however, KLF14 expression in blood of brain cancer patients was lower than that observed in SOL patients. www.nature.com/scientificreports/ Strong scientific evidence indicate the overexpression of TPD52 in many types of cancer 9,22,25,36,66 . However, no evidence was previously available delineating its expression in brain cancer. In the current study, TPD52 expression in brain cancer was found to be up regulated which agrees with the previous study 67 that further strengthened the oncogenic role of TPD52. Furthermore, TPD52 expression increased with cancer grade and was also found to be associated with brain cancer primary metastasis. Previously, TPD52 expression up-regulation with cancer stage progression was reported in breast cancer 36 . Additionally, TPD52 expression was found to be lower in brain SOL than brain cancer but higher compared to healthy control. Increased levels of TPD52 in SOL of brain may be a key player that leads toward cancer. Taken together, TPD52 may be a potential biomarker and effective target to improve therapeutic strategies for better treatment of brain tumour.
A tumor suppressive role for miR-124 in different cancers including colorectal, hepatocellular and gastric carcinoma is well-established [68][69][70] . MiR-124 down-regulation in gliomas is also reported 26,34 . The current study showed reduced expression of miR-124 in whole blood of brain tumour patients. Expression of miR-124 is also down-regulated in different sub-types of brain cancer such as oligodendroglioma, astrocytoma and GBM [31][32][33] . Our findings broadened the knowledge on miR-124 contribution in brain cancer by demonstrating its decreased expression in secondary metastatic brain tumors. Xia et al. 29 showed that restored miR-124 expression in GBM inhibited cancer invasiveness. Further, higher reduction of miR-124 in high-grade cancer than in lower-grade Figure 6. Expression of TPD52, PKCε, KLF14 and miRNA124 in brain tumor subtypes. (A) Expression of TPD52 in astrocytomas, glioblastomas, and oligodendrogliomas patients compared to healthy controls; (B) Expression of PKCε in astrocytomas, glioblastomas, and oligodendrogliomas patients compared to healthy controls; (C) Expression of KLF14 in astrocytomas, glioblastomas, and oligodendrogliomas patients compared to healthy controls; (D) Expression of miRNA 124 in astrocytomas, glioblastomas, and oligodendrogliomas patients compared to healthy controls. There is an elevated expression of TPD52 and PKCε in glioblastoma compared to astrocytomas and oligodendrogliomas. KLF14 expression was lower in glioblastomas, and oligodendrogliomas in comparison to astrocytomas. miRNA 124 expression was higher in astocytomas in comparison to glioblastomas, and oligodendrogliomas but lower compared to healthy control. Representative data were presented as mean ± SEM of triplicate experiments. Statistical significance was measured by one-way ANOVA (****P < 0.0001).  26 also reported down-regulation of miR-124 in all grades and pathologic types of gliomas. It is possible that reduced expression of miR-124 contributes to the transition from low to high-grade cancer e.g. GBM. Similarly, the down-regulated levels in SOL of brain relative to health control but up-regulated levels relative to brain cancer suggest miR-124 expression loss might be part of the transformation process of SOL into tumor. Our results opened a new avenue for elucidating the importance of miR-124 in normal functioning of brain and how loss of miR-124 contributes to brain malignancy. Expression of PKCε was also investigated in brain tumours. Consistent with the notion that PKCε is upregulated in many human cancers e.g. prostate, lung and breast cancer 71 , it was found to be over-expressed in brain tumours as well. Our work is in agreement with the previous study in which it was found to be overexpressed in different types of brain tumours 38 . Extensive study has been focused to find out its expression in different clinical features of brain tumours. Elevated expression of PKCε leads toward the metastasis of brain tumours to different organs 38,72,73 . In this study, PKCε levels were observed to be lower in grade I + II compared to advanced grade III + IV brain tumour. This data supports that continuous upregulation of PKCε is associated with higher grade cancer, confirming PKCε as a key player in brain tumours. Further this study provides novel data for a role of PKCε in SOL of brain. It was observed to be upregulated in space occupying lesions (SOL) of brain compared to healthy control. PKCε expression was higher in cancer patients compared to SOL brain, again providing a gradient of increased levels from healthy brain to SOL and finally to brain cancer. Our results confirm the role of PKCε as an oncogene in brain tumours; PKCε poses a potential biomarker for this disease, which warrants verification in further studies. It may be an effective target to improve therapeutic strategies for better treatment of brain tumour.
Pathway analysis indicated the interaction of under-studied genes with major signalling cascades, including the Akt pathway and KRAS signalling. KLF14 negatively modulates the functioning of oncogenic KRAS 74 , whereas PKCε brings about activation of KRAS 75 . KLF14 is present downstream of PKCε and its down regulation in cancer supports cell proliferation and survival. In prostate cancer, elevated KLF14 expression promotes cancer survival by initiating mechanisms attenuating metabolic processes induced oxidative stress 17 . KLF14 also modulates activity of PLK4 and suppresses amplification of centrosomes 76 . However, KLF14 loss enhances PLK4 that leads to carcinogenicity 13 . TPD52 acts as a bridge between receptor tyrosine kinases and Akt and promotes pro-survival signalling through Akt pathways 66 . Contrary to TPD52, expression of miR-124 attenuates signalling through Akt and associated downstream pathways 10 . Differential expression analysis of these genes www.nature.com/scientificreports/ in brain cancers and their molecular interaction in cellular pathways helped us in gaining insight of possible crosstalk to these genes. Previous work revealed the participation of PKCɛ in activating the GPCR coupled Ras/Raf pathway that facilitates growth of neuronal cells that are associated with memory 77 . A regulatory link of PKCε with STAT3 has also been established in prostate adenocarcinoma 78 . A recent study ascertains the activation of STAT3 via TPD52 in neuroblastoma 79 . Hence transcriptional activity of STAT3 is regulated by PKCε and TPD52 as well as Rho-kinases. PKCε involvement was also found in Rho signalling; specifically, PKCε mediated activation of Rho GTPase to facilitate metastasis in lung cancer 80 . Evidence from literature also indicated that ERK phosphorylation in Ras/Raf pathway is due to activation of downstream target of PKCε i.e. Rho GTPases 81,82 .
Involvement of PKCε in the Akt pathway was also revealed by our pathway analysis. PKCε is located upstream of TPD52 and both genes activate Akt signaling that promotes tumor proliferation and invasion. Phosphorylation of Akt at serine 473 induces its activation 83 . Akt regulates proliferation and cell cycle by targeting cyclin D1, p21, p53 and p27 84,85 . Forkhead box O (FOXO) is the transcription factors and serve as downstream targets of Akt (protein kinase B). Akt inhibits FOXO by phosphorylating it and hence promotes cell survival, growth, and proliferation. Similar to PI3K/Akt signaling, TPD52 and PKCε block the transcriptional activity of FOXOs (specifically FOXO1, 3, and 4), activate cyclin D and inactivate p27 (a negative regulator of cell cycle), leading to enhanced cellular proliferation 86 .
Studies indicate the role of miR-124 as a tumour suppressor gene which plays an important role in cell apoptosis. Evidence show that increased expression of miR-124 in cancer cells blocks proliferation by inhibiting the KRas pathway 69 . miR-124 also causes the inhibition of cyclin dependent kinase 4 (CDK4), an activator of a Figure 9. Phylogenetic analysis of KLF family. The evolutionary history was inferred using the UPGMA method. The optimal tree with the sum of branch length = 7.54632578 is shown. The tree is drawn to scale, with branch lengths in the same units as those of the evolutionary distances used to infer the phylogenetic tree. The evolutionary distances were computed using the Poisson correction method and are in the units of the number of amino acid substitutions per site. This analysis involved 15 amino acid sequences. All ambiguous positions were removed for each sequence pair (pairwise deletion option). There was a total of 602 positions in the final dataset. Evolutionary analyses were conducted in MEGA X 42 . www.nature.com/scientificreports/ pro-survival transcription factor E2F1, thus promoting cell senescence and apoptosis 68 . Further the dysregulated expression of miR-124 leads toward the increased expression of SLUG. Its role is to bind with the promoter region of E-Cadherin that causes cell invasion 87 .
The outcomes of the current study highlighted the diagnostic potential of co-expression of KLF14, PKCε, TPD52 and miR-124 in brain cancer. Moreover, evaluation of these genes at the protein level will further validate their efficacy as blood-based biomarkers for the diagnosis and prognosis of brain cancers. Differential expression of these on a large cohort size of different subtypes of brain cancer can further unravel the subtype-specific diagnostic efficacy of these genes.

Conclusions
In the current study, we investigated the co-expression profile of KLF14, TPD52, PKCε and miR-124 and found up-regulated expression of TPD52 and PKCε and down-regulated expression of KLF14 and miR-124 in peripheral blood of representative solid brain tumour and SOL patient samples. Dysregulation of these genes has been found to be associated with disease progression. These findings reveal the important role of these genes in brain tumour and SOL of brain, highlighting their role in brain functioning. Also, our results suggest their importance as a potential biomarker and therapeutic target for brain tumours. Further validation of co-expression of these genes in blood of brain cancer patients will be advantageous by providing less invasive means for early diagnosis of the disease.

Data availability
All the data is contained in the manuscript. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.