AHNAK2: Comparison
Please note this is a comparison between Version 1 by Mohamed zardab and Version 2 by Peter Tang.

AHNAK2 is a relatively newly discovered protein. It can interact with many other proteins. This protein is increased in cells of variety of different cancers. AHNAK2 may play a vital role in cancer formation. AHNAK2 may have a role in early detection of cancer.

  • AHNAK2
  • cancer
  • biomarker
  • screening
  • AHNAK

1. Introduction

AHNAK nucleoprotein 2 (AHNAK2), initially called C14orf78, was discovered in 2004 while exploring the function of its sister protein AHNAK nucleoprotein (AHNAK), by knocking out the Ahnak gene in mice through homologous recombination [1]. It is a large protein, whose canonical sequence is approximately 616 kDa, comprising 5795 amino acids. Human AHNAK2 is found on chromosome 14q32 with a 15 kb open reading frame (ORF) [1]. Two further isoforms have been identified, isoforms 2 and 3, with masses of 85 kDa and 605 kDa, respectively [2][3][2,3].
AHNAK2, like AHNAK, is a tripartite protein predicted to be composed of 24 highly conserved repeat segments approximately 165 amino acids long, starting from a short non-repetitive N-terminal segment and ending with a C-terminal segment approximately 100-kDa in size [1][4][1,4]. It is predicted that the repeat structure consists of a basic framework of linked antiparallel β-strands with interconnecting peptide loops (characterised as β-turns) [1]. Although not proven experimentally, a self-optimised method of secondary structure prediction (Network Protein Sequence Analysis, France) predicts it to contain 12 β-strands per repeat sequence, with 28 antiparallel β-strands packed into a disk-shaped seven-bladed (propeller-like) structure [1]. The predicted structure, due to the similarity in length and distribution, is likely that of a propeller protein, such as those found in RCC1, G β-protein, and clathrin, which have been confirmed with x-ray crystallography [5][6][7][5,6,7]. These repeating sequences also include SH3 binding sites and translocation and assembly modules (TamB). The short N-terminus has a ~90–100 amino acid segment that functions as a PSD-95/Discs-large/ZO-1 (PDZ) domain. The long C-terminal region has a nuclear localisation signal [4][8][4,8].
As AHNAK2 shares many structural characteristics with AHNAK, and in light of AHNAK2-specific studies, we can hypothesise possible functions, locations, and interactions [4][8][4,8]. First, PDZ domains are one of the most common modular interaction domains, usually consisting of 80–90 amino acids arranged in six β-strands and two α-helices [9]. The PDZ domain at the N-terminus allows AHNAK2 to attach to other PDZ domains, lipids, internal peptide sequences, and C-termini of a variety of proteins [9][10][9,10]. Furthermore, the β-propeller structure they share would allow AHNAK2, like AHNAK, to interact with numerous ligands and proteins, thus forming a part of large multi-protein complexes and scaffolding networks [10][11][12][10,11,12].
AHNAK and AHNAK2 localise to the nucleus, cytoplasm, and plasma membrane of various cell types [8][10][13][14][15][8,10,13,14,15]. AHNAK2 can reportedly localise together with AHNAK to the sarcolemma, transverse T tubules, and Z-band regions of mouse cardiomyocytes [1]. There was colocalisation of anti-AHNAK antibody in both wild-type and AHNAK knockout mice with anti-RyR (Sarcoplasmic reticulum), anti-dihydropyridine (DHP) receptor (T-tubules and sarcolemma), and anti-α-actinin (Z-band regions) antibodies [1]. However, in contrast, a separate study of skeletal muscle showed no colocalisation of AHNAK or AHNAK2 with dihydropyridine receptors [14]. Through SDS/PAGE immunoblotting of sub-cellular fractions of mouse myocardium, the AHNAKs were found to be co-sedimented, mostly with the bulk of Z-band material with nuclei and myofibrillar aggregates and the DHP receptor (membrane vesicular fraction). Both AHNAK proteins are linked to the T-tubule membranes through interactions with the β2 subunit of cardiac L-type calcium channels [1][16][1,16]. This link suggests a role in the excitation and contraction coupling mechanism [1]. The role, function, and location of AHNAK, given its homology to AHNAK2, may help us understand the role of AHNAK2.
AHNAK, through its C-terminus (aa 5262–5643), interacts with β2 and β1a isoforms of native cardiac L-type Ca2+ channels (Ca 1.2, 1.1) [16]. In cardiomyocytes, it is phosphorylated by protein kinase A (PKA), which is activated by cyclic adenosine monophosphate (cAMP), which in turn is activated by β adrenoreceptors [13][16][17][13,16,17]. Normally, AHNAK limits the influx of Ca2+ into the cardiomyocyte. In the immune system, when phosphorylated, AHNAK allows an influx of Ca2+ resulting in an action potential [13]. However, upon AHNAK knockdown, T-cells and osteoblasts malfunction due to the lack of Ca2+ influx [17]. In fact, AHNAK-knockout mice were shown to be susceptible to Bartonella henselae infection because of CD4+ T cell inactivation [18]. AHNAK also attaches to both G-actin and cytoskeletal F-actin, which have an important role in the maintenance of calcium channels in cardiomyocytes, smooth muscle cells, and osteoblasts [11].
It has been proposed that AHNAKs may act as both a tumour suppressor and promoter [8][19][20][8,19,20]. Transforming growth factor-β (TGFβ) is known to mediate tumour metastasis through the activation of the epithelial to mesenchymal transition (EMT) [21][22][21,22]. AHNAK was found to promote TGFβ/SMAD3-induced EMT and cancer metastasis [23]. Actin-dependent pseudopodal protrusion and tumour cell migration, known determinants of EMT, were found to be reliant on four proteins; AHNAK, Septin-9, elF4E and S100A11 [24]. Furthermore, AHNAK was associated with migration and invasion in mesothelioma [25] and hepatocellular carcinoma [26]. AHNAK is part of a gene panel with inflammatory markers that predict poor survival in laryngeal carcinoma [27]. Interestingly, an abundance of AHNAK and annexin A2 were found in extracellular vesicles released by mammary cancer cells towards non-cancer mammary fibroblasts, indicating a role for AHNAK in vesicular communication promoting cancer progression [28]. In a different context, AHNAK may have a tumour suppressor role. In seven of eight triple negative breast cancer cell lines, AHNAK mRNA expression was downregulated, with expression being inversely correlated with tumour status, lymph node status, lymph node infiltration, TNM staging, and prognosis [29]; this tumour-suppressive effect was associated with both the AKT/MAPK and Wnt/β-catenin pathways [29].
The large structure of AHNAK allows it to function as part of large multi-protein scaffolding networks, particularly in cell-to-cell contact. Upon reaching confluency, AHNAK in canine kidney cells relocates from the cytoplasm to the plasma membrane, associating as a hetero-tetrameric complex with actin and the annexin2/S100A10 complex [8][30][8,30]. In a calcium dependent environment, cell-to-cell contact forms through the phosphorylation of AHNAK by AKT [30][31][30,31], which then changes the localisation of AHNAK from cytoplasm to plasma membrane, where it attaches to a complex with annexin2/S100A10 and actin, specifically at the adherens junctions [8][16][30][8,16,30]. This implicates AHNAK, and by extension AHNAK2, in cell-to-cell contact, membrane ruffling, endocytic events, and cytoskeletal scaffolding and structural reinforcement [30][32][33][34][30,32,33,34].
Initially, knockout of AHNAK in murine models created no phenotypic change, suggesting the possibility that AHNAK2 may be compensating for the loss of AHNAK in various functions [1]. It has now been noted that AHNAK-deficient mice have a leaner phenotype, with resistance to high-fat diet induced obesity. They exhibited impaired glucose tolerance and higher fasting glucose levels with decreased Akt phosphorylation and cellular glucose transporter (Glut4) levels [35]. A strong link between AHNAK and bone development and metabolism in mice was discovered, with a newer long-term study of AHNAK-deficient mice showing reduced growth of the skeleton with more fragile and fracture-prone bones. Micro-CT scans of these mice at various intervals confirmed shortened, weaker bones and a difference in morphology of facial bones [36].

2. Function

2.1. AHNAK2 in Normal Tissue

AHNAK2 has been proven to be a structurally integral part of costameres in skeletal muscles, together with AHNAK [8][14][14][8,14,14]. Mouse skeletal muscle fibres were stained for these two proteins, as well as α-actinin, myomesin and vinculin [14], leading to the conclusion that AHNAK2 colocalises with vinculin and α-actinin at the end of z-discs connecting costameres to sarcomeres in skeletal muscle [14]. AHNAK2 and AHNAK are both part of the costameric network, located between Z-disks and the sarcolemma, which functions as a mediator of lateral transmission of force from sarcomeres across the sarcolemma to the extracellular matrix [14][37][14,37]. A murine study looking at wound healing following femoral artery wire injury showed AHNAK expression in the endothelium, with AHNAK2 exclusively expressed in the cytoplasm of the neointimal and medial cells [38]. On knockdown of AHNAK, there was a delay in healing after injury with no change in expression of AHNAK2, providing evidence that both proteins were not redundant in naïve arteries, with a novel localisation of AHNAK2 in smooth muscle cells [38]. Although AHNAK has been localised to myelinating Schwann cells and is important for their adhesion to laminin, little is known about AHNAK2 in the nervous system [39]. Periaxin (PRX), with an important role in myelination, is abundant in the peripheral nervous system, and structural studies have shown homodimerization of AHNAK2 with PRX through domain-swapping of their PDZ-like domains [10]. PRX is most closely related to the AHNAK proteins, specifically AHNAK2, with a 57% sequence identity in the PDZ domain [10]. AHNAK2 shares in the PRX PDZ configuration, with six major β strands (instead of 6) and two α helices in one monomer, although due to the configuration of the β2 and β3 strands the peptide binding pocket is larger in AHNAK2 [10]. Corresponding strands form a 6-stranded anti-parallel β sheet between the two intertwined sheets of AHNAK2 and PRX PDZ domains [10]. Although not proven in vivo, experiments concluded that both their PDZ domains exhibited uniquely intertwined dimers with extensive three-dimensional domain-swapping [10]. This means that PRX and AHNAK2 are likely to add stability and order to large molecular complexes through this interaction. With this dimerisation, it can be hypothesised that, like AHNAK and PRX, AHNAK2 may have a myelination maintenance role and that mutations in AHNAK2 could disrupt this specific interaction with PRX [8][10][40][8,10,40]. Furthermore, PRX mutations result in various demyelinating peripheral neuropathies, such as Charcot-Marie-Tooth (CMT) disease and Dejerne-Sottas disease, again reflecting PRX’s major role in the myelination of peripheral nerves [10][41][10,41]. Whole exome sequencing of patients with autosomal recessive CMT in a Malaysian family showed gene mutation and significantly reduced expression of AHNAK2 at both the mRNA and protein level in fibroblasts [40]. It isWe know that AHNAK and PRX are involved in large protein complexes with the plasma membrane, as well as the cell cytoskeleton; they are also involved in the packing of lens fibres and the development of the peripheral nervous system [8][39][42][8,39,42]. This novel link between PRX and AHNAK2 may open a window into a wide array of possible functions of AHNAK2, especially functions related to its ability to form stable multiprotein complexes through its PDZ domain. AHNAK2 may be involved in the development of the peripheral nervous system but this has yet to be proven in vivo.

2.2. AHNAK2 in Disease

Systemic Lupus Erythematosus (SLE) is an autoimmune disease affecting the skin, joints, kidneys, and other organs, with considerable genetic predisposition [43]. A large-scale exome-wide study of 5004 SLE patients and 8179 healthy controls in a Han Chinese population identified three novel coding variants and four new susceptibility regions for the disease [44]. AHNAK2 with LCT, TPCN2 and TNFRSF13B encompassed three novel mis-sense variants and two non-coding variants with genome-wide significance (p < 0.001) [44]. This may present new insight into the biological mechanism of SLE; it was the first study to comment on a gene aberration of AHNAK2 [44]. Interestingly, AHNAK was found to be a novel auto-antigen in SLE [45]. A study looking at methylating substrates of SMYD2 found that the central repeating units of both AHNAK and AHNAK2 are mono-methylated by SMYD2 at multiple sites [46].

2.3. AHNAK2 in Cancer

AHNAK2 over-expression has been identified in various cancer cohorts, including pancreatic ductal adenocarcinoma, a largely incurable, aggressive, and silent malignancy [47]. A multi-gene biomarker panel (TMPRSS4, AHNAK2, POSTN, ECT2, and SERPINB5) for diagnosing PDAC has been proposed, by analysing various transcriptomic microarray repositories with micro-dissected and whole tissue samples (the Gene Expression Omnibus (GEO) database, the ArrayExpress repository, and the Stanford Microarray Database [48]). This study concluded that the 5-gene panel can attain 95% sensitivity and an 89% specificity in five separate validation sets for differentiating between PDAC (n = 137) and all of the normal tissues (n = 197), PDAC precursor lesions (n = 15), and chronic pancreatitis (n = 9) (Table 1) [48].
Table 1. A collection of all oncological studies of AHNAK2 with associated findings. (GC: gastric cancer; BC: bladder cancer; LUAD: lung adenocarcinoma; PTC: papillary thyroid cancer; TC: thyroid carcinoma; CPTAC: clinical proteomic tumour analysis consortium; and GTex: genotype-tissue expression).

Cancer

Type of Study

Experimental Environment

Findings

Reference

PDAC

mRNA microarray analysis

In silico analysis of microarray data from PDAC datasets

1. Part of a 5-gene panel that differentiated between PDAC, early precursor lesions, and non-malignant tissue.

Bhasin et al. 2016 [48]

PDAC

mRNA microarray analysis with in-vitro studies

In silico analysis of microarray data from PDAC datasets with qPCR and immunohistochemistry

1. Part of a 17-gene panel that discriminated between PDAC and non-tumour tissue in FFPE and fresh frozen tissue.

2. High protein expression in PDAC versus non-tumour tissue.

Klett et al. 2018 [49]

PDAC

mRNA microarray analysis

In silico analysis of microarray data from PDAC datasets

1. Part of a 7-gene panel differentiating between PDAC and normal tissue with a significant association with poor prognosis.

Almeida et al. 2020 [50][51]

ccRCC

mRNA microarray analysis with in-vivo and in-vitro studies

qPCR of ccRCC cell lines versus non-tumour cell lines with knockdown studies of EMT, hypoxia, and fatty acid synthesis

1. High expression in ccRCC samples versus non-tumour tissue.

2. Increased tumour proliferation, tumorigenesis, colony formation, and migration. 3.Upregulation in hypoxia with increased EMT and increased lipid droplets, compared to knockdown.

Wang et al. 2017 [51][52]

UM

mRNA microarray analysis with in-vitro studies

qPCR of UM cell lines versus non-tumour cell lines and knockdown studies

1. High expression associated with shorter overall survival time in UM with inhibition of the PI3K signalling pathway and increased proliferation, migration, and invasiveness of cell lines versus knockdown.

Li et al. 2019 [52][53]

GC

DNA methylation analysis

Immunohistochemistry and DNA methylation status of gastric cancer cell lines

1. Higher methylation in EBVGC cells compared to normal GC with a connection to 5-fluorouracil and cisplatin resistance.

Ohmura et al. 2019 [53][54]

BC

Proteomics study

Label-free Fourier transform infrared liquid chromatography-tandem mass spectrometry proteomic analysis of bladder cancer, urocystitis, and reactive urothelial atypia tissue.

1. Potential biomarker for bladder cancer, which can differentiate between urocystitis and low-grade carcinoma with invasive high-grade bladder carcinoma when tissue is stained for AHNAK2.

Witzke et al. 2019 [54][55]

LUAD

mRNA microarray analysis with in vitro studies

In silico analysis of microarray data for LUAD at TCGA, CPTAC, GEO and GTEx datasets of lung tissue samples with in vitro studies

1. AHNAK2 expression is upregulated in tumour samples.

2. Silencing AHNAK2 inhibits migration, invasion, and EMT in lung adenocarcinoma cells by repressing the TGF-β/Smad3 pathway.

Liu et al. 2020 [55][56]

LUAD

mRNA microarray analysis

In silico analysis of microarray data from lung adenocarcinoma datasets with tumour-immune estimation resource analysis.

1. Significantly overexpressed in lung adenocarcinoma and found to be an independent prognostic factor.

2. Negatively correlated to activated B cells and CD8+ T cells, while positively correlated to CD4+ T cells and tumour-associated macrophages.

Zheng et al. 2021 [56][57]

PTC

mRNA microarray analysis with in vitro studies

In silico analysis of microarray data from GEO, Oncomine, TCGA, and HPA datasets. IHC staining analysis and tumour immune estimation resource analysis.

1. Upregulation is significantly correlated to poor survival, advanced stage and grade.

2. Positive correlation with and B-cell, CD4+ T cell, macrophage, neutrophil, and dendritic cell infiltration.

Zheng et al. 2021 [57][58]

TC

mRNA microarray analysis with in vitro studies

In silico analysis of microarray data from TCGA. In-vitro studies of TC cell lines.

1. AHNAK2 is associated with a poor clinical outcome.

2. Inhibition of AHNAK2 suppresses the NF-κB pathway.

Ye et al. 2021 [58][59]

Video Production Service