Unveiling the Physical and Functional Niches of FAM26F by Analyzing Its Subcellular Localization and Novel Interacting Partners.

The knowledge of a protein's subcellular localization and interacting partners are crucial for elucidating its cellular function and associated regulatory networks. Although FAM26F (family with sequence similarity 26, member F) has been recognized as a vital player in various infections, stimulation studies, cancer, and immune pathogenesis, the precise location and function of FAM26F are not well understood. The current study is the first to focus on functional characterization of FAM26F by analyzing its subcellular localization and identifying its novel interacting partners using advanced proteome approaches. The immunofluorescence and confocal microscopy results revealed FAM26F to be largely localized within the Golgi apparatus of the cell. However, its minor presence in endoplasmic reticulum (ER) pointed toward the probable retrograde transfer of FAM26F from Golgi to ER during adverse conditions. Moreover, co-immunoprecipitation and MS/MS results demonstrated a total of 85 proteins, 44 of which significantly copurified with FAM26F. Interestingly, out of these 44 MS/MS identified proteins, almost 52% were involved in innate immunity, 38.6% in neutrophil degranulation, and remaining 10% were either involved in phosphorylation, degradation, or regulation of apoptosis. Further characterization through Ingenuity Pathway Analysis showed that majority of these proteins was involved in maintaining calcium homeostasis of cell. Consequently, the validation of selected proteins uncovered the key interaction of FAM26F with Thioredoxin, which essentially paved the way for depicting its mechanism of action under stress or disease conditions. It is proposed that activation and inhibition of the cellular immune response is essentially dependent on whether FAM26F or Thioredoxin considerably interact with CD30R.

Introduction

In reference to the various proteins known for their potential roles in the immune system, family with sequence similarity 26, member F (FAM26F) is a relatively new name that has gained much significance in the past few years as being critical in modulating diverse immune responses. Previously termed as INAM [IRF-3-dependent natural killer (NK)-activating molecule],[1] FAM26F or CALHM6 (calcium homeostasis modulator protein 6) is a 315 amino acid long, reasonably conserved, stable protein with a 34.258 kDa molecular weight. It comprises of 3–5 transmembrane helices as well as an immunoglobin-like fold, emphasizing its significance as an immune molecule.[2] So far, there are only three studies that provide a brief overview of the FAM26F’s function. In 2010, FAM26F was recognized as a toll-like receptor (TLR) signal-derived membrane molecule by Ebihara et al. The molecule was found to modulate mDC–NK contact-mediated NK activation. Consequently, it was suggested and emphasized that owing to the NK cells activation, INAM possesses the capability to serve as a therapeutic for the tumors that are NK sensitive.[1] Another study carried out by the same group revealed that the expression of FAM26F on the surface of immune cells facilitates the production of interferon-gamma (IFN-γ) through the NK cells, thus anticipating FAM26F to be a novel target molecule for immunotherapy against IFN-γ suppressible tumors.[3] Moreover, investigating its function in SIV infection showed that pre-infection levels of FAM26F correlate inversely with general viral load of plasma, and thus, FAM26F can be regarded as one of the earliest prognostic markers which, in the infection’s early stage, can give us information related to the pace and strength of antiviral’s immune response.[4]

Apart from these, numerous whole transcriptome analyses have detected differential expression of FAM26F. The examples include a range of clinical studies primarily associated with inflammatory response[5−8] in melanoma patients[9] and in hepatitis C virus clearance.[10] Upregulation of FAM26F can occur as a result of the interaction among various signaling pathways, including stimulation of TLR3 via polyI/C or TLR4 receptor,[1,11,12] stimulation of Dectin-1 pathway,[13] upon exposure to IFN-β,[12] upon exposure to IFN-γ alone[4,11,14] or by the combined stimulation of IFN-γ with either lipopolysaccharide or IFN-β,[11,15] and after infection with murine cytomegalovirus.[16] Moreover, deletion of mice IFN-α and IFN-β receptors retracted the Poly I:C stimulated induction of FAM26F.[3] FAM26F expression in dendritic cells/macrophages was also lost or significantly reduced as a result of deletion of IRF-3 and TICAM-1/TRIF[1] or IRF-5[13] which consequently led to inadequate activation of NK cells and thus affected their cytolytic function.

One of the major challenges of the post genomic era is to functionally characterize all of the cellular proteins. Proteome analysis seeks to reliably annotate these proteins for determining their interaction partners and functionalities in the cellular environment.[17] A significant and foremost step in this regard is to determine each protein’s subcellular localization in order to demonstrate its operating environment within a cell. It impacts protein function by governing the availability and access to various molecular interaction partners. Therefore, understanding protein localization along with its interacting partners is important to characterize the cellular functions of both the hypothetical proteins as well as the newly discovered proteins.[18]

Although much is known now about the differential expression of FAM26F based on various infections, stimulation, and immune-related studies, the exact localization of FAM26F as well as its involvement in modulatory pathways which can shed light on its specific function is still unidentified. Thus, the current study was aimed to determine FAM26F’s subcellular localization and to find its interacting partners in order to decipher the particular pathway which is regulated when FAM26F is expressed in a cell. For this purpose, FAM26F was transiently expressed within the Human Embryonic Kidney (HEK293) cells and its localization was determined through confocal laser scanning microscopy following the immunofluorescence staining assay. Moreover, co-immunoprecipitation and MS analysis were performed to identify the interacting partners of FAM26F, which were subsequently visualized through immunofluorescence. FAM26F’s main locality was found to be within the Golgi apparatus of a cell, though it is moderately apportioned in the ER as well. Additionally, FAM26F largely interacts with proteins mediating calcium homeostasis of a cell, particularly with Thioredoxin, and consequently regulates the immune response.

Results

Determining Optimal Time Period for Maximum FAM26F Expression

Before proceeding with the localization or immunoprecipitation experiments, it is important to determine the optimal time post transfection at which minimum cell death and maximum protein expression is obtained. For this purpose, HEK293 cells were transiently transfected with FAM26F plasmid. Expression of FAM26F was determined at 3, 6, 12, 24, 36, and 48 h post transfection by performing biochemical tests including the [3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(-4-sulfophenyl)-2H-tetrazolium] (MTS) assay and Caspase-3 activity assay. The results of both the assays showed maximum expression of FAM26F at 24 h post transfection (Figure ). The cell viability increased with increasing time and peaked at 24 h post transfection, after which it declined and remained almost constant thereafter. In the case of cell death assay, minimum number of dead cells were observed at 24 h, the time point which was then selected for further experiments.

Figure 1

Relationship between the absorbance measured with the MTS and caspase assay and the time course post transfection. The results of (a) MTS assay and (b) caspase-3 activity assay confirmed that HEK293 cells had maximum viability and minimum cell death at 24 h post transfection. The OD values of samples were normalized to the values obtained with untreated control cells and then divided by 10 to adjust/fit with the standard range.

Subcellular Localization of FAM26F

To determine the localization of FAM26F within the cell, HEK293 cells were transiently transfected with GFP-tagged FAM26F plasmid (NM_001010919), followed by costaining with antibodies specific for respective endogenous markers of various cellular organelles and compartments. The localization of FAM26F was checked 24 h post transfection in endoplasmic reticulum (ER), Golgi apparatus, cytoskeleton, endosomes, and in the cell nucleus. Antibody staining revealed FAM26F to be majorly localized within the Golgi apparatus of the cell, whereas its fair presence could also be detected in the ER. The localization within Golgi apparatus was confirmed by using two different Golgi-specific antibodies, staining different portions of the Golgi apparatus. No colocalization of FAM26F was observed with the endosomes or nucleus, and it was insignificant in the case of cytoskeleton as well. The results are illustrated in Figure .

Figure 2

Localization of FAM26F in Golgi apparatus of HEK293 cells. (a) Confocal laser scanning microscopy images of HEK293 cells depicting transfected FAM26F (green) costained with various organelle specific markers (red channel), including antibodies against β actin (cytoskeleton marker), VCP (ER marker), Rab9 (endosomal marker), Syntaxin 6, and Golgin (Golgi apparatus marker). Nucleus was stained using TOPRO3 iodide (blue channel). Scale bar: 10 μm. (b) Densitometric analysis from 25 different images clearly revealed FAM26F to be majorly localized in the Golgi apparatus of the cell.

Co-immunoprecipitation of FAM26F and MS/MS Analysis

To determine the interacting partners of FAM26F, control (untransfected) and transfected HEK293 cell lysates (24 h) were immunoprecipitated with rabbit anti-FAM26F antibody. Negative controls (having beads + sample but no antibody) were also used in order to identify and eliminate the nonspecifically bound proteins. Eluates were resolved on sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), electrotransferred onto polyvinylidene difluoride (PVDF) membrane, and detected with FAM26F antibody (Figure ).

Figure 3

Co-immunoprecipitation and western blot analysis. FAM26F control (untransfected) and transfected HEK293 cell lysates (24 h) were immunoprecipitated with rabbit anti-FAM26F antibody. Eluates were resolved on SDS-PAGE, electrotransferred onto PVDF membrane, and detected with FAM26F antibody. The figure shows successful elution of the FAM26F complex (transfected) at 61 kDa. The endogenous protein expression can be observed at 34 kDa. The IgG bands at 25 and 50 kDa are also visible.

Once positive confirmation was attained, the residual eluate was 1-DE separated and then stained with Coomassie Blue. The eluate band was excised, in-gel digested, and proteins were classified through the Q-TOF MS/MS analysis. The proteins present abundantly in the negative control lanes were taken as background contaminants and were removed from the list of proteins obtained from control and FAM26F transfected eluates. Likewise, the trypsin-digested products were also removed as background hits. Moreover, exclusion criteria was also applied so that only those proteins were displayed which had a total calculated probability of ≥95% and had no fewer than two identified unique peptides that have at least 99% identification probability. Out of the protein list that remained, those proteins were considered for further analysis which were either differentially expressed between the control and transfected eluate, the peptide count being higher for transfected eluate as compared to the control eluate, or were novelly present in the transfected eluate. Total amount of such proteins came out to be 44. Each of these 44 proteins was then manually checked for its function in UniProt (http://www.uniprot.org/; Table S1).

The proteins were further analyzed using the Reactome database in order to predict the pathways in which they were involved. Reactome results showed that more than half of the proteins (almost 52%) identified in the study were involved in the innate immune system, 38.6% in neutrophil degranulation, and the remaining 10–13% were either involved in phosphorylation, degradation, or regulation of apoptosis (Figure ).

Figure 4

Reactome distribution of identified proteins into functional groups. The largest group comprising 52% of the proteins is constituted by the innate immune system, followed by neutrophil degranulation with 38.6% of proteins. A low percentage (10%) of proteins was involved either in phosphorylation, degradation, or regulation of apoptosis.

To get further insight into these broader categories displayed by Reactome, the refined protein set was analyzed using the Ingenuity Pathway Analysis (IPA) software which revealed that majority of the identified proteins fall in the category of calcium-binding proteins and may hence be involved in maintaining calcium homeostasis of the cell (Figure ). The six proteins involved in calcium regulatory mechanism were finally selected for further validation. Table enlists the selected proteins along with their calcium specific/dependent functions.

Figure 5

IPA distribution of identified proteins into functional groups. IPA revealed that majority of the identified FAM26F interacting proteins fall in the category of calcium-binding proteins, and hence, FAM26F may be involved in maintaining calcium homeostasis of the cell. The other major interacting groups included oxidoreductase, signaling molecules, and hydrolases.

Table 1

Selected MS Identified Proteins and Their Functional Involvement in Calcium Regulation/Homeostasis

no.	accession number	protein name	function	localization (uniprot)	refs
1.	P07384	Calpain	calcium-regulated nonlysosomal thiol-protease that catalyzes limited proteolysis of substrates that are involved in cytoskeletal remodeling and signal transduction	cytoplasm; translocates to the plasma membrane upon Ca2+ binding	(19,20)
2.	P18206	Vinculin	a major platelet protein which undergoes Ca(+2)-dependent tyrosine phosphorylation during the platelet activation	plasma membrane;	(21)
				cytoskeleton;
				other: adherens junction; focal adhesion
3.	P31151	protein S100-A7	calcium-binding protein containing the EF hand motif that exhibits antimicrobial activities against bacteria and induces immunomodulatory activities	extracellular region or secreted by a non-classical secretory pathway;	(22)
				other: cytoplasm
4.	P10599	Thioredoxin	calcium-dependent oxidation of thioredoxin occurs during the initiation of cellular growth and stress conditions	extracellular region or secreted by a leaderless secretory pathway; cytoplasm; nucleus	(23)
5.	P32119	Peroxiredoxin-2	Prx2 plays a role in calcium-activated potassium transport through the Gardos channel, and calcium has been reported to increase membrane binding of Prx2	cytoplasm	(24,25)
6.	Q9NZT1	Calmodulin-like protein 5	a regulatory Ca2+-binding protein which transmits a transiently increased intracellular Ca2+ concentration to an activation of specific enzymes	extracellular region or secreted	(26)

Functional Affiliation of MS/MS Identified Interactors with FAM26F

To further determine the extent of functional relatedness of each of the selected proteins with FAM26F, proteins were immunostained and their colocalization with FAM26F was visualized using confocal laser scanning microscopy as described earlier. The extent of colocalization was determined by the Leica software. Colocalization in the fluorescence imaging is characterized by the amount of overlap displayed by two dissimilar fluorescent labels having dissimilar emission wavelengths. If the fluorescent signal from two separately labelled proteins is detected within the same 3D pixel, it means that the two proteins are present at the same physical location or are extremely close to each other. In this study, it was demonstrated that FAM26F majorly interacts with Thioredoxin (Trx), as was evident from its high colocalization frequency. Moreover, some colocalization was also seen with peroxiredoxin, whereas the remaining proteins did not show significant results (Figure ). Interestingly, both Thioredoxin and Peroxiredoxin have significant roles in the Thioredoxin system which detoxifies the reactive oxygen species (ROS) and hence maintains the cells in a reduced environment.[27]

Figure 6

Colocalization of FAM26F with its various identified interactors. (a) Confocal laser scanning microscopy images of HEK293 cells depicting transfected FAM26F (green) costained with various identified interactors (red channel), including Calmodulin, Calpain, Peroxiredoxin, Vinculin, protein S-100 A7, and Thioredoxin. Nucleus was stained using TOPRO3 iodide (blue channel). Scale bar: 10 μm. (b) Densitometric analysis from four different images (±SD). The results showed FAM26F to be majorly interacting with Thioredoxin protein, and to a lesser extent with Peroxiredoxin protein, both of which are critical proteins of the Thioredoxin system. Colocalization with other interactors was not significant.

Discussion

Although numerous studies have reported FAM26F to have played a critical role in immunity, infection, cell differentiation, and as an antitumor agent, and have demonstrated it to be regulated by different cytokines/interferons such as IFN-γ, TNF, and so forth, the precise function and modulatory pathways of FAM26F are yet unknown. Hence, in the current study, the functional characterization of FAM26F was accomplished by analyzing its subcellular localization and identifying its interacting partners in HEK293 cells. Molecular techniques such as transient transfection, immunofluorescence, western blot analysis, and co-immunoprecipitation, as well as advanced tools such as confocal laser scanning microscopy and IPA were employed to increase the precision and significance of the outcomes.

The immunofluorescence results revealed that FAM26F is majorly expressed inside a cell’s Golgi apparatus. Golgi apparatus has two types of protein populations: Golgi transient proteins and Golgi resident proteins. The proteins that undergo post-translational modifications while passing through the Golgi apparatus and are then selectively targeted to various organelles are known as Golgi transient proteins while the population of Golgi proteins performing these functions are the Golgi resident proteins.[28] FAM26F comes in the latter category. This is evident from the fact that FAM26F does not contain any “leader sequence” (signal peptide) attached to it and also does not undergo N-glycosylation,[2] omitting its chance to be targeted to other organelles or to the extracellular space. The Golgi resident proteins may in turn either be integral membrane proteins (embedded within the membrane) or peripheral membrane proteins present on the Golgi’s cytoplasmic face. FAM26F contains 3–5 transmembrane helices,[2] which readily makes it an integral membrane protein.

Moreover, there are various retention signals/mechanisms which are accountable for the localization of resident proteins within the Golgi apparatus. The resident proteins may possess either one or several of these attributes within their sequence or structure which can contribute to their steady-state retention in a particular Golgi subcompartment.[29] One of the key retention signals which on its own is adequate to confer Golgi localization of a protein is the presence of a single transmembrane domain (TMD) with a small portion of N-terminal cytoplasm.[28] The significance of TMD with reference to Golgi localization has previously been observed in some glycosyltransferases, Golgi-resident SNARE proteins, for example, SedSp and Sftlp3r4, and with certain viral proteins destined to the Golgi apparatus.[30,31] Interestingly, FAM26F has also been found to possess a single transmembrane “calcium homeostasis modulator” domain,[2] preceded by a short cytoplasmic portion at the N-terminal, which confers its localization to the Golgi apparatus.

Additionally, some membrane proteins have Golgi localization signals in their cytoplasmic domains or within the sequences flanking the TMD. Some well-known examples are Furin, TGN38, and cation-dependent and cation-independent mannose 6-phosphate receptors (MPRs) in which cytoplasmic domain not only aids their localization in the Golgi[32−38] but also has the ability to interact with cytosolic transport proteins.[39−41] In these protein, there are certain tyrosine or serine-containing signals which are responsible for their trans Golgi network (TGN) localization.[38] In MPR, a serine (Ser) residue in the cytoplasmic tail having the tendency to get phosphorylated by casein kinase II (CK-II), is responsible for Golgi localization of the protein and is also closely linked with release from the TGN.[40,42−44] Consistently, FAM26F also contains a Ser residue in its cytoplasmic tail at the position 311 which is predicted to be phosphorylated by CKII[2] and which might also aid in the transport activities of FAM26F. Hence, in view of these results, FAM26F is affirmed to be localized in the Golgi apparatus, owing to its possession of TMD and the Ser-containing sequence, just like some of the other Golgi-retained proteins possessing both of these signals.[45,46]

Intriguingly, FAM26F was previously proposed to exist on the plasma membrane of the immune cells.[1] We also obtained similar results from the in silico prediction of FAM26F localization through CELLO.[2] However, the results of current study indicate FAM26F as being a Golgi resident protein. The difference in the proposed/predicted and experimental results can be because of the nonclassical secretion of FAM26F from Golgi to the plasma membrane, which might be aided by its Ser-containing sequence in the cytoplasmic domain as mentioned above. As a matter of fact, this phenomenon of protein cycling between organelles/compartments has long been observed in the case of other Golgi retained proteins as well. Furin and TGN38 are both primarily located in the TGN, but they cycle between TGN and the plasma membrane.[38,47−49] MAL is predominantly located in perinuclear vesicles arising from the TGN, but is also an itinerant protein cycling between the TGN and the plasma membrane.[50] MPRs are predominantly TGN- and endosome-localized proteins, and they also recycle between these compartments and the cell surface.[32,33] C11ORF24 is localized on the TGN, yet it cycles to the plasma membrane via endosomes.[51]

In order to perform its function in vivo, a protein hardly operates alone.[52] In fact, more than 80% of the proteins have been found to be acting in complexes.[53] Moreover, the proteins carrying out similar cellular processes often interact with each other to perform their function.[54] Thus, a protein with known function can serve as a tool to determine the function of its interacting proteins whose function is yet unidentified. Our study identified the interacting partners of FAM26F using co-immunoprecipitation and immunofluorescence techniques in order to determine the cellular process regulated by this protein. HEK293 cells were transiently transfected with FAM26F plasmid, cells were lysed, and FAM26F along with its interacting partners were co-immunoprecipitated using Dynabeads. The eluates were subjected to MS analysis to identify the interacting proteins, which were then scrutinized to determine the cellular processes they belong to. It was observed that majority of the proteins were involved in the innate immune system. Further analysis through IPA software showed that most proteins were regulating the calcium homeostasis pathway of the cell. Hence, proteins linked with calcium signaling were selected and subsequently analyzed by inspecting their colocalization with FAM26F using immunofluorescence and confocal laser scanning microscopy.

Colocalization of FAM26F was visualized with various identified interactors including Calmodulin, Calpain, Peroxiredoxin, Vinculin, protein S-100 A7, and Thioredoxin. The results demonstrated FAM26F to be highly interacting with Trx, whereas interaction with Peroxiredoxin was also seen to some extent. The term interaction here and in the subsequent statements does not necessarily means physical interaction; rather it provides a notion of functional relatedness of two proteins based on their colocalization. Trx is a small dithiol-disulfide oxidoreductase that exists in all living cells.[55] It belongs to the Thioredoxin system, one of the principal antioxidant systems in mammalian cells which ensure a cell’s reduced environment by detoxifying ROS.[27] Peroxiredoxin, on the other hand, is thiol-specific thioredoxin-dependent peroxidase that traps the hydrogen peroxide and thus protects the cell against apoptosis.[56] Overall, peroxiredoxins react with hydrogen peroxide and get oxidized; Trx then reduces these peroxidases to enable the entrapment of ROS.[27]

FAM26F is considered as a pore-forming component of a voltage-gated ion channel having a calcium homeostasis modulator domain and is expected to assist in calcium homeostasis through its transport activity.[2] The question is how do FAM26F and Trx interact to regulate the calcium homeostasis of a cell and contribute to the innate immune system? This is dependent on the mutual interaction between calcium and ROS, the key signaling molecules of FAM26F and Trx, respectively. Calcium (Ca2+) is an essential second messenger involved in both intracellular and extracellular signaling cascades to regulate a variety of cellular functions through the action of buffers, pumps, and exchangers present on the plasma membrane as well as in the internal organelles.[57] Among the various signaling pathways with which Ca2+ interacts is ROS, which includes hydrogen peroxide (H2O2), superoxide anion (O2–•), and hydroxyl radicals (HO•). ROS interacts with redox-sensitive signaling molecules such as protein kinases, protein tyrosine phosphatases, transcription factors, and ion channels thereby altering their biological activity and resulting in the regulation of cellular processes including hypoxic signal transduction, growth factor signaling, autophagy, stem cell proliferation and differentiation, and immune responses.[58,59] There is a bidirectional interplay among ROS and Ca2+ signaling, where ROS can regulate cellular Ca2+ signaling by modulating the activity of various Ca2+ channels, pumps, and exchangers, while Ca2+ signaling is necessary for the production of ROS.[60] Thus, enhanced levels of Ca trigger ROS-generating enzymes to form free radicals.[57] Malfunctioning in either of the systems might consequently disturb the functioning of the other system, resulting in harmful effects that may lead to the pathogenesis of a variety of diseases.[57]

Proposed Signaling of FAM26F (Mechanism of Action)

In view of our colocalization results and owing to the interplay between Ca2+ and ROS, FAM26F is proposed to follow undermentioned analogous signaling mechanism. The immune cells usually respond to environmental stimulation or stress conditions by inducing extracellular Ca2+ influx as a primary signaling response to trigger the effector functions.[61] This Ca2+ activates protein kinase C which in turn activates NADPH oxidase.[62] This activated NOX (NADPH oxidase) is known as the “respiratory burst” enzyme, and being part of the innate immune system mediates the release of excessive ROS[61] by generating inositol 1,4,5-trisphosphate (IP3) by the membrane receptors. The IP3, in turn, activates IP3Rs, and Ca2+ is released from the intracellular stores (ER and Golgi).[63,64] Increase in intracellular Ca2+ is important for immune response.[63,65,66] This cytosolic Ca2+ employs dual ways to exert cellular oxidative response; one by activating NADPH oxidase respiratory burst and, second, by inducing oxidation of Trx.[23] The oxidized Trx detoxifies the ROS and maintains the reduced environment of the cell. Moreover, under stress conditions, the mammalian cell Trx1/TrxR1 system normally present in the cytoplasm can either migrate into the nucleus to induce the transcription of certain genes, or it can be secreted through unconventional secretory route into the extracellular environment, where it contributes to the immune system network.[67−70] This unconventional mode of secretion holds true for several other immune related proteins as well, including proangiogenic fibroblast growth factor 2 (FGF2),[71] tumor-mediated immune suppressive galectin 1 (gal-1), inflammatory cytokines such as IL-1β,[72] IL-1α,[73] IL-33,[74] protein high-mobility group box 1 (HMGB1),[75] and macrophage migration inhibitory factor.[76]

Our results concluded FAM26F to be localized within a cell’s Golgi apparatus. However, it can be translocated back to the ER following the retrograde transport. Retrograde transport is the process by which certain fusion and export proteins such as v-SNAREs or Vma21,[77,78] misfolded proteins,[79,80] escaped ER proteins,[81,82] or even Golgi resident proteins[83] are recycled back from Golgi apparatus to ER, either for maintaining the organelles’ steady-state composition[84] or under stress conditions.[85] In the case of recycling Golgi resident proteins, one can speculate that the Golgi protein residency is generally distributed amid the Golgi and ER.[86,87] This justifies the apportioned presence of FAM26F in ER visualized during the localization experiments of FAM26F. Further, it has been previously reported that this retrograde transport is dependent on the Ca2+ gradient present between the cytosol and the lumen of the Golgi apparatus and ER.[84]

Once in ER, the stress and discomposure of ER caused by regulated Ca2+ and ROS levels, and activation of Trx may lead to the secretion of FAM26F from the ER by unconventional means as a part of innate immune response. This secretion from ER may be through direct transportation to the plasma membrane, or directly to the extracellular space with or without the use of secretory lysosomes.[88] Although both FAM26F and Trx are secreted because of immune activation and by the immune cells, both have entirely different functions. Trx1 is a powerful cell survival and growth factor which has been observed to be considerably elevated in numerous types of cancers[89−93] and is usually associated with tumor aggressiveness, immune system inhibition,[94] and decreased survival in tumor patients.[95] On the contrary, FAM26F although upregulated in cancers has been shown to have therapeutic potential against NK-sensitive[1] and IFN-γ-suppressible[3] tumors and has also been associated with clinical benefits in metastatic melanoma.[9]

Interestingly, in the extracellular space, Trx1 has been shown to catalytically interact with a single target protein expressed on immune cells (B, T, monocytes, NK cells, dendritic cells, granulocytes, and eosinophils), the CD30 cell membrane receptor (CD30R), which holds great clinical significance.[96−98] This is because the resulting CD30R’s redox state governs its capability to bind to its cognate ligand CD30L and also to transduce signals,[96] contributing toward an effective immune response. However, abnormally elevated levels of Trx1 result in a deficit of Th1-cell function[68,99−101] and have also been observed in cancers.[102,103] In line with these findings, it can be suggested that FAM26F may also have the ability to interact with CD30R either directly or indirectly (by binding with Trx and blocking its activity) to bring about the diverse immune responses and tumor regression attributed to it. Nonetheless, whether the interaction leads to the inhibition or activation of immune response depends on whether FAM26F or Trx will associate with CD30R. This proposition evidently requires thorough investigation through experimental procedures before it can be held true or significant. Graphical representation of the proposed mechanism of action of FAM26F is illustrated in Figure .

Figure 7

Proposed signaling mechanism of FAM26F. Environmental stimulation or stress conditions induce the entry of extracellular Ca2+ into the cell to activate the effector functions. This Ca2+ activates protein kinase C which in turn activates NADPH oxidase. This activated NOX (NADPH oxidase) is known as the “respiratory burst” enzyme and mediates the discharge of surplus amounts of ROS by generating IP3 by the membrane receptors, which in turn activates IP3Rs and releases Ca2+ from the ER and Golgi through either IP3Rs or ryanodine receptors (RyR). Increased cytosolic Ca2+ induces oxidation of Trx which scavenges the ROS and maintains the reduced environment of the cell. Moreover, the activated mammalian cell Trx1/TrxR1 system normally present in the cytoplasm can either migrate into the nucleus where it induces the transcription of certain genes, or it can be secreted into the extracellular environment where it contributes to the immune system network. On the other hand, the stress and discomposure of ER caused by regulated Ca2+ and ROS levels can induce retrograde transfer of FAM26F back to the ER from Golgi and its subsequent secretion from ER either directly to the plasma membrane, or to the extracellular space with or without the use of secretory lysosomes. In the extracellular space, Trx1 catalytically interacts with CD30R expressed on immune cells. It is suggested that FAM26F may also have the ability to interact with CD30R either directly or indirectly to bring about the diverse immune responses and tumor regression attributed to it. Nonetheless, whether the interaction leads to the inhibition or activation of immune response depends on whether FAM26F or Trx will associate with CD30R.

Conclusions

This study is the first one to shed light on the much needed physical and functional niches of FAM26F by analyzing its subcellular distribution and complexes under native conditions in HEK293 cells. Confocal laser scanning imaging, subcellular localization, and co-immunoprecipitation experiments provide valuable insights into (1) apportioned distribution of FAM26F within the Golgi apparatus and ER of the cell and (2) identification of the interaction of FAM26F with calcium homeostasis proteins, particularly with Trx. Consequently, the study opened new perspectives of a possible mechanistic link between FAM26F and Trx, regulating the immune response. The potential of FAM26F and Trx for use as targets and biomarkers for various pathological conditions and diseases have been described and can be inferred from literature, as can be the alterations in the physiological pathways regulating the redox and Ca2+ and hence the immunological systems. However, we believe that targeting FAM26F and Trx simultaneously would be more effective and beneficial for optimizing redox regulation and the functioning of immune system. Moreover, further knowledge of the mechanisms that regulate the ROS and Ca2+ levels in different cell organelles and the subsequent regulation of FAM26F may result in novel therapeutic strategies for the different diseases that are affected by the dysfunctional balance between Ca2+ and ROS.

Experimental Procedures

Cell Culture

HEK293 cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM) (Sigma-Aldrich Chemie, Steinheim, Germany), supplemented with 10% fetal bovine serum (FBS) (Biochrom AG, Berlin, Germany), and 1% penicillin/streptomycin (Biochrom AG, Berlin, Germany) at 37 °C in a humidified atmosphere of 5% CO2.

Transient Transfection

The HEK293 cells were transfected transiently with GFP-tagged FAM26F plasmid encoding the full-length FAM26F gene, named FAM26F (NM_001010919) human tagged ORF clone (RG222648, OriGENE) using the Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA) as per the manufacturer’s instructions. The cells were later harvested at 3, 6, 12, 24, 36, and 48 h time intervals post transfection.

Cell Viability Assay

Cell viability was determined by the MTS proliferation assay (Promega, Madison, WI, USA), which measures the reduction of MTS to a water soluble formazan salt, which occurs only in metabolically active cells. In brief, 1 × 105 cells per well were seeded in 24-well plates (Nunc, Roskilde, Denmark) and permitted to grow for 12 h at 37 °C. Thereafter, the cells were transfected with C terminus GFP-tagged FAM26F plasmid for variable times (3, 6, 12, 24, 36, and 48 h). The cell viability was quantitatively assessed by using the MTS reagent in the presence of phenazine methyl sulfate (PMS). The culture media was changed prior to MTS treatment. After addition of combined MTS/PMS solution into each well, plates were incubated in a humidified atmosphere containing 5% CO2 for 1 h at 37 °C for color development. Multiscan plate reader (Labsystems, Manassas, VA, USA) and Accent software 2.6 were used to record the absorbance values at 490 nm. The final absorbance value was achieved by subtracting the cell-free background absorbance of the medium incubated with the MTS reagent from the sample wells. The OD values of samples were normalized to the values obtained with untreated control cells. All MTS assays were performed in triplicates.

Caspase-3 Activity Assay

The caspase-3-activity assay quantitatively measures alterations in caspase-3 (DEVDase) protease activity, which is an early apoptotic event. Caspase-3 activity assay kit was used as per the manufacturer’s protocol. In brief, both untreated control cells and C-terminus GFP-tagged FAM26F transfected cells were lysed inside the cell lysis buffer for 15 min at 4 °C. A centrifugation at 10,000g was conducted next. Thereafter, protein concentration in the supernatants was assessed. Then, 50 μg of the total cell lysate was incubated with 50 μM of caspase-3 specific substrate DEVD-pNA for a period of 4–5 h at 37 °C. The caspase-3 mediated release of pNA was measured through absorbance at 405 nm. The background absorbance from the controls (untreated cells) was then deducted from the final absorbance value obtained for the samples. The OD values of samples were normalized to the values obtained with untreated control cells. All assays were performed in triplicates.

Antibodies and Fluorescent Probes

The primary antibodies used include rabbit anti-FAM26F (Abcam, Cambridge, U.K.), rabbit anti-VCP (Abcam, Cambridge, U.K.), rabbit anti-Rab9 (Cell Signaling technology, Frankfurt, Germany), mouse IgG actin cytoplasmic 1 (Sigma, Steinheim, Germany), rabbit anti-Syntaxin 6 (Abcam, Cambridge, U.K.), rabbit anti-Golgin (Abcam, Cambridge, U.K.), mouse anti-mu-Calpain (Invitrogen, Carlsbad, CA, USA), mouse anti-S100-A7 (Sigma-Aldrich, Steinheim, Germany), mouse anti-Vinculin (Sigma-Aldrich, Steinheim, Germany), rabbit anti-Thioredoxin (Epitomics, Abcam, UK), mouse anti-Peroxiredoxin (Abcam, Cambridge, U.K.), and rabbit anti-Calmodulin (Abcam, Cambridge, UK). The secondary antibodies used consisted of HRP-conjugated rabbit anti-mouse pAb (IBA, Göttingen, Germany), goat antimouse cy3-conjugated (Dianova, Hamburg, Germany), antirabbit pAb (Santa Cruz Biotechnology, Santa Cruz, CA, USA), goat antirabbit (Alexa 555-conjugated), and antimouse (Alexa 555-conjugated).

Co-immunofluorescence and Confocal Laser Scanning Microscopy

HEK293 cells grown on glass cover slips in 24-well plates (Nunc, Roskilde, Denmark) were transfected with FAM26F plasmid. After 24 h of transfection, the cells were immobilized with 4% paraformaldehyde solution for 20 min, followed by three washes with phosphate-buffered saline (PBS) of 5 min each. Thereafter, the cells were permeabilized with 0.2% Triton X-100 in 1× PBS for 10 min and then blocked for 30 min in a solution containing 10% FBS in 1% bovine serum albumin (BSA) solution in PBS. Cells were incubated with the primary antibody at 1:100 dilution in 1% BSA in PBS and kept overnight at 4 °C. Cells were then incubated with the secondary antibody at 1:200 dilution with 1% BSA in PBS and finally counterstained with TOPRO-3 iodide for 1 min to stain the nuclei. The coverslips were then mounted onto the glass slides using mounting media Fluoromount (DAKO, Hamburg, Germany). All the mentioned steps were conducted in dark, and each reaction was stopped by washing the coverslips three times with 1× PBS. The slides were then kept in dark at 4 °C until visualized. Confocal laser scanning microscopy was performed using a Leica laser-scanning microscope (Zeiss, Gottingen, Germany) with appropriate excitation/emission filter pairs. Leica Application Suite X (LAS X) software was used to analyze the individual images separately for determining the colocalization.

Immunoblot Analysis

HEK293 cells in a concentration of 4.8 × 105 cells per well were plated in 6-well plates and permitted to grow for 24 h at 37 °C till they were 70–80% confluent. Thereafter, the cells were transfected with C terminus GFP-tagged FAM26F plasmid and again kept at 37 °C for 24 h, after which they were harvested and lysed using Tris-Triton lysis buffer (50 mM Tris-HCl pH 8.0, 1% Triton X-100, 0.5% CHAPS, 1 mM DTT) supplemented with protease and phosphatase inhibitors (Roche). The cell debris was cleared from the lysates by centrifugation at 14,000 rpm for 30 min at 4 °C. The supernatant was collected, and Bradford Assay (Bio-Rad) was used to determine the protein concentration. Equivalent protein concentrations in the lysates were resolved by 12% SDS PAGE. Standard protocol was used to electrotransfer the proteins from the gel onto a PVDF membrane (Millipore).[104] A blocking solution of 5% nonfat milk diluted in PBS containing 0.05% Tween20 (PBS-T) was then used to soak the blot in. Primary antibody anti-FAM26F (1:1000) was added to the blot which was then incubated overnight at 4 °C. Thereafter, the membranes were washed in 1× PBS-T and incubated with the rabbit horseradish peroxidase-conjugated secondary antibody (diluted 1:10,000) for 1 h at room temperature. Immunoreactivity was then detected by incubating the membranes in enhanced chemiluminescence (ECL) solution, and images were visualized in ChemiDoc (Bio-Rad). The band intensities were determined by densitometry using the ImageLab (Bio-Rad) data analyzer software. Then, the membrane was reblotted with glyceraldehyde-3-phosphate dehydrogenase as a loading control.

Co-immunoprecipitation

Cell lysis and protein extraction was carried out in the manner described above. Immunoprecipitation was performed using protein G Magnetic Dynabeads (Invitrogen) following the manufacturer’s instructions. Typically, 6 μg of the anti-FAM26F antibody diluted to 1:50 in PBS was added to 30 μL of Dynabeads and incubated for 30 min at 4 °C. A quantity of 500 μg of protein lysate was then added to the antibody-Dynabead complex and left overnight at 4 °C. The next day, beads were rinsed thrice with 0.3% CHAPS in water, and 20 μL of 2× Laemmli buffer (doi: http://10.1101/pdb.rec10878 Cold Spring Harb Protoc, 2007) was used to elute the immunoprecipitated proteins from the beads–antibody–antigen complex. The elute was then cooked for 5 min at 95 °C and run onto 12% SDS-PAGE, followed by immunoblot analysis as described above.

In-Gel Tryptic Digestion and MS/MS Analysis

The eluates were run on 12% SDS-PAGE for 5–10 min to get a 5 cm run window. The gel was then stained with Coomassie Blue; the stained blue protein spots corresponding to the labeled proteins in the western blot were manually excised from the gel and washed with distilled water for 15 min. The gel pieces were destained by washing twice for 10 min with a solution containing 100 mmol/L of ammonium bicarbonate/acetonitrile (1:1, v/v) and then for a third time until all visible blue dye was removed. In-gel digestion with trypsin was carried out according to a protocol described previously.[105] The extracted peptides were then dissolved in 0.1% formic acid (FA) for ESI-QTOF MS/MS. One microliter of tryptic-digested peptide solution was introduced using a CapLC auto sampler (Waters) onto a μ-precolumn cartridge C18 pepMap (300 μm 5 mm; 5 μm partical size) and further separated through a C18 pepMap100 nano Series (75 μm 15 cm; 3 μm partical size) analytical column (LC Packings). The mobile phase consisted of solution A (0.1% FA in 5% ACN) and solution B (0.1% FA in 95% ACN). The single sample run time was set for 60 min. The chromatographically separated peptides were then analyzed on a Q-TOF Ultima Global (Micromass, Manchester, U.K.) mass spectrometer equipped with a nanoflow ESI Z-spray source in positive ion mode. The data acquisition was performed using the MassLynx (v 4.0) software on a Windows NT PC, and data were further processed on the Protein-Lynx-Global-Server (v 2.1), (Micromass, Manchester, UK). The processed data were searched against MSDB and Swiss-Prot databases through the Mascot search engine using a peptide mass tolerance of (0.5 Da) and a fragment mass tolerance of (0.5 Da). The search criteria were set up to maximum one missed cleavage allowed by trypsin and protein modifications set to methionine oxidation and carbamidomethylcysteine, when appropriate.

Computational Analysis

To determine the signaling pathways in which the identified proteins or interacting partners of FAM26F were involved, Reactome Pathway Database (https://reactome.org/) and IPA software (https://www.qiagenbioinformatics.com/products/ingenuity-pathway-analysis/) were used. Reactome is a pathway database which allows the visualization, interpretation, and analysis of pathway knowledge by employing intuitive bioinformatics tools.[106] IPA, on the other hand, is a powerful analysis and search tool that is capable of revealing the significance of ‘omics data by identifying the specific biological system in which the query proteins are involved.[107]

Statistical Analysis

All the results from this study were acquired on the basis of four individual experiment sets. Descriptive statistics was used to express the results as mean ± S.D. All the confocal images were quantitatively analyzed and assessed using the dedicated Leica software (https://www.leica-microsystems.com/products/microscope-software/) installed within the confocal laser scanning microscope system. ImageLab (Bio-Rad) software was used to perform the densitometric analysis of the 1-DE gels. All the graphs were prepared by GraphPad PRISM (GraphPad Inc.).