Masteroppgaver levert i 2012

N-terminal acetylation by NatC does notgenerally influence the subcellularlocalization of its substrates

Camilla Osberg (T. Arnesen, H. Aksnes)

Among the many different protein modifications occurring in eukaryotic cells, N-α-terminal acetylation (Nt-acetylation) is one of the most common. The addition ofacetyl groups onto the N-termini of proteins is catalyzed by one of six N-α-terminalacetyltransferases (NATs), NatA-NatF. A general function of Nt-acetylation has not beendetermined, but there are several studies of individual proteins suggesting variousprotein functions such as lifetime, protein-protein interactions and subcellularlocalization. As several of the localization-affected proteins described are NatCsubstrates,we questioned whether the Nt-acetyl group has a general role in thesubcellular localization of NatC-substrates.

In this thesis study, we have performed a microscopic investigation involving 11putative NatC-substrates. Using Saccharomyces cerevisiae GFP-strains, we have deletedthe gene encoding the catalytic subunit of NatC, NAA30, by homologous recombinationof a deletion-cassette. This allowed us to study protein localization in the absence of theNt-acetyl modification. Visualization of the target proteins was accomplished byfluorescence microscopy due to the GFP-tag already genomically incorporated at the Cterminalend of the proteins.

An initial fluorescence microscopy screen did not detect localization shifts formost of the investigated proteins. However, for some proteins a possible elevation of protein levels due to the naa30-deletion was observed and further investigated.

The outer mitochondrial membrane protein Tom70p was given more attentionbased on the apparent shift in protein level and localization pattern observed in thescreen. In the absence of Naa30p, Tom70p exhibited an aggregation-like localizationpattern. The mitochondria appeared to be normal after deletion of NAA30, based on thelocalization patterns of other GFP-tagged mitochondrial proteins. Further studiesdisplayed Tom70p in association with mitochondria in both wild type and mutantstrains, thus arguing against a localization shift as an explanation for the changedpattern. In attempts to further characterize the possible effect of naa30-deletion onTom70p, we carried out several follow-up experiments, but finally had to conclude thatthe initially observed aggregation-like effect was likely to be caused by an experimentalartifact and not the absence of Naa30p. Taken together, our results disfavor thehypothesis that Nt-acetylation by NatC is a main determinant of the subcellularlocalization its substrates.

Investigating the activity of the N-terminal Acetyltransferase Naa60p towards α-MSH and β-Endorphin

Alexander Kirkeby Eieland (T. Arnesen, K. Hole)

Nα-terminal Acetyltransferases (NATs) catalyze the transfer of an Acetyl-moiety from Acetyl-Coenzyme A (Ac-CoA) to the α-amino group of the nascent polypeptide co-translationally at the ribosome, or post-translationally independently of the ribosome. Recent studies indicate that about 85% of proteins in higher eukaryotes are partially or completely acetylated, and several of the NATs responsible for these modifications have been discovered. However, much work remains to elucidate the specific purpose and effects of N-terminal acetylation.

The activities of the hormones α-MSH and β-Endorphin are regulated by post-translational Nt- acetylation. However, the NAT responsible for the Nt-acetylation of peptide hormones is not identified. The human protein Naa60p has recently been accredited as a functional N-terminal acetyltransferase and its activity is denoted NatF. NatF is the first NAT to be demonstrated to be localized to organelles and is probably associated with secretory endosomes; a common trait shared with several secreted hormones. The catalytic activity of NatF has not been completely elucidated, and although some sequence specificities have been demonstrated its potential relation to secreted peptides like α-MSH and β-Endorphin remains to be fully investigated.

In this study a specific approach has been utilized to investigate the ability of hNatF (hNaa60p) to N-terminally acetylate the two peptide hormones α-MSH and β-Endorphin. Cells have been cultured and proteins have been isolated from un-transfected, control-transfected or hNaa60p-transfected cells, and the isolated proteins were assayed towards acetylation of the peptide hormones in an in vitro acetylation assay. This assay has been repeated several times by using proteins from total cell lysates, isolated organelles, immunoprecipitates and MBP-hNaa60p purified from E. coli. No activity towards α-MSH or β-endorphin was detected in great amounts. Some interesting data were collected, indicating that the purified NAT may exhibit low- activity or indicating that the methods used might need extensive optimization to investigate this specific NAT activity. A large degree of optimization has also been performed to utilize the aforementioned methods in these experiments. Expression of POMC in relation to NatF-expression was also investigated, but the results attained so far are inconclusive.

Characterisation of DNA Topoisomerase II alpha as a nuclear phosphoinositide-effector protein

Rhîan Gaenor Morgan (A. Lewis, V. Ardawatia)

Phosphoinositides (PIs) are second messengers composed of an inositol ring head group and two fatty acid tails. There are a total of 7 species of PIs which occur from different patterns of mono- di- or tri-phosphorylation on the inositol ring head group. As more work is done on the role of PIs in the nucleus, it is becoming more evident that PIs form a vast and complicated signalling network affecting numerous processes such as cell cycle regulation, apoptosis, pre-mRNA splicing (to name only a few). Although they have been implicated in regulation of nuclear processes, the precise mechanism by which they do still remain unknown. In a previous study, Lewis et al (2011) identified and validated DNA Topoisomerase II α (Topo II α) as a PtdIns(4,5)P2 binding partner, and showed this interaction can inhibit the decatenation activity of Topo II α. Moreover, it was suggested that the binding between Topo II α and PtdIns(4,5)P2 might occur via an electrostatic interactions between the negatively charged phosphate groups on PtdIns(4,5)P2 and the 7 highly basic arginine and lysine (K/R) motifs (consensus: K/R-X (n= 3-7) –K-K/R-K/R) concentrated in the C-terminal domain (CTD) of Topo II α. Therefore, the purpose of this thesis was to further characterize the interaction between the CTD of Topo II α and PIs. Different deletion constructs of Topo II α CTD lacking specific K/R motifs were assessed for their binding abilities to various PI species in lipid overlay assays. In general, deleting the K/R motifs led to a decrease in the binding of the mutant CTDs to the PIs. However, the specific deletion of K/R motifs M4-5 and 7 resulted in the most significant reduction in PI-binding, this suggests that more than one K/R motif may be involved in binding to PIs. Furthermore, individual peptides corresponding to the K/R motifs M2, M3, M4-5, M6 and M7, where found to be able to relive PtdIns(4,5)P2 inhibition of Topo II α decatenation activity in competition studies. Taken together, these results help validate K/R motifs in the CTD of Topo II α as PI-binding sites. In addition, when the sub-nuclear localisation of GFP-fused Topo II α CTD mutant (with K/R motifs M4-5 deleted) was analysed, it was found to localise significantly less often to the nucleoli as compared to the wild type Topo II α CTD. This observation suggests that the K/R motif M4-5, at least in part, is required for the nucleolar localisation of Topo II α. Overall, the findings from this study demonstrate the importance of K/R motifs in mediating Topo II α / PI interactions and indicate that sequences within K/R motif M4-5 might influence the sub-nuclear localisation of Topo II α.

Functional Analysis of the human N-terminal acetyltransferase hNaa40p

Monica Dalva (T. Arnesen, K. Hole)

Human orthologous of the yeast N-terminal acetyltransferases (NATs) Naa10p, Naa20p, Naa30p and Naa50p have been identified. Recently, the human orthologue of yeast Naa40p was also identified. This novel protein was denoted hNaa40p, and its activity hNatD. hNaa40p contains a conserved Acetyl Coenzyme A (Ac-CoA) binding motif, hence indicating that this protein may possess acetyltransferase activity as is the case for yeast Naa40p and other NATs. The ORF (open reading frame) encoding hNAA40 was also recently found to contain an alternative translation initiation site, which results in expression of a truncated hNaa40p variant, namely hNaa40p-M22. This variant lacks the first 21 amino acids at the N-terminal of the full-length hNaa40p, but contains the conserved Ac-CoA binding motif.

Functional analyses of hNaa40p with focus on acetylation activity, localization and functional roles were the initial aims of this study, however, during the thesis work functional analyses of hNaa40p-M22 were included. To investigate whether the two hNaa40 proteins possesses NAT activity toward histone H4, a substrate known to be acetylated by yNaa40p, recombinant MBP-hNaa40p and MBP-hNaa40p-M22 were purified and subjected to an in vitro acetylation assay. The results showed that both hNaa40p variants N-terminally (Nt) acetylated the N-termini of human histone H4 peptides. Ectopically expressed hNaa40p was also found to Nt-acetylate the yeast histone H4 in vivo. Deletion of yNAA40 in yeast strains has previously been shown to result in minor phenotypes in the ynaa40-Δ strains. To elucidate if ectopically expressed hNaa40p could have an effect on the previously reported phenotypes, a complimentary ynaa40-Δ-dependent phenotype analysis were conducted. No results were found, neither for the ynaa40-Δ strains nor for the strains ectopically expressing hNaa40p.

In the search of potential functional roles of the two hNaa40p variants, the localization of hNaa40p and hNaa40p-M22 was investigated using immunofluorescence and sub cellular fractionation studies. Based on the experiments, both hNaa40p variants were distributed to the nucleus and to the cytoplasm. By quantification of transfected cells in the immunofluorescence analysis, it was shown that hNaa40p and hNaa40p-M22 were more distributed to the nucleus. The functional roles of the two proteins in the nucleus and the cellular functions of the hNaa40 proteins remain to be further investigated.

Molecular Characterization of Mice Depleted of the Schizophrenia Susceptibility Gene Csmd1

Siri Merete Ratvik (B. Håvik, V. M. Steen)

Schizophrenia is a highly hereditary psychiatric disorder with a mostly unclear etiopathogeneses. Evidence is now pointing towards a link between psychiatric disorders, like schizophrenia, and genes expressed by the immune system. One of these genes is CSMD1, coding a CUB and sushi multiple domain protein, an inhibitor of the classical complement cascade (Lau and Scholnick, 2003).

In order to study the effects of the schizophrenia susceptibility gene Csmd1 on genotypes in neurological functions and behaviors, a novel mouse model (Csmd1 KO mice) has been produced by Lexicon (USA) after designs from the Dr. Einar Martens research group (Bergen).

To verify the knock down of Csmd1 in the mouse model, a western blotting protocol for detection of CSMD1-protein has been established for use in the laboratory. Findings from western blotting and qPCR confirmed a significant knock down of Csmd1. However, some residue product was detected and attributed to splice variants of the gene.

Csmd1 may contribute in a gene regulatory network that includes the expression of Csmd2 and mir-137, which also were identified as strong susceptibility factors in schizophrenia (Havik et al., 2011, Ripke et al., 2011). An effect of Csmd1 knock out on Csmd2 and mir-137 was not found by qPCR analysis.

A routine for cultivation of cortical primary neuronal cultures was set up and is now in use in the laboratory (succeeding this project). In order to gain insight into the function of Csmd1, its subcellular expression in primary cortical neurons was analyzed with immunocytochemistry. The staining pattern of a custom made goat anti-CSMD1 antibody was checked by comparison with a previously verified chicken anti-CSMD1 antibody (Kamal et al., 2010). The localization of CSMD1-protein in cortical neurons compared with CSMD2-protein and PSD95-protein has been assessed with a double-staining protocol. CSMD1-protein was partially co-localized with CSMD2, but did not co-localize with PSD95.

Role of the Parkinson’s disease-related protein DJ-1 in oxidative stress-induced cell death

Thaddaeus Mutugi Nthiga (K. E. Fladmark, L. V. Hjørnevik)

Loss of function mutations in the DJ-1 protein cause early on-set Parkinson disease (PD). This suggests that DJ-1 plays a role in neuroprotection, but how DJ-1 dysfunction triggers pathogenesis is not understood. Multiple functions of DJ-1 have been proposed, including regulation of oxidative stress and cell death. The precise mechanistic action of DJ-1 is however unknown. Its cellular localisation and importance hereof is debated. We have used a neuroblastoma SH-SY5Y cell line to investigate oxidative stress response and its effect on DJ-1 localisation and protein interactions. The non-sumoylatable K130R and the pathogenic L166P DJ-1 mutants were also investigated. Under basal conditions, endogenous DJ-1 was predominantly present in the cytoplasm. This cellular localisation was not altered by oxidative stress exposure. Oxidative stress induced an up-regulation and accumulation of acidic isoforms of the endogenous DJ-1. While the non-sumoylatable K130R mutant DJ-1 had no effect on the expression of DJ-1 compared to the wild-type protein, the expression of the pathogenic L166P mutant was substantially reduced due to degradation by the proteasome. Based on prior quantitative proteomics data we should confirm proposed DJ-1 interactions to potassium channel tetramerisation domain protein 5 (KCTD5) and Janus kinase 1 (JAK1). By co-immunoprecipitation studies, an endogenous interaction between KCTD5 and DJ-1 was confirmed, but the proposed DJ-1/JAK1 interaction was found to be the result of a general down-regulation of JAK1 in oxidative stress cells.

Influence of Polycystic Ovary Syndrome on serum lipids and fatty acids under Basal and Hyperinsulinemic conditions
Effect of Pioglitazone

Ojukwu Emeka Everest (R. Berge, P. Bohov, J. Skorve)

Polycystic ovary syndrome (PCOS) is a common, chronic endocrine condition affecting young women of reproductive age. It is characterised by hyperandrogenaemia, and profound menstrual and ovulatory dysfunction with consequent sub-fertility. Metabolic complications commonly include insulin resistance, disordered lipid metabolism and chronic low-grade inflammation. Overweight and obesity, as well as a degree of adipose tissue dysfunction, are often present in a large proportion of women with PCOS.

Lifestyle modification should be used as primary therapy for the treatment of metabolic abnormalities. If cardiovascular risk factors persist despite lifestyle management, the administration of lipid-lowering, insulin-sensitizing and other agents, targeting cardiovascular risk factors, is recommended. Pioglitazone should be used in insulin-resistant or obese PCOS women who do not tolerate or do not respond to metformin therapy. In lifestyle intrerventions, dietary fats are now recognized as biological regulators that influence various aspects of metabolic health. Polyunsaturated fatty acids (PUFA), particularly those of the n-3 family, are perhaps the most potent fatty acid (FA) regulators of metabolic function and are implicated in a diverse range of processes in vivo. In PCOS subjects, however, fatty acid profile of blood and other tissue lipids is not yet recognized.

Thus, the aim of the study was to evaluate serum fatty acid composition of PCOS patients under basal and insulin stimulated conditions. The effect of treatment with pioglitazone on serum FA levels was also estimated.

The results showed that serum FA composition of obese insulin resistant PCOS subjects, matched for age and obesity to the controls differs from the controls. An increased level of serum palmitic acid (C16:0) and some trans FA in PCOS subjects is atherogenic and may also impair an insulin sensitivity. On the other hand, decreased PUFAn-3, especially EPA (C20:5n-3) was found. After adjustment for serum triacylglycerols to the controls, these PCOS patients revealed also increased proportion of arachidonic acid (C20:4n-6) in serum. Such a trend in PUFA is associated with a greater circulating androgen concentrations and may impact steroidogenesis.

Insulin stimulated metabolism of FA during hyperinsulinemic euglycemic clamp decreases serum lipid classes and affect the serum FA composition in favor of PUFA whereas the percentage of SFA, MUFA and trans FA declines. This trend is almost identical in both,9controls and PCOS, and indicates that maximal insulin effect on the regulation of serum lipids and FA is preserved in PCOS subjects.

The serum lipids and FA levels were not changed after pioglitazone treatment, but under insulin stimulated conditions a higher utilization of serum MUFA and some trans-FA in pioglitazone treated subjects were found. This may be due to an up-regulation of mitochondrial β-oxidation of MUFA for energy requirement after pioglitazone treatment. This needs, however, a further evaluation.

Investigating Potential Functional Effects of Glucose-Dependent Dynamic N-Terminal Acetylation in Saccharomyces cerevisiae

Sylvia Varland (T. Arnesen, H. Aksnes)

In the light of recent findings by Wellen et al., showing that reduction of acetyl-CoA levels diminishedlysine acetylation of histones, we hypothesized that Nt-acetylation might also be metabolicallyregulated. Initially, the Arnesen and Gevart laboratories performed full-scale proteomic analysis onwhole-cell yeast extracts obtained at distinct metabolic conditions. An interesting observation fromthe resulting proteomic dataset was that a subset of proteins experienced fluctuations in their Ntacetylationstatus (Arnesen et al., unpublished data). The observed dynamicity in Nt-acetylationseemed to be dependent upon glucose/acetyl-CoA levels. The present thesis study was initiated asan effort to verify some of the observed fluctuations in Nt-acetylation. In addition, we wanted toassess potential protein-specific functional implications, i.e. localization, stability or function, causedby altered Nt-acetylation levels. For this objective, we wanted to utilize yeast strains expressing GFPfusionproteins.

This study was started by the evaluation of potential candidate proteins. Considering variousparameters, such as the absolute shift in Nt-acetylation, a group of particularly interesting proteinswas selected for in-depth analyses. In order to study these proteins in the absence of the Nt-acetylgroup, we endeavored to delete all the respective catalytic NAT-subunits. Ultimately, we were ableto transform all the NatA and NatC candidates, using homologous recombination of a kanMX4deletion cassette. Next, we set out to study the potential effects of the suggested metabolicallyregulatedNt-acetylation. Interestingly, several of the candidate proteins underwent a shift in theircellular concentration following deletion of the corresponding catalytic NAT-subunit. However, thesefindings are limited due to variations between independent experiments. Fascinatingly, Smy2p, aprotein of enigmatic function, was consistently regulated as shown in independent experiments.Also, this protein presented the most dramatic shift in Nt-acetylation status, changing from 33.7 % inactive to 100 % in stationary cells. Thus, Smy2p represent a candidate protein that presumably isfunctionally regulated by both growth phase and the presence of a functional NatC complex, andthus became a protein of great focus in the present thesis. Here, we suggest two alternative workingmodels for the functional implications of NatC-mediated Nt-acetylation of Smy2p. First, Nt-acetylatedSmy2p may act in ER-to-Golgi transportation through the involvement in COPII vesicle formation, butis recruited for the assembly of P-bodies upon glucose deprivation. Alternatively, Nt-acetylation maydestabilize Smy2p. The models can be used as a basis for future studies of the exact link between glucose metabolism, Nt-acetylation status and cellular function of Smy2p.

Taken together, the data presented here suggests that individual proteins are actuallyaffected by the fluctuations in Nt-acetylation induced by glucose deficiency during transition tostationary phase. However, more research needs to be undertaken before the interplay betweenmetabolism and Nt-acetylation is more clearly understood.

Specificity of CW domains for modified histone peptides

Øyvind Strømland (R. Aasland, V. Hoppmann)

Chromatin packaging plays a central role in the regulation of gene expression not only as transient signal‐response, but also throughout cell generations, in what is known as chromatin‐mediated epigenetic inheritance. Chromatin structure, and thus gene expression, is partially regulated by chemical modifications on histone tails. Recognition domains often situated in effector proteins, which can alter chromatin architecture, bind to these modifications. The CW domain was recently identified as a recognition domain showing specificity for histone H3 methylated at lysine 4. In humans, seven proteins harbouring CW domains have been identified, but hitherto the binding specificity of only two has been characterized. The aim of this project has been to characterize the ligand specificity of the remaining human CW domains, and to further characterize the different features of the domain leading to its binding specificity. ZCWPW2‐CW was found to bind to histone H3 methylated at lysine 4, while MORC3 was found to bind both histone H3 methylated at lysine 4 and 9. During this study, a novel CW domain was found in the nematode C. elegans. The MORC-like CW domain was cloned, expressed and shown to have novel histone-binding properties. This is the first time CW domains have been reported outside plants and chordates.

Heterogeneous Ribonucleoprotein U interacts with Phosphoinositides

Amanda Jayne Edson (A. Lewis, V. Ardawatia)

Phosphoinositides (PIs) are important molecules in the cell. They are found in both the cytoplasm and the nucleus of the cell. Often they are found embedded in lipid membranes however in the nucleus they also exist unattached to the nuclear envelop. These PIs act as vital signaling molecules to the cell either as second messengers themselves or as metabolites for signaling pathways. The PIs have a vast variety of functions from vesicle trafficking, gene expression, cell proliferation and survival, protein localization, signal transduction along and new functions are being discovered all the time. In an effort to elucidate the functions of these molecules in the nucleus the proteins with which they interact are often studied. Previous work has suggested that heterogeneous ribonucleoprotein U (hnRNP U) may be a nuclear PtdIns(4,5)P2-binding protein by a lipid pull down assay and mass spectrometric identification. This 120 kDa protein has several functions including mRNA stabilization, X-chromosome inactivation, and acting as nuclear scaffolding protein. Currently there is no knowledge of hnRNP U as a nuclear PI binding protein. In this study it was found that endogenous hnRNP U was capable of binding all PI species in vitro. Also recombinant protein fragments generated from the N-terminal, SPRY domain and C-terminal domains were all capable of binding PIs. While several areas of the protein may be sites of PI binding, a binding site was identified in the N-terminus of the protein consisting of patch of basic residues known as a K/R motif. A deletion of this region in the recombinant protein abrogated PI binding and was thus shown as a critical region of binding. Lastly nuclear staining of HeLa cells showed that hnRNP U was an extremely abundant and homogenous distributed protein in the nucleus (excluding the nucleolus) and thus while it did not colocalize specifically to the nuclear speckles containing PtdIns(4,5)P2 it is available in the region for interaction. Additionally preliminary staining results suggest that there might be two pools of hnRNP U which have individual localization patterns. As a whole these results not only suggest hnRNP U as a PI effector protein but also further validate the hypothesis that the unstructured K/R motif acts as a possible PI binding site in some nuclear effector proteins.

Characterization of the transcription factor NF-­E2p18/MafK as a nuclear phosphoinositide effector protein

Hilde Kristin Grung-Berle (A. Lewis, V. Ardawatia)

Phosphatidylinositol is a unique phospholipid of great interest, considering its ability to generate seven biological distinct phosphoinositides (PIs) by phosphorylation of its inositol head group. Multiple different kinases, phosphatases and phospholipases are responsible for PI modifications, implicating the vastness of the cellular functions of these phospholipids. PIs are signalling molecules found to play essential roles in a variety of cellular processes, such as differentiation, proliferation, gene expression, stress response etc., either by direct site-specific binding or as second messengers. PIs insert their two acyl chains in the hydrophobic membrane of the plasma membrane and other organelles, including the nucleus. At present, it is believed that there is more than one pool of PIs in the nucleus, including some that are not embedded in the nuclear envelop. The question of how these lipids shield their hydrophobic tail from the hydrophilic environment is still unanswered. One theory describes the prospect of carrier proteins enclosing the hydrophobic tails.

The complexity of these lipids and their interaction capacity suggests a wide array of possible effector proteins bound by either structured domains or unstructured basic motifs. A recent study in the laboratory published a list of possible PtdIns(4,5)P2 binding proteins of nuclear  origin, amongst them the transcription factor MafK. MafK is involved in the transcription of the beta-globin gene, encoding a part of the adult hemoglobin protein, by dimerizing with the NF-E2p45 transcription factor. It has proven to be a dual--‐function molecule shifting from a repressing to an activating state depending on its dimerization partner.

In this study MafK demonstrated binding to PtdIns3P, PtdIns4P, PtdIns5P, PtdIns(3,5)P2 and PtdIns(4,5)P2 in vitro. MafK was also displaced out of the nucleus by neomycin, a competitive agent with high affinity for PIs. Together with preliminary immunostaining results from MEL cells with indications of partial co--‐localization between MafK and PtdIns(4,5)P2, these results suggest that MafK may function as a PI effector protein in the nucleus. The functional implications of such an interaction within the nucleus are yet to be understood and serve as a motivation for further research.

The study of potential hNaa30p substrates EHD1 and eIF2D in hNaa30p depleted cells

Javier Sánchez Reyes (T. Arnesen, K. Starheim)

Nt-acetylation is one of the most common protein modifications in eukaryotic cells and it has been related to diverse cellular processes. Our group has identified the human protein Nt-acetyltransferase complex C (hNatC) of which hNaa30p is the catalytic subunit. Little is known about the function of hNatC and there are many potential substrates for hNaa30p. It was suggested that Naa30p has independent enzymatic activity of Naa35p and Naa38p in higher eukaryotes. Human Naa30p was found to have a role in Golgi organization in human cells. hNaa30p influences organelle organization through acetylation of GTPases that work on organelle trafficking. Depletion of NatC subunits leads to human cell death and decrease cell growth, and it also leads to the tRNA-specific m22G methyltransferase displacement from the inner nuclear membrane. Besides, depletion of hNaa30p also alters the subcellular localization of hArl8.

In the present project, we addressed the identification of functionally important hNaa30p substrates under the depletion of the hNatC catalytic subunit hNaa30p. To reach this purpose, our first objective was to clone the ORFs of the potential substrates into EGFP-N1 expression vector. As for the second objective, we planned to study the recombinant GFP-proteins in hNaa30p knockdown cells by doing siRNA transfection, plasmid transfection, Western blotting and immunofluorescence analyses and protein subcellular localization under fluorescence microscope.

Our hypothesis was: the hNaa30p knockdown phenotypes are due to the hNaa30p loosing acetylation, and this acetylation is important for its function.

In this study, we found that there was partial hNaa30p depletion by sihNAA30-1 and sihNAA30-2 in HeLa cells. No change in EHD1-GFP localization was observed in hNaa30p-depleted cells. When expressed in HeLa-cells, EHD1-GFP took a speckled distribution, in contrast to the tubular distribution that has been shown in previous studies of EHD1. Moreover, apparent perinuclear localization was shown in cells transfected with siRNA Ct and siNAA30s being slightly greater in cells transfected with hNAA30-1. As for eIF2D, a perinuclear localization was identified for the cells transfected with siRNA Ct, siNAA30-1 and siNAA30-2.