Received 21 March 2015; accepted 25 July 2015; published 28 July 2015

ABSTRACT

The Dictyostelium discoideum AMP-activated protein kinase (AMPK) snfA subcellular localization was studied in AX2 and stable HPF strains by use of AMPK antipeptide antibody and goat anti- rabbit Alexa-Flour 488-conjugated IgG antibody. The AMPK exhibited cytosolic localization patterns and uniform focalised concentrations in wild type and the strains alike. Constitutive activation and attenuation of the α subunit expression did not affect subcellular distribution of AMPK. However, snfA expression was more intense in strains in which AMPK was constitutively active compared with the AX2 but lesser in attenuation strains. The localisation of the snfA reinforced the putative standing that it had a plethora of cytoplasmic functions. Moreover, the oxidative cellular function would require a ubiquitous system and might coordinately regulate responses to metabolic requirements. Furthermore, the developmental phases of the life cycle would support the cytosolic localization; and since organelles were potentially reorganized or removed entirely during the transition from vegetative living to fruiting body morphology. This study provided insight into the subcellular distribution of AMPK in Dictyostelium discoideum. We demonstrated that AMPK localization was steady in AX2 and derived strains whether constitutively active or anti- sense inhibited depicting extreme genetic states.

Keywords:

AMPK α Subunit, Dictyostelium discoideum, Subcellular Localisation, AX2, Mitochondrial Assay

1. Introduction

Sucrose non-fermenting protein (snf)A/AMP-activated protein kinase (AMPK) ortholog are eukaryotic serine/ threonine kinases that mediate cellular energy homeostasis [1] [2] . The enzyme cascade regulates cell growth, transcription and the cellular response to nutrient limitation and stress [3] - [5] . In Dictyostelium, each subunit isoform is encoded by a single gene [6] [7] . The Dictyostelium AMPK α subunit gene ortholog snfA, encodes a protein of 727 amino acids with a molecular weight of 81638.9 Da; and is located on chromosome 3 (position 589784 - 592434) of the Dictyostelium genome [6] . The D. discoideum AMPK α subunit is made up of 5 exons and 4 introns [7] . The Asparagine-rich domains such as that within the AMPK α sequence are prevalent in Dictyostelium proteins, but their functions are unknown [7] . The D. discoideum AMPK α subunit is structurally partitioned into four major domains namely: the catalytic domain, also known as serine-threonine kinase domain (S_TKc), composed of 254 residues (positions 31 - 284), contains the phosphorylation residue, Thr188 on the α-subunit, which can be mediated by upstream regulatory kinases; the KIS binding domain (residues 330 - 389, 630 - 727) that interacts with the β subunit and is interrupted by an intervening asparagine-rich (241 residues) domains inserted 98 amino acids upstream of the C-terminus) and the autoinhibitory region extending to the C-terminus [8] . The KIS binding domain and autoinhibitory regions are required for assembly of the α subunit into the heterotrimeric holoenzyme and regulation by the γ subunit respectively assembled on β subunit [7] . These domains of AMPK α subunit are features shared by AMPK α subunit or their homologues in other eukaryotes showing that the AMPK α subunit is highly conserved in eukaryotes [9] . To date, little is known about precise subcellular localization of snfA within cell and how its intracellular distribution is controlled. This knowledge is important, because it will set the stage to identify specific AMPK functions in compartments that are dictated by physiological changes.

2. Materials and Methods

2.1. Dictyostelium discoideum: Strains and Culture Conditions

All experiments were conducted with D. discoideum AX2 wild-type parental strain [10] , and strains derived from it [7] . Each strain carried multiple copies of the following plasmid constructs: 1) pPROF362 (the AMPK α (snfA subunit antisense inhibition constructs) in HPF456-462; 2) pPROF361 (its corresponding RNA control construct in HPF466-468; or 3) pPROF 392 (the AMPK α³⁸⁰ overexpression construct) in HPF434-442 strains. All strains were isolated using the Ca(PO₄)₂/DNA coprecipitation method [11] and selected as isolated, independent colonies growing on Micrococcus luteus lawns on SM agar supplemented with 20 μg/mL geneticin (G-418) (Promega Corporation, Madison, WI). Strains of Dictyostelium were maintained as axenic cultures at 21˚C ± 1˚C in HL5 medium supplemented with 100 µg/mL ampicillin and 20 µg/mL streptomycin or on SM plates with Klebsiella aerogenes as a bacterial food source. The selective agent Geneticin (G418 @ 20 µg/mL) was added to HL5 medium for all strains. All localization experiments were performed on vegetative, exponentially growing cells.

2.2. Gene Cloning and Sequence Analysis

AMPK cloning strategies and vectors were as described previously [7] . The snfA [8] ; EMBL/GenBank ID AF118151) encoding the AMPK α subunit was amplified from genomic DNA template using the gene-specific primers PAMKF1 P5’ > GCGCTCTAGATTCGAAAAAATCATGAGTCCATATCAACAATAATCCCATT > 3’and PAMKR1 5’ > GCGCTCTAGACTCGAGTTAAACTACAAATATCAAAAATATGAATATTTCACC > 3’ [7] and cloned into pZerO₂ (Invitrogen, Carlsbad, CA). The amino acid sequences of AMPK α subunit gene used in this study had been earlier characterised [7] .

2.3. Generation of Constitutively Active Form of AMPK α Subunit

The cDNA encoding the catalytic domain of AMPK α subunit consisting of 380 amino acid residues was amplified using PACDNAF1 5’-GCGCTCTAGAAGCTTCTCGAGTTCGAAATGAGTCCATATCAACAAAATCC- CATTGG-3’ and PACDNAR1A 5’-GCGCTCTAGACTCGAGCCCGGGAATTCTTATTGGCCTCT-GGGGA- GCACTGACAT-3’ primers by reverse transcription PCR (RT-PCR) from RNA extracted from vegetative AX2 using the RNAzol (Life Technologies Inc., Grand Island, NY). The amplified catalytic domain comprising 1140 bp cDNA fragment designated AMPK α³⁸⁰ was cloned into the XbaI-linearized pZErO^TM-2 thereafter subcloned into the ClaI-XhoI sites of the pA15GFP [12] [13] replacing the resident green fluorescent protein (GFP) gene, creating a fusion of the actin15 promoter-AMPK gene to yield pPROF 392 for overexpression of a constitutively active, truncated AMPK α subunit. Localization of a fluorescently tagged protein expressed by the Act15 promoter with the markers has been used extensively to define subcellular localization [14] [15] .

2.4. Creation of the AMPK α Subunit Antisense/Sense Constructs

Antisense inhibition offers partial inhibition of expression of essential genes producing sublethal phenotypes. Plasmid constructs for expression of antisense RNA and and the corresponding sense RNA control were created from the full length genomic clone (pPROF348) template by amplifying a fragment using primers PAMPKF10 [5’GCGCTCTAGAATTCCCTATGGATGAAAAGATTAGAAGA3’] and PAMPKR10 [5’GCGCTCTAGAA- TTCTCCATGCTATTGCTATTGGTGG3’], cloned into XbaI site of pZErO^TM-2 and subcloned into the EcoRI site of the Dictyostelium expression vector, pDNeo2 under the control of the actin 6 promoter [16] in both orientations. The resulting AMPK α sense (pPROF361) and antisense (pPROF362) constructs express 1172 bp of the AMPK α antisense RNA along with the corresponding sense RNA control. Both fragments would be expressed under the control of the Dictyostelium actin 6 promoter.

Sequences of PCR products and clones were verified by sequencing conducted through the Australian Genome Research Facility (AGRF), Brisbane, Australia.

2.5. Determination of Plasmid Copies and Estimation of RNA Expression in Strains by ECF Signal Amplification Method

Equal amounts of DNA (digested with EcoRI) or RNA from AX2 and each strain was used. The DNA and RNA resolution, transfer and hybridization for quantitative Southern and northern blotting were performed and employing the fluorescein-labelled DNA probe and liquid block. For enhanced chemifluorescent detection, the blot was incubated in 5000-fold dilution of anti-fluorescein AP-conjugate containing 0.5% (w/v) BSA in buffer A1 (0.32 mL/cm²) with gentle agitation at room temperature for 1 hr and washed 3 × 10 min in 0.3% (v/v) Tween 20 in buffer A1 at room temperature with agitation and blot was scanned in the Storm860^TM Fluoroimager (Amersham Biosciences, Castle Hill, Sydney, Australia) using the 530 nm filter in fluorescence mode. The RFU values from each strain and serial dilutions of linear plasmid of known concentration as well as the RFU values from AX2 gDNA were analysed using the ImageQuant tool (TLv2003.03) software as the ratio of the RFU values of respective strain to the RFU values of AX2. The quantitative figures from the Southern blot enabled the computation of plasmid copy numbers for all strains tested whereas the signals from the northern blots were used for the computation of RNA expression levels in strains and AX2.

2.6. Mitochondrial “Mass” Estimation

MitoTracker green FM (Molecular Probes™ Inc., Eugene, OR, USA) selectively stains mitochondria in a manner that is independent of mitochondrial membrane potential. MitoTracker green fluorescence was therefore used as a measure of mitochondrial “mass”. Axenically growing cells at the exponential phase was spun at 3000 x g for 15 sec, washed once and incubated in Lo-Flo HL-5 (3.85 g/L glucose, 1.78 g/L proteose peptone, 0.45 g/L yeast extract, 0.485 g/L KH₂PO₄ and 1.2 g/L Na₂HPO₄∙12H₂O; filter sterile) buffer for 2 hrs. An aliquot was resuspended in Lo-Flo HL-5 containing 200 nM MitoTracker green FM alongside an unstained control and incubated for an hour in the dark. The unbound MitoTracker green was removed by washing the cells 3 times with Lo-Flo HL-5 for 10 min on an orbital shaker (150 rpm). The cells were resuspended in Lo-Flo HL-5 and fluorescence was measured in duplicate in a fluorometer (Turner Biosystems Modulus™ Inc., Mary Ave. Sunnyvale CA, USA) using the Blue Module. The MitoTracker green fluorescence per million cells was calculated after subtraction of the background fluorescence as in the unstained cells. The correlation of MitoTracker green fluorescence intensity was tested against plasmid copies in the in overexpression and antisense-inhibited strains relative to AX2.

2.7. Mitochondrial Staining

Mitochondria location within the cell was assayed by use of MitoTracker Red (CMX-Ros, Molecular Probes™ Inc., Eugene, OR, USA). Briefly, 5 - 7 mL axenically growing culture cells at 2 × 10⁶ cells/mL was deposited on sterile coverslips in the 6-well plates (Nalge Nunc™, Naperville, IL) and allowed to settle and adhere to the coverslip over a period of 4 - 6 hr. The cells were gently washed once in fresh Lo-Flo H-5 medium for 2 hrs. One well was incubated only Lo-Flo HL5 as an unstained control. The cells adherent to the coverslip were stained with 200 nM MitoTracker Red (CMX-Ros, Molecular Probes™ Inc., Eugene, OR, USA) in LoFlo HL-5 for 2 hours in the dark. Excess stain solution was decanted and unbound MitoTracker Red was washed off the cells with 4 gentle washes with the Lo-Flo HL-5 over a 2 hr period in the dark with shaking on an orbital shaker (150 rpm) between washes. Lo-Flo HL-5 also reduces autofluorescence from the HL-5 medium [17] . After washing the stained cells, they were fixed for 15 min in 3.7% paraformaldehyde in PBS (12 mM Na₂HPO₄, 12 mM NaH₂PO₄, pH 6.5) and the fixed cells were washed with PBS 4 times for 5 min. The cells were mounted on clean glass slides with mounting solution (2.6% Dabco in 90% Glycerol and 10% PBS) and sealed. The mounted slide was viewed in a conventional fluorescence microscope (Olympus BH-2 or BX-50) and confocal microscope at the Confocal Microscopy Facility, Deakin University, Melbourne, Australia and images of AX2 and HPF representative cells were captured and analysed.

2.8. AMPK Subcellular Localisation Assay

Our present work focuses on determining the cellular localization of the Dictyostelium AMPK using fluorescent fusion proteins driven by the heterologous actin 15 (Act15) or Dictyostelium actin 6 promoter promoter. Axenically growing AX2 or strains vegetative cells were harvested, washed and fixed for 15 min in 3.7% paraformaldehyde in PBS. The fixed cells were washed with PBS 4 × 5 min. The cells were then permeabilised by incubating them in 0.1% - 0.5% Triton X-100 in PBS for 5 min and washed 4 × 5 min in PBS. The fixed and permeabilised cells were blocked for 3 hr at 4˚C, incubated overnight in primary antibody 500-fold dilution of AMPK antipeptide antibody in blocking buffer (Roche Diagnostics, Mannheim Germany) at 4 ˚C and washed 4 × 5 min at 21˚C ± 2˚C in PBS containing 0.05% Tween20. The cells were then treated for 45 min at 21˚C ± 2˚C with goat anti-rabbit Alexa-Flour 488-conjugated IgG antibody (Molecular Probes™ Inc., Eugene, OR, USA) using a 1:2000 dilution in blocking buffer. Finally, the cells were washed 4 × 5 min with PBS containing 0.05% tween 20, air dried for 5 min, mounted on clean glass slides with mounting solution (2.6% Dabco in 90% Glycerol and 10% PBS) and sealed. The mounted slide was viewed in a conventional fluorescence microscope (Olympus BH-2 or BX-50) and images of AX2 and HPF representative cells were captured and analysed.

2.9. Statistics

Results are shown as means ± SD. Bonferroni tests for multiple statistical comparisons and online supplment and Student’s t-test (two-tailed) for unpaired samples were carried out to identify significant differences. For each experiment, all test results were compared with the control. R² is the coefficient of variation and equals the square of the Pearson product-moment correlation coefficient. The significance probability is the probability of the observed results occurring under the null hypothesis that the correlation coefficient was zero.

3. Results

The AMPK α subunit in Dictyostelium is encoded by a 2.6 kb single gene annotated as Q9XYP6 (www.expasy.org), AF118151 (www.ncbi.nlm.nih.gov/entrez) and DDB0215396 (http://www.dictybase.org/blast); comprised of five exons and four introns [9] . The N-terminus holds a highly conserved catalytic core, S_TKc domain and identical APE and DFG motifs with other eukaryotic cells (Figure 1).

3.1. Plasmid Copies and RNA Expression

The form of the AMPK α subunit that was over expressed contained the entire catalytic domain but was truncated within the putative region responsible for autoinhibition and binding to the β subunit. The truncation of the AMPK α subunit in the catalytic domain created a constitutively active form of AMPK. Southern blot analysis for the strains over expressing truncated AMPK α subunit and strains in which AMPK is antisense inhibited yielded bands that varied in their intensities (Figure 2(A) and Figure 3(A)). This indicates different plasmid copies of the respective strains. The least prominent band showed intensity higher than AX2 the wild type strain (Figure 2(A) and Figure 3(A)).

The transformants exhibited a characteristic, stable level of RNA expression and the differences reflected the expression vector copy numbers insertions in the genome. Both plasmid constructs affected expression in the

expected copy number-dependent manner. A steady state level of AMPK α³⁸⁰ mRNA in quantitative northern blots tightly correlated with the plasmid copies of the AMPK α³⁸⁰ expression construct insertions in the strain genomes (Figure 2). Overexpression mutants showed strong RNA expression signals of increasing intensities as the copy number of AMPK α³⁸⁰ construct increased (Figure 2). The sense RNA fragment in pPROF361, was also expressed at high levels and tightly correlated with the plasmid copies (Figure 3(B)). By comparison with the sense RNA controls the equivalent antisense RNA levels were dramatically reduced even at the lowest plasmid copies of the antisense construct (Figure 3). Interestingly, the extent of RNA reduction correlated with the

plasmid copies of the antisense RNA-expression construct (Figure 3 and Figure 4). This indicates probable degradation in the antisense RNA expressing strains of both the antisense and the native mRNA. This would mean lower expression levels of the native mRNA. Additionally, the extent of elevation (overexpression) or reduction (antisense) of RNA levels tightly correlated with the plasmid copy number (Figure 4). Accordingly, the plasmid copies of the corresponding constructs were used as AMPK α subunit expression index.

3.2. Mitochondrial Mass Assay

The mitochondria mass and localisation in the strains tested was assayed with MitoTracker green and MitoTracker red. The MitoTracker red stained only the mitochondria; and more stain intensity with higher mitochondrial mass (Figure 6). The overexpression of the AMPK α catalytic domain resulted in more mitochondrial signals that were more intense per cell, whereas AMPK α antisense inhibition exhibited lesser signal compared to that of AX2 cells (Figure 5 and Figure 6). Additionally, there was 2 - 3 fold increase in mitochondrial signals in strains overexpressing AMPK α subunit compared to AX2 (Figure 6). The mitochondrial fluorescence signal increase in AMPK α³⁸⁰ overexpression strains in relation to AX2 was assayed using the 530 nm filter in a Storm860^TM Fluoroimager (Amersham Biosciences, Castle Hill, Sydney, Australia) and analysed using the Image Quant tool (TLv2003.03) software (data not shown). The MitoTracker Green stain showed increased mitochondrial fluorescence in AMPK α³⁸⁰ overexpression strains but reduced in AMPK α subunit antisense strains compared AX2 (Figure 5). Increased intensity of MitoTracker Green stain in AMPK α³⁸⁰ strains signify elevated mitochondrial mass per cell. The higher the plasmid copies in the constitutively active AMPK strain the more the fluorescence from the assayed cells (Figure 5 and Figure 6). The increase in mitochondrial mass is indicative of mitochondrial proliferation. In the same vein, the severe the antisense inhibition, the more apparent the mitotracker fluorescence decrease became (Figure 5).

3.3. AMPK Subcellular Localisation in Dictyostelium

The AX2 and HPF strains were used to study the subcellular localization of AMPK. AMPK α was cytosolic in AX2 and the HPF strains (Figure 6). The AMPK α subunit distribution showed sequestration smooth homogeneity within the cytosol, less densely to the cell membranes and does not seem to localize to particulate cellular sites in AX2 cell (Figure 6(a)) and HPF strains (Figure 6(b), Figure 6(c)). Truncation of the α subunit at the 380 amino acid residue position to yield a constitutively active segement or at 510 position did not affect distribution or whether experimental tools discriminate between the cytoplasmic and nuclear pools of AMPK (Figure 6(b)). Constitutive activation or antisense inhibition of the subunit did not affect the subcellular localization (Figures 6(b) and Figure 6(c)). The same localization patterns were seen in growing cells of both the wild type and strains. The α subunit was not present in the mitochondria. This is evident in that the overlay did not portray this fact (Figure 6). This finding supports the earlier hypothesis that if AMPK was the proximal regulator of mitochondrial biogenesis it then would not be present in the mitochondria [9] .

4. Discussion

AMPK has been the focus of increasing attention for its fundamental roles in cellular energy homeostasis in healthy cells and in a variety of pathological situations, most notably diabetes, cancer [1] [18] and mitochondrial diseases [7] . The activated cascade regulates metabolic pathways and promotes the adaptation of eukaryotic cells to the ever changing microenvironment [1] [19] . Thus, AMPK functions as a metabolic masterswitch, or more accurately, “low fuel warning system” in mammalian [1] [18] and other eukaryotic cells [20] - [22] . Thus places AMPK holoenzyme at a central control point in maintaining cellular energy homeostasis. This pivotal role places AMPK in an ideal position to also play a wider role in regulating whole-body energy metabolism [23] . Furthermore, AMPK regulates the expression of a large number of genes, the stability of several mRNAs,

cell polarity, and mitosis [24] - [26] . Access to and modification of these substrates requires cytoplasmic localization of AMPK. Moreover, the oxidative cellular function will require a ubiquitous system and may coordinately regulate response to metabolic requirements.

In Dictydstelium, AMPK cascade have putative heteromeric forms and cytosolic localization patsterns. AMPK α remain cytosolic in location. Moreover, the developmental phases of the life cycle will support the cytosolic localization; and since the organelles are potentially reorganized or removed entirely during transitions from vegetative living to fruiting body morphology. Cytosolic subcellular localization of Dictyostelium snfA is not surprising given the extensive number of roles AMPK having been documented to play in this organism. The observed intracellular distribution of the snfA provides a means to maintain appropriate intracellular and organellar distribution of AMPK. The less complexity of Dictyostelium will make do with one gene coding for each subunit and the cytosolic localization. In the mammalian cells, AMPK shuttles between the nucleus and the cytoplasm under normal growth conditions, a process that depends on the nuclear exporter Crm1 [27] . However, many factors control the distribution of AMPK [27] [28] . For instance, environmental stress, heat shock or oxidant exposure drive more nuclear localization of AMPK [27] [29] . Given their apparently common localisation, AMPK tends to stay in the cytosol whether modified or wild type. In mammals, the AMPK subunits are encoded by multiple genes and preferentially and differentially localise to either the cytoplasm or nuclei [30] [31] . However, nucleocytoplasmic shuttling does not take place in high-density cell cultures, for which AMPK is confined to the cytoplasm [27] . This is typical state of the vegetative cells during vigorous proliferation and attaining high density frequently.

The localization of AX2 and HPF strains of the α subunit are similar. However, snfA expression ias more intense in constitutively active strains than that observe in the AX2 but lesser in attenuation strains. Constitutive activation of AMPK via truncation of the catalytic domain with significant increased in activity of the AMPK approximates the persistent AMPK activation such as from stressors [7] [32] . Strains expressing the antisense construct contained varied plasmid copies and corresponding RNA expression profile depicting dimunition of RNA as the plasmid copies increased; portrays a scenario of snfA attenuation by mutations. This indicates degradation in the antisense RNA expressing transformants of both the antisense and the native mRNA. This will mean lower expression levels of the native mRNA. During successful antisense RNA inhibition both strands of the mRNA, antisense RNA duplex is degraded by the RISC complex [33] . Moreover, strains with higher plasmid copies of antisense plasmids show severe growth and development defects [7] . The localisation of the α of snfA suggests that it may have plethora of cytoplasmic functions. Together, this intracellular localization provides a unique set of tools to rapidly adjust the distribution of AMPK to changes in cell physiology. The importance of these concepts is becoming ever more evident as mechanisms of hormone and drug action are resolved to the subcellular location of their target molecules [34] .

However, further studies needed to understand the importance and the dynamics of the compartmentalisation of the c2-AMPK complexes, focusing on specific interactions in each location. The need for isoform-selective activation of AMPK to develop cardioprotective therapies has been highlighted in a recent review article [35] and protein―protein interaction studies could result in new drug design that acts specifically on c2-AMPK complexes to ameliorate the disease caused by the PRKAG2 mutations.

Our study provides insight into the subcellular distribution of AMPK. Our results demonstrate that AMPK localization was steady in AX2 and derived strains whether constitutively active or antisense inhibited depicting extreme states of transformation.

Acknowledgements

We gratefully acknowledge the support by La Trobe University Postgraduate Research Scholarship (LTUPRS) and the Australian Government International Postgraduate Research Scholarship Scheme (IPRSS). The present study cannot have been completed without it.

Cite this paper

Paul B.Bokko,AfsarAhmed,Paul R.Fisher, (2015) AMPK Subcellular Localisation in Dictyostelium discoideum. American Journal of Molecular Biology,05,105-116. doi: 10.4236/ajmb.2015.53009

References