Desmin’s conformational modulation by hydrophobicity

Identification of proteins by mass spectrometry

Protein lysates from zebrafish skeletal muscle tissue was used for proteomic analysis. All procedures performed in studies involving animals were in accordance with the ethical standards of the Hacettepe University Animal Experimentations Local Ethics Board (18.02.2104; 2014/07-08). Mass spectrometry analysis was performed at Koç University, Proteomics Facility (İstanbul, Türkiye). Briefly, lysates from co-immunoprecipitation assays, one using an antibody against desmin (D8281, SIGMA, Missouri, USA) and one without non-specific IgG [11] were run and excised from Coomassie blue stained gels. Protein bands were digested in-gel. Samples were analyzed with Q Exactive Quadrupole-Orbitrap Mass Spectrometer with nano Liquid Chromatography (n-LC MS/MS) (Thermo Fisher Scientific, Massachusetts, USA) in duplicates with 300 nL/min flow rate for 50 min. False Discovery Rate (FDR) was 0.01 %. Raw data was analyzed using MaxQuant [16]. Peptides identified using zebrafish protein list downloaded from UniProt (UP000000437) [17]. The final protein list analyzed on PantherDB using statistical overrepresentation test on GO cellular compartment [18, 19]. The proteins listed under the term ‘nuclear’ than analyzed on Enrich R [20], [21], [22]. The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium via the PRIDE [23] partner repository with the dataset identifier PXD047121.

Cell culture

Human cervical cancer cells (HeLa) cells were grown in DMEM (Capricorn, Germany) containing 10 % foetal bovine serum (FBS) (Capricorn, Germany). Immortalized human skeletal myoblasts (T0033) were grown in proliferation media as described by Thorley (2016) [24]. Immortalized human skeletal muscle cells were differentiated as described by Thorley (2016) [24].

Plasmid construction and transient transfections

For construction of eGFP tagged desmin expression vector (pcDNA3.1(+)-hDES-eGFP), desmin sequence of human origin was used as template. Total RNA was extracted according to Pattern (2019) [25]. RNA concentrations quantified using NanoDrop™ spectrophotometer (Thermo Fisher Scientific, Massachusetts, USA).

cDNA was synthesized from total RNA isolated with iScript cDNA synthesis kit (Bio-RAD, California, USA) according to the manufacturer’s instructions.

The open reading frame (ORF) of DES was amplified by PCR with a 5′ primer 5′-acagcgGAATTCCCtcgcccgcgccgtcaccatga-3′ and a 3′ primer 5′-acagcgGCGGCCGCggtctctggtggcagagggtc-3′, which creates restriction sites for KpnI and BamHI enzymes at the 5′ and 3′ ends of DES, respectively. The product was digested with the respected enzymes and cloned into the pcDNA3.1(+) eGFP to create pcDNA3.1(+)-hDES-eGFP. All enzymes used for cloning procedures were purchased from New England BioLabs (Massachusetts, USA).

For transfection, cells were grown to 70–80 % confluency and transfected with pcDNA3.1(+)-hDES-eGFP with Lipofectamine 3000 Transfection Kit (Invitrogen, Massachusetts, USA) according to the manufacturer’s instructions. After 24 h, media was changed into fresh media.

Confocal imaging

Images were acquired using Zeiss LSM 880 laser scanning confocal microscope equipped with a 63×/1.4 Apochromat Oil DIC objective (Carl Zeiss, Germany). Cells were imaged using an argon laser with an excitation at 488 nm for eGFP and a diode laser at 405 nm for DAPI. Z-stacks were acquired with 1–2 μm intervals. Colocalization analysis was performed using FIJI with Coloc 2 plug in [26].

Surface hydrophobicity measurements

BSA (Thermo Fisher Scientific, USA) and recombinant desmin (GenScript, New Jersey, USA) proteins were dissolved in PBS containing 0–55 % v/v ethanol (EtOH) and 10 μM 4,4′-Bis(phenylamino)-[1,1′-binaphthalene]-5,5′-disulfonic acid dipotassium salt (Bis-ANS) (R&D Systems, USA) and equilibrated for 15 min at room temperature. Fluorescence was quantified using a microplate reader (SpectraMax M2, Molecular Devices, California, USA) with excitation and emission wavelengths of 405 and 460 nm, respectively.

Statistical analysis

Student’s t-test performed on Microsoft Excel, and graphics created on GraphPad Prism v.5 (GraphPad Software Inc., California, USA). The results were considered significant when p-value <0.05. All error bars were presented as mean±SD.

Desmin’s binding partners at the nuclear periphery

We previously co-precipitated desmin and nup214, a nup located at the cytoplasmic filaments of the NPC, and that desmin and nup214 co-precipitates [14].

To further explore the extends of this relationship and identify additional interactors, we conducted a proteomic analysis. The protein list obtained after initial analysis was subsequently analyzed on EnrichR [20], [21], [22] (Figure 1). Intriguingly, the terms nuclear pore cytoplasmic filaments, cytoplasmic periphery of the nuclear pore complex and cytoplasmic side of the nuclear pore were found to be enriched (Figure 1). Among other findings, a protein was particularly of interest: nup153. Nup153 resides in the nuclear basket of the NPC, known to shuttle between the nuclear and cytoplasmic faces of the NPC, and plays a role in protein transport through the NPC [27]. The results of proteomic analysis strongly suggest a functional relationship between the NPC and desmin.

Figure 1:

Analysis of the proteomic data. (A) Ontology analysis on EnrichR representing the cellular localization of nuclear partners of desmin. The X-axis (the length of the bar) represents the significance of the term, and the lighter color indicates more significant results. Starting from the top, the p-values are: 7.87E-04; 3.50E-03; 4.89E-03; 6.28E-03; 7.67E-03; 1.25E-02; 1.39E-02; 1.39E-02; 1.53E-02; 1.60E-02, respectively. (B) The relationship between desmin and nups as generated by STRING. Red dashed lines indicate the interactions of desmin that we identified.

Additionally, we performed a protein interaction network analysis on The Biological General Repository for Interaction Datasets (BIOGRID) [28, 29]. This analysis revealed that desmin also has an interaction with nup88 [30].

We then use STRING [31] to search for a pattern between nups and desmin (Figure 1). The human nups [5] and desmin were analysed for interaction. As expected, nups displayed strong interconnections, and our analysis revealed the interaction between desmin and nups.

When this data is combined with our interaction data (PXD047121) of nups and desmin it forms a strong case for nup assisted karyopherin-independent nucleocytoplasmic transport of desmin.

Desmin can be localized in the nucleus of human skeletal myoblasts and epithelial cells

The proteomic analysis was conducted using protein lysates from zebrafish skeletal muscle tissue. While previous studies have demonstrated the nuclear presence of desmin in cell lines from various origins, such as renal fibroblasts, skeletal myoblast of mouse origin, embryonic cardiac stem cells of mouse origin and others [3, 9, 32], there was no record of a study in human originated myoblasts. In addition to that, most of the studies dated before 2010s. Therefore, prior to investigating the hydrophobic features of desmin, which we believe to be a key feature that lets desmin into and out of the nucleus, we first demonstrated that desmin can be localized in the nucleus of human skeletal myoblasts and epithelial cells with taking advantage of the new imaging and analysing technologies. We transiently overexpressed the eGFP tagged desmin in two different cell lines one of which expresses (T0033) desmin and the other (HeLa) not 24 h after transfection, the skeletal myoblast proliferation media were changed to differentiation media and T0033 cells were allowed to differentiate for two days and then prepared for confocal imaging. HeLa cells were fixed 24 h after the addition of fresh proliferation media for confocal imaging.

We performed colocalization analysis with nuclear staining (DAPI) and desmin signal to determine desmin’s nuclear localization, and calculated the Pearson correlation coefficient [33]. According to colocalization analysis using FIJI [26], desmin was detected in 48 % nuclei of differentiating myoblasts and 68 % of epithelial cells (Figure 2, Supplemental Figures 1 and 2). We allow the myoblasts to start differentiate because desmin expression is relatively higher after the induction of differentiation [34]. Furthermore, we have showed that desmin enters the nucleus during early myogenesis (unpublished data), consistent with the literature. Inhibition or reduction of desmin interferes with the expression of muscle-specific genes, MYOD and MYOG, which are particularly important in the determination skeletal myoblast linage [35] and differentiation [36, 37], respectively.

Figure 2:

Localization of desmin in nuclei of myoblast (T0033) and epithelial (HeLa) cells. Representative confocal immunofluorescence micrographs of myoblast (T0033) and epithelial cells (HeLa). Desmin, green; nucleus, blue. Images at the right are merged images.

Whilst, desmin has been reported to be found in the nucleus of cells of various origins [3, 9, 32], our study presented here is the first report on desmin’s nuclear localization in human skeletal myoblasts.

Desmin can increase its surface hydrophobicity in hydrophobic environment

Proteins can enter the nucleus with direct interaction with nups via their amphipathic alpha helix [38] and they tend to increase their surface hydrophobicity in hydrophobic solutions [1]. Considering the relationship of desmin with the components of the NPC, we asked ourselves whether desmin has the ability to change its surface hydrophobicity to pass through nuclear pore complex. To investigate if desmin can change its surface hydrophobicity, we used Bis-ANS dye, which binds noncovalently to hydrophobic surfaces. Binding of Bis-ANS to a hydrophobic surface result in an increase in fluorescence. We used BSA as a control since its hydrophobic characteristics are well characterized [39, 40]. When equal amounts of BSA and desmin were diluted in PBS containing 10 μM of Bis-ANS in the absence and presence of 55 % ethanol, the signal intensity for desmin was significantly higher, as is for BSA, indicating that desmin increases its surface hydrophobicity in a hydrophobic environment (Figure 3A). This means that desmin can modulate its conformation to adapt the hydrophobic nature of the NPC and to interact with nups. We also observed that the signal intensity of desmin in an alcohol-free environment was significantly higher than that of BSA, which is probably due to the exposed hydrophilic regions on the surface and may potentially be antigenic.

Figure 3:

Conformational changes of desmin in hydrophobic environment. (A) Fluorescence signal intensities for desmin (DES) and BSA in the presence of 0 %, 55 % EtOH. Values of three different measurements are represented as the mean±SD (*: p≤0.05; **: p≤0.01). (B) Conjectural helical wheel representations of the desmin coils^† (RFU, relative fluorescence unit). Hydrophobic amino acids are indicated in yellow. Grey, red, cyan represent the polar, acidic, and basic amino acids, respectively. ^†Full amino acid sequence of coil 1a and coil 2a and, the hydrophobic regions according to Kyte–Doolittle [41, 42] of coil 1b (208–230 amino acids) and coil 2b (385–400 amino acids) was presented (Supplemental Figure 3).

Next, we created a helical wheel representation of desmin’s coils to illustrate the position of hydrophobic amino acids [43]. The amphiphilic rod domain of desmin consist of four coils namely coil 1a, coil 1b, coil 2a and coil 2b. To generate the helical wheel representations of the desmin coils, full amino acid sequence of coil 1a and coil 2a and, the hydrophobic regions according to Kyte–Doolittle [41, 42] of coil 1b (208–230 amino acids) and coil 2b (385–400 amino acids) was used (Supplemental Figure 3). As shown in Figure 3, hydrophobic regions of desmin coils are directed inwards in such a manner that as if they form a corridor (Figure 3B, Supplemental Figure 3). These regions hidden inside the coil, can be exposed to interact with nups during translocation through the NPC.