Mutation of leucine 170 alters the subcellular distribution, neurite outgrowth and three-dimensional structure of dystrophins Dp71 and Dp40

Abstract

Background

In addition to progressive muscle wasting, one-third of patients with Duchenne muscular dystrophy (DMD) exhibit varying degrees of cognitive impairment, which are associated with mutations of the DMD gene within the region coding for dystrophins Dp71 and Dp40. A previous study reported that the deletion of leucine 3238 in dystrophin Dp427 induces cognitive impairments without muscular dystrophy. This mutation has implications for all dystrophins, including Dp71 and Dp40.

Objective

This study aimed to evaluate the effect of the deletion of leucine 170 (leucine 3238 in the full-length dystrophin Dp427) on the subcellular localisation, neurite outgrowth and three-dimensional (3D) structure of Dp71 and Dp40 isoforms.

Methods

PC12 Tet-On cells were transiently transfected with pTRE2pur-Myc/Dp71 or Dp40 vectors, and recombinant Myc-Dp71/Dp40 proteins were expressed by adding doxycycline. The analysis of undifferentiated and nerve growth factor- (NGF) differentiated cells was conducted by cell morphology, immunofluorescence staining and confocal microscopy. The length of the neurites of differentiated cells was obtained using ImageJ software. Analyses of 3D models were carried out by Alphafold 2.0 and TM-align tools.

Results

Our research revealed that the deletion of leucine 170 (ΔL170) disturbs the membrane localisation and promotes the cytoplasm and nuclear accumulation of Dp71 and Dp40 in undifferentiated and NGF-differentiated PC12 Tet-On cells. Furthermore, this mutation altered the neurite outgrowth of PC12 Tet-On cells. It is important to note that both the deletion of leucine 170 (ΔL170) and the mutation of leucine 170 to proline (L170P) result in alterations to the 3D structures of these dystrophins.

Conclusions

The loss or mutation of leucine 170 disturbs the subcellular localisation, neuronal functions and 3D structures of Dp71 and Dp40. This may lead to cognitive deficits in patients with DMD.

Supplementary Information

The online version contains supplementary material available at 10.1007/s11033-026-11646-9.

Introduction

Dystrophins Dp71 and Dp40 are expressed from a common promoter located in intron 62 of the DMD gene [1]. The cognitive impairments present in some Duchenne muscular dystrophy (DMD) patients have been largely associated with mutations affecting dystrophin Dp71 [2], while very few studies have addressed the possible role of Dp40 in these pathologies. Dystrophin Dp71 and Dp40 are expressed in several human tissues, including the brain [1], and the expression of Dp71 increases during brain development from embryonic stage to adulthood [3–5], while Dp40 increases during postnatal brain development [6]. Dp71 is expressed and participates in the neuronal differentiation of PC12 cells [7–10] and interacts with different members of dystrophin-associated proteins (DAPs) in these cells [11, 12], hippocampal neurons [13] and in brain and retina [14], while Dp40 interacts with presynaptic proteins such as vesicle-associated membrane protein 2 (VAMP2), syntaxin-1, and synaptosome-associated protein 25 (SNAP25) in synaptic vesicle fractions [15]. Furthermore, Dp40 is expressed and may form a complex with several DAPs in neurons, astrocytes, oligodendrocytes and photoreceptors [14].

Dp71 and Dp40 have a unique N-terminal of seven amino acids, derived from the specific exon 1, fused to the amino acids of exons 63 to 79 for Dp71 and exons 63 to 70 for Dp40, corresponding to the cysteine-rich domain (half of the WW domain, the EF1/2 and ZZ domains) and the full C-terminal domain in Dp71 and the first 48 amino acids of the C-terminal domain in Dp40 (Supplementary Figure S3e). Furthermore, Dp71 transcripts undergo different alternative splicing events in the exons 71 to 74, 78 and intron 77 leading to multiple Dp71 isoforms classified into three main groups: Dp71d, Dp71f, and Dp71e. Based on the alternative splicing of exon 78 or intron 77, each Dp71 group has a specific C-terminal end: Dp71d group (plus exon 78) contains 13 specific amino acids at the C-terminal encoded by exon 78 and 79, Dp71f group (without exon 78) has 31 specific amino acids encoded from an elongated exon 79, and Dp71e group carries out 10 specific amino acids derived from the insertion of part of intron 77 [16–19]. Alternative splicing of one or more exons from 71 to 74, at different combinations, does not change the open reading frame; however, these splicing events together with specific C-terminal end determine the subcellular localisation of Dp71 isoforms in PC12 cells. Dp71 isoforms lacking exon 71 (Dp71d_Δ71, Dp71f_Δ71 and Dp71e_Δ71) are mainly localised in the cell membrane and cytoplasm, while the Dp71 isoforms lacking exons 71 to 74 (Dp71d_Δ71−74 and Dp71e_Δ71−74) are almost exclusively localised in the membrane [17, 20]. Additionally, splicing of exons 71 to 74 may disrupt the interaction of Dp71 isoforms with syntrophin, an important component of the DAP complex. Moreover, Dp40 is localised at the periphery, nucleus, and excitatory post-synaptic spines of mouse hippocampal neurons [6] and its nuclear localisation is promoted during the neuronal differentiation of PC12 cells [21]. A previous study has shown that a 3-base pairs deletion within the DMD gene, leads to leucine 3238 loss in full-length human dystrophin (Dp427) and induces cognitive impairment in DMD patients without muscular dystrophy [22]. This deletion is located in exon 67 of the DMD gene, within the EF-2 hand region, affecting all the dystrophin family proteins containing this motif, including Dp71 and Dp40. Leucine 3238 in the dystrophin Dp427 corresponds to leucine 170 in the dystrophins Dp71 and Dp40. In a previous study, we found that a point mutation from leucine 170 to proline in Dp40 (Dp40_L170P) favors its nuclear accumulation in undifferentiated and neuronal differentiated PC12 cells [21] and inhibits the neuronal differentiation of PC12 Tet-On cells [23]. Importantly, deletion of leucine 170 in dystrophin Dp71d (Dp71d_ΔL170) disturbs its localisation at the membrane promoting its nuclear accumulation and the loss of interaction with β-dystroglycan [24]. Therefore, in this study, we investigated the effect of leucine 170 deletion by site-directed mutagenesis on dystrophin Dp71 and Dp40 isoforms during neuronal differentiation of PC12 Tet-On cells. In undifferentiated PC12 Tet-On cells, we showed that deletion of leucine 170 in Dp71 isoforms (Dp71d_Δ71ΔL170, Dp71d_{Δ71−74ΔL170}, Dp71f_Δ71ΔL170, and Dp71e_Δ71ΔL170) and Dp40 (Dp40_ΔL170) disturbs the membrane localisation of Dp71 and Dp40 and promotes the nuclear localisation mainly of, Dp71d_{Δ71−74ΔL170} and Dp40_ΔL170, while in differentiated PC12 Tet-On cells, a cytoplasm mis-accumulation of Dp71 mutants was observed. In addition, deletion of leucine 170 on Dp71 and Dp40 alters the neurite outgrowth of PC12 Tet-On cells and disturbs the three-dimensional folding of Dp71 and Dp40 proteins, suggesting that both Dp71 and Dp40 have a relevant role on the neuronal functions.

Materials and methods

Cell culture

PC12 Tet-On cells (Clontech, Mountain View, CA, USA), a tetracycline inducible system model, were grown in Dulbecco´s modified eagle medium supplemented with 10% heat-inactivated horse serum, 100 U/mL penicillin, 1 mg/mL streptomycin, 0.25 µg/mL mycostatin (medium and supplements were from Gibco, Rockville, MD, USA), 5% foetal tetracycline-free calf serum (Clontech), and 150 µg/mL geneticin (Invitrogen, Carlsbad, CA, USA) at 37 °C with 5% CO₂.

Vector construction and in vitro site-directed mutagenesis

The cDNA for Dp71d_Δ71, Dp71d_Δ71−74, Dp71f_Δ71, Dp71e_Δ71, Dp40 and Dp40_L170P were previously cloned into pTRE2pur-Myc vector (pTRE2pur-Myc/Dp71/Dp40), a doxycycline inducible system, using MluI and NotI restriction sites [17, 23]. The deletion of the leucine 170 codon (CTG) in pTRE2pur-Myc/Dp71/Dp40 vectors was performed by site-directed mutagenesis using the primers 5′-CGTAGACTGGGTCTTCTTCATGATTCTATTCAAATCC-3′ and 5′-GGATTTGAATAGAATCATGAAGAAGACCCAGTCTACG-3′, and the QuickChange Lightning Site-Directed Mutagenesis Kit (Agilent Technologies, Santa Clara, CA, USA) following the manufacturer´s instructions. All Dp71 and Dp40 mutant vectors were confirmed by Sanger DNA sequencing using the Dye Deoxy Terminator Cycle Sequence Kit (Applied Biosystems, Foster City, CA, USA).

Cell transfection, Western blot and immunofluorescence assays

Cell transfection, Western blotting, immunofluorescence, and confocal analyses were performed as previously described [17, 20]. The expression of recombinant Myc-Dp71 and Myc-Dp40 proteins was induced using 1 µg/mL of doxycycline for 24 h. Following this period, the undifferentiated cells were analysed using Western blot and immunofluorescence assays. To induce neuronal differentiation, the induction medium (containing 1 µg/mL doxycycline) was replaced with fresh medium (containing 1 µg/mL doxycycline and 50 ng/mL nerve growth factor (NGF)) to maintain the expression of the recombinant Dp71 and Dp40 proteins and to induce neuronal differentiation in the PC12 Tet-On cells. The differentiation medium (containing doxycycline and NGF) was changed after three days. The six-day differentiated cells were analysed using an immunofluorescence assay. Recombinant proteins were detected using an anti-Myc monoclonal antibody (Santa Cruz Biotechnology, Dallas, TX, USA). Immunostained cells were analysed using a Leica confocal microscope, and eight to fifteen images of each experiment were captured (n = 2). An equatorial stack of confocal micrographs of the transfected cells was selected to demonstrate the subcellular localisation of the wild-type and mutant Dp71 and Dp40 proteins.

Quantification of subcellular localisation and neurite lengths

The fluorescence intensity of Dp71 and Dp40 wild-type and Dp71 and Dp40 mutants was measured from the equatorial stack of confocal micrographs using ImageJ software. For this, the total fluorescence intensity was obtained from all transfected cells, ranging from 8 to 28 cells per experiment (n = 2) and then, based on a standard width from the cell edge to the nucleus, the cell was divided into three main sections: membrane, cytoplasm and nucleus. The percentage of fluorescence intensity of each section was calculated based on the relative intensity per unit area. Finally, neurite lengths were measured from the maximum projection of each differentiated cell (10 to 15 cells per experiment (n = 2)) using the Leica Las X software tools. The maximum projection was used because the cell body and neurite extensions are typically in different z-stacks.

Three-dimensional modeling of Dp71 and Dp40 proteins

The amino acid sequences of Dp71 and Dp40 isoforms identified in PC12 cells (Rattus norvegicus), were obtained from the GenBank database, including Dp71d_Δ71, (GenBank: AY326947), Dp71d_Δ71−74 (GenBank: AY326949), Dp71f_Δ71 (GenBank: AY326948), Dp71e_Δ71 (GenBank: JF510048) and Dp40 (GenBank: KF154977) [17, 21]. For Dp71 and Dp40 mutants (ΔL170), leucine 170 was deleted from the wild-type sequences and a change of leucine 170 to proline (L170P) was incorporated in Dp71 and Dp40 isoforms. Dp71 and Dp40 wild-type and mutant proteins were modeled with Alphafold 2.0. Five models per execution were generated for each protein and models were classified based on the confidence estimated by Alphafold 2.0 according to the predicted local distance difference test (PLDDT) and PLDDT average on a scale of 0 to 100 [25, 26]. After two or more independent executions per sequence, the model with the highest PLDDT in all runs was selected to represent the three-dimensional (3D) structure for each protein. Several virtual CPU and GPU units with different RAM sizes were tested to compare the confidence of the prediction of the same protein. Consecutive model executions were partial and independent.

For comparative analyses, 3D models of Dp71 and Dp40 wild-type were aligned with Dp71 and Dp40 mutants using the online TM-align tool [27]. Values of the root-mean-square deviation (RMSD) vary from 5 Å to 0 Å, in which a low score indicates highly similar structures between both proteins. In addition, TM score was also computed for each alignment. The TM score can be from 0 to 1 and values close to 1 indicate more confidence. Together, RMSD and TM scores were used to determine the structural similarity between Dp71 and Dp40 wild-type and mutant proteins. The results were loaded into the UFSC ChimeraX program for the visualisation and comparative analyses of Dp71 and Dp40 structures.

Statistical analysis

Data show the mean plus standard deviation from two independent experiments. Statistical analyses were performed with GraphPad Prism 5 software using Student’s t-test. P values ≤ to 0.05 were considered statistically different.

Results

Site-directed mutagenesis of Dp71 and Dp40 isoforms

To investigate the effect of the deletion of leucine 170 in the localisation and function of Dp71 and Dp40 isoforms, we deleted the codon CTG that codes for leucine 170 on pTRE2pur-Myc/Dp71/Dp40 wild-type vectors by site-directed mutagenesis to obtain the vectors with the mutation: pTRE2pur-Myc/Dp71d_Δ71ΔL170, pTRE2pur-Myc/Dp71d_{Δ71−74ΔL170}, pTRE2pur-Myc/Dp71f_Δ71ΔL170, pTRE2pur-Myc/Dp71e_Δ71ΔL170 and pTRE2pur-Myc/Dp40_ΔL170. All Dp71 and Dp40 wild-type isoforms are expressed in PC12 cells and have been previously characterized [16, 17, 21]. Dystrophin Dp40_L170P, a punctual mutation of leucine 170 to proline, previously described [21, 23], was also analysed. The new mutant Dp71 and Dp40 vectors were analysed by Sanger DNA sequencing and compared with wild-type sequences. All mutagenised vectors showed the deletion of CTG codon for leucine 170 (ΔL170) or mutation of CTG to CCG codon to change leucine 170 to proline (L170P) (Fig. 1a). To test the correct expression of non-mutated and mutated Dp71 and Dp40 isoforms, the pTRE2pur-Myc/Dp71d_Δ71ΔL170, pTRE2pur-Myc/Dp71d_{Δ71−74ΔL170}, pTRE2pur-Myc/Dp71f_Δ71ΔL170, pTRE2pur-Myc/Dp71e_Δ71ΔL170, pTRE2pur-Myc/Dp40_ΔL170, pTRE2pur-Myc/Dp40_L170P as well as pTRE2pur-Myc/Dp71/Dp40 wild-type and pTRE2pur-Myc (empty vector) vectors were transiently transfected into PC12 Tet-On cells and the expression of Myc-Dp71 and Myc-Dp40 proteins was induced by adding doxycycline. The recombinant Myc-Dp71 and Myc-Dp40 proteins were analysed by Western blot assays with anti-c-Myc antibody. All Dp71 and Dp40 wild-type and mutant proteins were expressed at the expected molecular weight of 71 and 40 kDa, respectively (Fig. 1b). These results demonstrated that the deletion (ΔL170) or the mutation (L170P) of leucine 170 does not prevent the expression of mutant Dp71 and Dp40 isoforms.

Fig. 1.

Cloning and site-directed mutagenesis of dystrophin Dp71 and Dp40 isoforms. cDNA sequences of Dp71 and Dp40 isoforms were cloned into pTRE2pur-Myc vector and site-directed mutagenesis were carried-out in pTRE2pur-Myc-Dp71/Dp40 vectors to eliminate leucine 170 (ΔL170) as described in Materials and Methods. a Comparative schemes showing nucleotide (upper panel) and amino acid (lower panel) sequence alignments of Dp71/Dp40 (wild-type), Dp40_L170P, and Dp71/Dp40_ΔL170. Mutations are indicated by the dotted rectangle. b Western blot showing the expression of recombinant Dp71 and Dp40 proteins. PC12 Tet-On cells were transfected with pTRE2pur-Myc (empty vector) and wild-type and mutant pTRE2pur-Myc-Dp71/Dp40 vectors. Expression of recombinant Dp71 and Dp40 proteins was induced with 1 µg/mL doxycycline as described in Materials and methods. Western blot was performed with anti-c-Myc antibody. Beta-actin detection was used as loading control. Molecular weight in kilodaltons (kDa) is indicated

Subcellular distribution of Dp71d_Δ71ΔL170, Dp71d_{Δ71−74ΔL170}, Dp71f_Δ71ΔL170 and Dp71e_Δ71ΔL170 and neurite growth reduction of PC12 Tet-On cells. 

It has been reported that the deletion of leucine 3238 (L3238) from the dystrophin Dp427 (leucine 170 in Dp71 and Dp40) results in a cognitive deficit without the occurrence of muscular dystrophy [22]. Moreover, deletion of leucine 170 in Dp71d (Dp71d_ΔL170) disturbs its membrane localisation [24]. Here, we deleted the leucine 170 (ΔL170) on the family of Dp71 and Dp40 including Dp71d_Δ71ΔL170, Dp71d_{Δ71−74ΔL170}, Dp71f_Δ71ΔL170, Dp71e_Δ71ΔL170 and Dp40_ΔL170 isoforms to evaluate their subcellular distribution in undifferentiated and NGF-differentiated PC12 Tet-On cells. According to the fluorescence intensity, wild-type Dp71 isoforms derived from the Dp71d, Dp71f and Dp71e groups (Dp71d_Δ71, Dp71d_Δ71−74, Dp71f_Δ71 and Dp71e_Δ71) are mainly localised in the membrane (51%-62%) and cytoplasm (28%-37%), with a low localisation in the nucleus (8%-14%) of undifferentiated PC12 Tet-On cells, respectively (Fig. 2a-d), as previously reported [17, 20]. In contrast, all Dp71 mutans, Dp71d_Δ71ΔL170, Dp71d_{Δ71−74ΔL170}, Dp71f_Δ71ΔL170 and Dp71e_Δ71ΔL170, showed a delocalisation. The Dp71 and Dp40 mutants decreased in the membrane (31%-34%) and increased in the cytoplasm (43%-49%) and nucleus (17%-25%) (Fig. 2a-d). Table 1 summarises the subcellular localisation of Dp71 and Dp40 wild-type and mutant proteins in undifferentiated PC12 Tet-On cells as well as the subcellular localisation of Dp40 wild-type and mutants in NGF-differentiated PC12 Tet-On cells. Importantly, in NGF-differentiated PC12 Tet-On cells, localisation of Dp71d_Δ71ΔL170, Dp71d_{Δ71−74ΔL170}, Dp71f_Δ71ΔL170 and Dp71e_Δ71ΔL170 was completely disturbed, while wild-type Dp71 isoforms were mainly localised in the membrane, cytoplasm and neurite extensions (Fig. 3a-b), as previously described [17, 20], the Dp71 mutants were delocalised in the membrane and mis-aggregated in the cytoplasm close to the nucleus (Fig. 3a-b). Notably, staining of Dp71 mutants was diminished in the neurite extensions, which was only observed in the maximal projection compared to Dp71 wild-type isoforms that were observed throughout the neurite extensions (Fig. 3a-b). In some cells, Dp71d_{Δ71−74ΔL170} was mainly localised in the nucleus of undifferentiated and differentiated PC12 Tet-On cells (Supplementary Figure S1).

Fig. 2.

Subcellular localisation of wild-type and mutant Myc-Dp71 proteins in undifferentiated PC12 Tet-On cells. PC12 Tet-On cells were transfected with wild-type and mutant pTRE2pur-Myc-Dp71 vectors. Expression of recombinant Dp71 proteins was induced with 1 µg/mL doxycycline. Immunostaining for Myc-proteins was performed with anti-c-Myc antibody (green) and nuclei were stained with DAPI (blue). a Subcellular localisation of Dp71d_Δ71 and Dp71d_Δ71ΔL170. b Subcellular localisation of Dp71d_Δ71−74 and Dp71d_{Δ71−74ΔL170}. c Subcellular localisation of Dp71f_Δ71 and Dp71f_Δ71ΔL170. d Subcellular localisation of Dp71e_Δ71 and Dp71e_Δ71ΔL170. Nuclei (left) and merged (right) images of nuclei and Myc-proteins are shown at the lower part of each panel. Graphs in each panel show the mean plus standard error of fluorescence intensity from two independent experiments. Significant differences (Student’s t-test) between Dp71 wild-type and Dp71 mutant (ΔL170) proteins in membrane, cytoplasm and nucleus are shown. Scale bar is indicated for each panel. ***P < 0.001

Table 1.

Percentage of fluorescence intensity of wild-type and mutant Dp71 and Dp40 proteins

Dp71 isoforms	Membrane			Cytoplasm			Nucleus
	Wt	ΔL170	L170P	Wt	ΔL170	L170P	Wt	ΔL170	L170P
Dp71d Δ71	51	32*	ND	37	48*	ND	12	20*	ND
Dp71d Δ71−74	62	32*	ND	28	43*	ND	10	25*	ND
Dp71f Δ71	61	34*	ND	31	49*	ND	8	17*	ND
Dp71e Δ71	53	31*	ND	33	47*	ND	14	22*	ND
Dp40	54	22*	21*	32	32NS	32NS	14	46*	47*
Dp40 NGF	52	22*	24*	35	38NS	38NS	13	40*	38*

^*Statistically significant (P < 0.001). ^NSNo statistically significant. ^NGFDifferentiated cells at 6 days. Wt: wild-type Dp71 and Dp40; ΔL170: deletion of leucine 170; L170P: mutation of leucine 170 to proline. ND: Not developed

Fig. 3.

Subcellular localisation of wild-type and mutant Myc-Dp71 proteins in NGF-differentiated PC12 Tet-On cells. PC12 Tet-On cells were transfected with wild-type and mutant pTRE2pur-Myc-Dp71 vectors. Expression of recombinant Dp71 proteins was induced with 1 µg/mL doxycycline. Immunostaining for Myc-proteins was performed with anti-c-Myc antibody (green) and nuclei were stained with DAPI (blue). a Subcellular localisation of Dp71d_Δ71 and Dp71d_Δ71ΔL170. b Subcellular localisation of Dp71d_Δ71−74 and Dp71d_{Δ71−74ΔL170}. c Subcellular localisation of Dp71f_Δ71 and Dp71f_Δ71ΔL170. d Subcellular localisation of Dp71e_Δ71 and Dp71e_Δ71ΔL170. Merged images of nuclei and Myc-proteins (left) and maximum projection (right) of Myc-proteins are shown at the lower part of each panel. Maximum projection was used to show neurite extensions of differentiated PC12 Tet-On cells. e Morphometric analyses of NGF-differentiated PC12 Tet-On cells. Maximum projection of confocal images was used to measure neurite lengths of transfected PC12 Tet-On cells. Graphs show the mean of neurite lengths plus standard error from two independent experiments. P values are shown to indicate significant differences (Student’s t-test) between Dp71 wild-type and Dp71 mutant proteins. Scale bar is indicated for each panel

Interestingly, a reduction in the length and number of neurites per cell was observed in PC12 Tet-On cells transfected with Dp71 mutants compared with Dp71 wild-type and this was significantly different in cells expressing Dp71d_{Δ71−74ΔL170} and Dp71e_Δ71ΔL170, and somehow different in Dp71d_Δ71ΔL170 (P = 0.0502) (Fig. 3c). Additionally, the differentiation index, measured as percentage of cells with long neurites, was notably diminished in PC12 cells transfected with Dp71 mutants compared with their wild-type Dp71 counterparts; conversely, the number of cells with short neurites increased substantially in PC12 Tet-On cells transfected with Dp71d_Δ71ΔL170, Dp71d_{Δ71−74ΔL170} and Dp71e_Δ71ΔL170 (Supplementary Figure S2) similar to that observed in the neurite lengths (Fig. 3c). Importantly, the PC12 Tet-On cells transfected with Dp71d_Δ71−74, Dp71f_Δ71 and Dp71e_Δ71 wild-type isoforms showed several and branched neurites outgrowth while cells transfected with Dp71d_Δ71 wild-type and mutants Dp71d_Δ71ΔL170, Dp71d_{Δ71−74ΔL170}, Dp71f_Δ71ΔL170 and Dp71e_Δ71ΔL170 were differentiated as bipolar cells. Although some cells expressing Dp71f_Δ71 exhibited larger and more branched neurites than the mutant Dp71f_Δ71ΔL170 cells, no significant differences were observed. All these results demonstrated that the deletion of leucine 170 (ΔL170) disturbs the subcellular distribution of Dp71 isoforms in undifferentiated and NGF-differentiated PC12 Tet-On cells reducing its presence in the membrane and affects the neuronal differentiation process of PC12 Tet-On cells.

Subcellular distribution of Dp40, Dp40_ΔL170and Dp40_L170Pand neurite growth reduction of PC12 Tet-On cells

In a previous work, we showed that the mutation of leucine 170 to proline (L170P) in the dystrophin Dp40 (L3238P in the full-length dystrophin Dp427) induces its nuclear localisation [21] and alters the neuronal differentiation of PC12 Tet-On ells [23]. To determine the effect of the deletion of leucine 170 (ΔL170) in the subcellular localisation of Dp40 protein, we also generated the dystrophin Dp40_ΔL170. Expression and localisation of Dp40_ΔL170 were compared with Dp40 wild-type and Dp40_L170P. In undifferentiated PC12 Tet-On cells, Dp40 was mainly localised in the membrane (54%) and cytoplasm (32%) than in the nucleus (14%) (Fig. 4a), as previously described [20, 21]. Interestingly, the deletion of leucine 170 in Dp40 (Dp40_ΔL170) decreased its membrane localisation (22%) and was predominantly distributed in the nucleus (46%) whereas no change was observed in the cytoplasm localisation (32%) (Fig. 4a). Importantly, the membrane localisation of Dp40_L170P was lost (21%), and accumulated in the nucleus (47%) while no change was observed in the cytoplasm (32%). Interestingly, this subcellular distribution was very similar to that observed for Dp40_ΔL170 (Fig. 4a). In NGF-differentiated PC12 Tet-On cells, Dp40 wild-type had a high localisation in the membrane (52%) and cytoplasm (35%) as compared with the nucleus (13%) (Fig. 4b). As expected, the localisation of Dp40_ΔL170 decreased in the membrane (22%) and increased in the nucleus (40%), while no change was observed in the cytoplasm localisation (38%) (Fig. 4b). The subcellular localisation of Dp40_L170P decreased in the membrane (24%), accumulated in the nucleus (38%) and maintained in the cytoplasm (38%) (Fig. 4b). All these results were similar in both undifferentiated and NGF-differentiated PC12 Tet-On cells (Table 1). Notably, PC12 Tet-On cells transfected with Dp40 wild-type showed a well-differentiated morphology while cells transfected with Dp40_ΔL170 and Dp40_L170P showed a reduction of neurite lengths (Fig. 4b), similar to the stable expression of Dp40_L170P which reduce the differentiation index of PC12 Tet-On cells compared to the PC12 Tet-On cells expressing Dp40 [23]. Additionally, we observed a reduction in fluorescence intensity throughout the neurite extension, like the one observed for Dp71 mutants. All together these results suggest that both deletion (ΔL170) and/or mutation of leucine 170 to proline (L170P) alter, in a similar way, the subcellular distribution and neuronal functions of Dp40 isoform.

Fig. 4.

Subcellular localisation of wild-type and mutant Myc-Dp40 proteins in undifferentiated and NGF-differentiated PC12 Tet-On cells. PC12 Tet-On cells were transfected with wild-type and mutant pTRE2pur-Myc-Dp40 vectors. Expression of recombinant Dp40 proteins was induced with 1 µg/mL doxycycline. Immunostaining for Myc-proteins was performed with anti-c-Myc antibody (green) and nuclei were stained with DAPI (blue). a Subcellular localisation of Dp40, Dp40_ΔL170 and Dp40_L170P in undifferentiated PC12 Tet-On cells. Nuclei (upper panel) and merged (lower panel) images of nuclei and Myc-proteins are shown at the right part of each panel. b Subcellular localisation of Dp40, Dp40_ΔL170 and Dp40_L170P in NGF-differentiated PC12 Tet-On cells. Merged images of nuclei and Myc-proteins and maximum projection of Myc-proteins are shown at the right part of each panel. Graphs in each panel show the mean plus standard error of fluorescence intensity from two independent experiments. Significant differences (Student’s t-test) between Dp40 wild-type and Dp40 mutants (ΔL170 and L170P) proteins in membrane and nucleus are shown. Scale bar is indicated for each panel. ***P < 0.001

Comparative modeling of Dp71 and Dp40 wild-type and mutants

To explore the impact of deletion (ΔL170) or mutation (L170P) of leucine 170 on the tertiary (3D) structure of Dp71 and Dp40 proteins, we performed the 3D structural modeling of Dp71 and Dp40 wild-type and mutant proteins using the Alphafold 2.0 and TM-align tools. The 3D structure models of wild-type Dp71 isoforms have different folding depending on the alternative splicing of exon 71 or exons 71 to 74, and the specific C-terminal end is sufficient to produce a different global folding (Supplementary Figure S3a). Notably, a significant 3D structural change was observed in all Dp71 mutant forms, the deletion of leucine 170 (ΔL170) resulted in a marked redistribution of structural domains, leading to a more rigid and compact arrangement of the C-terminal domain (Supplementary Figure S3b), while the mutation of leucine 170 to proline (L170P) caused a spatial inversion of the C-terminal domain in Dp71d_∆71L170P and Dp71e_∆71L170P, an expected outcome given the rigidity and cyclic structure of proline; however, this was not observed in Dp71d_{∆71−74L170P} and Dp71f_∆71L170P, in which the spatial orientation of the C-terminal domain remained unchanged (Supplementary Figure S3c). Furthermore, no significant changes of the global folding of Dp40, Dp40_∆L170 and Dp40_L170P were observed (Supplementary Figure S3d).

To verify the structural differences between Dp71 and Dp40 wild-type and Dp71 and Dp40 mutants, a structural alignment was performed using TM-align tool. The alignments of Dp71 wild-type with Dp71 mutants (∆L170) showed similar RMSD scores (3.37 Å to 4.04 Å); however, the Dp71d_{∆71−74ΔL170} exhibited the shortest displacement with a TM score of 0.77 that indicates a much more confidence than Dp71d_∆71∆L170, Dp71f_∆71∆L170 and Dp71e_∆71∆L170 mutants which showed the large displacements with TM scores between 0.55 and 0.65 (Fig. 5a). The L170P mutation also showed displacement for all Dp71 mutants and the Dp71d_{∆71−74L170P} exhibited a short displacement as well with a TM score of 0.72 (Fig. 5b). Similarly, the comparison between Dp71 ∆L170 and L170P mutants showed a notable displacement from each other confirming that they are differentially structured (Fig. 5c). As expected, the comparison of Dp71d_{∆71−74ΔL170} and Dp71d_{∆71−74L170P} showed the best TM score (0.77), despite the orientation of their C-terminal end was different, compared with the other Dp71 mutant alignments with TM scores between 0.57 and 0.67 (Fig. 5c). Moreover, structural alignments of Dp40, Dp40_∆L170 and Dp40_L170P proteins were also performed. Overall, no significant changes were detected in the topological and global folding of Dp40 wild-type and mutants, except for a slight displacement of N-terminal region in the alignment of Dp40 wild-type with Dp40_∆L170 and Dp40_L170P, and the C-terminal region in the alignment of Dp40 wild-type with Dp40_∆L170 and Dp40_L170P, and Dp40_∆L170 with Dp40_L170P (Fig. 5d). The comparison of Dp40 proteins presented the best TM scores of 0.97 to 0.98 (Fig. 5d). As expected, no significant difference was detected between Dp71d_∆71 and Dp40 alignment with a TM score of 0.96; however, the displacement of N- and C-terminal regions of Dp40 was much more evident and was reflected in the RMSD score of 1.83 Å (Fig. 5e) compared to the scores of Dp40 protein alignments that were from 0.93 Å to 1.06 Å (Fig. 5d).

Fig. 5.

Comparative modeling of wild-type and mutant Dp71 and Dp40 proteins. Structural alignments of Dp71 and Dp40 proteins were performed with TM-align tool and analysed with ChimeraX. a Structural alignment of wild-type Dp71d_Δ71, Dp71d_Δ71−74, Dp71f_Δ71 and Dp71e_Δ71 versus mutants Dp71d_Δ71ΔL170, Dp71d_{Δ71−74ΔL170}, Dp71f_Δ71ΔL170 and Dp71e_Δ71ΔL170, respectively. b Structural alignment of wild-type Dp71d_Δ71, Dp71d_Δ71−74, Dp71f_Δ71 and Dp71e_Δ71 versus mutants Dp71d_Δ71L170P, Dp71d_{Δ71−74L170P}, Dp71f_Δ71L170P and Dp71e_Δ71L170P, respectively. c Structural alignment of mutants Dp71d_Δ71ΔL170, Dp71d_{Δ71−74ΔL170}, Dp71f_Δ71ΔL170 and Dp71e_Δ71ΔL170 versus mutants Dp71d_Δ71L170P, Dp71d_{Δ71−74L170P}, Dp71f_Δ71L170P and Dp71e_Δ71L170P, respectively. d Structural alignment of Dp40 versus Dp40_ΔL170 and Dp40_L170P, and Dp40_ΔL170 versus Dp40_L170P, respectively. e Structural alignment of Dp40 versus Dp71d_Δ71. Scores of RMSD and TM are indicated for each panel. Scale bar is equal to 0.10 Å

Finally, a detailed analysis of aligned 3D structures, by zooming in the region of interest, revealed a displacement in the α-helix in all Dp71 and Dp40 mutants, either ∆L170 or L170P mutation. Specifically, Dp71d_{∆71−74ΔL170} and Dp71e_∆71ΔL170 mutants resulted in a more prominent displacement than the Dp71d_∆71ΔL170 and Dp71f_∆71ΔL170. In all cases, the leucine 168 of Dp71 wild-type isoforms skips its homologous residue in the Dp71 mutants causing a displacement to allow the alignment of leucine residues 169 and 170 of Dp71 wild-type with leucine residues 168 and 169 of Dp71 mutants (Fig. 6a). For the mutation L170P, the α-helix displacement was less pronounced, and the most significant displacement was observed in Dp71f_∆71L170P mutant (Fig. 6b). Notably, the difference prevailed in the alignment of Dp71 mutants (∆L170/L170P), in which the α-helix of the L170P mutants presented a deep twist to fit with Dp71 structures lacking leucine 170 (∆L170) (Fig. 6c). Similarly, the Dp40_ΔL170 mutant showed the most prominent displacement in the α-helix structure as compared with Dp40_L170P, and the Dp40_L170P also showed a deep twist to align with Dp40_∆L170 (Fig. 6d). Importantly, a close analysis of the alignment of wild-type dystrophins Dp40 and Dp71d_∆71 did not show a substantial difference (Fig. 6e) and the alignment of Dp40 with other wild-type Dp71 isoforms showed similar results (data not shown). Overall, these modeling results demonstrated that the deletion (ΔL170) and mutation (L170P) of leucine 170 in Dp71 isoforms disturb the global and local folding of Dp71 proteins, while these mutations disturb the local folding of Dp40, close to the mutated region. In addition, the alternative splicing of exons 71 to 74 as well as the specific C-terminal end, participate on the global folding of Dp71 isoforms.

Fig. 6.

Zoom analyses of the structural alignment between wild-type and mutant Dp71 and Dp40 proteins. Zoom of TM aligned models was performed with ChimeraX. a Dp71d_Δ71, Dp71d_Δ71−74, Dp71f_Δ71 and Dp71e_Δ71 versus Dp71d_Δ71ΔL170, Dp71d_{Δ71−74ΔL170}, Dp71f_Δ71ΔL170 and Dp71e_Δ71ΔL170. b Dp71d_Δ71, Dp71d_Δ71−74, Dp71f_Δ71 and Dp71e_Δ71 versus Dp71d_Δ71L170P, Dp71d_{Δ71−74L170P}, Dp71f_Δ71L170P and Dp71e_Δ71L170P. c Dp71d_Δ71ΔL170, Dp71d_{Δ71−74ΔL170}, Dp71f_Δ71ΔL170 and Dp71e_Δ71ΔL170 versus Dp71d_Δ71L170P, Dp71d_{Δ71−74L170P}, Dp71f_Δ71L170P and Dp71e_Δ71L170P. d Dp40 versus Dp40_ΔL170 and Dp40_L170P, and Dp40_ΔL170 versus Dp40_L170P. e Dp40 versus Dp71d_Δ71. L170 is marked in dark blue (wild-type Dp71 and Dp40) and P170 in red (L170P mutants of Dp71 and Dp40) indicated by black arrows. Amino acid residues near to the mutation zone are indicated (Q163, R164, R165, L166, G167, L168, L169, L170, H171, D172, S173, I174, Q175 and I176). The loss of leucine 170 changes the position of L170 to H170 as well as the number of the next amino acids. The amino acid L168 in Dp71 and Dp40 proteins is indicted in red circle that is displaced due to the deletion of leucine 170 (ΔL170) in Dp71 and Dp40 mutants. Scale bar is equal to 0.10 Å

Discussion

An accumulative effect has been reported in the cognitive disabilities of patients with DMD due to mutations of C-terminus of the full-length dystrophin Dp427, which also disturbs the expression of dystrophin Dp260, Dp140, Dp71, and Dp40 [28–31]. Additionally, it has been shown that in-frame deletions or duplications lead to a mild phenotype of cognitive impairment [32]. Although these mutations affect the expression of both Dp71 and Dp40, according to the DMD (dp71/dp40) gene structure, very few studies have shown the effect of dystrophin Dp40 in the central nervous system (CNS). Particularly, a study showed a correlation between the severity of cognitive deficits in DMD patients and the DMD gene mutations that lead to the loss of dystrophin Dp71 expression [2] as well as Dp40 in which mutations are located within the exons 63 to 70. Both Dp71 and Dp40 play an essential role in the neuronal differentiation of PC12 cells [7, 10, 23]. Additionally, the social and emotional alterations in mice lacking the short dystrophins Dp71 and Dp40 have been recently described [33]. All these results demonstrate the importance of Dp71 and Dp40 in the proper functions of the CNS.

In this study, we evaluate the effect of the deletion of leucine 170 (ΔL170) on dystrophin Dp71 and Dp40 based on the previous interesting finding reported by de Brouwer et al., 2014 [22]. They identified six men in a family with a deletion of three base pairs in the DMD gene that results in the deletion of leucine 3238 in the dystrophin Dp427 protein (leucine 170 in Dp71 and Dp40). These patients had varying degrees of intellectual disability but did not present muscular degeneration. This suggests that the deletion of leucine 170 in these patients does not alter the function of dystrophin Dp427 in the skeletal muscle or brain, while the functions of brain dystrophins Dp140, Dp71 and Dp40, in which Dp71 and Dp40 are highly expressed in the CNS, are strongly altered with this deletion, suggesting that the intellectual disability is mainly due to the loss of both Dp71 and Dp40 dystrophins. This hypothesis is supported by our results, since we found that all Dp71 and Dp40 mutants (ΔL170) are delocalised from the membrane, relocalised or mis-accumulated in the cytoplasm and increased in the nucleus of undifferentiated and neuronal differentiated PC12 Tet-On cells. The disturbance in the localisation of Dp71 and D40 mutants may alter the canonical functions including the loss of Dp71- and Dp40-DAP complex in the membrane and the loss of their membrane localisation might inhibit the correct membrane composition required for neuronal differentiation of PC12 cells and brain development. Similarly, a delocalisation of Dp71 mutants, both C272Y mutation and E299 deletion, was observed in neuronal SH-SY5Y and N1E-115 cells [34]. This hypothesis is also supported by a previous study, in which we showed that the mutation L170P in Dp40 (Dp40_L170P) located in the EF-2 hand motif, which has the same position reported by de Brouwer et al., 2014 [22], promotes the nuclear accumulation of Dp40, probably due to the loss of a putative nuclear export signal within the EF-2 hand region [35]. In contrast, the mis-accumulation of Dp71 mutants in the cytoplasm of differentiated cells is likely due to aberrant retention in endoplasmic reticulum. Moreover, Dp71 mutants alter the neurite outgrowth of PC12 Tet-On cells probably due to the loss of the correct cytoplasmic/nuclear transport of these proteins [20, 36] and consequently the post-translational modifications that are required for their distribution and canonical functions. Importantly, the stable expression of Dp40_L170P decreased the percentage of PC12 cells with neurites and neurite lengths compared to Dp40 wild-type [23]. Here, we noticed that both Dp40_ΔL170 and Dp40_L170P showed a similar subcellular localisation; therefore, these results suggest that Dp40_ΔL170 may also affect the neuronal differentiation of PC12 Tet-On cells similar to that observed with Dp40_L170P revealing, the relevance of this mutation in the cognitive disabilities of DMD patients.

Recently, a Dp71 complex with β-dystroglycan and other key proteins of DAP members has been reported in both embryonic and adult mouse brain [37]. Furthermore, another recent study reported that the deletion of leucine 170 in the Dp71d isoform (Dp71d_ΔL170) changes its localisation from the cytoplasm to the nucleus and disrupts its binding to β-dystroglycan and phosphorylation [24], supporting our current hypothesis that the deletion of leucine 170 (ΔL170) in Dp71 and Dp40 brake-down the interaction with β-dystroglycan and other members of DAP complex causing cognitive deficits. In addition, it has been suggested that the presence of EF-1 and EF-2 hand motifs is necessary for the correct folding of the WW domain and the interaction with β-dystroglycan [38]. The results showed structural alterations in all Dp71 and Dp40 mutants (ΔL170 and L170P) as well as in putative L170P mutation of Dp71 isoforms, located within the EF-2 hand region. The structural alteration of Dp71 mutants (ΔL170 or L170P), described in this work, was more evident for Dp71 mutant isoforms than for Dp40_ΔL170 and Dp40_L170P, probably due to the presence of full C-terminal domain in Dp71. However, in both cases, changes in the subcellular distribution of Dp71 and Dp40 proteins demonstrate that this deletion or mutation disrupts the correct folding and likely the neuronal function of Dp71 and Dp40 isoforms. In addition, it has been recently described that both C272Y mutation and E299 deletion disturb the 3D structure of Dp71 protein and promote the formation of Dp71 aggregates in neuronal SH-SY5Y and N1E-115 cells [34], suggesting that all these mutations drastically disturb the functions of Dp71 and Dp40 isoforms. This hypothesis is strongly supported by previous studies, in which the overexpression of Dp71e_Δ71 [10] and Dp40 [23], stimulates the neuronal differentiation; in contrast, the overexpression of Dp40_L170P mutant delayed and inhibited the neuronal differentiation of PC12 Tet-On cells [23].

Previously, we have described that Dp71 and Dp40 isoforms are differentially regulated in the adult mouse brain and retina and during the neural stem cells differentiation [18, 39], and members of the DAPs are also differentially expressed during the neural stem cells differentiation [40]. More recently, we have reported that the expression of Dp71-splice variants is temporally regulated during rodent brain development [5] forming distinct and specific Dp71- and Dp40-DAP complexes in different cell types of the mouse brain and retina [14]. Interestingly, Dp71f group of isoforms are mainly expressed in the embryonic brain, as compared with expression of these isoforms in the post-natal and adult brain; in contrast; Dp71d isoforms are found at low levels in the embryonic stages and increase during the brain development until adult brain. Of these, Dp71f isoform is highly expressed in the embryonic stage (E10.5) as compared with its expression in the adult brain (P60), whereas the Dp71d_Δ71−74 isoform is mainly expressed during the post-natal development of mouse hippocampus and cortex (P1, P7, P14, P21 and P60), and in early post-natal stages of cerebellum (P1 and P7) [5]. All these results suggest that each Dp71 isoform has a different and temporary specific role in the embryonic and post-natal development of the CNS and support the hypothesis that Dp71 variants and the specific Dp71- and Dp40-DAP complexes may have different/opposite and complementary roles during the CNS development and the expression of mutants Dp71d_Δ71ΔL170, Dp71d_{Δ71−74ΔL170}, Dp71e_Δ71ΔL170 and Dp40_ΔL170 disrupt the neuronal differentiation of PC12 Tet-On cells.

Conclusion

In this study, we demonstrate that both deletion of leucine 170 (ΔL170) on dystrophin Dp71 and Dp40 and mutation of leucine 170 to proline (L170P) on Dp40 drastically disturb the subcellular distribution and alter the three-dimensional structure of Dp71 and Dp40 proteins. The mis-localisation of these mutant Dp71 and Dp40 isoforms seems to alter the neuronal differentiation of PC12 Tet-On cells probably due to the loss of their membrane localisation inhibiting the neurite extension and breaking-down the Dp71- and Dp40-DAP interaction. These findings support the current hypothesis that the cognitive impairment observed in DMD patients without muscular dystrophy [22] is mainly due to the loss of the function of Dp71 and Dp40 isoforms in the CNS. This hypothesis is also supported by a previous study in DMD patients [2] and by a recent study in which social and emotional alterations have been observed in a mouse model lacking dystrophins Dp71 and Dp40 [33]. However, further studies detailing the function of Dp71 and Dp40 are needed to completely understand the role of these dystrophins in cognitive deficits. The conclusions here are derived from transient expression of Dp71 and Dp40 mutants; thus, a stable expression of these dystrophin mutants will help us to further characterise the neuronal functions of Dp71 and Dp40 isoforms in this model. Finally, the loss of Dp71 and Dp40 alters the social and emotional behavior of DMD mice models [33]; therefore, the role of dystrophin Dp140 mutants, at the same residue, and Dp71 and Dp40 needs to be evaluated in animal models and screened in DMD patients to understand or discard the participation of these dystrophins in the cognitive disorders.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

Study conception and design were performed by Cecilia Montanez, Jorge Aragón and César García-Cruz. Data collection was carried out by Jorge Aragón, César García-Cruz, Luis Sanchez-Perez, Fernanda Herrera-Rojas, Manuel Posadas-Trejo, Víctor Ceja and Alejandra Sánchez-Trujillo. Analysis and interpretation of results were performed by Cecilia Montanez, Jorge Aragón, César García-Cruz, Luis Sanchez-Perez, Fernanda Herrera-Rojas, Manuel Posadas-Trejo and Candelaria Merino-Jiménez. Supervision: Cecilia Montanez and Jorge Aragón. Resources, funding acquisition and project administration: Cecilia Montanez. The drafts of the manuscript were written by Jorge Aragón, César García-Cruz and Cecilia Montanez. All authors commented on the manuscript.

Funding

This work was supported by Consejo Nacional de Humanidades, Ciencias y Tecnologías (CONAHCyT), Mexico: Fellowship 282052 to C. García-Cruz, and Grants CB-2017-2018-24868 and SEP-CONAHCyT ECOS-NORD-276330 to C. Montanez.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Competing interests

The authors declare no competing interests.

Conflict of interest

The authors declare that they have no conflict of interest. All authors have read and approved the final manuscript to be published.