Background
Frontotemporal lobar degeneration (FTLD) is the primary cause of early onset dementia after Alzheimer’s disease (AD) and it is characterized by progressive alterations in behaviour, personality and language [
1‐
3]. Mutations in
tau (
MAPT) and
progranulin (
GRN) genes and a repeat expansion in
C9orf72 are the most common genetic alterations observed in FTLD patients [
4‐
7]. Histochemically, FTLD can be subdivided according to the major component of the protein inclusions deposited in the brain. Around 50 % of the patients can be assigned to the subgroup named FTLD-TDP, because the pathological protein present in the inclusions is TDP-43 (transactive response DNA-binding protein 43 KDa). In the most of the cases, FTLD-TDP is associated with
GRN mutations [
4,
5]. Deposition of TDP-43 has been also been detected in some patients with amyotrophic lateral sclerosis (ALS) [
8], in consonance with the fact that these two diseases share some clinical and genetic features such as mutations the in
TARDBP,
FUS and
C9orf72 genes [
9].
TDP-43 is an evolutionarily conserved nuclear protein that can bind to DNA and RNA, repress transcription, and initiate exon skipping [
10]. Under physiological conditions TDP-43 is a predominantly nuclear protein. Its pathology is characterized by hyperphosphorylation, ubiquitination, cleavage of C-terminal fragments, and nucleus-to-cytoplasm translocation [
8,
11], and its pathogenesis may involve both loss of normal function in the nucleus and toxic gain of function in the cytoplasm [
12].
The phosphorylation of TDP-43 at tandem serines 409 and 410 characterizes all TDP-43 proteinopathy cases and therefore it is considered a hallmark of pathological TDP-43 [
13,
14]. It is known that phosphorylation of site Ser 409/410 of TDP-43 leads to oligomerization and fibril formation in vitro [
13]. Phosphorylation of TDP-43 may also play a role inhibiting the ubiquitin–proteasome system mediated degradation, contributing to the formation of aggregates [
15]. On the other hand, the mutation of serines 409 and 410 to aspartic acid reduces the TDP-43 aggregation [
16].
Casein kinases 1 and 2 (CK-1 and CK-2) were shown to phosphorylate TDP-43 in vitro [
13]. However, antibodies raised against TDP-43 label in histological sections of FTLD and ALS brains show strong reactivity only for phosphorylated epitopes generated by CK-1 [
13]. In addition, it was demonstrated that the products of CK-1 phosphorylation in vitro had similar electrophoretic mobility than hyperphosphorylated TDP-43 present in brain inclusions in FTLD patients [
17]. Together, these observations suggest that CK-1-mediated TDP-43 phosphorylation play a role in disease pathogenesis.
CK-1 is a Ser/Thr protein kinase that is ubiquitously expressed in eukaryotic organisms [
17]. At least seven isoforms (α, β, γ1 − 3, δ, and ε) and various splice variants have been characterized in different organisms [
18]. Among them, CK-1δ has been determined to phosphorylate many different sites on TDP-43
in vitro [
19]. Recently, we have developed a number of potent, very selective and brain permeable CK-1δ inhibitors. These compounds are benzothiazolyl derivatives that showed a selectivity index “S” score of 0.04 after being tested on a wide panel of more than 450 different protein kinases [
20]. We have demonstrated that CK-1δ inhibition prevents TDP-43 phosphorylation in vitro decreasing its neurotoxicity in
drosophila models [
20]. The present work was undertaken to further explore the potential of these CK-1δ inhibitors to overcome main pathologic features of cells derived from FTLD-TPD patients. Our previous work highlighted the role of the CDK6/pRb pathway controlling cell fate survival/death of lymphoblasts from carriers of a loss-of-function
GRN mutation, c.709-1G > A [
21,
22]. It was suggested that an aberrant activation of this cascade could have pathogenic significance in PGRN deficiency-linked FTLD, as it is believed that unscheduled cell cycle entry underlies neuronal loss in neurodegenerative disorders [
23‐
26]. The re-entry of quiescent neurons into the cell cycle may result in a mitotic failure and cell death [
27‐
29]. Moreover, we found accumulated TDP-43 in the cytoplasm of these PGRN-deficient lymphoblasts [
21,
30]. Therefore, it appears that these cell lines from patients, easily accessible, could represent a suitable platform to search novel disease-modifying drugs. Here, we report the effects of two brain penetrant CK-1δ inhibitors, (IGS-2.7 and IGS-3.27), in TDP-43 phosphorylation levels, cytoplasmic TDP-43 accumulation, loss of TDP-43 nuclear function, and proliferative activity of immortalized lymphocytes from FTLD-TDP patients. Both compounds were able to normalize the aberrant cell cycle control and pathological distribution of TDP-43 of PGRN deficient lymphoblasts. Furthermore, our results show a neuroprotective effect of these drugs in a neuronal model of induced TDP-43 phosphorylation. Finally, an
in vivo pharmacokinetic study of IGS-2.7 confirms the brain penetration in mice after i.p. and oral administration. It is suggested that these drugs can be considered promising candidates for novel treatments for FTLD associated to
GRN mutations and others pathologies in which TDP-43 is involved.
Discussion
To date there is no specific pharmacological treatment for FTLD-TDP, being the most frequently occurring dementia in the presenile population. In the majority of the cases it is associated with mutations in the GRN gene. The study of PGRN haploinsufficiency and its influence altering important signaling pathways, as well as the insights into pathological processing of TDP-43 open new avenues for the identification of appropriated targets and the discovery of effective drugs.
Previously we described a cell cycle control failure in lymphoblasts from FTLD patients harboring a single pathogenic splicing mutation in the
GRN gene (c.709-1G > A) [
22]. The enhanced proliferative activity of PGRN deficient cells was accompanied by accumulation of TDP-43 in the cytosolic compartment [
22,
30]. Therefore we concluded that these lymphoblastoid cell lines from FTLD-TDP patients could be a useful platform to test novel disease-modifying drugs, as they recapitulate at least two pathogenic mechanisms thought to be involved in the neurodegenerative process in FTLD, such as reactivation of cell cycle and alteration of TDP-43 subcellular distribution.
TDP-43 pathological processing includes translocation from nucleus to cytoplasm, truncation, hyperphosphorylation and ubiquitination. It is believed that abnormal phosphorylation of TDP-43 at the Ser 409/410 is a critical step in FTLD-TDP and other neurodegenerative diseases such as amyotrophic lateral sclerosis (ALS) [
8,
33] Alzheimer’s disease (AD) [
34], and Parkinson’s disease (PD). Since it was reported that CK-1δ is likely to be involved in TDP-43 phosphorylation
in vivo [
13], the search for specific inhibitors of this enzyme has become a challenge for the treatment of these proteinopathies [
35]. Recently, we developed a number of potent CK-1δ inhibitors able to prevent TDP-43 phosphorylation in vitro and neurotoxicity in vivo [
20]. Two of these benzothiazolyl amides, named IGS-2.7 and IGS-3.27 are able to cross the BBB with IC
50 values in the nM range and great selectivity on a panel of more than 450 different kinases. Here we have evaluated the efficacy of these two drugs in normalizing the survival pattern of lymphoblasts harboring a loss-of-function
GRN mutation. These lymphoblastoid cell lines were previously obtained from patients of FTLD-TDP and asymptomatic individuals with the mutation. For this purpose, we investigated the effects of these two drugs on the increased proliferative activity, TDP-43 phosphorylation, TDP-43 cytosolic accumulation, as well as their effects on CDK6 expression levels. These are distinct features of these PGRN-deficient lymphoblastoid cell lines [
22,
30]. Moreover the potential neuroprotective role of these CK-1δ inhibitors was studied in a neuronal cell model of induced TDP-43 phosphorylation driven by ethacrynic acid treatment [
32]. Our results show that IGS-2.7 and IGS-3.27 prevented the enhanced serum-mediated proliferation of PGRN deficient lymphoblasts, affecting very little the normal basal rates of proliferation in control cells. Moreover these CK-1δ inhibitors blunted the stimulated CDK6 protein and transcript levels observed in mutant lymphoblasts, thus providing an explanation for the antiproliferative effect of these drugs. At the doses used in the proliferative experiments, both IGS-2.7 and IGS-3.27 inhibited significantly the endogenous TDP-43 phosphorylation. Interestingly, the inhibition of TDP-43 phosphorylation was accompanied by reduced cytosolic TDP-43 accumulation in lymphoblasts carrying the
GRN mutation. Taken together these results suggest that CK-1δ-mediated phosphorylation of TDP-43 may play an important role in controlling the trafficking of TDP-43 protein from the nucleus to the cytosol. This finding is in consonance with previous work showing similar effects of inhibitors of cyclin-dependent kinases (CDKs) on cytosolic TDP-43 accumulation [
36]. It is worth to highlight the apparent relationship between TDP-43 phosphorylation, cytosolic TDP-43 accumulation, and cell cycle regulatory proteins. Our data showing that the PGRN deficiency-induced increased proliferative activity and CDK6 levels are accompanied by TDP-43 cytosolic accumulation, and the previous reports showing the involvement of cell division cycle 7 (CDC7) kinase on TDP-43 pathology [
37] provide further support to the idea that altered cell cycle regulatory proteins may play a role in abnormal TDP-43 processing under pathological conditions.
The important question as to whether cytosolic TDP-43 accumulation implicates the gain of a new toxic function for TDP-43 and/or the concomitant reduced TDP-43 nuclear levels may represent the loss of an essential function, cannot be fully ascertained with the present data. However our results indicate that the well-known nuclear effect of TDP-43, inducing repression of
CDK6 expression [
31] is blocked in PGRN-deficient lymphoblasts. These results are in consonance with a recent report indicating the loss in brains of FTD/ALS individuals, of other nuclear function of TDP-43 acting as splicing repressor of nonconserved cryptic exons [
38]. Thus loss of TDP-43 nuclear functions may have pathogenic significance. Interestingly, our data demonstrate that CK-1δ inhibitors can rescue the nuclear transcriptional regulation of
CDK6 gene by preventing phosphorylation and cytosolic exportation of the TDP-43 protein.
We didn’t observe significant differences between c.709-1G > A mutation carriers, either asymptomatic or with clinical signs of dementia, regarding proliferative activity and cytosolic TDP-43 accumulation. However, it was reported that clinically asymptomatic carriers show poorer neuropsychological performance reflecting a prodromal phase of the disease [
39]. Because most of the asymptomatic carriers are younger than the patients it was suggested that these alterations might be early manifestations of the disease. Thus it is tempting to speculate that CK-1δ inhibitors could hopefully be useful to slow disease progression in the early stages of disease.
Although FTLD-TDP-associated changes detected in lymphoblasts from patients may reflect those occurring in brain, it is also shown that our CK-1δ inhibitors, IGS-2.7 and IGS-3.27, were also effective preventing death in a neuronal cell model of induced TDP-43 phosphorylation with a concomitant decrease of TDP-43 phosphorylation, TDP-43 cytosolic accumulation, and blocking the increase in CDK6 levels.
Conclusion
Our data indicate that the brain penetrant CK-1δ inhibitors are able to normalize the increased proliferation of FTLD-TDP lymphoblasts and to prevent aberrant TDP-43 cytosolic accumulation. Considering the pathogenic role of aberrant TDP-43 homeostasis and cell cycle control failure in FTLD-TDP brain, it is suggested that CK-1δ could be potentially a novel therapeutic target for the treatment of FTLD-TDP and other TDP-43 proteinopathies, with special mention to ALS, considering that a clinical, genetic and neuropathological overlap exists between FTLD-TDP and ALS. Moreover, the CK-1δ inhibitor, IGS-2.7, with excellent pharmacokinetic properties, emerges as a new drug candidate for the future treatment of neurodegenerative diseases where TDP-43 is involved.
Methods
Materials
All components for cell culture were obtained from Invitrogen (Barcelona, Spain). The
N-benzothiazolyl-2-phenyl-acetamides derivatives, CK-1δ inhibitors, IGS-2.7 and IGS-3.27 were synthesized as previously described (compound 20 and 46) [
20]. The chemical structure, IC
50 values regarding CK-1δ inhibition together with effective permeability values, that predict their ability to cross the blood brain barrier (BBB) assessed by PAMPA (Parallel artificial membrane permeability assay) [
40], are provided in Table
2. In the PAMPA-BBB assay the permeability of compounds IGS-2.7 and IGS-3.27 were compared with that of 10 commercial clinical drugs for experiment validation [
20]. Ethacrynic acid was obtained from Sigma (Alcobendas, Spain). Antibodies against human TDP-43 (10782-2-AP) and phospho (409/410)-TDP-43 (22309-1AP) were obtained from Proteintech (Mancheser, UK). Antibodies against CDK6 (sc-177), β-actin (sc-81178) and α-tubulin (sc-23948) were obtained from Santa Cruz Biotechnologies (Santa Cruz, CA, USA) and anti-LaminB1 was purchased from Calbiochem (Billerica, MA, USA).
Table 2
Overview of the compounds used in this study
IGS-2.7 | | >60 | 0.023 ± 0.002 | 11.3 ± 2.0 | CNS+ |
IGS-3.27 | | >60 | 0.047 ± 0.005 | 4.4 ± 2.9 | CNS+ |
Cell lines
Lymphoblastic cell lines
Peripheral blood samples of all the individuals enrolled in this studio were taken after written informed consent of the patients or their relatives (demographic information is presented in Table
3) to establish the lymphoblastoid cell lines as previously described [
41], by infecting peripheral blood lymphocytes with the Epstein Barr virus (EBV). All study protocols were approved by the Donostia Hospital and the Spanish Council of Higher Research Institutional Review Board and are in accordance with National and European Union Guidelines. Lymphoblastoid cells lines were grown in suspension in T flasks in an upright position, in approximately 8 ml of RPMI-1640 medium that contained 2 mM L-glutamine, 100 μg/ml streptomycin/penicillin and 10 % (v/v) fetal bovine serum (FBS) and maintained in a humidified 5 % CO
2 incubator at 37 °C. Fluid was routinely changed every 3 days by removing the medium above the settled cells and replacing it with an equal volume of fresh medium.
Table 3
Characteristics of individuals enrolled in this study
Age (years) | 51.8 ± 4.3 | 52.8 ± 4.3 | 65.3 ± 2.3 |
Sex, female, % (n) | 50 % (5) | 50 % (6) | 100 % (7) |
Age at onset | - | - | 61 ± 0.6 |
Phenotype | Asymptomatic | Asymptomatic | FTD-bv; CBS |
Family history, % (n) | 70 % (7) | 54.5 % (6) | 57.1 % (4) |
Neuronal cell culture
The human neuroblastoma SH-SY5Y cell line was propagated in Dulbecco’s Modified Eagle Medium (DMEM) containing L-glutamine (2 mM), 1 % non-essential amino acids, 1 % penicillin/streptomycin and 10 % fetal bovine serum (FBS) under humidified 5 % CO2. On attaining semiconfluence, cells were treated with ethacrynic acid (EA) (20 μM) for 12 h. Some cultures were pretreated for 1 h with the CK-1δ inhibitors (5 μM). After treatment, cell viability was assessed by MTT, pTDP-43 and CDK6 levels by Western blotting and the subcellular localization of TDP-43 was visualized under a confocal microscopy.
Determination of cell proliferation, cell viability and cell cycle
Cell proliferation was determined by total cell counting, using a TC10™ Automated Cell Counter, Bio-Rad Laboratories, S.A. (Madrid, Spain). EBV-immortalized lymphocytes from control and
GRN mutation carriers were seeded at an initial cell concentration of 1 × 10
6 cells × mL
−1 and enumerated everyday thereafter. Cells failing to exclude the dye were considered nonviable. Cell viability was determined by the MTT assay (3-[4,5-Dimethylthiazol-2-yl]-2,5-Diphenyltetrazolium Bromide), as previously described [
42]. Cell survival was estimated as the percentage of the value of untreated controls. Cell cycle phase distribution was routinely determined by cell permeabilization followed by propidium iodide (PI) staining and flow cytometry analysis using an EPICS-XL cytofluorimeter (Coulter Científica, Móstoles, Spain).
Immunoblotting analysis
To prepare whole-cell extract, cells were harvested, washed in PBS and then lysed in ice-cold lysis buffer as previously described [
43]. To separate the cytosolic and nuclear fractions, cells were harvested, washed in PBS and then lysed in ice-cold hypotonic buffer as previously described [
44]. After extraction on ice for 15 min, 0.5 % Nonidet P-40 was added and the lysed cells were centrifuged at 4,000 rpm for 10 min. Supernatants containing cytosolic proteins were separated and pellets were resuspended in hypertonic buffer to lysate the nucleus [
44]. The protein content of the extracts was determined by the Pierce BCA Protein Assay kit (Thermo Scientific). 50–100 μg of protein were fractionated on a SDS polyacrylamide gel, and transferred to Poly (vinylidene) fluoride (PVDF) membranes (Millipore, Billerica, MA, USA). The membranes were then blocked with 5 % Bovine Serum Albumin (BSA) (Sigma) and incubated, overnight at 4 °C, with primary antibodies in the following concentrations, TDP-43 (1:1000); phospho-(S409/410)-TDP-43 (1:500); CDK6 (1:1000); β-actin (1:500); α-tubulin (1:1000) and Lamin B1 (1:1000). Signals from the primary antibodies were amplified using species-specific antisera conjugated with horseradish peroxidase (Bio-Rad) and detected with a chemiluminiscent substrate detection system ECL. Protein band densities were quantified using Image J software (National Institutes of Health, Bethesda, Maryland, USA) after scanning the images with a GS-800 densitometer from Bio-Rad.
RNA preparation and quantitative real-time PCR
Total RNA was extracted from cell cultures using Trizol reagent (Invitrogen, Alcobendas, Madrid, Spain). RNA yields were quantified spectrophotometrically and RNA quality was checked by the A260/A280 ratio and on a 1.2 % agarose gel to determine the integrity of 18S and 28S ribosomal RNA. RNA was then treated with DNase I Amplification Grade (Invitrogen, Alcobendas, Madrid, Spain). One microgram was reverse transcribed with the Superscript III Reverse Transcriptase kit (Invitrogen, Alcobendas, Madrid, Spain). Quantitative real-time polymerase chain reaction (PCR) was performed in triplicates using TaqMan Universal PCR MasterMix No Amperase UNG (Applied Biosystems, Alcobendas, Madrid, Spain) reagent according to the manufacturer’s protocol. Real time quantitative PCR was performed in the Bio-Rad iQ5 system using a thermal profile of an initial 5-min melting step at 95 °C followed by 40 cycles at 95 °C for 10 s and 60 °C for 60 s. Primers were designed using the Universal ProbeLibrary for Human (Roche Applied Science, Madrid, Spain) and used at a final concentration of 20 μM. The sequences of the forward and reverse primers used are the following: for CDK6 5′-tgatcaactaggaaaaatcttggac-3′ and 5′-ggcaacatctctaggccagt-3′; for β-actin, 5’-ccaaccgcgagaagatga-3’ and 5’-ccagaggcgtacagggatag-3’. Relative messenger RNA (mRNA) levels of the genes of interest were normalized to β-actin expression using the simplified comparative threshold cycle delta-delta CT method (2-[ΔCT CDK6 -ΔCT Actin]).
Confocal laser scanning microscopy
Cells (1 × 106 × ml−1 for lymphoblast and 300,000 for SH-SY5Y) were fixed for 30 min in 4 % paraformaldehyde in PBS, and blocked and permeabilized with 0.5 % TritonX-100 in PBS-0.5 % BSA for 60 min at room temperature. Then, cells were incubated overnight with anti-TDP43 polyclonal antibody. After removing the primary antibody, cells were washed with PBS and incubated with Alexa Fluor 488-conjugated anti-rabbit antibody alone or in combination with phalloidin for cytoskeleton staining. For nuclear staining, the preparations were mounted on ProLong® Gold Antifade Reagent with DAPI (Thermo Fisher) and visualized with the LEICA TCS-SP5-AOBS confocal microscope system (Heidelberg, Germany). Quantification of TDP-43 was performed using Image J software. Data is expressed as the ratio of the fluorescence intensity of cytosolic TDP-43 vs the intensity of the fluorescence of the nuclear protein.
Pharmacokinetic in vivo study
A group of 48 male BALB/c mice (8–12 weeks old) weighting between 25 and 35 g following a single intraperitoneal (i.p) and oral dose (o.p) administration were used following the guidelines of the Institutional Animal Ethics Committee (IAEC). Mice were divided into two groups (Group 1: i.p. and Group 2: p.o.) with each group comprising of 24 mice. Animals in Group 1 were administered with derivative 11 solution formulation in 5 % NMP, 5 % solutol HS in normal saline intraperitoneally at a dose of 2 mg.Kg−1. The dosing volume administered was 10 mL.Kg−1. Animals in Group 2 were administered orally with derivative 11 in suspension formulation in 0.1 % Tween 80, 0.5 % NaCMC in water at a dose of 10 mg/Kg through oral gavage using a 22-G oral feeding needle. The dosing volume administered was 10 mL.Kg−1. Blood samples (approximately 60 mL) were collected from retro-orbital plexus under light isoflurane anesthesia from retro orbital plexus of three mice at each time point: 0.08, 0.25, 0.5, 1, 2, 4, 8 and 24 h (i.p.) and 0.25, 0.5, 1, 2, 4, 6, 8 and 24 h (p.o.). Samples were collected into labeled micro-tubes, containing 20 % K2EDTA solution as an anticoagulant. Plasma samples were separated from the whole blood by centrifugation at 4000 rpm for 10 min at 4 ± 2 °C and stored below −70 °C until bioanalysis. Immediately after collection of blood, brain samples were collected from each mouse at 0.08, 0.25, 0.5, 1, 2, 4, 8 and 24 h (i.p.) and 0.25, 0.5, 1, 2, 4, 6, 8 and 24 h (p.o.). Brain samples were homogenized using ice-cold phosphate buffer saline (pH 7.4) and homogenates were stored below −70 °C until analysis. All samples were processed for analysis by protein precipitation using acetonitrile. Concentrations of IGS-2.7 in mouse plasma and brain samples were determined by fit-for-purpose LC-MS/MS method using the following equipment and parameters: MS System Used: AB Sciex API-4000; Software Version: Analyst 1.5; Ion Source: Turbo spray; Mobile Phase: A: 0.1 % Formic acid in acetonitrile, B: 10 mm Ammonium formate; Flow Rate: 0.8 mL/min; Column Used: Waters Xterra, 50 × 3.0, 5 μm. The final bioanalytical method developed has a lower limit of quantification (LLOQ) of 2.01 ng/mL in plasma and 10.07 ng/mL in brain. Non-Compartmental-Analysis tool of Phoenix WinNonlin® (Version 6.3) was used to assess the pharmacokinetic parameters. Peak plasma concentrations (Cmax) and time for the peak plasma concentrations (Tmax) were the observed values. The areas under the concentration time curve (AUClast and AUCinf) were calculated by linear trapezoidal rule. The terminal elimination rate constant, k was determined by regression analysis of the linear terminal portion of the log plasma concentration time curve. MRT was calculated by using formula as MRT = AUMCinf/AUCinf.
Statistical analysis
Statistical analyses were performed with Graph Pad Prism 6 (La Jolla, CA, USA). All the statistical data are presented as mean ± standard error of the mean (SEM). Normality was checked with the Shapiro-Wilk test. Parametric tests were therefore used in the statistical analysis. Based on the expertise achieved on previous works [
21,
22] we can expect that, with the sample size we used and the significance level we fixed, the variability within groups will be low enough and the differences between groups to detect will be high enough to ensure a statistical power above 0.9. Statistical significance was estimated by both one-way and two-way analysis of variance (ANOVA) followed by the Bonferroni’s test for multiple comparisons. A value of
p < 0.05 was considered significant.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
AM-R and AM conceived the study, and participated in its design and coordination. CA and AdlE carried out the experiments with cell lines, CG and DIP participated in the design and synthesis of the CK-1 inhibitors, and IGS carried out the chemical synthesis and the PAMPA assay. FM and ALdM recruited the FTLD-TDP patients, carried out the genetic analysis, and helped to draft the manuscript. CA, IGS, and AdlE performed the statistical analysis and prepared the figures. AM-R, CA, and AM wrote the manuscript. All authors critically discussed results, revised and approved the manuscript. All authors read and approved the final manuscript.