Effects of Transplanted Umbilical Cord Blood Mononuclear Cells Overexpressing GDNF on Spatial Memory and Hippocampal Synaptic Proteins in a Mouse Model of Alzheimer’s Disease

Abstract

Background/Objective:

Alzheimer’s disease (AD) is a progressive incurable neurodegenerative disorder. Glial cell line-derived neurotrophic factor (GDNF) is a prominent regulator of brain tissue and has an impressive potential for use in AD therapy. While its metabolism is still not fully understood, delivering neuropeptides such as GDNF via umbilical cord blood mononuclear cells (UCBMCs) to the sites of neurodegeneration is a promising approach in the development of innovative therapeutic avenues.

Methods:

UCBMCs were transduced with adenoviral vectors expressing GDNF and injected into AD transgenic mice. Various parameters including homing and survival of transplanted cells, expression of GDNF and synaptic proteins, as well as spatial memory were evaluated.

Results:

UCBMCs were observed in the hippocampus and cortex several weeks after transplantation, and their long-term presence was associated with improved spatial memory. Post-synaptic density protein 95 (PSD-95) and synaptophysin levels in the hippocampus were also effectively restored following the procedure in AD mice.

Conclusions:

Our data indicate that gene-cell therapy with GDNF-overexpressing UCBMCs may produce long-lasting neuroprotection and stimulation of synaptogenesis. Such adenoviral constructs could potentially possess a high therapeutic potential for the treatment of AD.

Keywords

Alzheimer’s disease glial cell line-derived neuro-trophic factor post-synaptic density protein 95 synaptophysin umbilical cord blood mono-nuclear cells

BACKGROUND

Alzheimer’s disease (AD) is progressive incurable neurodegenerative disorder that is the most prevalent cause of dementia, accounting for about 70% of all dementia cases. Currently used pharmacotherapies provide only transient symptomatic effects. The absence of a cure presently available for AD makes the search for novel effective approaches for AD treatment one of the most important directions in current biomedical science.

AD is characterized by two main histopathological hallmarks: extracellular deposition of amyloid-β plaques and intracellular aggregates of hyperphosphorylated tau protein, which were observed in the hippocampus, cerebral cortex, and other regions of the brain. A large body of evidence supports the theory that AD may be primarily a synaptic disease [1]. Hippocampal synapses begin to decline in patients with mild cognitive impairment (MCI) [2]. It was shown that synaptic dysfunctions in AD precede amyloid-β plaque deposition and profound neuronal death [3]. Numerous previous studies demonstrated a progressive loss of synapses, impaired basal neurotransmission and long-term potentiation, and decreased amounts of synaptic proteins (synaptophysin, PSD-95, dynamin-1, etc.) in the hippocampus and cerebral cortex of AD patients as well as in AD mouse models [4 –6].

Alterations in brain levels of neurotrophins is one of the important pathophysiological mechanisms underlying the development of AD and other neurodegenerative disorders. Glial cell line-derived neurotrophic factor (GDNF) is a secreted protein playing important roles in synapse formation, growth, differentiation and survival of neurons [7].

Despite the fact that several studies implicated GDNF regulation in AD pathogenesis [7, 8], it remains poorly understood. Nevertheless, mature GDNF peptide was found to be downregulated in postmortem middle temporal gyrus of AD patients [9], and the GDNF expression level was reduced in the hippocampus of 3xTg-AD model mice [10]. Moreover, two studies demonstrated that the serum GDNF levels are reduced in AD patients [11, 12].

The usage of umbilical cord blood mononuclear cells (UCBMCs) for delivery of neurotrophic factors to the sites of neurodegeneration is a perspective approach for the development of therapies of neurodegenerative diseases [13 –16]. The main goals of such intervention are to support the survival of neurons, restore synaptic functions and plasticity, and to improve the blood supply in affected brain regions. Notably, we demonstrated earlier that a recombinant GDNF gene can be successfully delivered to sites of neurodegeneration using gene-cell constructs based on UCBMCs and viral vectors [16 –18].

This study was aimed at transplanting UCBMCs transduced with adenoviral vectors expressing GDNF to AD-modeling transgenic mice (APP/PS1 line) with a subsequent evaluation of spatial memory, ability of grafted cells for homing, survival and expression of a therapeutic gene in the brain, and hippocampal expression of synaptic proteins.

METHODS

Production and analysis of gene-cell constructs

Human embryonic kidney cells (HEK293A cell line, Invitrogen, USA) were cultivated at 37°C in conditions of humidified atmosphere with 5% CO₂ content, in Dulbecco’s modified Eagle’s medium (DMEM) with 10% fetal bovine serum (FBS, HyClone, EU), 1% antibiotic mixture of penicillin and streptomycin, and 2 mM L-glutamine. Transfection of HEK293 cells with genetic constructs (pAd-EGFP, pAd-GDNF) was performed using the transfection reagent TurboFect (Fermentas Inc, Canada) in accordance with procedure recommended by the manufacturer.

Expression of recombinant genes was confirmed 48 h after transfection. Intensity of enhanced green fluorescence protein (EGFP) expression was evaluated by fluorescent microscopy. Intensity of GDNF expression was assessed by immunofluorescent analysis. Transfected HEK293A cells were fixed with pre-chilled methanol with following incubation at -20°C for 10 min. All wash steps were performed with Tris-buffered saline (TBS; 150 mM NaCl, 50 mM Tris, pH 7.6). Cell membranes were permeabilized by 0.1% solution of Triton X -100 (Helicon, Russia). Cells were incubated with primary anti-GDNF antibodies (Ab) in TBS for 1 h, then washed with TBS and incubated with secondary Alexa 488 Ab for 1 h. The cell nuclei were stained with 4’,6-diamidino-2-phenylindole (DAPI; 10μg/ml, Invitrogen, USA). Results were analyzed using AxyoObserver Z1 fluorescent microscope (Carl Zeiss, Germany).

To produce recombinant adenoviruses Ad5-EGFP and Ad5-GDNF, HEK293A cells were transfected with purified predominantly linearized plasmid DNA pAd-GDNF and pAd-EGFP digested with PacI enzyme. After transfection cultural media was replaced every 2-3 days with fresh one until the formation of visible cytopathic regions characterized by changed cell morphology. On the 10th day after transfection, cell suspension was collected and subjected to several freeze/thaw cycles followed by centrifugation to prepare a crude viral lysate. Viral stock was kept at -80°C.

To obtain preparative amounts of adenovirus encoding egfp and gdnf genes, HEK293A cells were infected with derived virus. After 72 h cell lysate was collected and subjected to several freeze/thaw cycles. Virus concentration and purification was performed by two-step centrifugation in CsCl density with following dialysis and determination of titer of the virus by optical density and plaque assay.

Umbilical cord blood was taken after obtaining informed consent from pregnant woman and prenatal screening for contraindications to blood donation. Blood was collected in CPDA-1 250 GG plastic containers (Terumo, Japan) and delivered to laboratory. Isolation of nuclei containing red blood cells was performed according to procedure published elsewhere [19]. After purification, a fraction of mononuclear cells from umbilical cord blood was cultivated in RPMI-1640 medium with addition of 10% FBS and mixture of penicillin and streptomycin (100 U/ml, 100 mg/ml) (PanEco, Russia). Immediately after isolation, UCBMC were seeded in 10 cm culture dish and transduced with adenoviruses Ad5-GDNF and Ad5-EGFP with multiplicity of infection 10. Cells were incubated for 12–16 h in humidified atmosphere with 5% CO₂ content at + 37°C.

Transduction efficacy and expression of target proteins in UCBMC was evaluated by western blotting. Adenovirus infected cells after 72 h of incubation were lysed in×1.5 sample buffer (10% glycerol, 50 m? Tris-HCl (pH 6.8), 2 m? EDTA, 2% sodium dodecyl sulphate (SDS), 144 m? 2-mercaptoethanol, 0.0084% bromphenolic blue) and analyzed by SDS polyacrylamide gel electrophoresis. Proteins were transferred (55 min, 118 mA) onto polyvinylidene fluoride (PVDF) membranes using Trans-Blot® SD semi-dry electrophoretic transfer cell (Bio-Rad, USA). Before immunological staining, PVDF cellulose membranes containing proteins were placed for 2 h at room temperature in phosphate-buffered saline (PBS) with 0.1% Tween 20 and 5% of skimmed milk in order to block nonspecific Ab binding. Then membranes were incubated (18 h, 4°C) with anti-β-actin Ab and anti-GDNF Ab or anti-GFP Ab. Then the membranes were washed in PBS with 0.1% Tween 20 and incubated with horseradish peroxidase (HRP)-conjugated anti-mouse, anti-rabbit, anti-goat Ab for 2 h. The Ab-stained membrane was placed in the mix of reagents from the ECL Western blotting substrate kit (Bio-Rad, USA) at room temperature for 5 min. The results were visualized with a ChemiDoc XRS + system analyzer (Bio-Rad, USA).

Before transplantation UCBMCs were precipitated by centrifugation and diluted in sterile saline to the concentration of 2×10⁶ cells/. Cells in the amount of 2 million per animal were injected by syringe into the retro-orbital venous sinus of mice.

Experimental groups of animals

Transgenic mice with AD model expressing human mutant genes of amyloid precursor protein and presenilin 1 (genotype B6C3 – Tg(APP695)85Dbo Tg(PSENI)85Dbo) were purchased from the Jackson Laboratory (USA) and housed in the Puschino animal facility within the Bioresource Collection of the Branch of RAS Institute of Bioorganic Chemistry (Moscow Region). Four experimental groups of animals were analyzed: wild type – control (WT); transgenic mice – control (Alz); transgenic mice after transplantation of EGFP-expressing UCBMCs (Alz-EGFP); transgenic mice after transplantation of GDNF-expressing UCBMCs (Alz-GDNF). Animals were 8-9-month-old at the moment of cell transplantation (non-treated animals were age-matched with Alz-EGFP and Alz-GDNF groups).

Animal treatment protocol was approved by Ethic Committee of Kazan State Medical University.

Behavioral test

T-maze setup (Open Science, Russia) was utilized to study spatial memory in four experimental groups of mice: WT (n = 40), Alz (n = 33), Alz-EGFP (n = 9), Alz-GDNF (n = 9). The mice were trained on a rewarded alternation task using a conventional T-maze [20]. The preparation period of experiment was started on day 7 after UCBMCs transplantation. Food consumption was limited during an experiment. The learning of 14 days duration started after the preparation period of 7 days duration. Mice received 6 pairs of training trials on each experimental day for 14 days. At the first (forced) trial of each pair one of the goal arm doors was closed and the mouse was constrained to selecting the opposite arm where food was placed. On the second (free-choice) trial, both goal arm doors were opened, but only the arm opposite the one chosen in the first trial was baited. The criterion for a mouse having learned the task was 3 consecutive days of at least 5 correct responses of the 6 free trials. Mice that did not reach the learning criterion during the entire period of training were assigned with value of 14. Latency to make an arm choice was measured in all trials.

Immunofluorescence and western blot analysis of brain sections and tissue

Expression of synaptic proteins synaptophysin (SYP) and postsynaptic density protein-95 (PSD-95) was evaluated in the hippocampus. SYP and PSD-95 are commonly used as synaptic markers for quantification of synapses [21 –23]. SYP is synaptic vesicle glycoprotein which level reflects the number and density of synapses [21]. PSD-95 is the major scaffolding protein in the excitatory postsynaptic density [24, 25].

Anesthetized mice were decapitated and the brains were removed on day 9 or day 48 after UCBMCs transplantation in 4 experimental groups: WT (n = 5), Alz (n = 4), Alz-EGFP (n = 4), Alz-GDNF (n = 4). The hippocampus from one cerebral hemisphere of each mouse decapitated on day 9 after UCBMCs transplantation was frozen in liquid nitrogen and placed in a deep-freeze chamber at -80°C (for a further quantitative protein assay with western blotting). The hippocampus from another cerebral hemisphere was fixed for immunofluorescence staining in a 4% paraformaldehyde solution for 24 h and then placed in a 30% sucrose solution in PBS with addition of 0.02% sodium azide. To prepare cryostat sections, the tissue was placed into a Neg 50 frozen section medium and frozen for 2 min. Sections were placed in PBS, washed in 0.1% Triton-X100 solution in PBS, and incubated in PBS supplemented with 5% donkey serum and 0.1% Triton-X100 at room temperature for 45 min.

To visualize transplanted GDNF-overexpressing UCBMCs, the sections were incubated with anti-human nuclei antigen (HNu) Ab and anti-GDNF Ab for 1 day at 4°C and washed in PBS. Then the sections were incubated with proper secondary Alexa 647 and Alexa 488 Ab followed by cell nuclei staining with propidium iodide (PI, 5μg/ml in PBS, Sigma, USA).

To assess PSD-95 and SYP expression levels, the sections of brains from mice decapitated on day 9 were incubated with anti-PSD-95 Ab or anti-SYP Ab at 4°C for 2 days, washed in PBS, and cultured with secondary Alexa 647 Ab at room temperature in dark for 2 h and washed in PBS. To visualize nuclei the sections were additionally stained with DAPI (10μg/ml, Sigma) and washed in PBS. The stained sections were embedded in a Shandon ImmuMount medium and examined with LSM 510-Meta confocal scanning microscope (Carl Zeiss, Germany). The dentate gyrus (DG), and the CA1, CA3, and CA4 areas of hippocampus were studied.

Fluorescence intensity of immunostaining was analyzed using the ImagePro software. Images were acquired in the eight-bit grayscale regimen, and the fluorescence intensity (brightness) of each pixel was determined in a range from 0 (absolute black) to 255 (absolute white). The fluorescence intensity of a selected region was calculated in relative units as an average brightness of all pixels of the analyzed region.

Western blot technique was used for the quantitative analysis of synaptic proteins in the hippocampus. Hippocampal tissue was homogenized in radioimmunoprecipitation assay buffer (150 mM NaCl, 50 mM Tris, 1.0% Triton X-100, 0.5% sodium deoxycholate, 0.1% SDS, pH 8.0). Protein concentration measurements were carried out under standard Lowry protocols. A protein concentration was adjusted to 40μg/10μl. A commercial Express Plus RAGE Gel (Genscript, USA) was used for electrophoresis. 10μl of a standardized homogenate was added to every well with gel. Electrophoresis was carried out at 200 V. Proteins were transferred from gel to a PVDF membrane on a semi-dry transfer cell device (Bio-Rad, USA) at 17 V for 40 min. Nonspecific Ab binding was blocked by placing the PVDF membrane in PBS with 0.2% Tween 20 and 5% skimmed milk (a blocking buffer) at 4°C for 24 h Then the membrane was incubated in the blocking buffer with primary anti-β-actin Ab and anti-PSD-95 or anti-SYP. Then the membrane was washed in PBS with 0.2% Tween 20 and incubated for 2 h with secondary Alexa 555 and Alexa 647 Ab. Fluorescent images were visualized with a Typhoon FLA 9500 scanner. The ImageJ Launcher software was used to analyze fluorescent images. The sum of pixels intensity values in a protein lane (RawIntDen) was calculated. The relative expression of a studied protein was determined by dividing RawIntDen of the studied protein to RawIntDen of β-actin and normalizing obtained results to the WT mice group values taken as 100%.

Statistical analysis

Data are presented as mean±standard error of the mean (SEM). In the box-and-whisker charts, the box represents SEM, the whisker is a standard deviation, the horizontal line represents median and the dot in the center—a mean value. The numeric data were processed with the Origin 8.0 software using the one-way ANOVA test with post-hoc Bonferroni corrections. A Fisher’s exact test was used to analyze dichotomous variables in experiments assessing spatial memory in T-maze. Differences were considered significant at p < 0.05.

RESULTS

In vitro study of recombinant genes expression

Western blot analysis of UCBMCs transduced with recombinant adenovirus Ad5-GDNF at 3-day post-infection has revealed a band (19 kDa) specific for N-epitope of GDNF (Fig. 1). We have also detected a specific band (28 kDa) intrinsic to EGFP in cell lysates of UCBMCs transduced with Ad5-EGFP (Fig. 1). These observations confirm the efficiency of transduction of cells with recombinant adenoviruses Ad5-GDNF and Ad5-EGFP.

Fig.1

Analysis of GDNF and EGFP expression in genetically modified UCBMCs. Western blot analysis of cell lysates for UCBMCs transduced with Ad5-EGFP, UCBMCs transduced with Ad5-GDNF and non-transduced UCBMCs (control). Panels represent bands for GDNF (19 kDa), EGFP (28 kDa), β-actin (42 kDa). PL, protein ladder.

Assessment of spatial memory in T-maze

We found that WT mice learned in T-maze for 10.4±0.6 days, with 30% (12 animals) and 72.5% (29 animals) of mice from the total number of animals (40) in the group learned by days 8 and 14, respectively (Fig. 2). The latency to make the arm choice reduced from 28.3±2.1 s to 10.2±1.9 s (by 63.9%) in the process of learning. APP/PS1 transgenic mice (the Alz group) learned significantly longer comparing to WT mice, for 13.3±0.3 days. A percentage of learned mice in Alz group was significantly lower comparing to the WT group: 6.1% and 15.2% animals (2 and 5 of 33 mice) learned by days 8 and 14, respectively. On day 1 of learning, the latency of arm choice in Alz group was 38.5±3.6 s, and it was reduced by 54.3% to 17.6±2.1s by day 14. Thus, the latencies of arm choice at first and last training days were significantly higher in Alz group as compared to the WT group.

Fig.2

Effects of transplantation of UCBMCs overexpressing GDNF on spatial memory performance in APP/PS1 transgenic mice. The results of T-maze experiments are shown for different experimental groups of mice (black symbols – WT, red symbols – Alz, green symbols – Alz-EGFP, blue symbols – Alz-GDNF). Graph represents the progress of learning (x-axis - days of training, y-axis - amount of learned animals as a percentage of the total number of animals in the group). Inlet represents the duration of learning in days. An asterisk (*) indicates values that differ significantly from those in the Alz mice group.

The transplantation of UCBMCs expressing EGFP or GDNF significantly improved spatial memory performance in APP/PS1 mice (Fig. 2). Mice of the Alz-EGFP group learned for 12.2±0.8 days, with 11.1% of them (1 of 9 mice) by day 8 (no significant difference comparing to Alz group), and 44.4% (4 of 9 mice) of them by day 14 (a significant increase compared to Alz group). In the process of learning, the latency of arm choice was reduced from 31.5±2.9 s to 9.9±1.6 s in the Alz-EGFP mice (by 68.6%) which was similar to that in the WT group.

Mice of the Alz-GDNF group learned for 12.3±0.9 days, with 11.1% (1 of 9 mice) (no significant difference comparing to Alz group of mice) and 44.4% (4 of 9 mice) (a significant increase comparing to Alz group of mice) of animals learned by days 8 and 14, respectively (Fig. 2). The latency of arm choice reduced from 46.8±4.2 s to 23.6±4.1 s (by 49.6%) during learning.

Thus, the transplantation of UCBMCs expressing EGFP or GDNF significantly improved spatial learning and memory in transgenic mice modeling AD. Notably, we did not observe significant differences between effects of EGFP- or GDNF-expressing UCBMCs transplantation on the spatial memory performance.

Detection of grafted cells in brain sections from APP/PS1 transgenic mice treated with UCBMCs overexpressing GDNF

To verify the successful cell transplantation and recombinant gene expression in genetically modified UCBMCs, we performed triple fluorescent staining (Fig. 3). Using immunofluorescent staining, we observed HNu⁺/GDNF⁺ cells in the cortex and hippocampus of Alz-GDNF mice both on day 9 and day 48 after UCBMCs transplantation.

Fig.3

Visualization of grafted cells in the brains of APP/PS1 transgenic mice. Triple staining of cortical and hippocampal slices of APP/PS1 transgenic mice on day 48 after transplantation of GDNF-expressing UCBMCs (Alz-GDNF group) with antibodies to HNu (red) and GDNF (green). Nuclei are stained with PI (violet). Scale bar: 20μm.

Thus, genetically modified UCBMCs were detected in the hippocampus and cerebral cortex of APP/PS1 mice at different time points after cell transplantation. Their ability to express a recombinant GDNF gene for long period after the transplantation was also confirmed.

Localization and intensity of PSD-95 and SYP immunoreactivity in the hippocampus

PSD-95 immunoreactivity in hippocampal slices from WT mice was observed predominantly in perikarion of neurons and to a lesser extent in their processes. Imaging of hippocampal slices from Alz and Alz-EGFP, but not Alz-GDNF mice showed very low PSD-95 expression in neuronal processes compared to WT mice (Fig. 4). PSD-95 immunofluorescence in WT hippocampal slices revealed the highest immunostaining intensity in the CA3 (22.85±2.91) and CA4 (33.93±3.14) regions comparing to DG and CA1 (7.84±1.49 and 7.28±1.53, respectively). Alz and Alz-EGFP mice showed significantly lower PSD-95 immunostaining intensity at DG hippocampal regions compared to WT. Meanwhile there were no significant differences in PSD-95 immunofluorescence intensity in Alz-GDNF mice comparing to WT in all hippocampal regions tested (Fig. 4).

Fig.4

PSD-95 immunoreactivity in the hippocampus. Representative fluorescent images demonstrate PSD-95 immunoreactivity at DG, CA4, CA3, CA1 hippocampal areas (shown at (A) of WT, Alz, Alz-EGFP and Alz-GDNF mice (C). PSD-95 immunostaining intensity (ISI) box-and-whisker charts are shown at (B). An asterisk (*) indicates values that differ significantly from those in the Alz mice group. Scale bar: 50μm.

SYP positive boutons in WT hippocampal slices were observed predominantly in processes of neurons and to lesser extent in their perikarions that was similar to other experimental groups (Fig. 5). WT hippocampal slices showed higher SYP immunostaining intensity in CA4 (22.87±2.83) compared to other hippocampal regions (17.37±2.15, 12.97±2.72 and 17.16±1.89 in DG, CA1, and CA3, respectively). Mostly there were no significant differences in SYP immunostaining intensity between WT and other experimental groups (Fig. 5).

Fig.5

SYP immunoreactivity in the hippocampus. Representative fluorescent images demonstrate SYP immunoreactivity in DG, CA4, CA3, CA1 hippocampal areas (shown in (A) of WT, Alz, Alz-EGFP and Alz-GDNF mice (C). SYP immunostaining intensity (ISI) box-and-whisker charts are shown at (B). An asterisk (*) indicates values that differ significantly from those in the Alz mice group. Scale bar: 50μm.

Thus, morphological immunofluorescent analysis revealed that PSD-95 immunoreactivity is significantly reduced in the hippocampus of APP/PS1 mice, whereas transplantation of GDNF-expressing UCBMCs (but not EGFP-expressing UCBMCs) effectively restored PSD-95 expression. Meanwhile, SYP expression under given experimental conditions did not differ significantly in Alz group comparing to WT, as well as not affected by UCBMCs transplantation.

Western blot analysis of PSD-95 and SYP levels in the hippocampus

Representative images of western blots are shown in Figure 6A. PSD-95 relative expression was slightly decreased in Alz and Alz-EGFP comparing to WT (differences are not significant). Meanwhile PSD-95 expression in Alz-GDNF group was 160.10±19.20%, which is significantly higher comparing to all other experimental groups (Fig. 6B).

Fig.6

Western blot analysis of PSD-95 and SYP levels in the hippocampus. A) Representative images of Western blots showing PSD 95, SYP and their corresponding control – β-actin. PL, protein ladder. B, C) PSD-95 and SYP relative expression box-and-whisker charts, correspondingly. An asterisk (*) indicates significant difference between values.

Table 1

Antibodies for immunological studies

Antigen	Host	Dilution	Manufacturer	Application
Primary antibodies
GDNF	Rabbit Rabbit Goat	1 : 100 1 : 150 1 : 500	Santa Cruz Santa Cruz Sigma	HEK293 cells, IF Brain sections, IF UCBMCs, WB
GFP	Rabbit	1 : 300	Santa Cruz	UCBMCs, WB
β-actin	Mouse	1 : 5000 1 : 6000	GenScript	UCBMCs, WB Hippocampus, WB
HNu	Mouse	1 : 150	Millipore	brain sections, IF
PSD-95	Rabbit	1 : 500 1 : 200	Abcam	Hippocampus, WB Brain sections, IF
SYP	Rabbit	1 : 500 1 : 200	Abcam	Hippocampus, WB Brain sections, IF
Secondary antibodies
Anti-rabbit Alexa 488	Donkey	1 : 1000 1 : 200	Life Technologies Thermo Fisher Scientific	HEK293 cells, IF Brain sections, IF
Anti-rabbit Alexa 555	Donkey	1 : 2000	Thermo Fisher Scientific	Hippocampus, WB
Anti-mouse Alexa 647	Donkey	1 : 2000 1 : 200	Thermo Fisher Scientific	Hippocampus, WB Brain sections, IF
Anti- rabbit Alexa 647	Donkey	1 : 200	Thermo Fisher Scientific	Brain sections, IF
Anti-rabbit HRP-conjugated	Goat	1 : 2000	Sigma	UCBMCs, WB
Anti-goat HRP-conjugated	Donkey	1 : 2000	Sigma	UCBMCs, WB
Anti-mouse HRP-conjugated	Donkey	1 : 2000	Sigma	UCBMCs, WB

WB, western blot; IF, immunofluorescence.

SYP relative expression was insignificantly changed in Alz and Alz-EGFP mice comparing to WT mice. However, SYP relative expression in Alz-GDNF mice (227.79±24.08%) was dramatically increased compared to other tested experimental groups (Fig. 6C).

Western blot analysis showed that SYP and PSD-95 relative expression had tendency to decrease in APP/PS1 mice hippocampus compared to WT, which is in line with published results [5], even not confirmed by significant differences. GDNF- (but not EGFP-) expressing UCBMCs transplantation significantly increased relative expression of SYP and PSD-95 in hippocampus.

DISCUSSION

The GDNF family of ligands and their receptors are one of the major neurotrophic networks in the nervous system serving important functions for the development, maintenance and survival of a variety of neurons and glial cells [8]. GDNF has a relatively high specificity for dopaminergic neurons which determines it’s significant potential for the treatment of Parkinson’s disease [26]. However, GDNF has been also shown to have effects on sensory and autonomic ganglia, cerebellar Purkinje cells, hippocampal neurons, as well as noradrenergic, serotoninergic, and cholinergic neurons [26]. Only a handful of studies explored the therapeutic potential of GDNF for AD. GDNF exposure was shown to alleviate cognitive deficits in aged rats [27]. GDNF was found to reduce free radical production in the hippocampus [28], and GDNF gene therapy was shown to be neuroprotective and it prevented the development of cognitive loss in 3xTg-AD mice [10].

In our study, we demonstrated that the transplantation of GDNF-overexpressing UCBMCs resulted in significant improvement of spatial memory in APP/PS1 mice, which was supported by presence of grafted cells and their GDNF expression in the hippocampus and cortex for several weeks after transplantation. This improvement of spatial memory in APP/PS1 mice after GDNF-expressing UCBMCs transplantation was similar to those observed after transplantation of EGFP-overexpressing UCBMCs questioning the positive therapeutic impact of GDNF overexpression. However, long-term effects of brain exposure to neurotrophin may not be obvious in the spatial memory behavioral test which is often associated with a high variability of measured readouts and may reflect only one cognitive function (memory). Evaluating levels of synaptic proteins expression in the brain may provide more generalized and quantifiable estimate of treatment effects. Synaptic dysfunction occurs early in AD; brains of AD patients and mice with AD model have been shown to contain abnormally low levels of synaptic proteins [2, 29]. A reduced expression of SYP indicates synapse loss and is correlated with memory dysfunction in individuals with AD [30]. Reduction of PSD-95 level in the inferior temporal cortex was found to correlate with the AD pathology severity [29]. We assessed expression of pre- (SYP) and post-(PSD-95) synaptic proteins by immunofluorescent and western blot methods in the hippocampus, which is directly involved in memory mechanisms and AD pathogenesis. We found that GDNF-, but not EGFP-overexpressing UCBMCs transplantation effectively increased PSD-95 and SYP levels at hippocampus of APP/PS1 mice. The observed effects of GDNF may be explained by enhanced synaptogenesis and enhanced synaptic strength in the hippocampus. Thus, prolonged (for several weeks) expression of GDNF by grafted cells may provide sustainable therapeutic effects in affected brain areas.

The advantages of UCBMCs for the use in proposed gene-cell therapy include their safety, availability, low immunogenicity, homing capacity, ability to penetrate the blood-brain barrier, and ability to migrate to the sites of neurodegeneration [13 , 14–31]. It was shown recently that the use of human UCBMCs for the treatment of neurodegenerative diseases may not require HLA matching or immunosuppression [32, 33]. UCBMCs are shown to secrete cytokines, chemokines, growth factors and neurotrophic factors which may possess additional therapeutic effects [34 –36].

Conclusions

In summary, our results suggest that gene-cell therapy with UCBMCs overexpressing GDNF may produce long-term neuroprotection and stimulation of synaptogenesis in AD mouse brains. Therefore, gene-cell constructs based on UCBMCs and GDNF-overexpressing adenoviral vectors may have a high therapeutic potential for the treatment of AD.

Footnotes

ACKNOWLEDGMENTS

This study was supported by joint grant of Russian Foundation for Basic Research (RFBR) and Government of Republic of Tatarstan #18-415-160016 and RFBR grant #17-04-02175A. EOP and AVL are supported by scholarships of the President of the Russian Federation for young scientists and PhD students SP-2331.2018.4 and SP 255.2016.4, respectively. Some aspects of methodology for the development of gene-cell approaches to the treatment of neurodegenerative disorders were implemented with the support by the RSF grant #14-15-00847-Π. YOM, IIS, EEG, MSK and AAR were supported by the Russian Government Program of Competitive Growth of Kazan Federal University.

We are grateful to dr. Vadim Bolshakov for critical reading of the manuscript and valuable suggestions.

Authors’ disclosures available online ().

References

Querfurth

, LaFerla

(2010) Alzheimer’s disease. N Engl J Med 362, 329–344.

Scheff

, Price

, Schmitt

, DeKosky

, Mufson

(2007) Synaptic alterations in CA1 in mild Alzheimer disease and mild cognitive impairment. Neurology 68, 1501–1508.

Selkoe

(2002) Alzheimer’s disease is a synaptic failure. Science 298, 789–791.

Sivanesan

, Tan

, Rajadas

(2013) Pathogenesis of Abeta oligomers in synaptic failure. Curr Alzheimer Res 10, 316–323.

Forner

, Baglietto-Vargas

, Martini

, Trujillo-Estrada

, LaFerla

(2017) Synaptic impairment in Alzheimer’s disease: A dysregulated symphony. Trends Neurosci 40, 347–357.

Mukhamedyarov

, Zefirov

(2013) [The influence of beta-amyloid peptide on the functions of excitable tissues: Physiological and pathological aspects]. Usp Fiziol Nauk 44, 55–71.

Budni

, Bellettini-Santos

, Mina

, Garcez

, Zugno

(2015) The involvement of BDNF, NGF and GDNF in aging and Alzheimer’s disease. Aging Dis 6, 331–341.

Ibanez

, Andressoo

(2017) Biology of GDNF and its receptors - Relevance for disorders of the central nervous system. Neurobiol Dis 97, 80–89.

Airavaara

, Pletnikova

, Doyle

, Zhang

, Troncoso

, Liu

(2011) Identification of novel GDNF isoforms and cis-antisense GDNFOS gene and their regulation in human middle temporal gyrus of Alzheimer disease. J Biol Chem 286, 45093–45102.

10.

Revilla

, Sunol

, Garcia-Mesa

, Gimenez-Llort

, Sanfeliu

, Cristofol

(2014) Physical exercise improves synaptic dysfunction and recovers the loss of survival factors in 3xTg-AD mouse brain. Neuropharmacology 81, 55–63.

11.

Forlenza

, Miranda

, Guimar

, Talib

, Diniz

, Gattaz

, Teixeira

(2015) Decreased neurotrophic support is associated with cognitive decline in non-demented subjects. J Alzheimers Dis 46, 423–429.

12.

Straten

, Saur

, Laske

, Gasser

, Annas

, Basun

, Leyhe

(2011) Influence of lithium treatment on GDNF serum and CSF concentrations in patients with early Alzheimer’s disease. Curr Alzheimer Res 8, 853–859.

13.

Islamov

, Rizvanov

, Fedotova

, Izmailov

, Safiullov

, Garanina

, Salafutdinov

, Sokolov

, Mukhamedyarov

, Palotás

(2017) Tandem delivery of multiple therapeutic genes using umbilical cord blood cells improves symptomatic outcomes in ALS. Mol Neurobiol 54, 4756–4763.

14.

Islamov

, Rizvanov

, Mukhamedyarov

, Salafutdinov

, Garanina

, Fedotova

, Solovyeva

, Mukhamedshina

, Safiullov

, Izmailov

, Guseva

, Zefirov

, Kiyasov

, Palotás

(2015) Symptomatic improvement, increased life-span and sustained cell homing in amyotrophic lateral sclerosis after transplantation of human umbilical cord blood cells genetically modified with adeno-viral vectors expressing a neuro-protective factor and a neural cell adhesion molecule. Curr Gene Ther 15, 266–276.

15.

Mukhamedyarov

, Leushina

, Tikhonova

, Petukhova

, Garanina

, Ben Taleb

, Kaligin

, Mukhamedshina

, Rizvanov

, Zefirov

, Islamov

(2018) Intravenous transplantation of human umbilical cord blood mononuclear cells overexpressing nerve growth factor improves spatial memory in APP/PS1 transgenic mice with a model of Alzheimer’s disease. Bionanoscience 8, 473–480.

16.

Petukhova

, Mukhamedshina

, Vasilieva

, Aksenova

, Solovyeva

, Garanina

, Rizvanov

, Zefirov

, Islamov

, Mukhamedyarov

(2015) Stimulation of neurogenesis at hippocampus in Alzheimer’s disease. Genes Cells 10, 54–59.

17.

Mukhamedshina

, Shaymardanova

, Garanina

, Salafutdinov

, Rizvanov

, Islamov

, Chelyshev

(2016) Adenoviral vector carrying glial cell-derived neurotrophic factor for direct gene therapy in comparison with human umbilical cord blood cell-mediated therapy of spinal cord injury in rat. Spinal Cord 54, 347–359.

18.

Mukhamedshina

, Shaymardanova

, Mukhitov

, Garanina

, Rizvanov

, Chelyshev

(2014) Phenotypic modifications and quantitative analysis of glial cells in the area of spinal cord injury at the cell-mediated and direct gene delivery gdnf. Genes Cells 9, 217–222.

19.

Hawley

, Herbert

, Eaker

, Hawley

(2004) Multiparameter flow cytometry of fluorescent protein reporters. Methods Mol Biol 263, 219–238.

20.

Deacon

, Rawlins

(2006) T-maze alternation in the rodent. Nat Protoc 1, 7–12.

21.

Calhoun

, Jucker

, Martin

, Thinakaran

, Price

, Mouton

(1996) Comparative evaluation of synaptophysin-based methods for quantification of synapses. J Neurocytol 25, 821–828.

22.

Roque

, Guizzetti

, Costa

(2014) Synaptic structure quantification in cultured neurons. Curr Protoc Toxicol 60, 122211–32.

23.

Zhang

, Gu

, Zhang

, Liu

, Zhang

, Guo

, Wang

, Gong

, Xu

(2017) NSCs promote hippocampal neurogenesis, metabolic changes and synaptogenesis in APP/PS1 transgenic mice. Hippocampus 27, 1250–1263.

24.

Hunt

, Schenker

, Kennedy

(1996) PSD-95 is associated with the postsynaptic density and not with the presynaptic membrane at forebrain synapses. J Neurosci 16, 1380–1388.

25.

Migaud

, Charlesworth

, Dempster

, Webster

, Watabe

, Makhinson

, He

, Ramsay

, Morris

, Morrison

, O’Dell

, Grant

(1998) Enhanced long-term potentiation and impaired learning in mice with mutant postsynaptic density-95 protein. Nature 396, 433–439.

26.

Allen

, Watson

, Shoemark

, Barua

, Patel

(2013) GDNF, NGF and BDNF as therapeutic options for neurodegeneration. Pharmacol Ther 138, 155–175.

27.

Pertusa

, Garcia-Matas

, Mammeri

, Adell

, Rodrigo

, Mallet

, Cristofol

, Sarkis

, Sanfeliu

(2008) Expression of GDNF transgene in astrocytes improves cognitive deficits in aged rats. Neurobiol Aging 29, 1366–1379.

28.

Cheng

, Fu

, Guo

(2004) Ability of GDNF to diminish free radical production leads to protection against kainate-induced excitotoxicity in hippocampus. Hippocampus 14, 77–86.

29.

Proctor

, Coulson

, Dodd

(2010) Reduction in post-synaptic scaffolding PSD-95 and SAP-102 protein levels in the Alzheimer inferior temporal cortex is correlated with disease pathology. J Alzheimers Dis 21, 795–811.

30.

Sze

, Troncoso

, Kawas

, Mouton

, Price

, Martin

(1997) Loss of the presynaptic vesicle protein synaptophysin in hippocampus correlates with cognitive decline in Alzheimer disease. J Neuropathol Exp Neurol 56, 933–944.

31.

Garbuzova-Davis

, Willing

, Zigova

, Saporta

, Justen

, Lane

, Hudson

, Chen

, Davis

, Sanberg

(2003) Intravenous administration of human umbilical cord blood cells in a mouse model of amyotrophic lateral sclerosis: Distribution, migration, and differentiation. J Hematother Stem Cell Res 12, 255–270.

32.

Passweg

, Baldomero

, Bregni

, Cesaro

, Dreger

, Duarte

, Falkenburg

, Kroger

, Farge-Bancel

, Gaspar

, Marsh

, Mohty

, Peters

, Sureda

, Velardi

, Ruiz de Elvira

, Madrigal

, European Group for Blood and Marrow Transplantation (2013) Hematopoietic SCT in Europe: Data and trends in 2011. Bone Marrow Transplant 48, 1161–1167.

33.

Yang

, Zhang

, Wu

, Min

, Minev

, Zhang

, Luo

, Ramos

, Ichim

, Riordan

, Hu

(2010) Safety evaluation of allogeneic umbilical cord blood mononuclear cell therapy for degenerative conditions. J Transl Med 8, 75.

34.

Harris

, Rogers

(2007) Umbilical cord blood: A unique source of pluripotent stem cells for regenerative medicine. Curr Stem Cell Res Ther 2, 301–309.

35.

Neuhoff

, Moers

, Rieks

, Grunwald

, Jensen

, Dermietzel

, Meier

(2007) Proliferation, differentiation, and cytokine secretion of human umbilical cord blood-derived mononuclear cells in vitro. Exp Hematol 35, 1119–1131.

36.

Fan

, Zhang

, Tang

, Han

, Wang

, Han

(2005) Human umbilical cord blood cells express neurotrophic factors. Neurosci Lett 380, 322–325.