Quantitative Immunofluorescence Mapping of HSP70’s Neuroprotective Effects in FUS-ALS Mouse Models

Abstract

Amyotrophic lateral sclerosis (ALS) is a progressive neurodegenerative disease, often linked to mutations in the FUS gene, leading to toxic protein aggregates. This study investigates the role of HSP70, a molecular chaperone, in mitigating neurodegeneration in FUS-ALS mouse models. Using quantitative immunofluorescence microscopy, we mapped cellular changes in the primary motor cortex of double transgenic FUS/HSP70 mice and compared them to single FUS-transgenic controls. Our results reveal that double transgenic mice exhibit significantly reduced neuronal damage and increased levels of mature neuronal (NeuN) and microglial (Iba1) markers, indicating a protective effect of HSP70. Intracellular HSP70 expression proved more effective than extracellular release, suggesting that targeted HSP70 delivery to neurons may offer a promising therapeutic avenue for ALS. This study underscores the potential of quantitative immunofluorescence for mapping neuroprotective pathways and highlights HSP70’s impact on mitigating FUS-related pathology in ALS.

1. Introduction

Amyotrophic lateral sclerosis (ALS) is one of the most common neurodegenerative diseases, develops rapidly and results in fatal outcomes [1]. Despite its clinical symptoms and important regulatory points in skeletal muscles, ALS is primarily neuronal damage [2]. A large number of both familial and, to a lesser extent, sporadic ALS cases refer to mutations of the fused in sarcoma/translocated in liposarcoma gene (FUS). For instance, FUS-associated disease is second most common among ALS cases in China [3] and fourth in Western Europe [4], although there are certain regions where FUS mutations in ALS are seldom found [5,6].

Molecular chaperones maintain protein homeostasis in cells, thereby taking part in any reaction to neuronal damage, including that in ALS [7]. In FUS-associated ALS, the damage involves aggregation of mutant FUS gene product, aberrant FUS RNA binding protein, and is also managed by molecular chaperones [8]. Among the diverse group of molecular chaperones, heat shock 70 kDa proteins (HSP70) are distinguished for their crucial role of protein quality control in cell populations of nervous tissue [9,10]. HSP70 have already been shown to prevent aberrant FUS aggregation in affected neurons [8]. Therefore, employing HSP70 upregulation for preventing tissue pathology development in aberrant FUS presence is an aim of high relevance.

Plausible mechanisms of HSP70–FUS interaction are not comprehensively elucidated yet. Pathogenic FUS leaves its normal nuclear localization for cytoplasm [11]. There, aberrant FUS joins to stress granules [12], i.e., cytoplasmic condensates in the cytoplasm, blocking their functioning [13] or turning into solid state from liquid droplets of FUS, as shown in vitro [14]. Normally, HSP70 should prevent FUS from this transition due to its phase separation potential [15]. However, it is not clear why this potential of normal HSP70 cannot protect neurons in case of FUS-associated ALS development.

Such a failure of molecular chaperoning can result from a generally low activity of HSP70 in ALS because natural heat shock factor (HSF1) mediated HSP70 upregulation is compromised [16]. Here, we decided to address if the enhancement of the HSP70 expression may be beneficial in FUS-associated pathology. In this regard, we used a murine model with human FUS ^[1-359] gene insertion [17,18] co-expressed with human HSP70.

Nevertheless, HSP70 overexpression can be equivocal as different HSP70 isoforms are situated in various cellular locations and act differently in ALS in particular. In cytosol, Hsp70 member 1A (HSPA1A) is considered as the key isoform, and its role for protein quality control was already reported [10]. Further, Serlidaki and colleagues showed HSPA1A to be efficient against development of SOD1-associated ALS [19]. We designed this study to regard cellular changes in nervous tissue due to FUS ^[1-359]–HSPA1A interaction in vivo when both genes are expressed together.

Current research belongs to a series of studies in FUS-related ALS pathogenetic mechanisms in various structures of the central nervous system.

2. Materials and Methods

2.1. Animal Experiments

We used mice of six breeds (n = 6 in each group). Wild-type C57Bl/6 animals were a control group (further referred to as “Control”). Single HSP70 transgenic mice had either human HSPA1A expression with intracellular accumulation (“HSP70 (in)”) or extracellular secretion (“HSP70 (out)”), two models reported previously by Gurskiy and colleagues with positive impact shown for high chaperone activity [11]. Single FUS-transgenic mice expressed mutant human FUS [1-359] with slightly noticeable cytoplasmic mislocalization of its aberrant product (“FUS [1-359]”), a model reported earlier by Shelkovnikova and colleagues [20]. In this model, aberrant FUS changes its localization from nuclear to cytoplasmic with further pathological aggregation there. Wild-type and single transgenic animals were received from Sechenov University vivarium. Further, double transgenic FUS [1-359] Hemi/HSPA1A mice were bred out, whereas “FUS [1-359]/HSP70 (in)” were with intracellular HSP70 expression and “FUS [1-359]/HSP70 (out)” were with its extracellular secretion. We provided PCR control of offspring genotype after the coupling of single transgenic animals (Figure 1).

Mice of all the groups were anesthetized by xylazine 5 mg/kg (Interchemie, Venray, The Netherlands) with tiletamine/zolazepam 40 mg/kg (Virbac, Carros, France) and sacrificed by decapitation for studying on Week 20 after reaching their average body mass at 32 ± 2.5 g, whereas both male and female animals were studied. We housed animals under a 12 h light–dark cycle with a room temperature and relative humidity of 55–60%. Mice got free access to water and granulated feeding ad libitum. The animal experiments were approved by the Local Ethics Committee of Sechenov University (Protocol No. 04-23, 2 March 2023). All the manipulations are provided with respect to 3R principles [21].

2.2. Immunofluorescent Study

Immediately after the animals were sacrificed, we fixed their brains in 10% neutral buffered formalin (ErgoProduction, Saint-Petersburg, Russia), dehydrated them with increasing concentration of isopropanol (Biovitrum, Saint-Petersburg, Russia) and paraffinized, further obtaining coronal sections of the brain (d = 5 μm) on silane adhesive (Minimed, Bryansk, Russia). After specimen drying and dewaxing (One-step Dewaxing/Antigen Retrieval Buffer, pH 9.0, 20×, lot XF05RT4N9592, Elabscience, Wuhan, China), we blocked non-specific protein reactions by 2% bovine serum albumin (BSA) (lot PM-T1725/1000, Biosera, Cholet, France) diluted by phosphate saline buffer.

We used the following primary antibodies: anti-NeuN, monoclonal antibody, diluted as 1:1000 (clone SR45-07, No. ET1602-12, lot H661803001; Huabio, Bejing, China); anti-GFAP, monoclonal antibody, diluted as 1:500 (clone SA03-04, No. ET1601-23, lot HO0913; Huabio, Bejing, China); anti-Iba1, polyclonal antibody, diluted as 1:200 (No. ER1802-20, lot HL0213; Huabio, Bejing, China); anti-S100-β, monoclonal antibody, diluted as 1:100 (clone SC57-02, No. ET-1610-3, lot H661380007; Huabio, Bejing, China); and anti-SOX9, monoclonal antibody, diluted as 1:100 (clone R06-7F6, No. E-AB-81453, lot KL00H0265902; Elabscience, Wuhan, China).

After the primary antibody staining, we incubated the slides with 2% BSA and polyclonal secondary fluorescent antibody anti-rabbit-TRITC, diluted as 1:100 (No. E-AB-1053, 22038; Elabscience, Wuhan, China). All the antibodies were used according to standard manufacturer protocols. For counterstaining, we employed 4′,6-diamidino-2-phenylindole (DAPI) to visualize cell nuclei (No. E-IR-R103; Elabscience, Wuhan, China). After buffer washing and dehydration, we mounted the specimens under coverslips.

2.3. Cell Detection

In the cerebrum of mice at obtained histological slides, we detected zones of the primary motor cortex, which is the most affected by ALS pathogenesis [22]. The zones of interest were found as MoP and M1 according to Paxinos and Franklin stereotaxic atlas and Allen stereotaxic atlas, correspondingly [23,24], considering the differences between them [25]. We obtained microphotographs with Axio Imager. A1 microscope and Axiocam 305 color camera in Zen 3.10 software package (Zeiss, Baden, Germany). For microphotographs of fluorescent images, we employed fluorescent light with λ of 540 to 620 nm to visualize secondary antibodies subtracting the autofluorescence spots.

We calculated the number of antibody-positive cells per field of view in layers II-VI of the primary motor cortex at the sections. The number of analyzed sections per each animal in each group was selected in order to have appropriate statistical significance and power values and varied for each calculation within sections performed at 0.2 mm forward from bregma. We employed objective magnification of 10× for calculation of positive cells in staining by antibodies to NeuN, GFAP, S100β, and SOX9 markers, and objective magnification of 20× for anti-Iba1 staining.

For anti-NeuN staining, we used machine learning software QuPath 0.5.0 (Queen’s University Belfast, Belfast, UK) as reported by Bankhead and colleagues [26], calculating the total number of NeuN-positive cells and their ratio to DAPI-stained cells at the same fields of view with threshold of intensity coefficient at 25 units and cell edge distribution to 5 pixels. For other antibodies, we employed manual calculation, considering a cell to be antibody-positive if cell body/processes (if applicable) were stained with autofluorescence subtracted.

2.4. Data Curation

We processed the data obtained for any animal with statistical software package OriginPro 2024 (OriginLab, Northampton, MA, USA). For survival of animals with mutant FUS expression, we employed log-rank test. For cell calculation, the normality of distribution was estimated by Shapiro–Wilk test considering the type of data and sample sizes. To compare average group means in normal distribution (detected for NeuN and SOX9 staining) we used single-factor analysis of variance (ANOVA), as the most appropriate test for morphological parameters registered independently of time measures, with Tukey test for post hoc difference estimation in pairs of group samples considering the statistics’ results and sample size. As many groups had no normal distribution, we used the Kruskal–Wallis test with Dunn’s test as post hoc criterion to compare samples in total instead comparison between median values [27] as the Kruskal–Wallis is more credible for comparison of samples in their integrity and not simply of median values. For all tests, we considered statistical difference as significant for p value less than 0.01 and power more than 0.95. Sample sizes and the number of fields of view per animal were sufficient accounting for statistical power and 3R principles [21].

3. Results

Mice of FUS [1-359], FUS [1-359]/HSP70 (in), and FUS [1-359]/HSP70 (out) populations, not randomized to the experiment, lived for 14–27 weeks since birth on average, whereas the populations of all single HSP70 mice and wild-type animals lived much more than 30 weeks (Figure 2). In FUS [1-359]/HSP70 (in) mice, the overall survival is greatly higher than in FUS [1-359] mice, but even a lower duration of overall survival was shown for FUS [1-359]/HSP70 (out) mice. However, all the FUS-expressing mice were symptomatic for muscle atrophy independently of their survival terms.

Figure 3A shows images of the primary motor cortex stained by anti-NeuN antibodies for mice from six groups with the corresponding boxplots at Figure 3B,C. The weakest level of fluorochrome luminance is found in the primary motor cortex of FUS [1-359] mice. At Figure 3B, the boxplot illustrates a significant and considerable decrease in average mean value for the number of NeuN-positive cells in FUS [1-359] animals compared to both groups of single transgenic HSP70-expressing mice and controls, whereas double transgenic mice have no statistical difference of the value from all the groups except FUS [1-359] animals.

The differences at the NeuN/DAPI ratio calculations are less prominent between the groups. FUS [1-359] mice show a lower ratio of NeuN-positive to DAPI-stained cells compared to all animals lacking aberrant FUS expression (Figure 3C), although no increase occurs in double transgenic animals compared to FUS [1-359] mice. The distribution for NeuN-stained cells is normal, and for the NeuN/DAPI ratio is not, so not the average means but general samples are compared for the NeuN/DAPI ratio.

The number of GFAP-positive cells has almost no difference between the groups, excluding FUS [1-359]/HSP70 (out) animals with a slight statistically significant increase in GFAP + cells compared to HSP70 (in), their counterparts with intracellular HSPA1A accumulation, as well as with FUS [1-359] mice (Figure 4A,C). However, Figure 3B and the relevant graph at Figure 4D reveal a colossal increase in number of S100β-positive cells in all groups expressing FUS compared to control wild-type animals, what is well shown by the dynamics of fluorescent pattern at Figure 3B. Single FUS [1-359] transgenic animals also have a significant difference in the number of S100β-positive cells to HSP70 (out) group, whereas this pattern is not seen for anti-GFAP staining.

Figure 5A shows visual differences between the groups for Iba1 antibody staining. Statistically significant differences are visible after calculations, represented at the corresponding boxplot (Figure 5C). FUS [1-359] animals have less Iba1-positive cells than double transgenic mice with intracellular HSPA1A expression, FUS [1-359]/HSP70 (in), as well as with the animals of HSP70 (in) group. Further, we observe control animals to have less Iba1 expression than all the groups with HSPA1A overexpression.

Figure 5B,D visualize data for comparison of average means for SOX9-positive cells in the primary motor cortex between the groups. We observe here a considerable divergence between FUS [1-359]/HSP70 (in) mice and all other groups. Meanwhile, FUS [1-359]/HSP70 (out) mice have a significant difference of SOX9 + cells from both groups of single HSPA1A transgenic animals but no difference with either controls or double transgenic mice revealed.

Double transgenic mice with human FUS and Hsp70 member 1A (HSPA1A) expression have less structural signs of brain tissue damage than FUS-transgenic mice without transgenic insertion of HSPA1A. Interestingly, the results differ significantly between two types of HSPA1A expression, with intracellular accumulation and extracellular release.

4. Discussion

In our study, we bred out the mice carrying aberrant form of the human FUS gene with the mice overexpressing human HSPA1A, both intra- and extracellular. We revealed animal survival to be even shorter in double FUS [1-359]/HSP70 (out) transgenic mice than in FUS [1-359] animals. However, increased terms of survival were characteristic for FUS [1-359]/HSP70 (in) transgenic mice, thus pretending HSPA1A overexpression with intracellular accumulation is more beneficial. It has been reported that not only survival but body mass also decreases in animals with FUS-related ALS pathology [28], although no body mass decrease occurs in HSP70 overexpression [29].

As ALS is principally a motor neurons degeneration, the primary zone of our interest in histological study included the motor cortex. We preferred NeuN over other possible neuronal molecular markers as it is the most universal one for mature cerebral neurons [30]. We observed a sharp decline in NeuN-positive cells for FUS-transgenic animals compared to all other groups. In this study, we reveal for the first time a neuroprotective impact of both intracellular and extracellular HSPA1A overexpressed together with FUS [1-359] in motor neurons of the primary motor cortex.

For the NeuN/DAPI ratio, we did not mention a significant difference between FUS [1-359] mice and double transgenes. This may be explained as an evidence of cell damage reduction in HSP70-affected animals not only in neuronal population. If total number of mature neurons increases, and no considerable increase in ratio to all cells occurs, this suggests that the total number of cells also increases, so total cellularity increases. However, there may also be a different interpretation. For instance, increased cellularity is specific for tumorigenic phenomena, whereas HSP70 is a well-known anti-apoptotic and thereby pro-oncogenic agent [31]. Despite the lack of significant difference between two double transgene groups for the NeuN/DAPI ratio, we can easily notice a larger peak in Tg FUS + HSP70 (out), i.e., with extracellular secretion (Figure 3C). This is probably due to a lower inhibiting impact of extracellular HSP70 on apoptotic enzymatic cascades inside the cytosol in cells other than NeuN-positive ones.

Other markers elucidate the HSP70-related potential of glial cell populations in the cerebral cortex. For astrocytes, we used three markers. GFAP is specific for cytoskeleton of astrocytes, so its content increases in GFAP activation during defensive reactions [32], e.g., in neuroinflammation, neurodegeneration, and traumas. However, all FUS-overexpressing groups have normal number of GFAP-positive cells. In contrast, single Tg HSP70 (out) mice have larger GFAP expression than even single FUS [1-359] transgenic mice. We explain this by the fact that not all the neurodegenerative processes lead to the same level of considerable changes outside the cells, and aberrant FUS stays in cytoplasm with no aggregation in extracellular matrix. At the same time, extracellular HSP70 binds Toll-like receptors (TLR) on astroglial cells [33], thereby invoking innate immune reactions and therefore astrocytic non-specific activation.

Further, GFAP is more characteristic for white matter and not for the cortex. Meanwhile, an increase in S100β-positive cells, which are usually grey matter astrocytes [32], is observed for all FUS-containing groups. Thus, these data show a slight upregulation of marker expression for grey matter astrocytes (S100β) and almost no changes in white matter astrocytes marker (GFAP). The original study of FUS [1-359] transfection model in mice showed an increase in GFAP content but the analysis was performed with specimens of the spinal cord and brainstem there [20], whereas we observed the primary motor cortex.

Third marker studied for astrocytes, SOX9, was included as reported in previous studies being upregulated in rodents with ALS modeled, but for SOD1 and not FUS models [34]. We, however, did not observe any difference involving transgenic FUS mice for this marker. Curiously, FUS [1-359]/HSP70 (in) animals got a strong rise of SOX9 expression in the primary motor cortex compared to controls, and this phenomenon is not typical for other study groups. This fact deserves a supplementary investigation due to its probable role in explaining molecular mechanisms of intracellular interaction between HSPA1A and FUS in stress granules directly inside astrocytes or in neuron–astrocyte intercommunication.

To estimate differences in microglial cell populations between the groups, we studied the number of Iba1-positive cells, as protein Iba1 is evenly distributed in the cytoplasm and processes of microglial cells [35]. Iba1 was also shown to increase sharply in single FUS [1-359] transgenic animals before [36]. FUS [1-359]/HSP70 (in) group attracts the most attention, being significantly different in sample distribution from both controls and FUS [1-359] animals. Moreover, single HSP70 (in) mice have a similar pattern of difference from controls and FUS [1-359] animals.

These data for Iba1 are consistent with those for SOX9 marker and are probably explained by a more efficient phagocytosis of damaged cellular structures after chaperone-launched signaling in neuron–astrocyte intercommunication. As the double transgenic animals, FUS [1-359]/HSP70 (in), had also no difference from FUS [1-359] animals in the NeuN/DAPI ratio, this probable phagocytosis, if exists, does not greatly affect the total cell number. As this phenomenon is not quiet typical, we consider it to be an independent finding, and the degree of its relation to HSP70 upregulation should be tested further.

Together, all our data mentioned require a more profound studying of neuronal and astroglial proteome in FUS [1-359]/HSP70 (in) animals to understand a molecular interplay between astrocytes and microglial cells. Our study showed some evident signs of neuroprotection according to immunofluorescent marker assessment, whereas NeuN level is principal, as we estimated it in the primary motor cortex, thereby showing neuronal survival actually at the cortical level of motor activity regulation. Neuroprotection studies may, however, include a plenty of additional approaches as we show further in perspectives.

In assessment of histological preparations, we preferred immunofluorescent (IF) assay over standard immunohistochemical staining (IHC) even though a risk for degrading of fluorochrome signal with course of time is obvious. However, for such experimental studies as we did, a diverse nature of markers estimated requires to allow multiplexing of slides if needed what is only achievable for IF and not IHC. For example, NeuN estimation solely without the NeuN/DAPI ratio is not logical and even speculating as the level of NeuN expression may depend on duration of fixation [30], thus these factors are two be counted at the same slides together.

In our study, we investigated the only protein aberration of ALS in combination with HSP70 overexpression. However, ALS pathology comprises a multifaceted RNA-protein participation [37] with RNA-dependent and RNA-independent steps of protein aggregation [38]. Further, there were reports about nuclear aggregation of FUS with even no mislocalization to cytoplasm [39]. Thus, different points of view are accessible to regard the hypothesis of interaction between molecular chaperones and aberrant RNA and proteins in ALS. A large body of evidence had been already demonstrated for HSP40-mediated HSP70 activity in ALS [40]. Novoselov et al. also showed double transgenic mice (Hsp40 and SOD1) to have a higher degree of motor neuron survival [41]. Further, Chen and colleagues reported Hsp40 expression to reduce signs of damage in TDP-43-associated ALS in vitro and in patients [17], and the consistent effect was achieved for HSP70 [42]. For another key player of ALS, C9orf72 protein, a presence of Hsp40 enhanced clearance of aberrant C9orf72 [43].

However, history of studying HSP70 interactions in ALS is not long in regard to FUS-associated pathology, and covers molecular studies and research of direct FUS-HSP70 intercommunication. For molecular studies, it was found that pathological stress granules with FUS are rather stable [13], although HSP70 and HSP40 were supposed to disintegrate such aberrant agglomerates [44]. Thus, Rozales and colleagues experimentally discovered an increased mobility of mutant FUS aggregates having HSP40- and therefore HSP70-dependent manner [45].

For direct HSP70–FUS interaction, there were attempts to regard the administration of HSP70 inducers in fusopathies. In 2020, Kuta and colleagues showed aberrant FUS^R525L expression to regress in vitro after treatment with arimoclomol, which provided an enhanced HSP70 binding to its upregulator, HSF-1 [46]. Nevertheless, excellent series of studies from the same team revealed almost no positive changes in motor neurons with mutant FUS^R521G isoform after HSP70 induction by arimoclomol [47,48]. Therefore, cells with different types of mutant FUS may experience a different degree of involvement into HSP70-driven changes. This ambivalence requires further understanding, as shown by our study, where we used a FUS ^{[1-359] Hemi} version of aberrant FUS protein.

Another approach is to test absolutely other molecular chaperones for their potential to resist to aberrant FUS. For example, HSP104 functioning requires no HSP70 engagement [49]. It was also found that HSP104-related machinery is highly efficient for disaggregating misfolded FUS but is only appropriate as external chemical because HSP104 is typical for yeast and not synthesized by metazoans [50].

Discussing our data, we should outline the difference of two HSPA1A isoforms tested. A theoretical thinking allows to consider extracellular HSPA1A usage to be worse. Exactly, HSP70/HSPA1A is a ligand for TLR [33]. TLR-mediated HSPA1A overexpression outside the cells can lead to generalized seizures as shown in HSP70-transgenic mice [51]. Interestingly, we observed no damaging effects of a direct HSP70 treatment to nervous tissue in our earlier study, so the phenomenon described by von Rüden and colleagues is not universal [52].

HSP70 takes part in many cellular processes during neuroinflammation and neuronal damage [53]. Thus, HSP70 obviously has side effects as its overexpression results not only in FUS modification but in changing a number of other cellular pathways it participates in. Even in ALS, HSP70 may act via different mechanisms including, e.g., histone deacetylase management [47]. This implies possible effects of excessive HSP70 on FUS models by different mechanisms. Therefore, we cannot claim a direct effect of HSP70 to FUS mislocalization and/or aggregation, but a beneficial role of HSP70 presence together with aberrant FUS expression is observed. One may also expect a potential of this presence to be studied not only in ALS but in other neurodegenerative diseases with a proven pathogenesis of fusopathies, such as frontotemporal dementia and Alzheimer’s disease.

Further, HSP70 is well-known for its tumorigenic potential [31]. Although this effect may even become a benefit in neurodegeneration, as there were reports about cell proliferation stimulated by HSP70 in hippocampal region [54], we cannot deny possible risks. The anti-apoptotic activity of HSP70 may also result in cardiac hypertrophy and not only in tumorigenesis [55]. Considering these risks of extracellular HSP70 and no significant benefit of its presence over intracellular HSP70 for the management of FUS-associated ALS in our study, we consider the usage of targeted HSPA1A delivery to the cells of nervous tissue to become a more promising therapeutic option in ALS than administration of HSP70 inducers.

A possible limitation may be supposed in our study considering that the overexpression of HSP70 had not been consistent with an upregulation of its co-chaperones so the resulting effect may be weaker. However, a set of classical works by team of Sapolsky [56,57] showed HSP70 is sufficient to be overexpressed alone with no supplementary co-stimulation of co-chaperones required for cellular changes including protecting ones. Probably, that is due to the multifaceted intracellular role of HSP70 as we demonstrated in our review last year [10]. There are meanwhile several limitations and perspectives for this study.

4.1. Limitations of This Study

• Murine models should be extrapolated to humans with caution, especially considering studies of this type to serve mainly for further pharmacological investigations.
• Body mass assessment has not been provided. We outline measurements of body mass dynamics for our double transgenic animals further to observe a greater number of animals for this biometric parameter.

4.2. Perspectives

• Study of neuroinflammation process with proinflammatory cytokines and enzymes of apoptotic cascades will potentially disclose machinery of HSP70 action in FUS-related changes in the primary motor cortex.
• Molecular and cellular approaches to assess disassembling of FUS aggregates and changes in aberrant FUS expression are promising to elucidate if there is a direct or indirect impact of HSP70 presence on FUS localization and fusopathy.
• Spinal cord and brainstem study is relevant for a comprehensive complex assessment of FUS expression, aggregation, and disassembling under HSP70 impact.
• HSP70 role in slowing ALS pathogenesis down can be shown in other in vivo models related to mutations in SOD1 and C9orf72 genes, as well as the roles of HSP70 co-chaperones and counterparts (i.e., HSP40 and HSP90).
• Physiological motor tests and electroencephalography of the motor cortex and striatum in our model of double FUS [1-359]/HSP70 transgenic mice are beneficial in future works to reveal the level of motor disorder in experimental animals.
• Brain proteome and transcriptome analysis are additional methods that would be appropriate at double transgenic mice with FUS [1-359]/HSP70 expression to assess the overall neuroprotection degree.

5. Conclusions

In this study, we demonstrated double transgenic FUS ^{[1-359] Hemi}/HSPA1A mice to have less signs of brain tissue damage than FUS-transgenic mice without insertion of HSPA1A. Our study underscores the potential of quantitative immunofluorescence for mapping neuroprotective pathways and highlights HSP70’s impact on mitigating FUS-related pathology in ALS. The most prominent recovery is observed for levels of NeuN marker in the primary motor cortex of both male and female animals, whereas NeuN-positive cells distinguish the population of mature neurons. Our data suggest that HSPA1A has a strong impact on tissue architectonics in aberrant FUS aggregation in vivo. The levels of markers considered do not differ between two types of HSPA1A expression, extra- and intracellular, in neuronal population, while there are differences in glial populations. Fluorescent approach instead of IHC study allowed us to visualize a difference between total cell number peaking relative to NeuN expression. Regarding both the evident efficacy of HSP70 overexpression to resist FUS-related damage in the primary motor cortex and possible adverse effects of extracellular HSP70, we suppose that studies of targeted HSPA1A delivery to motor neurons might be preferable direction of modern research compared to the development of non-tissue- and cell-specific HSP70 inducers. This work outlines perspectives of future HSP70 studies in ALS in vivo models.