Differential accumulation of tau pathology between reciprocal F1 hybrids of rTg4510 mice

Tau, a family of microtubule-associated proteins, forms abnormal intracellular inclusions, so-called tau pathology, in a range of neurodegenerative diseases collectively known as tauopathies. The rTg4510 mouse model is a well-characterized bitransgenic F1 hybrid mouse model of tauopathy, which was obtained by crossing a Camk2α-tTA mouse line (on a C57BL/6 J background) with a tetO-MAPT*P301L mouse line (on a FVB/NJ background). The aim of this study was to investigate the effects of the genetic background and sex on the accumulation of tau pathology in reciprocal F1 hybrids of rTg4510 mice, i.e., rTg4510 on the (C57BL/6 J × FVB/NJ)F1 background (rTg4510_CxF) and on the (FVB/NJ × C57BL/6 J)F1 background (rTg4510_FxC). As compared with rTg4510_CxF mice, the rTg4510_FxC mice showed marked levels of tau pathology in the forebrain. Biochemical analyses indicated that the accumulation of abnormal tau species was accelerated in rTg4510_FxC mice. There were strong effects of the genetic background on the differential accumulation of tau pathology in rTg4510 mice, while sex had no apparent effect. Interestingly, midline-1 (Mid1) was identified as a candidate gene associated with this difference and exhibited significant up/downregulation according to the genetic background. Mid1 silencing with siRNA induced pathological phosphorylation of tau in HEK293T cells that stably expressed human tau with the P301L mutation, suggesting the role of Mid1 in pathological alterations of tau. Elucidation of the underlying mechanisms will provide novel insights into the accumulation of tau pathology and is expected to be especially informative to researchers for the continued development of therapeutic interventions for tauopathies.

Tau, a family of microtubule-associated proteins, forms abnormal intracellular inclusions, so-called tau pathology, in a range of neurodegenerative diseases collectively known as tauopathies. The rTg4510 mouse model is a well-characterized bitransgenic F1 hybrid mouse model of tauopathy, which was obtained by crossing a Camk2α-tTA mouse line (on a C57BL/6 J background) with a tetO-MAPT*P301L mouse line (on a FVB/NJ background). The aim of this study was to investigate the effects of the genetic background and sex on the accumulation of tau pathology in reciprocal F1 hybrids of rTg4510 mice, i.e., rTg4510 on the (C57BL/6 J × FVB/NJ)F1 background (rTg4510_CxF) and on the (FVB/ NJ × C57BL/6 J)F1 background (rTg4510_FxC). As compared with rTg4510_CxF mice, the rTg4510_FxC mice showed marked levels of tau pathology in the forebrain. Biochemical analyses indicated that the accumulation of abnormal tau species was accelerated in rTg4510_FxC mice. There were strong effects of the genetic background on the differential accumulation of tau pathology in rTg4510 mice, while sex had no apparent effect. Interestingly, midline-1 (Mid1) was identified as a candidate gene associated with this difference and exhibited significant up/downregulation according to the genetic background. Mid1 silencing with siRNA induced pathological phosphorylation of tau in HEK293T cells that stably expressed human tau with the P301L mutation, suggesting the role of Mid1 in pathological alterations of tau. Elucidation of the underlying mechanisms will provide novel insights into the accumulation of tau pathology and is expected to be especially informative to researchers for the continued development of therapeutic interventions for tauopathies.
Tau belongs to a family of microtubule-associated proteins and was originally identified as a heat-stable protein essential for microtubule assembly 1 . Human tau is encoded by the MAPT gene on chromosome 17q21 2 and consists of six isoforms, 0N3R, 1N3R, 2N3R, 0N4R, 1N4R, and 2N4R, which are generated by alternative splicing of exon 2 (E2), E3, and E10 [3][4][5] .
Tau forms abnormal intracellular inclusions, so-called tau pathology, in neurons and glial cells in a range of neurodegenerative diseases collectively known as tauopathies, which include Alzheimer's disease, progressive supranuclear palsy, corticobasal degeneration, and Pick's disease. Tau pathology is characterized by abnormally hyperphosphorylated tau molecules that form bundles of paired helical filaments and straight filaments 6,7 . In 1998, missense mutations of tau were identified in patients with frontotemporal dementia associated with parkinsonism linked to chromosome 17 [8][9][10] . This discovery demonstrated that the pathological abnormalities of tau can be a primary or even the sole cause of tauopathy 7 .
rTg4510 is a well-characterized bitransgenic F1 hybrid mouse model of tauopathy. rTg4510 mice were originally produced by crossing a responder transgenic mouse (tetO-MAPT*P301L mouse line on a FVB/ NJ background) with an activator transgenic mouse (Camk2α-tTA mouse line on a 129S6 background). The responder carries a tetracycline response element placed upstream in a complementary DNA (cDNA) encoding human 0N4R tau with a P301L mutation that is linked to hereditary tauopathy, while the activator expresses a tetracycline-controlled transactivator (tTA) under the control of the Ca 2+ -calmodulin kinase II α (Camk2α) promoter 11,12 . Consequently, the hybrid F1 mice express human P301L mutant tau in the forebrain at a level approximately 13-fold higher than that of endogenous mouse tau and display progressive age-dependent accumulation of tau pathology, which can include abnormal tau phosphorylation and argyrophilic inclusions in the forebrain region, decreased dendritic spine density, neurodegeneration, and a variety of cognitive deficits [11][12][13][14][15][16][17] .
In a recent study, the original 129S6 background of the Camk2α-tTA mouse line was replaced with the C57BL/6 background, one of the most commonly used mouse strains, and reported that the introduction of the C57BL/6 background into the rTg4510 mouse background altered the presentation of tau pathology only minimally from the original phenotype and with no loss of fidelity 18 . In contrast, we previously reported a significant difference in the accumulation of tau pathology in male reciprocal F1 hybrids of rTg4510 mice 19 . In addition, Yue et al. reported a sex difference in tau pathology and memory decline in the original rTg4510 mice 20 . The importance of the genetic background and sex in the phenotype of transgenic mouse model is well known. Therefore, in the present study, the mechanisms underlying differential tau accumulation in rTg4510 mice were investigated, while focusing on the effects of the genetic background and sex. A deeper understanding of tau pathology will be useful for the development of therapeutic regimens for tauopathies.

Results
Differential accumulation of tau pathology between reciprocal F1 hybrids. Reciprocal F1 hybrids of rTg4510 mice consisting of two genetic backgrounds (Table 1) were used in this study: rTg4510 mice on a (C57BL/6 J × FVB/NJ)F1 (maternal strain × paternal strain) background (rTg4510_CxF) and rTg4510 mice on a (FVB/NJ × C57BL/6 J)F1 background (rTg4510_FxC). To investigate the accumulation of tau pathology, histological analysis was performed of coronal brain sections from male and female rTg4510_CxF and rTg4510_ FxC mice at 6 months of age. Immunohistochemical analysis of human tau (HT7) revealed a high level of immunoreactivity in the cerebral cortex and hippocampus in all groups although there were no significant differences (Fig. 1A). Immunohistochemical analysis of phosphorylated tau (AT8) and Gallyas silver staining unveiled a large number of positively stained structures in the cerebral cortex and hippocampus of rTg4510 mice (Fig. 1A). Quantitative analysis was performed of the AT8-immunoreactive areas and Gallyas-positive stained areas of the cerebral cortex of rTg4510 mice. Two-way ANOVA with the genetic background and sex as factors revealed that the genetic background had the most profound effect [F (1, 20) = 31.18, p < 0.0001], while sex had no effect [F (1, 20) = 3.37, p = 0.0813] in the AT8-immunoreactive areas of the cerebral cortex (Fig. 1B). Subsequent analysis with the Mann-Whitney U test indicated significantly greater AT8-immunoreactivity in rTg4510_FxC mice than rTg4510_CxF mice (p < 0.0001; Fig. 1C). Furthermore, two-way ANOVA with the genetic background and sex as factors revealed that the genetic background had the most profound effect [F (1, 20) = 45.74, p < 0.0001], while sex had no effect [F (1, 20) = 3.726, p = 0.0679] in the Gallyas-positive stained areas of the cerebral cortex (Fig. 1D). Gallyas-positive staining of rTg4510_FxC mice was significantly greater than that of rTg4510_CxF mice (p < 0.0001; Fig. 1E). Quantitative analysis also revealed that the genetic background, but not sex, had a significant main effect in the AT8-immunoreactive areas and Gallyas-positive stained areas of the hippocampus, and significantly higher levels of these areas in rTg4510_FxC mice, as compared with rTg4510_CxF mice (Supplementary Fig. S1, Table S1). These results suggest more severe accumulation of tau pathology in rTg4510_FxC mice than rTg4510_CxF mice. In addition, there was no significant sex difference in the accumulation of tau pathology detected in rTg4510 mice.

Soluble and insoluble tau accumulation differs according to the genetic background.
Western blot analysis of the TBS-soluble and sarkosyl-insoluble fractions extracted from the cerebral cortex of rTg4510 mice was performed to investigate the accumulation of soluble and insoluble tau in the brain. With the use of polyclonal (Tau) and monoclonal (HT7) antibodies, Tau was detected at 55 kDa as a major band and at around 64 kDa as a minor band in the TBS-soluble fraction ( Fig. 2A); however, sarkosyl-insoluble tau was only detected at around 64 kDa (Fig. 2B). Two-way ANOVA (genetic background × sex) of the densitometric analysis results revealed no effect of the genetic background [F (1,20) (Fig. 2D,H). The level of 64-kDa tau in the TBS-soluble fraction was significantly higher in rTg4510_FxC mice than rTg4510_CxF mice (p = 0.0005 for Tau, p = 0.0003 for HT7; Fig. 2E,I).
In the sarkosyl-insoluble fraction, the genetic background and sex had no effect on 64-kDa tau levels probed with the polyclonal antibody (Tau), as determined by two-way ANOVA (Fig. 2F). In contrast, two-way ANOVA (genetic background × sex) revealed that the genetic background had a significant main effect [F (1, 20) = 5.937, p = 0.0243], but not sex [F (1, 20) = 0.7476, p = 0.3975], on sarkosyl-insoluble tau levels probed with the HT7 antibody (Fig. 2J). The level of sarkosyl-insoluble tau probed with HT7 was significantly higher in rTg4510_FxC mice than rTg4510_CxF mice (p = 0.0197; Fig. 2K). These results suggest greater accumulation of abnormal tau species detected at around 64 kDa in rTg4510_FxC mice, as compared with rTg4510_CxF mice, although there Effect of genetic background on the abnormal phosphorylation of tau. Subsequently, western blot analysis of abnormally phosphorylated tau in the TBS-soluble and sarkosyl-insoluble fractions was performed. Phosphorylated tau probed with AT8 (pSer202 and pThr205), AT180 (pThr231), and AT270 (pThr181) in the TBS-soluble fraction was revealed as a major band at 64-kDa and a minor band at 55-kDa, while AT100 (pThr212 and pSer214) showed only a 64-kDa band (Fig. 3A). In the sarkosyl-insoluble fraction, phosphorylated tau probed with AT8, AT100, AT180, and AT270 was only detected at around 64 kDa (Fig. 3B).  Table S1). The Mann-Whitney U test revealed significantly higher levels of AT8-immunoreactivity (C) and Gallyaspositive staining (E) in samples obtained from the rTg4510_FxC mice, as compared with the rTg4510_CxF mice. The rTg4510_CxF and rTg4510_FxC groups contained six males and six females each (n = 12/group). Data are presented as the mean ± SEM. ***p < 0.001 (Mann-Whitney U test).   Table S1). The Mann-Whitney U test revealed significantly higher levels of Tau (E) and HT7 (I) at 64 kDa in the TBSsoluble fraction and HT7 (K) in the sarkosyl-insoluble fraction from rTg4510_FxC mice, as compared with rTg4510_CxF mice. The rTg4510_CxF and rTg4510_FxC groups contained six males and six females each (n = 12/group). Data are presented as the mean ± SEM. ***p < 0.001, *p < 0.05 (Mann-Whitney U test).    Table S1). The Mann-Whitney U test revealed significantly higher levels of AT8 (E), AT100 (H), AT180 (M), and AT270 (R) at 64 kDa in the TBS-soluble fraction and AT100 (J) and AT180 (O) in the sarkosyl-insoluble fraction of the rTg4510_FxC mice, as compared with the rTg4510_FxC mice. The rTg4510_CxF and rTg4510_FxC groups contained six males and six females each (n = 12/group). Data are presented as mean ± SEM. ***p < 0.001, **p < 0.01, *p < 0.05 (Mann-Whitney U test). www.nature.com/scientificreports/ determined by two-way analysis of variance (ANOVA) (Fig. 3I,N), while the genetic background and sex had no effect on phosphorylated tau in the sarkosyl-insoluble fraction with the use of the AT8 and AT270 probes (Fig. 3F,S). Probing of the sarkosyl-insoluble fraction with AT100 and AT180 revealed significantly higher levels of phosphorylated in rTg4510_FxC mice than rTg4510_CxF mice (p = 0.0377 for AT100, p = 0.0284 for AT180; Fig. 3J,O). We further analyzed the ratios of phosphorylated tau to total tau which were detected at 55 and 64 kDa in the TBS-soluble fraction and at 64 kDa in the sarkosyl-insoluble fraction ( Supplementary Fig. S2). There were no significant changes in the ratio of phosphorylated tau to total tau, which was detected at 55 kDa in the TBS-soluble fraction ( Supplementary Fig. S2). A two-way ANOVA (genetic background × sex) revealed an effect of genetic background, but no effect of sex on the ratios of AT8 ] in rTg4510 mice. The Mann-Whitney U test revealed significantly higher ratios of AT100 and AT180 to total tau in rTg4510_FxC mice compared with rTg4510_FxC mice (p = 0.0464 for AT8, p = 0.0038 for AT100; Supplementary Fig. S2). Taken together, these results suggest that the genetic background of rTg4510 mice may affect not only the increase in the levels of abnormal tau, but also the increase in the ratio of tau phosphorylation, in particular at the AT100 and AT180 sites, and in part at the AT8 site.
No difference in tau mRNA levels between reciprocal F1 hybrids. Next Identification of a gene related to the differential accumulation of tau in rTg4510 mice. We attempted to identify genes related to the difference in tau accumulation between rTg4510_CxF and rTg4510_ FxC mice. Transgene insertion has been reported to cause a 244-kb deletion on chromosome 14 of tetO-MAPT*P301L mice, which affects Fgf14 and a 508-kbp deletion on chromosome 12 of Camk2α-tTA mice, which affects Vipr2, Wdr60, Esyt2, D430020J02Rik, and Ncapg2 21 . We therefore attempted to determine if the genetic  Table S2). The WT_CxF and WT_FxC groups contained four males and four females each (n = 8/group). The rTg4510_CxF and rTg4510_FxC groups contained six males and six females each (n = 12/group). Data are presented as the mean ± SEM. www.nature.com/scientificreports/ background had any effect of the mRNA levels of particular genes in WT and rTg4510 mice. However, the results revealed no significant effects of the genetic background on the mRNA levels of the tested genes between reciprocal F1 hybrids ( Supplementary Fig. S3, Table S2, S3). Subsequently, we investigated whether there were differences in the mRNA levels of disease-associated tau kinases and select phosphatases, namely cyclin-dependent-like kinase 5, dual specificity tyrosine-phosphorylation-regulated kinase 1A isoform 1, glycogen synthase kinase-3 β, casein kinase 1 alpha 1, microtubule affinity-regulating kinase 1 (Mark1), Mark2, Mark3, Mark4, protein kinase cAMP-activated catalytic subunit alpha, and protein phosphatase 2 catalytic subunit alpha. However, the genetic background and sex had no effect of on the mRNA levels of these genes ( Supplementary Fig. S4, Table S2, S3). Next, qPCR analysis of 10 upregulated and downregulated genes was performed (Fig. 4, Supplementary  Fig. S5, S6, Table S4), based on the RNA-seq analysis results of male WT_CxF (n = 1) and male WT_FxC (n = 1) mice. The results revealed a significant difference in the mRNA levels of midline-1 (Mid1) between reciprocal F1 hybrids. Two-way ANOVA (genetic background × sex) revealed that the genetic background, but not sex, had a significant effect on  Fig. 5A]. The levels of Mid1 in WT_CxF and rTg4510_CxF mice were significantly higher than those in WT_FxC and rTg4510_FxC mice, respectively (p = 0.0122 for WT, p = 0.0007 for rTg4510; Fig. 5B).
Tau accumulation and Mid1 mRNA levels in rTg4510 mice at 3 months of age. Histological and western blot analyses of male and female rTg4510_CxF and rTg4510_FxC mice at 3 months of age were performed. Histological analysis showed no significant differences in AT8-immunoreactivity and Gallyas-positive staining between rTg4510_CxF and rTg4510_FxC mice (Supplementary Fig. S7). Western blot analysis of the TBS-soluble fraction showed significant differences in AT8 levels, but not levels of HT7, AT180, and AT270 ( Supplementary Fig. S8A-K). By two-way ANOVA (genetic background × sex), there was a main effect of the genetic background, but not sex, on AT8 levels at 55 kDa [F (1, 12) = 5.919, p = 0.0316 for genetic background, F (1, 12) = 0.5805, p = 0.4608 for sex] and 64 kDa [F (1, 12) = 12.9, p = 0.0037 for genetic background, F (1, 12) = 0.8891, p = 0.3643 for sex; Supplementary Fig. S8D and E], while the Mann-Whitney U test revealed significantly high levels of AT8 at 55 kDa (p = 0.0207) and 64 kDa (p = 0.0207) in rTg4510_FxC mice, as compared with rTg4510_CxF mice (Supplementary Fig. S8F and G). In contrast, two-way ANOVA (genetic background × sex) revealed no effects of the genetic background or sex on the levels of HT7, AT8, AT100, AT180, and AT270 in the sarkosyl-insoluble fractions between the rTg4510_CxF and rTg4510_FxC mice ( Supplementary  Fig. S8L-Q).

Wild-type rTg4510
CxF  Table S3). (B) The Mann-Whitney U test revealed significantly higher Mid1 mRNA levels in WT_CxF than in WT_FxC mice and in rTg4510_CxF as compared with rTg4510_FxC mice. The WT_CxF and WT_FxC groups contained five males and five females each (n = 10/group). The rTg4510_CxF and rTg4510_FxC groups contained six males and six females each (n = 12/group). Data are presented as the mean ± SEM. **p < 0.01, *p < 0.05 (Mann-Whitney U test).  (Fig. 6A). Western blot analysis of the TBSsoluble fraction showed no significant differences in the levels of HT7, AT180, and AT270 at 55 kDa between cells treated with negative control siRNA and MID1 siRNA (Fig. 6B-E). MID1-knockdown cells showed a significant increase in AT180 levels at 64 kDa (p = 0.0206, Student's t-test) and slight, but not significant, increases in the levels of HT7, AT8 and AT270 at 64 kDa ( Fig. 6F-I). Western blot analysis of the sarkosyl-insoluble fractions of MID1-knockdown cells was also performed; however, no bands were detectable with the use of the HT7, AT8, AT180, and AT270 antibodies (data not shown). We further analyzed the ratios of phosphorylated tau to total tau; however, there were no significant differences in these ratios between cells treated with negative control siRNA and MID1 siRNA ( Supplementary Fig. S10).
No relationship of the number of siblings, body weight, and brain weight with differential tau pathology. We finally investigated whether the genetic background resulted in changes to the number of siblings, body weight, and brain weight ( Table 2). Two-way ANOVA revealed no significant effect of the genetic  Based on these results, the number of siblings, body weight, and brain weight appear to have no effect differential tau accumulation in rTg4510 mice.

Discussion
We investigated tau pathology in reciprocal F1 hybrids of rTg4510 mice at 3 and 6 months of age. The results showed more severe pathological tau accumulation and phosphorylation in rTg4510_FxC mice than rTg4510_ CxF mice with no obvious sex difference at 6 month of age. In contrast, there were no obvious effects of the genetic background or sex on pathological alterations of tau in rTg4510 mice at 3 months of age. There was, however, a significant difference in Mid1 mRNA levels between the reciprocal F1 hybrids, and MID1 silencing with siRNA induced pathological phosphorylation of tau in HEK293T cells stably expressing human tau with the P301L mutation. These findings suggest that downregulation of Mid1 contributes to severe pathological alterations of tau in rTg4510_FxC mice, although the effect might be somewhat weak and will require a considerable amount of time to be apparent in the in vivo brain. The Mid1 gene is located on the X chromosome and encodes a 667-amino acid protein that belongs to the tripartite motif family and consists of a RING finger domain, two B-box domains, a coiled-coil region, a fibronectin type III domain, and B30.2 domains. The Mid1 protein has been shown to function as an E3 ubiquitin ligase targeting the microtubule-associated catalytic subunit of protein phosphatase 2A (PP2Ac) in ubiquitin-mediated degradation 22 . The Mid1 protein is, however, involved in signaling of mammalian target of rapamycin complex 1 through the regulation of PP2A activity 23 . PP2A is a phosphatase that targets tau and mTOR signaling is known to regulate tau phosphorylation and degradation 24 . Considering the previous studies cited above, Mid1 upregulation would be expected to increase tau phosphorylation and accumulation through the inhibition of PP2A degradation and the activation of mTOR, which contradicts the results of the present study. Therefore, it is of interest to investigate the difference in the level/activity of PP2A between reciprocal F1 hybrids of rTg4510 mice in future studies. However, it is difficult to predict the role of Mid1 in the accumulation of tau pathology, since it has been reported to serve other functions, such as the regulation of mRNA translation by direct association with the mRNA 25 . Furthermore, mutations to the Mid1 gene have been identified in patients with X-linked Opitz G/BBB syndrome, an inherited multiple-organ disorder primarily affecting midline structures 26,27 , although the pathological mechanisms remain unknown. Taken together, although the role of Mid1 in pathological alterations of tau in the in vivo brain remains to be elucidated, the results of the present study provide novel insight into the mechanism underlying the pathological alterations of tau in tauopathy.
Although the mechanisms underlying the difference in Mid1 levels between mice with the (C57BL/6 J × FVB/ NJ)F1 and (FVB/NJ × C57BL/6 J)F1 backgrounds are currently unclear, we expect that epigenetic mechanisms are involved, as it has been proposed that maternal care and environmental factors (e.g., diet, stress, etc.) may influence epigenetic changes in offspring 28,29 . Mice on the (C57BL/6 J × FVB/NJ)F1 background are delivered from mothers on a C57BL/6 J background, while mice on the (FVB/NJ × C57BL/6 J)F1 background are delivered from mothers on a FVB/NJ background. Therefore, maternal strain-dependent epigenetic changes may occur in offspring between the prenatal period and weaning; consequently, expression levels of genes, including Mid1, would differ depending on the maternal strain.
Mitochondrial DNA (mtDNA) has also been suspected in the differential accumulation of tau, because of its maternal inheritance. A previous study reported variations in mtDNA between strains C57BL/6 J and FVB/ NJ 30,31 . Specifically, the authors reported a G to T substitution at position 7778, which results in an amino acid substitution (aspartic acid to tyrosine) in adenosine triphosphate synthase 8. In addition, a T to C substitution at position 9461 resulted in no amino acid change (methionine to methionine) in NADH dehydrogenase 3. Furthermore, polymorphic variants contribute to some phenotypic differences in mice 32,33 . Mitochondria are the powerhouses of the cell, generating cellular energy in the form of ATP through oxidative phosphorylation (OXPHOS) 34 . Mammalian mtDNA encodes 13 polypeptides that are subunits of OXPHOS complexes, 22 transfer RNAs, and two ribosomal RNAs that are necessary for the translation of 13 polypeptides. Hence, mtDNA mutations affect energy production via the OXPHOS system [34][35][36] . Since maintaining electrochemical gradients and releasing and recycling synaptic vesicles are highly energy-demanding processes, neurons rely almost exclusively on the mitochondrial OXPHOS system to fulfill their energy needs 37,38 , which suggests that the OXPHOS system plays an important role in neural function. A recent study by Kimura et al. 39 reported that low-frequency stimulation induces the accumulation of sarkosyl-insoluble tau aggregates in the hippocampus of aged WT mice, suggesting an important contribution of synaptic impairment to the formation of tau pathology. Taken together, mtDNA variants possibly cause disturbances in neural function, such as synaptic impairment, which may result in enhanced formation of tau pathology. Furthermore, mitochondrial oxidative stress caused by sod2 knockout was reported to induce hyperphosphorylation of tau in a Tg2576 mouse model of Alzheimer's disease 40 . Since the mitochondrial OXPHOS system also generates reactive oxygen species, such as superoxide anion radicals, mitochondria have antioxidant defense systems to maintain redox homeostasis in the cell. However, according to the findings of previous studies, an imbalance in redox homeostasis would contribute to the accumulation of www.nature.com/scientificreports/ tau pathology. Based on these pieces of evidence, distinct mtDNA variants in rTg4510 mice are considered to potentially explain mechanisms underlying phenotypic differences, such as variations in levels of tau pathology. It is difficult to explain the mechanisms underlying gene expression patterns in rTg4510 mice. Fgf14 levels were expected to be reduced in rTg4510 mice due to transgene integrations. However, the results showed that Fgf14 levels in rTg4510 mice were much higher than in WT mice, which is in agreement with previous findings 41 . Dysregulation of Fgf14 mRNA splice variants has been suggested to contribute to the increase in Fgf14 levels 41 ; however, the detailed mechanism remains unknown. Considering the findings of Fgf14 expression in rTg4510 mice, it is probable that unknown mechanisms may induce the expression of other genes, such as Vipr2, Wdr60, Esyt2, D430020J02Rik, and Ncapg2, although the expression levels of these genes are expected to be decreased due to transgene integrations. In addition, the result showing that Hspa1b expression in rTg4510_FxC mice was significantly lower than in rTg4510_CxF mice suggests that the downregulation of Hspa1b may contribute to severe pathological accumulation of tau in rTg4510_FxC mice, because Hspa1b (HSP70) has been reported to inhibit and degrade tau aggregation 42,43 .
The results of the present study raise concerns about the use of rTg4510 mice as models for the evaluation of therapeutic interventions of tau pathology. No study, except for our previous work 19 , has investigated tau pathology levels in reciprocal F1 hybrids of rTg4510 mice. We used Camk2α-tTA mice on a C57BL/6 J background, rather than the original 129S6 background 11,12 . Accordingly, differences in tau pathology levels may only occur in Camk2α-tTA mice on a C57BL/6 J background. In contrast, sex had no effect on pathological tau accumulation in the present study, while there was a sex difference in pathological tau accumulation in the original rTg4510 mice, which was more severe in females than males 20 . Furthermore, it has been confirmed that there are differences in the effect of the Y chromosome, mtDNA, maternal care, and environmental factors between reciprocal hybrid offspring. Therefore, it is important to take the genetic background into consideration when evaluating tau pathology in hybrid rTg4510 mice.
The results of this study revealed a strong effect of the genetic background on the differential accumulation of tau pathology between reciprocal F1 hybrids of rTg4510 mice, while there was no evident significant effect of sex. Comparisons between the two hybrids indicate that Mid1 may be associated with the differential accumulation of tau pathology. Hence, elucidation of the mechanisms underlying differences in tau pathology levels would provide novel insights into the accumulation of tau pathology in tauopathy. Furthermore, such findings would be beneficial to researchers developing therapeutic interventions for tauopathies.

Animals.
All animal experiments complied with the ARRIVE guidelines, were carried out in accordance with the National Institutes of Health guide for the care and use of Laboratory animals (NIH Publications No. 8023, revised 1978), and were approved by the Animal Care and Use Committee at the Shiga University of Medical Science (approval no. 2018-6-5). All efforts were made to minimize animal suffering.
To obtain rTg4510 mice, the hemizygous Camk2α-tTA mouse line (on the C57BL/6 J background; 007,004; The Jackson Laboratory, Bar Harbor, ME, USA), carrying a tTA that consisted of the tet-off open reading frame placed downstream of Camk2α promoter elements, was crossbred with the hemizygous tetO-MAPT*P301L mouse line (on the FVB/NJ background; 015,815; The Jackson Laboratory) carrying human 0N4R tau cDNA with the P301L mutation placed downstream of a tetracycline-operon-responsive element. Consequently, mice harboring both the activator and responder genes were obtained and used as rTg4510 mice in the present study. The Camk2α-tTA mouse line on the C57BL/6 J background, rather than the original 129S6 background, was used in this study. Therefore, the mice obtained in this study had either the (C57BL/6 J × FVB/NJ)F1 or (FVB/ NJ × C57BL/6 J)F1 background (Table 1). Littermates expressing neither the activator nor the responder were used as WTs. Two to four mice were housed per standard laboratory cage on wood shavings and fed a standard chow diet and maintained at 23ºC under a 12-h light/dark cycle (lights on at 08:00-20:00 h) with free access to water and food in a specific pathogen-free animal facility.
Preparation of brain extracts. Mice were sacrificed at 3 and 6 months of age with an overdose injection of sodium pentobarbital (200 mg/kg, i.p.). Brains were quickly removed and cut into two hemispheres. Cerebral cortex tissues were isolated from the right hemisphere, snap frozen in liquid nitrogen, and stored at − 80 °C. Sarkosyl-insoluble tau was purified as described previously 44 with slight modifications. Briefly, brain tissues were homogenized in 10 volumes of Tris-buffered saline [TBS; 25 mM Tris-HCl (pH 7.5), 150 mM NaCl, 1 mM EGTA, protease inhibitor cocktail (Roche Diagnostics, Mannheim, Germany) and phosphatase inhibitor cocktail (Roche Diagnostics)] and centrifuged in a TLA110 rotor (Beckman Coulter, Brea, CA, USA) at 27,433 × g for 20 min at 4 °C. The supernatant was collected as the TBS-soluble fraction. The pellet was resuspended in five volumes of high salt/sucrose buffer [10 mM Tris-HCl (pH 7.5), 0.8 M NaCl, 10% sucrose, and 1 mM EGTA]. The sample was then centrifuged at 27,433 × g for 20 min at 4 °C. The supernatant was transferred to a new tube, incubated in 1% sarkosyl at 37˚C for 1 h, and centrifuged at 149,361 × g for 1 h at 4 °C. The pellet was resuspended in 50 µL of TBS, sonicated, and collected as the sarkosyl-insoluble fraction.
Immunohistochemical analysis. The brain hemisphere was fixed in 4% paraformaldehyde for 24 h at 4 °C, immersed in 0.1 M phosphate buffer (pH 7.4) containing 15% sucrose and 0.1% sodium azide for at least 2 days for cryoprotection, and subsequently cut into 20-µm coronal sections with the use of a cryostat. Immunohistochemical analysis was performed as described in previous studies 19,45 . Briefly, free-floating sections were incubated with biotinylated antibodies against phosphorylated tau (clone AT8; 1:1000; Thermo Fisher Scientific) and tau (clone HT7; 1:2000; Thermo Fisher Scientific) in PBS-T containing 0.2% bovine serum albumin overnight at 4 °C. The antibodies were detected using a Vectastain ABC Elite kit (1:1000; Vector Laboratories, Burlingame, CA, USA) with 3,3′-diaminobenzidine (Dojindo Laboratories, Kumamoto, Japan) with nickel ammonium sulfate. The sections were scanned with a camera and the immunoreactive areas were quantified using ImageJ software.
Simplified Gallyas silver staining. Simplified Gallyas staining was performed as described in a previous study 46 with some modifications 19,45 .

RNA isolation and quantitative polymerase chain reaction (qPCR).
Total RNA was extracted from the cerebral cortex using an RNeasy Plus Universal Mini Kit (Qiagen, Hilden, Germany) in accordance with the manufacturer's instructions. RNA samples were reverse transcribed into cDNA using SuperScript IV VILO Master Mix (Thermo Fisher Scientific) in accordance with the manufacturer's instructions. qPCR was conducted in duplicate 10-µL reactions containing 5 μl of THUNDERBIRD SYBR qPCR Mix (Toyobo, Osaka, Japan), 0.3 µM of each primer, and 0.5 ng of cDNA on a LightCycler 480 instrument (Roche Diagnostics) using the following cycling parameters: denaturation at 95 °C for 1 min, followed by 45 cycles of annealing at 95 °C for 15 s and extension at 60 °C for 30 s. Melting curve analysis confirmed that single amplicons were present and relative gene expression levels were determined using the ΔΔCT method with glyceraldehyde-3-phosphate dehydrogenase as the reference gene. Primer sequences are listed in Supplementary Table S5.
Plasmid construction and stable cell line development. Full-length human tau (0N4R) was cloned into the pcDNA3.1 vector acquired from human neuroblastoma cell line SH-SY5Y cDNA. A site-directed mutation at P301L was introduced in the construct using the QuikChange Lightning Multi Site-Directed Mutagenesis Kit (Agilent Technologies, Santa Clara, CA, USA). The sequence of human tau with the P301L mutation was then subcloned into the pIRESpuro3 vector (Clontech Laboratories, Mountain View, CA, USA). HEK293T cells were cultured in Dulbecco's modified Eagle's medium (Nacalai Tesque) supplemented with 10% (v/v) fetal bovine serum (Hyclone, Logan UT, USA), and kept at 37 °C under a humidified atmosphere of 5% CO 2 /95% air. Cells were seeded at 1 × 10 5 cells/well in a 24-well plate. At 24 h after plating, the cells were transfected with the pIRESpuro3 vector containing human tau with the P301L mutation using Lipofectamine 3000 reagent (Thermo Fisher Scientific). After a 24-h incubation period, the cells were transferred to a 6-cm culture dish, and stable cell lines were selected with 1 µg/mL puromycin (Nacalai Tesque) for 12 days. Single cells from resistant colonies were transferred into the wells of 96-well plates.
MID1 silencing with siRNA. HEK293T cells were cultured in Dulbecco's modified Eagle's medium (Nacalai Tesque) supplemented with 10% (v/v) fetal bovine serum (Hyclone, Logan UT, USA), and kept at 37 °C under a humidified atmosphere of 5% CO 2 /95% air. Cells were seeded at 1 × 10 5 cells/well in a 24-well plate. At 24 h after plating, the cells were transfected with the pIRESpuro3 vector containing human tau with the P301L mutation using Lipofectamine 3000 reagent (Thermo Fisher Scientific). After a 24-h incubation period, the cells were transferred to a 6-cm culture dish, and stable cell lines were selected with 1 µg/mL puromycin (Nacalai Tesque) for 12 days. Single cells from resistant colonies were transferred into the wells of 96-well plates.
Statistical analysis. Statistical analyses were performed using GraphPad Prism 7 software (GraphPad Software, La Jolla, CA, USA). Data are presented as the mean ± standard error of the mean (SEM). Two-way ANOVA with genetic background and sex as factors was conducted. If the results of the two-way ANOVA revealed a main effect of the genetic background or sex, statistical significance was subsequently determined with the Mann-Whitney U test for the corresponding single comparison. The Student's t-test was used to identify significant