A novel Huntington's disease mouse model to assess the role of neuroinflammation on disease progression and to develop human cell therapies

Abstract Huntington's disease (HD) is a fatal autosomal‐dominant neurodegenerative disease caused by a trinucleotide CAG repeat expansion of the huntingtin gene (HTT) that affects 1 in every 10 000 individuals in the United States. Our lab developed a novel immune deficient HD mouse strain, the YACNSG, from a commonly used line, the YAC128 mouse, to enable transplantation studies using engineered human cells in addition to studying the impact of the immune system on disease progression. The primary goal of this project was to characterize this novel immune deQficient HD mouse model, using behavioral assays and histology to compare this new model to the immune competent YAC128 and immune deficient mice that had engraftment of a human immune system. Flow cytometry was used to confirm that the YACNSG strain lacked immune cells, and in vivo imaging was used to assess human mesenchymal stem/stromal cell (MSC) retention compared with a commonly used immune deficient line, the NSG mouse. We found that YACNSG were able to retain human MSCs longer than the immune competent YAC128 mice. We performed behavioral assessments starting at 4 months of age and continued testing monthly until 12 months on the accelerod and in the open field. At 12 months, brains were isolated and evaluated using immunohistochemistry for striatal volume. Results from these studies suggest that the novel immune deficient YACNSG strain of mice could provide a good model for human stem‐cell based therapies and that the immune system appears to play an important role in the pathology of HD.

this project was to characterize this novel immune deQficient HD mouse model, using behavioral assays and histology to compare this new model to the immune competent YAC128 and immune deficient mice that had engraftment of a human immune system.
Flow cytometry was used to confirm that the YACNSG strain lacked immune cells, and in vivo imaging was used to assess human mesenchymal stem/stromal cell (MSC) retention compared with a commonly used immune deficient line, the NSG mouse. We found that YACNSG were able to retain human MSCs longer than the immune competent YAC128 mice. We performed behavioral assessments starting at 4 months of age and continued testing monthly until 12 months on the accelerod and in the open field. At 12 months, brains were isolated and evaluated using immunohistochemistry for striatal volume. Results from these studies suggest that the novel immune deficient YACNSG strain of mice could provide a good model for human stem-cell based therapies and that the immune system appears to play an important role in the pathology of HD.

| INTRODUCTION
Huntington's disease (HD) is a fatal autosomal-dominant neurodegenerative disease that currently has no cure. It is characterized by the progressive onset of chorea, behavioral and psychiatric changes, and cognitive decline. 1,2 The loss of efferent medium spiny neurons (MSN) in the striatum, located in the subcortical basal ganglia of the forebrain, results in a measurable decline in striatal volume, followed by general whole brain atrophy in HD patients. 3 HD is caused by a trinucleotide CAG repeat expansion in exon 1 of the huntingtin (HTT) gene, located on chromosome 4. 4 This leads to a mutated huntingtin (mHTT) protein which has an elongated stretch of glutamine at the Nterminus, resulting in a misfolded protein. 4 The misfolded HTT protein can fold and stick together in clumped, rigid aggregates. Once aggregates form, they tend to accumulate in inclusion bodies where they are sequestered. 3,5 Evidence suggests that aggregates interfere with normal cellular function, including axonal transport between the cell body and the synaptic terminal, 6 leading to disease progression.
Neuroinflammation was originally used to describe events such as ischemic stroke, traumatic brain injury, multiple sclerosis, or viral/bacterial infections that allowed infiltration of peripheral immune cells into the brain. Currently, the study of neuroinflammation has expanded to include neurodegenerative diseases such as Huntington's disease that do not attract inflammatory cells from the periphery but are characterized by cellular and molecular features of inflammation such as alterations of cytokine expression and microglia activation. 7 In postmortem HD brains, accumulation of reactive microglia and astrocytes has been observed in addition to PET/MR imaging, suggesting that neuroinflammation begins to accumulate in the basal ganglia of HD patients and could serve as a valuable biomarker and target for clinical interventions. 7,8 The role of neuroinflammation in HD is still a matter of debate. Neuroinflammation may play an important role early in disease progression as a mechanism to promote clearance of cell debris and protein aggregates. 9 However, as the disease progresses neuroinflammatory mechanisms may play a role in neuronal death which in turn would activate further neuroinflammation resulting in a vicious cycle of neurodegeneration and inflammation. 7,9 An important aspect of HD research is the development and use of transgenic mouse models that express the human mHTT protein.
The HD transgenic mouse model termed YAC128 has been characterized as an accurate replication of adult onset HD. It recapitulates the slow decline in motor and behavioral function and the progressive striatal loss and neuropathology that are seen in human HD. 10 The mutation in the YAC128 mice was created by microinjection into pronuclei. 10 The YAC128 mice have a human HTT gene with 128 CAG repeats on a yeast artificial chromosome, which is an autonomous chromosome. 10 The YAC128 mouse line expresses the full length human mutated HTT gene as well as the promoter, introns, upstream, and downstream regulatory elements. 10 YAC128 mice begin displaying HD symptoms at 3 months of age and are considered presymptomatic. 10 At this stage they display a hypokinetic phenotype, followed by motor deficits on the rotarod by 6 months of age and progression to hypokinesis by 12 months. 10 The symptoms indicate loss of muscle movement over time, similar to that which is seen in human HD patients. These behavioral changes correlate to neuronal loss.
Striatal atrophy is evident at 9 months of age with cortical atrophy by 12 months of age. 10 The YAC128 strain is an excellent model for HD. However, this mouse model must be immune suppressed in order to study human cell-based therapies. Past studies with the YAC128 mouse model have used immune-suppression drugs in order to reduce the immune response in mice, allowing human cells to survive longer in vivo to test potential therapies. A previous study from our lab showed that human mesenchymal stem cells (hMSC) could only survive in vivo for 7 days without immune-suppression drugs, but up to 28 days with them in the YAC128 strain. 11 hMSC have gained much interest in the field for their ability to modulate inflammation with several clinical trials aimed at autoimmunity. 12,13 However, these immune-suppression drugs have toxic effects on the mice over time [14][15][16][17] and do not result in a completely immune deficient animal, making the drugs inefficient for studying human stem cell based therapies for HD.
Here, we generated and characterized an immune deficient and xeno-tolerant mouse model of Huntington's disease with the human mutated HTT gene by cross-breeding the NOD.Cg-Prkdc scid Il2rγ tm1Wgl / SzJ (NSG) mouse, 18,19 a well-known immune deficient model, with YAC128 mice. This resulted in a novel immune deficient HD mouse model, referred to as YACNSG. In addition, we examined the role of the human immune system in contributing to the pathogenesis of HD by humanizing the YACNSG mice. It is hypothesized that the accumulation of activated microglia in response to mHTT protein aggregates plays a role in disease progression. The development of these mouse strains allows for the study of human stem cell products without the need for immunosuppression, as well as the study of the role of the immune system in the progression of disease in murine models of Huntington's disease.

| METHODS
All mouse studies were performed under UC Davis IACUC approval.
Umbilical cords designated for research were acquired from the

Significance statement
The present manuscript describes a novel, immunodeficient mouse model of Huntington's disease. This mouse strain was extensively characterized and is valuable to the research community due to the ability of not rejecting human cell products. The study also describes mice that had their immune system reconstituted with a human immune system. These studies further delineate the role of the immune system in neurodegenerative diseases.
California Umbilical Cord Blood Collection Program under IRB approval.

| Strain development
To generate the YAC128/NSG strain we performed 15 backcrosses to introduce the full length human HD transgene of the FVB-Tg (YAC128)53Hay/J (YAC128) mouse model of HD onto the xenotolerant NOD.Cg-PrkdcscidIl2rγtm1Wgl/SzJ (NSG).

| DNA/RNA/protein analysis of YACNSG strain
A 4 mm section of brain tissue was isolated for DNA/RNA analysis.
Genotyping was performed as previously described using published primers. 20 Genotyping primers were designed from the Jackson Laboratory genotyping protocol. Primers used were as follows: Transgene forward primer CCG CTC AGG TTC TGC TTT TA, transgene reverse primer TGG AAG GAC TTG AGG GAC TC. Western blot was performed using MAB2166 (Millipore).

| Flow cytometry of YACNSG strain
To determine the presence of immune cells in the YACNSG strain, red-blood-cell-lysed peripheral blood, spleen, and bone marrow were

| MSC isolation, transduction, and culture
Whole bone marrow was purchased commercially (Lonza, Walkersville, Maryland) and mesenchymal stem cells (MSCs) were isolated using a Ficoll-Paque density gradient as previously described. 2  Forty-eight hours prior to implantation, cells were moved to a 5% CO 2 /1% O 2 incubator for hypoxic preconditioning. 21,22 On the day of injection, the cells were lifted by trypsinization (HyClone) and were resuspended in Normosol-R (Hospira, Inc, Lake Forest, Illinois) at a final concentration of 50 000 cells/mL.

| MSC transplantation for IVIS imaging study
Animals used in this study were generated from the breeding colony at the UC Davis Institute for Regenerative Cures (Sacramento, California).
Prior to surgery, mice were anesthetized using isoflurane (2-3% in oxygen) and placed in a stereotaxic frame. The hair was removed using Nair and skin was cleaned three times with alternating betadine solution and alcohol. A small incision (<1 cm) was made on the scalp and a 1 mm burr hole was drilled into the skull at +0.5 mm AP, +1.7 mm ML relative to bregma.
Cells were injected into the striatum at a depth of −2.5 mm using a Hamilton syringe and automated microsyringe pump. The cells were injected at a rate of 0.5 μL/min with a total volume of 5 μL. Five minutes post injection, the needle was withdrawn and the incision sutured using 6-0 silk suture. Carprofen was administered by subcutaneous injection (5 mg/kg) at the time of surgery and again the following day.

| Humanization of YACNSG mice
Humanization of YACNSG mice followed established protocols. [23][24][25] Briefly, human CD34+ hematopoietic stem and progenitor cells (HSPC) were isolated from umbilical cord blood obtained from the UC Davis Umbilical Cord Collection Program by Ficoll-Paque (GE Healthcare, Logan, Utah) density gradient. They were further purified by CD34 magnetic bead column separation (Miltenyi Biotec, Auburn, California). Mice were sublethally irradiated with 125 rads 2-5 days after birth. Irradiated mice were intrahepatically injected with 2-5 × 10 5 purified CD34+ cells with up to 30 μL of volume using an insulin syringe. Twelve weeks postimplantation, blood was collected from the tail vein to test for human cell engraftment. Flow cytometry was conducted to identify the level of engraftment using a PE-CY7-conjugated anti-human CD45 antibody (BD Biosciences, Clone-HI30), a pan-leukocyte marker. Flow cytometry was performed using a Beckman Coulter FC-500. Transgenic and wildtype mice were chosen for the behavior study by matching engraftment levels. Ten male and 13 female transgenic, along with 10 male and 15 female wildtype mice, were used for behavioral testing.

| Functional assessment
For this study, mice were used from an established breeding colony at the UC Davis Institute for Regenerative Cures. YACNSG and YACNSG-humanized mice were tested once a month with YAC128 transgenic and wildtype littermate controls beginning at 4 months of age. All functional and histological assessments were performed blinded to both the genotype and engraftment status. At the conclusion of all data collection the experimenters were unblinded to perform data analysis.

| Rotarod
Mice were initially trained on the rotarod for 3 days; three trials/ day at a fixed speed of 5 rpm for 120 seconds. Mice that fell off the rod were placed back on for the duration of each trial. The final day of training was completed 3 days prior to the first testing day. Mice were tested on the accelerating rotarod (4-40 rpm over 360 seconds) three trials each testing day with a resting period between each trial. Total amount of time spent on the rod was recorded by Rotamex system software (Columbus Instruments).

| Histology
At the conclusion of behavior testing, mice were euthanized by CO 2 asphyxiation followed by bilateral thoracotomy. Mice were perfused with 10 mL PBS (HyClone) followed by 10 mL formalin (Fisher Healthcare, Pittsburg, Pennsylvania). Brains were harvested, fixed in formalin for 24 hours, then transferred to 30% sucrose (Fisher Chemical, Fair Lawn, New Jersey) for 24-48 hours at 4 C. Brains were then frozen for 3 minutes in a dry ice/100% isopropanol bath at −77 C and then stored at −80 C until further processing.

| Statistical analysis
SPSS v26 was used to perform all statistical analyses. Behavioral data was analyzed using a repeated measures analysis of variance (ANOVA) to measure changes between genotypes across timepoints.
When appropriate, an LSD post hoc analysis was completed. Oneway ANOVA was performed at the 12 month timepoint for striatal atrophy followed by an LSD post hoc analysis when appropriate.

| Flow cytometry
Mice were euthanized CO 2 and blood, spleen, and bone marrow were isolated. Blood (approximately 1 mL) was collected from the body cavity of the mouse using vacuum capillary tubes and incubated with 1X RBC lysis buffer (eBioscience) for 10 minutes. Following lysis, the sample was centrifuged at 3000 rpm for 3 minutes. The supernatant was removed, and the remaining pellet was washed with 1 mL PBS.
Resuspended cell pellets were divided evenly between several tubes and centrifuged at 3000 rpm for 3 minutes. Supernatant was removed and the sample was resuspended in 100 μL PBS and placed on ice until ready to label with antibodies.
The spleen was removed from the animal and placed onto a To examine the impact of the immune system on the progression of HD and to extensively characterize the new strain with the parental YAC128 mouse, a longitudinal functional assessment was performed.
A group of YACNSG were sub lethally irradiated at postnatal day 2-5 and intrahepatically transplanted with human CD34+ derived from umbilical cord blood to provide the immune deficient mice with a human immune system. 25 YACNSG were considered "humanized" if F I G U R E 4 A, Human cell engraftment in the YAC128/NSG mouse strain. Male and female YAC128/NSG transgenic and wildtype littermates were sublethally irradiated and intrahepatically transplanted with cord blood derived CD34+ cells. Twelve weeks posttransplantation blood was collected via the tail vein and analyzed for human specific CD45 using flow cytometry. Mice selected for behavioral analysis displayed between 14.8% and 65.7% engraftment. No significant differences were observed between groups for level of engraftment. B, Accelerating rotarod was performed monthly for all groups. An overall repeated measures ANOVA interaction was observed. Post hoc revealed significant main effects between YAC128 WT and Tg mice as well as overall significance between the nonhumanized YACNSG WT and Tg. C, Open field assessment was performed for 10 minutes one time per month. Overall repeated measures ANOVA interactions were observed for total distance, rest time, and vertical activity. A within-subject main effect was observed for stereotypic behavior and a between-group effect was observed for time in the center quadrant. An analysis of effect size for the accelerod at 5-, 9-, and 11-month time points revealed moderate to large effect size for each cohort of mice they demonstrated successful engraftment of greater than 25% human CD45+ at 3 months of age as measured by flow cytometry ( Figure 4A). Functional assessment took place starting at 4 months of age and was performed monthly until the mice were 12  F I G U R E 5 A, Striatal volume was assessed from coronal sections as previously described using a Darpp32 antibody. Slides were imaged using an AxioScan and volume was calculated using Zen Blue. Striatal volume was normalized to the wildtype littermate and atrophy was calculated as 1 − (volume/average volume of WT strain). B, An overall effect was observed with a one-way ANOVA. Tukey's post hoc revealed a significant phenotype between the YAC128 WT and Tg; the YACNSG H WT and Tg, but not between the YACNSG nH WT and Tg. This data demonstrates the role of the immune cells in the neurodegenerative process of the YAC128 mouse model significant to all YACNSG lines (P < .005). YACNSG WT nH significant to YACNSG WT H (0.025). YACNSG Tg nH significant to YACNSG Tg H (0.031) ( Figure 4C). The vertical activity time results suggest an impact of the NSG background as well as the humanization protocol.
While the novel YACNSG mice did not have phenotypic deficits at late stage of disease, trends toward significant differences were observed in early stages of disease progression.
At the conclusion of functional assessment, the mice were perfused and the brains were prepared for histology. Striatal volume was calculated following previously published protocols 26

| DISCUSSION
The present study describes a novel animal model of Huntington's disease that allows for the careful study of the role of the human immune system on disease progression, and allows for the development of human cellular products to be tested in a disease background without the concern of rejection due to a xenotransplantation immune response. A novel line of YAC128 transgenic mice on an immune deficient NSG background strain was successfully generated, the YACNSG. This mouse model demonstrated the presence of the mutant transcript and protein in the brain in conjunction with a lack of circulating T, B, and natural killer cells.
The novel YACNSG transgenic mouse strain allows for the development of human cell transplantation products in a disease context. [27][28][29][30][31][32][33] The results from this study suggest that adult human bone marrow derived mesenchymal stem (hMSC) cells engineered to express GFP and luciferase are retained longer than the YAC128 and similar to NSG mice. Our data demonstrates that cells are rapidly cleared from the CNS when using the parental YAC128 line without immune suppression as expected. Our previous work demonstrated that hMSC can survive for approximately 6 weeks when the mice are treated with a continuous administration via osmotic pumps of a combination of FK506 and rapamycin. While that approach does allow for the assessment of human cell efficacy over that time frame, the use of immune suppression is expensive, can result in abnormal functional efficacy, [34][35][36] and limits any longitudinal assessment to the timeframe the mice are receiving the drug, which is usually only 4 to 6 weeks due to logistics of the alzet pump. However, the YAC128 mouse strain presents a great model to assess novel therapeutic development due to the progressive onset of motor, cognitive and neuropathological deficits.
The novel YACNSG mouse strain allows for the careful assessment of the role of the immune system on disease progression. To assess the role of the immune system, a cohort of age-and sex-matched YACNSG mice were given a human immune system at postnatal day 2-5 following previously established protocols. [23][24][25] The humanized YACNSG were then longitudinally assessed alongside the parental YAC128 strain and nonhumanized YACNSG. Following assessment of the level of "humanization", functional assessment began at 4 months and was continued until the mice were 12 months old. The YACNSG mice were able to demonstrate reconstitution of human immune cells in circulation similar to previous reports. 37,38 Analysis of the accelerod, a measure of motor coordination, revealed that the YAC128 and nonhumanized YACNSG mice demonstrated a reproducible phenotypic difference, as expected. While the Humanized YACNSG did not reach an overall repeated measures significant effect, many individual timepoints reveal a robust motoric phenotype.
When assessing the effect size at three commonly used timepoints in the YAC128 (6-, 9-, and 11-months) a medium to large effect size was observed, suggesting that measurements of motor phenotypes in this strain serves as a valuable outcome measure. Open field was also used to assess spontaneous exploration and overall motoric function. Multiple measures in the open field are suggestive of overall phenotypes between the WT and Tg from each respective cohort with interesting findings demonstrating the role of the immune system on disease progression.
At the conclusion of functional assessments, the brains of the YAC128 or YACNSG mice were prepared for histology following previously published protocols. The YAC128 displayed a typical 30% striatal atrophy when compared with their WT littermates. Interestingly, the YACNSG nonhumanized mice, while still showing significant striatal atrophy when compared with their WT littermates, were much reduced compared with the parental strain. The humanized YACNSG mice also displayed significant striatal atrophy when compared with their WT littermates, and the level of atrophy more closely resembled the parental YAC128 strain than the nonhumanized YACNSG. This data is suggestive that, at least for neuropathology, the immune system plays a large role in the loss of medium spiny neurons and striatal atrophy. This data would be supportive of the preclinical and clinical development of immune targeting molecules, [39][40][41][42] and while the SIG-NAL and LEGATO-HD trials have mixed clinical outcomes, targeting the neuroinflammation associated with disease onset and progression may result in clinically meaningful outcomes.

| CONCLUSION
In summary, this study describes a novel mouse strain of Huntington's disease that allows for both the development of human cellular products for novel interventions and further describes the impact of the immune system on disease progression. Future studies will focus on characterizing the extent of humanization observed and the further study the role of neuroinflammation both in the presence and absence of cell or gene therapy.