UC Irvine

Ongoing studies have demonstrated that a high proportion of neurodegenerative risk genes are highly expressed by microglia, the resident immune cell of the central nervous system (CNS). Polymorphisms within many of these risk genes have further been shown to alter microglial response to neuropathology, impacting disease onset and progression. These findings highlight the importance of microglia in maintaining neural health and the pressing need to develop approaches to therapeutically target dysfunctional microglia in neurodegenerative disease. However, achieving brain-specific uptake of large therapeutic molecules remains largely elusive, limiting effective concentrations within the brain and often promoting peripheral side-effects. More invasive techniques including direct injection of peptides or viral gene therapy have also been examined. However, these methods often require multiple treatments for long-term therapeutic efficacy, provide limited diffusion beyond the injection site, and inefficiently target resident microglia.

Progress in the development of immune cell therapies (ICTs) has begun to offer a promising approach to treating a variety of peripheral blood diseases including lymphoma, leukemia, and sickle cell disease. When coupled with advancements in gene editing techniques, they offer long-term therapeutic efficacy and potentially address many of the limitations of previously developed gene therapy approaches. Most recently, researchers have begun to propose the use of bone marrow-derived hematopoietic stem cell (HSC) transplantation as a source of peripherally derived macrophage populations to supplement or potentially replace diseased microglia. These populations have been shown to infiltrate and reside within the CNS following injury and degeneration and are often described as having adopted a ‘microglial’ or microglia-like’ identity. However, recent studies have shown that even after long-term CNS engraftment, these myeloid populations remain transcriptionally and functionally distinct from endogenous microglia. Furthermore, large scale infiltration and engraftment of peripherally derived macrophage populations requires toxic bone marrow preconditioning treatments, which have been shown to cause considerable patient mortality.

As the resident immune cell of the brain, microglia stand out as the ideal candidate for ICT in the brain. They are long-lived, self-renew without contribution from the periphery, and can be readily differentiated from patient-derived induced pluripotent stem cells (iPSC). Studies have shown that iPSC-derived microglia (iMG) transcriptionally resemble ex vivo human microglia and respond to both injury and neuropathology when transplanted into xenotolerant mice. For this reason, iMG have proven to be an effective preclinical tool for the investigation of human microglia function and their role in disease progression. However, studies have also shown that human iMG transplanted into an adult murine brain engraft minimally and remain relatively near the injection site. This is due to the well-regulated “microglial niche” that manages microglia numbers and hampers donor microglial engraftment.

The focus of this dissertation is to develop a microglia replacement strategy that allows for robust engraftment into an occupied “microglial niche” and to explore the therapeutic potential of iMG to prevent and/or reverse neurodegenerative disease. To this end, we engineered an inhibitor-resistant CSF1R that enables CNS-wide microglial replacement. CSF1R is necessary for microglia viability and is implicated in regulating microglia homeostasis and function. Therefore, inhibition of CSF1R has been shown to deplete microglia in vivo. We identified a single glycine to alanine substitution at position 795 of human CSF1R (G795A) that confers resistance to multiple CSF1R inhibitors (CSF1Ri), including PLX3397 and PLX5622, with no discernable gain or loss of function. Xenotransplantation studies show G795A-iMG exhibit nearly identical gene expression to wildtype iMG, respond to inflammatory stimuli, and progressively expand under constant CSF1Ri treatment, replacing endogenous microglia to fully occupy the brain. Importantly, G795A-iMG remain in newly engrafted regions of the brain and return to a homeostatic state one month after cessation of CSF1Ri treatment. These findings, discussed in detail within Chapter 1, demonstrate a novel platform and robust approach to enable the investigation and future development of microglia replacement therapies with iMG.

Transplantation of microglia could have therapeutic implications for a variety of neurological diseases. However, primary microgliopathies, a group of rare diseases that specifically result from microglial dysfunction, represent the most logical indication in which to first develop and test this novel approach. One example of a primary microgliopathies is Adult-onset leukoencephalopathy with axonal spheroids and pigmented glia (ALSP). ALSP is a rare autosomal dominant neurodegenerative disease caused by mutations in CSF1R, a receptor that is critical for the development and survival of microglia. Thus far, the therapeutic potential of microglia replacement to treat ALSP has been limited by the unavailability of mouse models that recapitulate the diverse neuropathologies and reduced microglia numbers observed in patients. We therefore generated and examined a humanized xenotolerant mouse model lacking a conserved enhancer (fms-intronic regulatory element, FIRE) within the mouse Csf1r locus (hFIRE) that develops nearly all the hallmark pathologies including axonal spheroids, white matter abnormalities, reactive astrocytosis, and brain calcifications. Remarkably, transplantation of human iMG progenitors restores a homeostatic microglial gene signature in hFIRE mice and prevents the development of each of these ALSP-related neuropathologies. To further examine a potential autologous approach, we generated and CRISPR-corrected ALSP-patient-derived iPSCs. We found genetic correction of CSF1R rescues mutation-induced deficits in microglial proliferation and enables brain-wide microglial engraftment. Surprisingly, transplantation of CSF1R-corrected iMG reduces pre-existing spheroid, astrogliosis, and calcification pathologies within just 6 weeks. These results, which are discussed in detail within Chapter 2, provide compelling evidence that transplantation of iMG could offer a promising new therapeutic strategy for ALSP and perhaps other microglia-associated neurological disorders.

For the treatment of neurodegenerative diseases other than primary microgliopathies, replacing dysfunctional microglia with CRISPR-corrected microglia may not suffice to achieve therapeutically favorable outcomes. However, our ability to deploy genetically modified iMG in patients could permit the delivery of therapeutic peptides previously prevented by the functional qualities of the blood brain barrier. Therefore, as proof of principle, we CRISPR-engineered iMG to produce neprilysin, a well-characterized beta-amyloid degrading protease, under the control of the endogenous CD9 promoter for transplantation into a xenotolerant, amyloid-accumulating transgenic mouse model of Alzheimer’s Disease. We demonstrate CD9, a plaque-associated microglia marker, is capable of driving membrane-bound neprilysin (NEP) and secreted neprilysin (sNEP) specifically in response to plaque deposition without off-target degradation of homeostatic neuropeptides bradykinin and somatostatin. Biochemical analysis of the transplanted hippocampus and overlying cortex reveals both NEP and sNEP iMG reduce levels of soluble and insoluble amyloid proteins while only sNEP iMG prevents synaptic degradation and significantly reduces levels of astrogliosis in vivo. Importantly, no significant differences in these levels were observed by transplantation of human WT iMG.

To better understand if CNS-wide microglia replacement is necessary for a secreted payload like sNEP to achieve brain-wide therapeutic efficacy, sNEP-G795A iMG precursors were transplanted into a second cohort of mice treated with or without CSF1Ri. Remarkably, biochemical analysis of whole brain lysates reveals sNEP iMG without treatment not only significantly reduces amyloid protein levels, including Aβ oligomers, as effectively as CSF1Ri-treated mice, but also significantly prevents synaptic degradation, reduces levels of astrogliosis, lowers peripheral plasma NfL, and decreases levels of inflammatory markers in comparison to untreated mice. Taken together, these results, which are further discussed in Chapter 3, indicate iMG can be engineered ex vivo to produce and deliver therapeutics such as amyloid-targeting peptides in response to neuropathology and even achieve brain-wide therapeutically favorable outcomes with only partial microglial engraftment.

In conclusion, these studies provide compelling evidence that iPSC-derived microglia provide an optimal platform to deliver immune cell therapies to the central nervous system. We show that iMG can be deployed to replace dysfunctional microglia in vivo, prevent the development of ALSP-associated neuropathologies, reverse already formed neuropathologies, and be genetically modified to enable brain-wide delivery of therapeutic peptides. These advancements address many of the limitations currently faced by peripheral delivery of therapeutic peptides to the brain and offer a novel approach for the treatment of rare neurodegenerative diseases, like ALSP, which currently lack any FDA-approved treatment.