Drag-and-drop genome insertion of large sequences without double-strand DNA cleavage using CRISPR-directed integrases

Programmable genome integration of large, diverse DNA cargo without DNA repair of exposed DNA double-strand breaks remains an unsolved challenge in genome editing. We present programmable addition via site-specific targeting elements (PASTE), which uses a CRISPR-Cas9 nickase fused to both a reverse transcriptase and serine integrase for targeted genomic recruitment and integration of desired payloads. We demonstrate integration of sequences as large as ~36 kilobases at multiple genomic loci across three human cell lines, primary T cells and non-dividing primary human hepatocytes. To augment PASTE, we discovered 25,614 serine integrases and cognate attachment sites from metagenomes and engineered orthologs with higher activity and shorter recognition sequences for efficient programmable integration. PASTE has editing efficiencies similar to or exceeding those of homology-directed repair and non-homologous end joining-based methods, with activity in non-dividing cells and in vivo with fewer detectable off-target events. PASTE expands the capabilities of genome editing by allowing large, multiplexed gene insertion without reliance on DNA repair pathways.

Identifying Human Specific Adverse Outcome Pathways of Per- and Polyfluoroalkyl Substances Using Liver-Chimeric Humanized Mice

Background

Per- and polyfluoroalkyl substances (PFAS) are persistent organic pollutants with myriad adverse effects. While perfluorooctanoic acid (PFOA) and perfluorooctane sulfonic acid (PFOS) are the most common contaminants, levels of replacement PFAS, such as perfluoro-2-methyl-3-oxahexanoic acid (GenX), are increasing. In rodents, PFOA, PFOS, and GenX have several adverse effects on the liver, including nonalcoholic fatty liver disease.

Objective

We aimed to determine human-relevant mechanisms of PFAS induced adverse hepatic effects using FRG liver-chimeric humanized mice with livers repopulated with functional human hepatocytes.

Methods

Male humanized mice were treated with 0.067 mg/L of PFOA, 0.145 mg/L of PFOS, or 1 mg/L of GenX in drinking water for 28 days. Liver and serum were collected for pathology and clinical chemistry, respectively. RNA-sequencing coupled with pathway analysis was used to determine molecular mechanisms.

Results

PFOS caused a significant decrease in total serum cholesterol and LDL/VLDL, whereas GenX caused a significant elevation in LDL/VLDL with no change in total cholesterol and HDL. PFOA had no significant changes in serum LDL/VLDL and total cholesterol. All three PFAS induced significant hepatocyte proliferation. RNA-sequencing with alignment to the human genome showed a total of 240, 162, and 619 differentially expressed genes after PFOA, PFOS, and GenX exposure, respectively. Upstream regulator analysis revealed inhibition of NR1D1, a transcriptional repressor important in circadian rhythm, as the major common molecular change in all PFAS treatments. PFAS treated mice had significant nuclear localization of NR1D1. In silico modeling showed PFOA, PFOS, and GenX potentially interact with the DNA-binding domain of NR1D1.

Discussion

These data implicate PFAS in circadian rhythm disruption via inhibition of NR1D1. These studies show that FRG humanized mice are a useful tool for studying the adverse outcome pathways of environmental pollutants on human hepatocytes in situ.

Behavioral and cognitive performance of humanized APOEε3/ε3 liver mice in relation to plasma apolipoprotein E levels

Plasma apolipoprotein E levels were previously associated with the risk of developing Alzheimer's disease (AD), levels of cerebrospinal fluid AD biomarkers, cognition and imaging brain measures. Outside the brain, the liver is the primary source of apoE and liver transplantation studies have demonstrated that liver-derived apoE does not cross the blood-brain-barrier. How hepatic apoE may be implicated in behavioral and cognitive performance is not clear. In the current study, we behaviorally tested FRGN mice with humanized liver harboring the ε3/ε3 genotype (E3-human liver (HL)) and compared their behavioral and cognitive performance with that of age-matched ε3/ε3 targeted replacement (E3-TR) mice, the latter produces human apoE3 throughout the body whereas the E3-HL mice endogenously produce human apoE3 only in the liver. We also compared the liver weights and plasma apoE levels, and assessed whether plasma apoE levels were correlated with behavioral or cognitive measures in both models. E3-HL were more active but performed cognitively worse than E3-TR mice. E3-HL mice moved more in the open field containing objects, showed higher activity levels in the Y maze, showed higher activity levels during the baseline period in the fear conditioning test than E3-TR mice, and swam faster than E3-TR mice during training to locate the visible platform in the water maze. However, E3-HL mice showed reduced spatial memory retention in the water maze and reduced fear learning and contextual and cued fear memory than E3-TR mice. Liver weights were greater in E3-HL than E3-TR mice and sex-dependent only in the latter model. Plasma apoE3 levels were similar to those found in humans and comparable in female and male E3-TR mice but higher in female E3-HL mice. Finally, we found correlations between plasma apoE levels and behavioral and cognitive measures which were predominantly model-dependent. Our study demonstrates mouse-model dependent associations between plasma apoE levels, behavior and cognition in an 'AD-neutral' setting and suggests that a humanized liver might be sufficient to induce mouse behavioral and cognitive phenotypes.

Brain integrity is altered by hepatic APOE ε4 in humanized-liver mice

Liver-generated plasma apolipoprotein E (apoE) does not enter the brain but nonetheless correlates with Alzheimer's disease (AD) risk and AD biomarker levels. Carriers of APOEε4, the strongest genetic AD risk factor, exhibit lower plasma apoE and altered brain integrity already at mid-life versus non-APOEε4 carriers. Whether altered plasma liver-derived apoE or specifically an APOEε4 liver phenotype promotes neurodegeneration is unknown. Here we investigated the brains of Fah-/-, Rag2-/-, Il2rg-/- mice on the Non-Obese Diabetic (NOD) background (FRGN) with humanized-livers of an AD risk-associated APOE ε4/ε4 versus an APOE ε2/ε3 genotype. Reduced endogenous mouse apoE levels in the brains of APOE ε4/ε4 liver mice were accompanied by various changes in markers of synaptic integrity, neuroinflammation and insulin signaling. Plasma apoE4 levels were associated with unfavorable changes in several of the assessed markers. These results propose a previously unexplored role of the liver in the APOEε4-associated risk of neurodegenerative disease.

Extensive double humanization of both liver and hematopoiesis in FRGN mice

Preclinical research in animals often fails to adequately predict the outcomes observed in human patients. Chimeric animals bearing individual human tissues have been developed to provide improved models of human-specific cellular processes. Mice transplanted with human hematopoietic stem cells can be used to study human immune responses, infections of blood cells and processes of hematopoiesis. Animals with humanized livers are useful for modeling hepatotropic infections as well as drug metabolism and hepatotoxicity. However, many pathophysiologic processes involve both the liver and the hematolymphoid system. Examples include hepatitis C/HIV co-infection, immune mediated liver diseases, liver injuries with inflammation such as steatohepatitis and alcoholic liver disease.

We developed a robust protocol enabling the concurrent double-humanization of mice with mature hepatocytes and human blood. Immune-deficient, fumarylacetoacetate hydrolase (Fah^−/−), Rag2^−/− and Il2rg^−/− deficient animals on the NOD-strain background (FRGN) were simultaneously co-transplanted with adult human hepatocytes and hematopoietic stem cells after busulfan and Ad:uPA pre-conditioning. Four months after transplantation the average human liver repopulation exceeded 80% and hematopoietic chimerism also was high (40–80% in bone marrow). Importantly, human macrophages (Kupffer cells) were present in the chimeric livers.

Double-chimeric FRGN mice will serve as a new model for disease processes that involve interactions between hepatocytes and hematolymphoid cells.