Much of the research in our laboratory involves the study of interactions between the nervous and immune systems. Using knockout mice and over-expression in vivo with viral vectors, we are exploring the role of the neuropoietic cytokine leukemia inhibitor factor (LIF) in regulating neural stem cell proliferation and fate in the adult brain. LIF is also being tested in a gene therapy for a mouse model of multiple sclerosis. In the context of neuroimmune interactions during fetal brain development, we are investigating a mouse model of a known risk factor for autism and schizophrenia, maternal infection. How the maternal inflammatory response alters fetal brain development and subsequent behavior of the offspring is being examined, and potential therapeutic interventions are being explored. Huntington's disease is another focus, where we are investigating potential therapies using intracellular expression of antibodies (intrabodies) and also manipulating IKKβ activity in mouse models. We are also studying the regulation of MeCP2 by IKKα, because MeCP2 mutations are responsible for Rett syndrome, which frequently involves autism symptoms.
Huntington’s disease (HD) is an autosomal dominant, fatal neurodegenerative disease for which there is currently no cure. The intrabody Happ1, which is an intracellularly expressed antibody that binds to the proline-rich domain of huntingtin (Htt) exon 1 (HDx-1), reduces the toxicity and aggregation of HDx-1 in cell culture as well as in vivo. Happ1 increases the turnover of soluble and insoluble mutant HDx-1 (mHDX-1), but not wild type HDx-1 (wt HDx-1) (Southwell et al., 2008). In vivo, Happ1 treatment restores the motor and cognitive performance in 4 different transgenic mouse models. It also prevents the striatal neuron loss in a mHDx-1 lentiviral model and normalizes ventricular size in transgenic mouse models. This intrabody also increases body weight and survival of N171-82Q HD mice (Southwell et al., 2009). Therefore, the highly beneficial effects of Happ1 in all 5 mouse models tested indicates that it is a candidate therapeutic for HD.
Although Happ1 is highly effective in vivo, it cannot completely prevent HD symptoms in mice. If the affinity of Happ1 could be improved further, its efficacy would likely increase as well. Therefore, we are using yeast surface display technology to affinity mature Happ1. Resulting high affinity intrabodies would then be tested in cultured cells for their effects on mHDx-1 levels and their ability to ameliorate HD-associated pathology at a lower intrabody/HDx-1 ratio.
Proteolysis of huntingtin (Htt) plays a key role in the pathogenesis of Huntington’s disease (HD). The N-terminal fragments generated from mutant Htt have the propensity to form oligomers and amyloid structures that are neurotoxic. We have identified IKKβ as a regulator of Htt cleavage and neurotoxicity, as inhibitors of IKKβ prevent Htt cleavage and are able to reduce neurotoxicity of N-terminal Htt fragments. IKKβ may also regulate neuroinflammation in HD. Presymptomatic HD patients have elevated levels of proinflammatroy cytokines in their sera and CNS years before the onset of motor symptoms, and IKKβ is a prominent kinase that regulates cytokine production. Thus, IKKβ is likely to play a prominent role in HD pathogenesis. We are examining the role of IKKβ in HD pathogenesis in vivo. Towards this aim, we have generated an HD mouse model that lacks IKKβ in the CNS. These mice are being examined for neuropathology, motor deficits and expression of inflammatory markers in the CNS. We will also examine the effects of small molecule inhibitors of IKKβ on HD pathology in vivo. These studies should determine whether IKKβ is a therapeutic target for HD.
IKKα is a serine/threonine kinase and a component of the IB kinase complex (IKKα, IKKβ, and IKKγ), which regulates the NF-B pathway. IKKα is also a chromatin-modifying enzyme. In cells stimulated by cytokines, IKKα enters the nucleus and is able to phosphorylate several substrates including histone-3, CREB binding protein (CBP), and the silencing mediator of retinoid and thyroid hormone receptor (SMRT). This leads to the assembly of an active chromatin configuration. Although IKKα is expressed in the CNS, its functions are not well understood. We are using human neural stem cells to examine whether changes in the homeostasis of the IKKα affects neuronal differentiation and lineage commitment.
Homeostasis of methyl-CpG binding protein 2 (MeCP2) is critical for neuronal function and development. Mutations in the coding region of MeCP2 are responsible for most cases of Rett syndrome. The levels of MeCP2 are also reduced in the brains of a subset of idiopathic autism patients. MeCP2 is expressed at high levels in mature neurons and is suggested to play a role in dendritic growth and synaptic plasticity. MeCP2 is known to bind to methylated CpG islands in the promoters of many genes, and may influence the expression of up to 2,200 genes, including BDNF and CREB (Chahrour et al., 2008, Science, 320: 1224-9). The context and environmental signals that regulate the transcriptional activity of MeCP2 are poorly understood. Phosphorylation of MeCP2 at several serine residues is important for its biological function. Activity-induced phosphorylation of MeCP2 at Ser 421 by calcium/calmodulin-dependent protein kinase II (CaMK-II) in neurons induces expression from several promoters including BDNF exon-IV, enhances dendritic branching, and regulates complex behaviors such as seizures and circadian rhythms in animal models (Zhou et al., 2006, Neuron, 52:255-69). We have identified IKKα as a potential modulator of MeCP2 expression and activity in neurons. We are investigating whether IKKα/MeCP2 interaction regulates the expression of neuronal genes and affects neurodevelopment.
Perturbation of immune homeostasis is linked to several neuropsychiatric disorders. Inflammatory mediators have access to the CNS and influence brain development as well as function by various mechanisms. Cytokines such as IL-6, TNF-α, and IL-1 influence neurogenesis and synaptic plasticity (Deverman & Patterson, 2009, Neuron. 64:61-78). Elevated cytokines also exacerbate behaviors such as anxiety, feeding disorders, depression, motor and sleep abnormalities, and cognitive dysfunction (Derecki et al., 2010, Mol Psychiatry.15:355-63). Although Rett syndrome is thought to be an autonomous neuronal disease, recent studies demonstrate that lack of MeCP2 in astrocytes and immune-derived microglia promotes the secretion of neurotoxic factors (Ballas, et al., 2009, Nat Neurosci.12:311-7. Maezawa & Jin, 2010, J Neurosci. 30:5346-56). MeCP2 is also expressed in adaptive immune cells and influences the expression of important genes such as IL-6 and FOXP3, which are vital to the development of regulatory T cells (Dandrea et al., 2009, Nucleic Acids Res. 37:6681-90; Lai et al., 2009, J. Immunol. 182:259-73). We hypothesize that loss of MeCP2 in the immune system disturbs neuroimmune signaling pathways and may contribute to the pathogenesis of Rett syndrome. We are investigating the function of MeCP2 in immune cells and asking whether reactivation of MeCP2 in the immune system of a Rett mouse model can slow the progression of symptoms.
Maternal infection increases the risk for schizophrenia and autism in the offspring. In rodents, maternal influenza infection or maternal immune activation (MIA) with the double-stranded RNA, poly(I:C), causes behavioral, histological and transcriptional changes in adult offspring that are consistent with those seen in human schizophrenia and autism. We have found that the cytokine interleukin-6 (IL-6) is necessary and sufficient for the manifestation of a variety of behavioral and cellular abnormalities in the offspring of poly(I:C)-treated mothers. Thus, we are examining the mechanism by which this cytokine influences fetal development in vivo. We focus on the placenta as the site of direct interaction between the mother and fetus and as a principal modulator of fetal development. We also investigate the role of IL-6 in the developing brain itself as a direct mechanism of shaping neural development and behavior.
To search for the possible prenatal and postnatal therapeutic interventions in the maternal immune activation (MIA) model of autism and schizophrenia is a principal goal. One of our candidates is nicotinic acetylcholine receptor subunit alpha-7 (Chrna7). Clinically, the brains of schizophrenia subjects show low Chrna7 expression while compared to normal subjects. Moreover, association of Chrna7 expression level with P50 auditory gating provides convincing evidence that Chrna7 is associated with schizophrenia. In rodent studies, perinatal choline supplementation (a selective agonist of Chrna7) increases brain neurogenesis and alters rodent behaviors. Our lab uses poly(I:C) injection into pregnant female mice as an MIA model in which the offspring exhibit autism- and schizophrenia-like behaviors and neuropathology. We aim to investigate the effect of Chrna7 alternation on MIA-induced offspring phenotype in mice. Although there are as yet no direct links between MIA and Chrna7 activation, the convergent behavioral phenotype between the mouse model and the human disease makes it worthwhile to investigate the cross-talk of cytokines and the cholinergic receptor.
While autism spectrum disorders (ASD) are characterized by language and social deficits, increasing evidence suggests a role for the immune system in ASD pathogenesis. Studies have reported immune-related endophenotypes in the brain, peripheral tissues and gastrointestinal tracts of ASD individuals. Moreover, maternal infection (immune challenge) in a mouse model leads to the development of autism-related endophenotypes in the offspring. We are interested in characterizing immune alterations in peripheral, enteric and neural systems in offspring of immune-activated mothers, and exploring how early maternal immune activation may lead to priming of stem cells and progenitors in fetal hematopoietic niches.
There are reports that significant subsets of children with autism spectrum disorders display gastrointestinal (GI) abnormalities, including chronic inflammation of the colon, increased intestinal permeability and altered composition of GI microbiota. Moreover, antibiotic treatment and altered diet (gluten and casein) have been reported to ameliorate some behavioral symptoms in autistic individuals. We are exploring the potential connections between the gut, brain and immune system in the MIA mouse model.
Epidemiological studies have linked maternal infections, both viral and bacterial, to higher risk of developing schizophrenia and autism in the adult offspring. Factors common to all infections are thought to be responsible for the adverse effect on fetal brain development. The synthetic double stranded RNA poly I:C, which stimulates an antiviral inflammatory response without viral infection, is used to model maternal immune activation (MIA). Structural MRI reveals that adult MIA offspring have enlarged ventricles and smaller hippocampal volumes. These hallmarks of schizophrenia are not apparent in adolescent offspring, however. In a therapeutic intervention, The Weiner and Feldman groups find that giving the anti-psychotic drug clozapine during adolescence prevents the onset of both structural and behavioral changes in MIA offspring. Another neuropathological feature of prefrontal cortex in schizophrenia is a deficit in immunostaining for the calcium binding proteins calbindin and parvalbumin in subpopulations of GABA interneurons. We are examining these neurons in the maternal immune activation model.
Hallucinations are defined as the activation of the visual or auditory system in the absence of appropriate sensory input. A corollary is that such activity should be enhanced by drugs that are known to induce hallucinations in normal people and that exacerbate this symptom in schizophrenic subjects. Activation of 5-HT2A receptors (5-HT2ARs) is responsible for the psychomimetic properties of hallucinogens in humans. 5-HT2AR agonists such as 2,5-dimethoxy-4-iodoamphetamine (DOI) and lysergic acid diethylamide (LSD) stimulate head twitches in mice, which are not seen in 5-HT2AR null mutant mice. We find that DOI induces this stereotyped behavior in mice in a dose-dependent manner. At the molecular level, DOI activates the expression of the immediate early genes egr-1 and c-fos in the auditory, visual and somatosensory cortices. Thus, in the absence of external acoustic and visual input, the hallucinogen DOI activates surrogate markers of neuronal activity in the cortex.
The second stage of the project involves our MIA mouse model (Shi et al., 2003) that is based on the epidemiological finding that maternal infection increases the risk of schizophrenia in the offspring. We find that, compared to controls, the offspring of mothers whose immune systems were activated at mid-gestation show increased stereotypical behavioral responses (head-twitching, grooming and rearing) to DOI. This increased sensitivity to the hallucinogen raises the question of whether they also experience spontaneous, hallucination-like, neuronal activity. We are exploring the use of novel functional MRI methods to examine this question.
We are investigating the neurobehavioral development of mouse pups born to mothers whose immune systems were activated at mid-gestation. A behavioral assay that can be used very early in development involves ultrasonic vocalizations (USVs), which are important for mother–infant social interaction. We find that i.p. injection of the double-stranded RNA, poly(I:C), in C57BL/6J pregnant females at mid-gestation alters pup USVs in the isolation test. Ten-day-old pups from immune-activated mothers display a lower rate of USVs compared to pups born to saline-injected mothers. In addition, analysis of sonogram structure shows differences in the repertoire of calls emitted by male pups. Compared to controls, pups from immune-activated mothers emit significantly fewer composite calls and more short calls. As adults, males from poly(I:C)-injected mothers display a deficit in a social interaction test in which they are given a choice between interacting with another mouse or spending time in an empty chamber. In addition, these males show significantly fewer USVs in response to a female mouse stimulus. In sum, these results suggest that maternal immune activation yields males with poor social and communicative behavior, which are hallmarks of autistic-like behavior. This is consistent with our finding that the offspring of immune-activated mothers also display a neuropathology that is frequently found in autism, a spatially-restricted deficit in Purkinje cells (Shi et al., 2009).
Communication utilizes auditory, visual, tactile and chemical modes to influence and react in social situations. Mice, particularly males, use ultrasonic vocalizations (USVs) for intraspecific communication. To investigate the structure and function of these acoustic signals, we collected USVs from neonatal males during and after brief separation from their mothers, and from adults in several social situations. These USVs contain 10 distinct types of syllables. Analysis of syllable prevalence and duration indicates that the harmonic and frequency-step syllable types are most significant during neonatal isolation and male-female interactions, respectively, while upward syllables are most significant during male-male intrusions and juvenile reunions. These results suggest that syllable usage in USVs could convey emotional state and/or motives. We wish to investigate the possibility that not only the prevalence but also the ordering of syllables may contain information that is important at the species level, as it is in bird song. It is also possible that such information is relevant at the level of individual mice.
The mammalian CNS contains stem and progenitor that generate differentiated cell types through adulthood. While these stem and progenitor cells respond to injury and neurodegenerative and demyelinating diseases with enhanced proliferation and subsequent generation differentiated progeny, this response is typically inefficient and incomplete. We are attempting to stimulate these neural progenitor cells with exogenous factors delivered by recombinant viruses as a means to augment CNS repair. We are particularly interested in exploring the potential of the cytokine LIF to promote repair in models of demyelination. We previously showed that LIF enhances neural stem cell self-renewal and oligodendrocyte progenitor cell proliferation in the adult brain, and based on these findings, we are testing whether LIF can be used to promote oligodendrocyte generation and remyelination following demyelination.
Multiple sclerosis is a devastating inflammatory demyelinating and neurodegenerative disorder that lacks effective treatment. In animal models of the disease, several gp130 cytokines have beneficial effects on disease onset and severity. Because gp130 cytokines affect a wide range of cell types, including neural and immune cells, it is not clear whether these cytokines exert their beneficial effects by acting directly on neural cells or through effects on immune or vascular cells. To test this, we are genetically ablating the gp130 receptor in specific cell types and examining the consequences for oligodendrocyte generation and myelination during development, disease, and in response to therapeutic delivery of exogenous gp130 cytokines. Findings from these experiments will significantly expand our understanding of how gp130 cytokines influence oligodendrocytes and how they may be best used therapeutically to promote OL survival and enhance repair.
Learn More
Our Sponsors
Anne P. and Benjamin F. Biaggini Professor of Biological Sciences
Address: 1200 E. California Blvd. Pasadena, CA 91125
Mail code: 216-76
Office: 151B Beckman Behavioral Laboratories
Phone: (626) 395-6826
e-mail: php@caltech.edu
New book:
http://mitpress.mit.edu/catalog/item/default.asp?ttype=2&tid=12756
Book blog: https://infectiousbehavior.wordpress.com/
Another new book: http://www.cup.columbia.edu/book/978-0-231-15124-5/the-origins-of-schizophrenia
Independent, energetic postdoctoral fellows, Ph.D. students and undergraduates with a passion for discovery are encouraged to write to php@caltech.edu with a description of your experiences, interests and current and future goals.
