Newborn neurons in the adult hippocampus may help form new memories. If so, might neurogenesis play a role in neurological disorders and adult mental retardation? New research into Fragile X syndrome suggests so. This disorder is the most common form of inherited mental retardation, and is caused by a mutation that prevents production of a single gene product, Fragile X mental retardation protein (FMRP). Although most research on the protein has focused on its role in synaptic signaling, recent studies indicate that FMRP is necessary for normal neurogenesis as well. Now, in the April 24 Nature Medicine online, researchers led by Xinyu Zhao at the University of New Mexico, Albuquerque, present striking evidence that FMRP’s effects on adult neurogenesis tie directly to the learning deficits seen in a mouse model of the disorder.
To demonstrate this, Zhao and colleagues bred transgenic mice in which they could selectively turn the gene off or on in adult neural stem cells. They found that when stem cells lacked FMRP but all other brain cells expressed it normally, the mice learned hippocampal-dependent tasks poorly. Conversely, when the scientists restored FMRP expression only in adult neural stem cells in a mouse line without much FMRP, hippocampal-dependent learning returned to normal. The authors also confirmed previous findings that FMRP is necessary for normal neural stem cell proliferation and differentiation. Although many questions remain, the data hint that promoting neurogenesis might have therapeutic potential for people with Fragile X syndrome. In addition, other scientists in the field note that the genetic technique itself could have broad applications for studying not only Fragile X, but other neurological conditions as well.
“Having the capacity to do selective gain of function and loss of function is a technical triumph,” said Brian Christie at the University of Victoria, Canada, who was not involved in this study. “I think the technique is going to become widely used.”
Daniela Zarnescu at the University of Arizona, Tucson, agreed, noting, “This is a very elegant genetic study. It contributes greatly to our understanding of FMRP and may help shift the focus a little bit from synapses to stem cells.”
FMRP is an mRNA-binding protein that regulates translation. Having multiple mRNA targets, the protein seems to have varied effects depending on where and when it acts. Extensive studies have shown FMRP is needed at synapses for normal glutamate receptor signaling. In the last few years, a number of papers have also pointed to a critical role for FMRP in maintaining normal neurogenesis (for a review, see Callan and Zarnescu, 2011). Zhao’s group focused on FMRP’s effects on adult neurogenesis in mice. They previously found that FMRP knockouts have abnormal neurogenesis, with more stem cell proliferation, more newborn cells adopting a glial fate, and less production of neurons (see Luo et al., 2010). Since many studies suggest neurogenesis is important for learning (see, e.g., Snyder et al., 2009; Leuner et al., 2006; Leuner et al., 2006), Zhao and colleagues wondered what effect these changes might have on learning and memory.
First author Weixiang Guo developed a transgenic mouse in which he could selectively turn off FMRP in neural stem cells. He started with an animal created by coauthor Amelia Eisch that expressed inducible Cre recombinase under the control of a nestin promoter, thus confining Cre expression to neural stem cells, then crossed these mice with a FMRP conditional knockout strain, as well as a yellow fluorescent protein (YFP) conditional reporter strain, to create transgenic cKO;Cre;YFP mice. Injecting these triple transgenics with tamoxifen induced Cre in nestin-positive cells, where it snipped out the FMRP gene and activated YFP by excising a stop codon. As a result, stem cells and their progeny lost FMRP and became fluorescent, making them easy to identify. As the group’s earlier work had shown, neural stem cells without FMRP proliferated more, and their progenies’ fate changed from neurons to astrocytes. By eight weeks after FMRP deletion, the stem cells had given birth to about 50 percent more astrocytes and half as many neurons as normal.
To examine effects on learning and memory, coauthor Andrea Allan searched the literature for challenging, sensitive, hippocampal-dependent behavioral tasks. This was important because other studies have shown inconsistent behavioral results from FMRP knockouts. Allan adapted two tasks. The first was a trace conditioning task, in which mice are either exposed to a novel environment, or hear a tone 30 seconds before receiving a mild foot shock. After 24 hours, the researchers expose the mice to the tone or novel environment again to see if the mice freeze, indicating they associate the stimulus with the unpleasant sensation. Mice with hippocampal problems are much worse at making the association and therefore freeze less. The second task tests hippocampal-dependent spatial learning using a modified radial arm maze published by Fred Gage’s group (Clelland et al., 2009), in which mice must remember which arm previously contained food and select a different arm to find a new reward. In both tests, mice lacking FMRP only in neural stem cells did about as poorly as regular FMRP knockouts, and much worse than control littermates. The results demonstrated that merely removing FMRP from stem cells replicates the learning defects of complete knockouts.
Guo and colleagues completed the picture with a gain-of-function experiment. They used a mouse line, created by coauthors Ruiting Zong, David Nelson, and Ben Oostra, in which a neomycin gene interrupts the FMRP site, causing the mice to express hardly any protein and behave similarly to complete knockouts. The authors crossed these mice with both the nestin-Cre and YFP reporter mice. When they injected the adult triple transgenics with tamoxifen, Cre turned on YFP and deleted the neomycin cassette, restoring FMRP expression only in neural stem cells. These mice performed as well as control littermates, showing that having normal FMRP only in stem cells and their progeny is enough to rescue learning defects. Ongoing work is testing whether restoring FMRP expression in stem cells also normalizes neurogenesis in these mice, Zhao said.
Complementing this in-vivo work, the authors isolated adult neural stem cells from the dentate gyrus of conditional FMRP knockouts and induced FMRP loss by infecting the cultures with a Cre-GFP-containing retrovirus. Cultured stem cells lacking FMRP replicated defects in proliferation and differentiation seen in vivo. In addition, neurons born from FMRP-negative stem cells had abnormally simple, stunted neurites. Conversely, stem cells isolated from low FMRP-expressing mice proliferated and differentiated normally in culture once FMRP expression was restored.
Christie noted that, in addition to the two behavioral tasks the authors used, it would be informative to test these mice on context discrimination, where they must distinguish between two similar environments. This task is highly hippocampal dependent, Christie said, and shows robust differences in Fragile X model mice (see Eadie et al., 2010). Christie also suggested that learning defects in Fragile X mice are likely to result from both synaptic and neurogenesis problems. He pointed out that, although Zhao’s study targeted neural stem cells, it did not rule out synaptic effects. “When you alter the fate of neural stem cells, you are changing the way the whole system is interacting,” he said.
Zhao agreed, saying that abnormal synapses may be part of the problem in her conditional FMRP knockouts. Since cultured newborn FMRP-negative neurons have less complex neurites, she speculates that these cells in vivo form abnormal connections with the existing hippocampal circuit, disrupting normal function. “The learning deficit we see in these mice is not only because they generate fewer neurons, but also that those neurons are not normal,” she suggested. In June 2011 her lab will move to the Waisman Center and newly formed Department of Neuroscience at the University of Wisconsin, Madison, where she intends to collaborate with neuroscientists who have the tools to study this question.
Intriguingly for the Alzheimer’s field, previous research has suggested a link between Fragile X syndrome and AD. APP appears to be one of the targets regulated by FMRP (see Westmark and Malter, 2007; Sokol et al., 2011). APP has also been linked to neurogenesis (see, e.g., Ghosal et al., 2010; Crews et al., 2010). Zhao said she is interested in this question, and will work with Michael Wilhelm at UW-Madison to investigate possible connections between the two disorders.
Other neurological disorders are also revealing connections to neurogenesis, suggesting this could be a common theme. For example, huntingtin protein is essential for cell division, and neurogenesis defects are known to occur in a Huntington’s disease model (see Godin et al., 2010; Fedele et al., 2011), and methyl-CpG-binding protein 2, which is mutated in the developmental disorder Rett’s syndrome, has been shown to promote neuronal differentiation (see Tsujimura et al., 2009).
If these findings hold up, could enhancing neurogenesis, for example, through exercise, be therapeutic for people with Fragile X syndrome? Christie noted that physical activity has shown promise for treating other neuropathologies, but should start before the age of 12, when the brain is most dynamic. “If you get kids with autism, attention deficit hyperactivity disorder, and other early neuropathologies to exercise, you can dramatically improve their learning and cognitive capacity,” Christie said. He is about to test this treatment in children with fetal alcohol spectrum disorders. Since synaptic connection and other factors play a role in Fragile X syndrome, exercise may not be enough, however, Zarnescu cautioned, suggesting that intervening both at the synaptic and the stem cell level might be more effective.—Madolyn Bowman Rogers
- Callan MA, Zarnescu DC. Heads-up: new roles for the fragile X mental retardation protein in neural stem and progenitor cells. Genesis. 2011 Jun;49(6):424-40. PubMed.
- Luo Y, Shan G, Guo W, Smrt RD, Johnson EB, Li X, Pfeiffer RL, Szulwach KE, Duan R, Barkho BZ, Li W, Liu C, Jin P, Zhao X. Fragile x mental retardation protein regulates proliferation and differentiation of adult neural stem/progenitor cells. PLoS Genet. 2010 Apr;6(4):e1000898. PubMed.
- Snyder JS, Radik R, Wojtowicz JM, Cameron HA. Anatomical gradients of adult neurogenesis and activity: young neurons in the ventral dentate gyrus are activated by water maze training. Hippocampus. 2009 Apr;19(4):360-70. PubMed.
- Leuner B, Waddell J, Gould E, Shors TJ. Temporal discontiguity is neither necessary nor sufficient for learning-induced effects on adult neurogenesis. J Neurosci. 2006 Dec 27;26(52):13437-42. PubMed.
- Leuner B, Gould E, Shors TJ. Is there a link between adult neurogenesis and learning?. Hippocampus. 2006;16(3):216-24. PubMed.
- Clelland CD, Choi M, Romberg C, Clemenson GD, Fragniere A, Tyers P, Jessberger S, Saksida LM, Barker RA, Gage FH, Bussey TJ. A functional role for adult hippocampal neurogenesis in spatial pattern separation. Science. 2009 Jul 10;325(5937):210-3. PubMed.
- Eadie BD, Cushman J, Kannangara TS, Fanselow MS, Christie BR. NMDA receptor hypofunction in the dentate gyrus and impaired context discrimination in adult Fmr1 knockout mice. Hippocampus. 2012 Feb;22(2):241-54. PubMed.
- Westmark CJ, Malter JS. FMRP mediates mGluR5-dependent translation of amyloid precursor protein. PLoS Biol. 2007 Mar;5(3):e52. PubMed.
- Sokol DK, Maloney B, Long JM, Ray B, Lahiri DK. Autism, Alzheimer disease, and fragile X: APP, FMRP, and mGluR5 are molecular links. Neurology. 2011 Apr 12;76(15):1344-52. PubMed.
- Ghosal K, Stathopoulos A, Pimplikar SW. APP intracellular domain impairs adult neurogenesis in transgenic mice by inducing neuroinflammation. PLoS One. 2010;5(7):e11866. PubMed.
- Crews L, Rockenstein E, Masliah E. APP transgenic modeling of Alzheimer's disease: mechanisms of neurodegeneration and aberrant neurogenesis. Brain Struct Funct. 2010 Mar;214(2-3):111-26. PubMed.
- Fedele V, Roybon L, Nordström U, Li JY, Brundin P. Neurogenesis in the R6/2 mouse model of Huntington's disease is impaired at the level of NeuroD1. Neuroscience. 2011 Jan 26;173:76-81. PubMed.
- Tsujimura K, Abematsu M, Kohyama J, Namihira M, Nakashima K. Neuronal differentiation of neural precursor cells is promoted by the methyl-CpG-binding protein MeCP2. Exp Neurol. 2009 Sep;219(1):104-11. PubMed.
- mGluR5 Antagonist Crosses First Human Testing Hurdle for Fragile X
- Halving Glutamate Receptors Restores Balance in Fragile X Mouse
- One Gene, Two Diseases: The Long and Short of Fragile X
- Synaptic Rescue—Lowering Kinase Activity Sends Fragile X Symptoms PAKing
- AWOL AMPAs Suggest Therapy for Fragile X—Parallels for AD?
- Fragile X, MHC Proteins Shape Synapses
- Guo W, Allan AM, Zong R, Zhang L, Johnson EB, Schaller EG, Murthy AC, Goggin SL, Eisch AJ, Oostra BA, Nelson DL, Jin P, Zhao X. Ablation of Fmrp in adult neural stem cells disrupts hippocampus-dependent learning. Nat Med. 2011 May;17(5):559-65. PubMed.