22 Sep 2007

Loss of the Fragile X mental retardation protein (FMRP), whose gene is mutated in Fragile X syndrome, causes excessive trafficking of GluR1, one of the AMPA-type glutamate receptor subunits, according to researchers at Emory University in Atlanta, Georgia. The errant internalization of GluR1 seems related to the activity of the mGluR5 subunit of the metabotropic glutamate receptor. Reported in this week’s PNAS online, the finding not only helps explain increases in long-term depression seen in mouse models of Fragile X syndrome, but also suggests that mGluR5 antagonists may help correct for FMRP loss. In an interview with ARF, co-author Gary Bassell said that this latest research shows parallels with Alzheimer disease, although a benefit, if any, of mGluR5 antagonists for AD is debatable. “Clearly, Aβ accumulation leads to loss of AMPA receptors. But unlike in Fragile X, the mGluR pathway may not be altered in Alzheimer disease,” he said. Bassell’s group collaborated with the laboratory of Stephen Warren, also at Emory.

Loss of FMRP has previously been linked to increased long-term depression (LTD) in mouse models of this neurodevelopmental disorder, which strikes boys disproportionately. Work from the groups of Mark Bear at MIT and Eric Klann (then at Baylor College of Medicine, Houston, Texas, now at New York University) established that knocking out the FMRP gene (Fmr1) enhances mGluR-dependent LTD in mouse hippocampus (see Huber et al., 2002 and Hou et al., 2006). “That work predicts that there may be an AMPA trafficking effect,” said Bassell. But altered AMPAR trafficking has not been demonstrated in Fragile X syndrome (FXS) models until now.

First author Mika Nakamoto and colleagues took advantage of an established dual staining method to measure AMPA receptor dynamics. Dual staining reveals both cell-surface and internalized proteins and allows researchers to quantify receptor trafficking. The researchers first tested the method in wild-type neurons. When they stimulated primary rat hippocampal neurons with the group 1 mGluR-specific agonist DHPG ([RS] 3,5 dihydroxyphenylglycine), they found a reduction in cell-surface GluR1 and an increase in internalized GluR1. This DHPG-induced GluR1 internalization depended on protein synthesis, because three different translation inhibitors blocked it, whereas the transcription inhibitor actinomycin D had no effect.

Given that FMRP is a translational repressor, how might it affect GluR1 internalization? To find out, Nakamoto and colleagues used short interfering RNAs to knock down FMRP gene expression. They found that in cells expressing si-Fmr1, more GluR1 subunits became internalized even in the absence of mGluR stimulation. And because FMRP was knocked down, not out, the researchers were also able to measure a dose response: as FMRP levels fell, the level of internalized GluR1 and GluR2 subunits rose. “Thus, without exogenously applied agonist, FMRP deficiency directly correlated with AMPAR internalization,” write the authors. They note that this is the first demonstrated link between Fragile X syndrome and altered AMPAR trafficking.

If the increased AMPAR trafficking in FMRP-deficient neurons is due to metabotropic glutamate receptor signaling, then what is activating the mGluRs? The researchers found that the mGluR5-specific inverse agonist MPEP (2-methyl-6-phenylethynyl-pyridine) attenuated Fmr1 knockdown effects on AMPARS. MPEP blocks both constitutive and agonist-induced activation mGluR5 receptors. The finding supports the idea that excessive mGluR5 signaling stimulates AMPAR internalization, and suggests that mGluR5 antagonists may help ameliorate some of the symptoms of Fragile X syndrome. “Several start-up companies have licensed such antagonists with this goal in mind,” said Bassell.

The downstream targets linking mGluR5 signaling with AMPAR trafficking are unclear at present. Factors that may be involved include Erk, Ca2+, and post-synaptic density 95 (PSD95). “PSD95 is a potentially interesting link because we know that it is an FMRP target,” said Bassell. That comes from work at Bassell’s own lab (see Muddashetty et al., 2007) and also from Claudia Bagni’s lab in Rome (see ARF related news story).

Fragile X researchers have made great strides in the past several years, and Bassell pointed out reasons why this might be the case. “FRAXA [the Fragile X Research Foundation], the largest private funder of Fragile X research, purposefully tried to get plasticity people interested in Fragile X. I think they played a key role in getting leaders in LTD and LTP fields involved and gave out grants without requiring preliminary data because they wanted anyone in the plasticity field to work on this mouse model,” he said. He added that the Fmr1 knockout model was readily circulated around labs that were working on plasticity. Obtaining mouse models has been difficult at times for AD researchers, but see ARF pages on Jackson lab service.

In a separate line of research, PSD95 has also been linked to Aβ toxicity (see ARF related news story) and to AMPA receptor loss in AD models (see ARF related news story). Furthermore, Aβ precursor protein (APP) mRNA was recently suggested as a target for FMRP. Parallels between neuronal plasticity alterations in AD and Fragile X seem to be turning up at a rapid pace, but they only go so far. One key difference between AMPA loss in AD and Fragile X models is that in the latter, there appears to be little or no total loss of AMPAs, just heightened internalization. “We think that the receptors are just cycling in and out of the membrane and they are untethered,” said Bassell. “If you look at the steady state, there is some reduction, but it depends on the assay and the stage of development, and the data is not striking.” Next, researchers will study the physiological consequences of receptors that cycle in and out of the membrane more quickly.

Bassell also suggested that the dual-staining assay, which can quantify receptor internalization, would be a good technique to apply to AD models of disease. “With some of the APP transgenic mouse models you could look at the dynamics of AMPA receptor internalization to see if it is one of the early things that go awry.” He also suggested that it would be useful for the Fragile X and AD communities to come together at a conference to compare and contrast the latest developments.

In other AMPA news, Mike Ehlers and colleagues at Duke University and the University of North Carolina, Chapel Hill, report in the September 20 Neuron that endocytosis of AMPA receptors depends on dynamin-3 and Homer. According to the study, these proteins help bring the endocytic zone (EZ), which is necessary for receptor endocytosis, in close proximity to the post-synaptic density (PSD). Loss of Homer has been linked to AD pathology (see ARF related news story), and dynamin crops up in the AD literature at times, though a role has not been defined.

First author Jiuyi Lu and colleagues found that disruption of dynamin-3, a GTPase that helps sever clathrin-coated vesicles from the plasma membrane, uncouples the endocytic zone from the post-synaptic density. They also found that dynamin-3 and Homer must interact to position the EZ close to the PSD. It is not dynamin-3’s GTPase activity that is required for EZ localization; rather, protein-protein interactions, including oligomerization of dynamin-3 itself, appear crucial. When a mutant dynamin-3 lacking the domain required for oligomerization was expressed in hippocampal neurons, the researchers found that many more synapses lacked an endocytic zone.

Because endocytosis is a known facet of AMPAR trafficking, Ehlers and colleagues asked what would happen to these receptors if the EZ was uncoupled from the PSD. When the scientists disrupted synaptic EZ formation by introducing dynamin-3 mutants, they found a significant drop in cell surface GluR1 and a decrease in the AMPAR/NMDAR ratio at glutamatergic synapses. Their data suggest that dynamin-3, by linking the EZ to the PSD via Homer, plays a role in AMPAR trafficking.

“These findings indicate that spatially localized endocytosis and presumably local recycling acts to retain AMPARs, and potentially other membrane proteins, in the vicinity of the PSD by internalizing and reinserting them before they drift away by lateral membrane diffusion,” suggest Frederic Jaskolski, Stephane Martin, and Jeremy Henley, University of Bristol, England, in an accompanying Neuron preview. How this might relate to increased AMPAR internalization induced by FMRP loss, or Aβ toxicity, may be a future story.—Tom Fagan

Comments

No Available Comments

Make a Comment

To make a comment you must login or register.

References

News Citations

1. Fragile X, MHC Proteins Shape Synapses 16 Apr 2007
2. Amyloid-β Zaps Synapses by Downregulating Glutamate Receptors 19 Jul 2005
3. AMPA Receptors: Going, Going, Gone in Aβ-exposed Synapses, PSD95 Knockouts 13 Dec 2006
4. LTP and Memory Effects May Be the First Casualties in AD 8 Jul 2003

Paper Citations

1. Huber KM, Gallagher SM, Warren ST, Bear MF. Altered synaptic plasticity in a mouse model of fragile X mental retardation. Proc Natl Acad Sci U S A. 2002 May 28;99(11):7746-50. PubMed.
2. Hou L, Antion MD, Hu D, Spencer CM, Paylor R, Klann E. Dynamic translational and proteasomal regulation of fragile X mental retardation protein controls mGluR-dependent long-term depression. Neuron. 2006 Aug 17;51(4):441-54. PubMed.
3. Muddashetty RS, Kelić S, Gross C, Xu M, Bassell GJ. Dysregulated metabotropic glutamate receptor-dependent translation of AMPA receptor and postsynaptic density-95 mRNAs at synapses in a mouse model of fragile X syndrome. J Neurosci. 2007 May 16;27(20):5338-48. PubMed.

Other Citations

1. ARF pages

External Citations

1. Fragile X Research Foundation

Further Reading

News

1. Fragile X, MHC Proteins Shape Synapses 16 Apr 2007
2. Amyloid-β Zaps Synapses by Downregulating Glutamate Receptors 19 Jul 2005
3. AMPA Receptors: Going, Going, Gone in Aβ-exposed Synapses, PSD95 Knockouts 13 Dec 2006
4. LTP and Memory Effects May Be the First Casualties in AD 8 Jul 2003