Constipated endosomes are a hallmark of degenerating neurons in people with AD. Scientists blame crumbling retromers, complexes that normally flush proteins out of these little organelles. A study published January 22 in Molecular Neurodegeneration reported that stabilizing the retromer complex with a small molecule got endosomal trafficking moving again. Researchers led by Domenico Praticò of Temple University in Philadelphia reported that the compound also prevented Aβ deposition, tau hyperphosphorylation, synaptic loss, and memory deficits in a mouse model of amyloidosis.

  • R33, retromer chaperone, slowed Aβ deposition in 3xTg mice.
  • The small molecule lessened tau phosphorylation.
  • It prevented memory deficits in this strain.

Previous studies discovered that the retromer ushers amyloid precursor protein through endosomes, reducing APP’s exposure to amyloidogenic processing by endosomal BACE1 (Small et al., 2005; May 2008 news; Jun 2012 news). 

Subsequently, Scott Small of Columbia University teamed up with Dagmar Ringe of Brandeis University in Massachusetts and Gregory Petsko at Weill Cornell Medical College, New York, to develop small molecules that bolster the complex. R33 and R55 fortified the interface between two of the three proteins that form the core of the retromer, vacuolar protein sorting-associated protein 35 and VPS26. The compounds, which the researchers called small-molecule chaperones, squelched Aβ production and tau hyperphosphorylation in cultured neurons from J20 mice, and in stem-cell-derived neurons from healthy volunteers and people with sporadic AD (Apr 2014 news; Chu and Praticò, 2017; Young et al., 2018). 

In the new study, first author Jian-Guo Li and colleagues investigated whether these retromer “chaperones” could do the same in one strain of transgenic mice—3xTg, which express pathogenic versions of human APP, PSEN1, and MAPT genes. Starting when the animals were 3 months old, the researchers spiked their drinking water with R33 (also known as TPT-172), or with an inert control. Nine months later, R33 had lessened the mice’s deficits in working, associative, and spatial memory, the scientists reported. The compound even gave wild-type mice a memory boost.

Treated mice had slightly more synaptophysin, a marker of synaptic integrity, in their hippocampi than did untreated. R33 raised the concentration of VPS35, VPS26, and VPS29 to near normal levels, supporting the idea that the chaperone stabilized the retromer.

The compound roughly halved levels of Aβ, as well as BACE1-cleaved APP fragments sAPPβ and CTFβ. Phosphorylated tau levels were a quarter to a third lower than in untreated animals, depending on the epitope. The compound did not change levels of multiple kinases known to phosphorylate tau, including GSK-3β. Praticò told Alzforum that phospho-tau accumulates in endosomes, where it escapes degradation. The enhanced retromer function likely promotes phospho-tau degradation, he said.

“This recent study, along others that preceded it, validate what we had originally considered a remarkable observation: That stabilizing the binding of just two proteins in retromer’s cargo recognition core will enhance the overall function of a complicated, multimodular, trafficking machine,” wrote Small.

Petsko thinks the new data support the idea that the retromer could be targeted therapeutically. “Although the relevance of extreme transgenic models, like the one used here, to the human disease can (and should) be questioned, the ability of R33 to suppress the pathologies of the disease in vivo, including improving behavioral deficits, is impressive and suggests, as we have argued for many years, that endosomal trafficking in general, and retromer in particular, is a valid therapeutic target,” he wrote.

Petsko added that neither R33 nor R55 are likely to advance to clinical studies. Both are plagued by short half-lives and likely react with many other biomolecules. Praticò and Petsko both said they are working with medicinal chemists to find more suitable analogs for clinical development.—Jessica Shugart

Comments

  1. In 2014, as a joint effort with the Petsko and Ringe labs, we introduced a class of drugs called “retromer pharmacological chaperones” (Mecozzi et al., 2014). 

    Retromer’s core is a trimer, and one of the most reliable observations from the ever-growing literature on retromer chemical biology is that destabilizing the trimer accelerates its degradation and reduces retromer function. We wondered if the opposite was true. Through a series of steps, we identified two small molecules, R55 and R33, that stabilized the trimer and increased retromer levels. We showed in neuronal culture that the chaperones enhanced retromer’s trafficking function by finding that it increased the flow of APP and the retromer receptor Sorl1 out of the endosome, and reducing amyloid production.

    Retromer has been primarily implicated in AD and PD (Small and Petsko, 2015), and studies have been using these pharmacological chaperones in model systems to show that increasing retromer function can ameliorate PD-related pathology (e.g., Follett et al., 2016; Li et al., 2018). 

    This recent elegant paper by Dominico Particò’s lab adds to the growing number of studies that have used the chaperones to ameliorate AD-related pathology. In a previous study, they showed that APP mutations cause deficiencies in retromer and the retromer-receptor Sorl1, and in cell culture R55 restored retromer and sorl1 levels. In the current study they used R33 (also call TPT-172) in an in vivo mouse study. A previous study by Jessica Young’s lab (Young et al., 2018) used R33 in IPS human neurons, derived from patients with early onset and sporadic AD, to show that retromer can ameliorate amyloid pathology and tau pathology, and that the effect on tau pathology is independent of amyloid. The current study largely agrees with the Young et al. conclusions, showing that R33 can ameliorate amyloid and tau pathology, but since an in vivo mouse study, the main advance is that they show that the chaperone can reduce synaptic pathology and ameliorate cognitive dysfunction. I congratulate them on all the work that went into this long-term study, in which mice were administered R33 for nine months!

    More than the trimer, the “retromer” is actually a multimodular protein assembly, with the trimer acting as retromer’s “cargo recognition core.” This recent study, and the others that preceded it, together validate what we had originally considered a remarkable observation: That stabilizing the binding between just two proteins in retromer’s cargo recognition core will enhance the overall function of a complicated, multimodular, trafficking machine.

    References:

    . Pharmacological chaperones stabilize retromer to limit APP processing. Nat Chem Biol. 2014 Jun;10(6):443-9. Epub 2014 Apr 20 PubMed.

    . Retromer in Alzheimer disease, Parkinson disease and other neurological disorders. Nat Rev Neurosci. 2015 Mar;16(3):126-32. Epub 2015 Feb 11 PubMed.

    . Parkinson Disease-linked Vps35 R524W Mutation Impairs the Endosomal Association of Retromer and Induces α-Synuclein Aggregation. J Biol Chem. 2016 Aug 26;291(35):18283-98. Epub 2016 Jul 6 PubMed.

    . Phospholipase PLA2G6, a Parkinsonism-Associated Gene, Affects Vps26 and Vps35, Retromer Function, and Ceramide Levels, Similar to α-Synuclein Gain. Cell Metab. 2018 Oct 2;28(4):605-618.e6. Epub 2018 Jun 14 PubMed.

    . Stabilizing the Retromer Complex in a Human Stem Cell Model of Alzheimer's Disease Reduces TAU Phosphorylation Independently of Amyloid Precursor Protein. Stem Cell Reports. 2018 Mar 13;10(3):1046-1058. Epub 2018 Mar 1 PubMed.

  2. In our 2014 paper, Scott Small’s lab and ours collaborated to show that R33 (and a related compound, R55, which I think some companies call TPT-260) stabilized retromer in vitro and also increased retromer protein levels in neurons in culture. Treatment with these chaperones successfully lowered Aβ and APP CTF levels in cell culture of neurons from AD transgenic mice, and the reduction was dose-dependent. Many other labs have since used the chaperones in various cell culture studies of AD pathology, tau pathology, and synuclein pathology, and as far as I know, they are good tool compounds and always increase retromer and reduce pathology. A great example of their use is from Jessica Young’s lab (Young et al., 2018), which showed that R33 reduces both Aβ pathology and tau pathology in sporadic and familial AD patient-derived neurons, and the tau pathology reduction was independent of the presence of APP (a very important finding, since up to that point most people had thought tau pathology was always downstream of APP pathology).

    Scott, Dagmar Ringe, and I have been very gratified that there have been so many proofs of principle, by so many independent labs, that increase in retromer protein levels can rescue neurons from the pathologies of Alzheimer’s and other neurodegenerative diseases. Since they don’t act directly on plaques or tangles, these treatments also provide evidence that these pathologies may be the smoke, not the fire, in these disorders. Scott and I have written articles that advance the hypothesis that it is endosomal trafficking dysfunction that is the fire. Of course, smoke can still be bad for you, and it may be helpful to clear it, but the best way to deal with a fire is to put out the flames, because if you just get rid of the smoke you can still burn to death. 

    The principle was certainly established in vitro by our earlier work. This paper nicely extends that proof of principle to animal models. Although the relevance of extreme transgenic models of AD like the one used here to the human disease can (and should) be questioned, the ability of R33 to suppress the pathologies of the disease in vivo, including improving behavioral deficits, is impressive and suggests, as we have argued for many years, that endosomal trafficking in general, and retromer in particular, is a valid therapeutic target.

    That said, I’d be surprised if R33 or R55 were useful as human drugs. They are thiophene thioureas and as such are not ideally stable in vivo—R55 is particularly bad in this regard. They can react with nucleophiles to form covalent adducts and their metabolic fates in most organs are unknown. I suspect pharmaceutical chemists would want to find analogs that do not have these issues, and we (and others) have been doing that.

    References:

    . Stabilizing the Retromer Complex in a Human Stem Cell Model of Alzheimer's Disease Reduces TAU Phosphorylation Independently of Amyloid Precursor Protein. Stem Cell Reports. 2018 Mar 13;10(3):1046-1058. Epub 2018 Mar 1 PubMed.

Make a Comment

To make a comment you must login or register.

References

News Citations

  1. Mice, Flies Further Implicate Retromer in AD Pathogenesis
  2. Coming Into Vogue? Retromer in APP Processing, AD Pathogenesis
  3. Could Bolstering the Retromer Thwart Alzheimer’s?

Research Models Citations

  1. J20 (PDGF-APPSw,Ind)
  2. 3xTg

Paper Citations

  1. . Model-guided microarray implicates the retromer complex in Alzheimer's disease. Ann Neurol. 2005 Dec;58(6):909-19. PubMed.
  2. . The retromer complex system in a transgenic mouse model of AD: influence of age. Neurobiol Aging. 2017 Apr;52:32-38. Epub 2017 Jan 3 PubMed.
  3. . Stabilizing the Retromer Complex in a Human Stem Cell Model of Alzheimer's Disease Reduces TAU Phosphorylation Independently of Amyloid Precursor Protein. Stem Cell Reports. 2018 Mar 13;10(3):1046-1058. Epub 2018 Mar 1 PubMed.

Further Reading

Primary Papers

  1. . A pharmacological chaperone improves memory by reducing Aβ and tau neuropathology in a mouse model with plaques and tangles. Mol Neurodegener. 2020 Jan 22;15(1):1. PubMed.