Two heads may be better than one when it comes to solving problems, but two drugs may be better than one when it comes to shoring up a failing memory. Preclinical animal data reported in the June issue of the journal Neuropharmacology suggest that a combination of acetylcholinesterase inhibitor and GABA receptor antagonist might work better than either alone when it comes to improving spatial memory. Also just published is the first publicly reported data on the effect of γ-secretase inhibition in humans.

Michela Gallagher and colleagues at The Johns Hopkins University, Baltimore, Maryland, and the University of North Carolina, Chapel Hill, tested the effects of the cholinesterase inhibitor donepezil and the Saegis Pharmaceuticals Inc. GABA-B antagonist, SGS742, on memory and performance in rats. Donepezil is currently approved by the US Food and Drug Administration for the treatment of Alzheimer disease (AD), while SGS742 is in phase II clinical trials for AD and has just come through a phase II trial for mild cognitive impairment (MCI).

First author K. A. Helm and colleagues found that in rats, administration of a combination of the drugs led to statistically significant improvement in performance in a radial arm maze test. In the test, rats placed in the center of the maze are allowed to retrieve food rewards from one of up to 12 radial arms. Animals were first trained in the maze so that they made very few errors—an error being an attempt to retrieve food a second time from any arm. The memory test began with an “information phase,” where rats were placed in a maze that had several of the arms blocked off. After the animals had collected the available food, they were removed from the maze, the roadblocks removed, and then the animals were reintroduced into the maze after a short delay. Animals with the better memories would be expected to make fewer errors, visiting mostly the newly accessible arms containing the food and ignoring the empty, disappointing arms.

Helm and colleagues found that when they administered either donepezil or SGS742 alone to the animals about 40 minutes before their memory test began, performance improved significantly. The authors found that both drugs work about equally well, reducing errors from approximately four to an average of about 3.25. These results are not unexpected, as the two compounds have previously been shown to improve memory in mammals. But when Helm administered the two drugs in combination, there was even more improvement—errors fell to an average of about 2.75.

“An additive effect of combined pretrial treatment with the two compounds could reflect augmentation of different but complementary processes that contribute to optimal cognitive function,” write the authors. In this regard it is worth noting that when the drugs were administered 24 hours prior to the information phase of the test, donepezil had no effect, but SGS742 still led to a significant improvement, reducing errors to an average of about 3.75 compared to 4.5 in animals receiving placebo.

These findings fit with one of the proposed mechanisms of action of GABA-B receptor antagonists. In addition to having an inhibitory effect on neurotransmission, activation of GABA-B receptors has recently been shown to enhance the activity of CREB2, or cAMP response element binding protein 2 (see Nehring et al., 2000 and White et al., 2000). Because CREB2 is known to suppress the transcriptional activation of CRE- (or cAMP response element)- controlled genes that are required for laying down new memories, any GABA-B blocker, SGS742 included, would be expected to relieve that inhibition and enhance formation of new memory.

And indeed, when Helm and colleagues tested hippocampal extracts for proteins that bind to the CRE, they found that extracts from rats given SGS742 had significantly less CRE binding protein—only about 50 percent as much protein loaded on CRE as in extracts from animals given placebo. The finding suggests that “…augmentation of neurochemical release by SGS742 could provide an indirect mechanism for modulating cognition independent of, or in addition to, an action on postsynaptic GABA-B receptor induced alterations of CRE-binding activity,” according to the authors.

SGS742 is still in the development phase. Data from the phase II trial for MCI, as reviewed by Roger Bullock in last January’s Current Opinion in Investigational Drugs, showed no difference between placebo and drug groups in the primary endpoint, the Hopkins Verbal Learning Test, after 8 weeks of treatment. But there were significant improvements in additional endpoints, including reaction time on the Cambridge Neuropsychological Test Automated Battery and in rapid visual information processing. “Hence the SGS742 arm demonstrated improvements in multiple cognition domains, including psychomotor speed, attention, and memory,” writes Bullock. The phase II trial for AD is expected to be completed in the third quarter of 2005.

Meanwhile, the first published data on the effects of a γ-secretase inhibitor in humans have just come out from Patrick May and colleagues at Eli Lilly and Company. γ-secretase is the protease responsible for releasing the notorious amyloid-β from its precursor protein. But the protease is also required for processing of other proteins, among them Notch, which plays a key role in multiple developmental and regulatory pathways (see ARF related news story). So safety and side effect concerns continue to hang over development of γ-secretase inhibitors.

At the 7th International Conference on AD/PD, held last March in Sorrento, Italy, May reviewed some of the data on the second-generation γ-secretase inhibitor LY 450139 (see ARF related news story), and this data has now been published in the May-June issue of Clinical Neuropharmacology.

First author Eric Siemers and colleagues conclude, as May had suggested in Sorrento, that further dose optimization studies will be needed before this drug can be taken to a large study designed to determine its effect on disease progression. This is partly because while plasma Aβ levels fell in a dose-dependent manner in volunteers (37 healthy men and women, aged 45 or older), once the nadir was reached, levels then rose to around three times baseline before falling again. However, levels did not get back to baseline within the 24-hour gap between doses of the drug. This biphasic response has also been noted in animal studies.

What causes the biphasic response is unclear, as is where the additional Aβ comes from. “While a complete explanation of the biphasic response in plasma Aβ shown in this study will await additional investigations, these data are consistent with the concept that trafficking of Aβ in brain, CSF, and periphery is complex and may be highly regulated,” write the authors. Though the authors did measure Aβ in the CSF of the volunteers, they found no differences between baseline levels and those at time of measurement, 6 hours after administration of the drug, at which point the plasma Aβ had already started to increase following the initial drop.—Tom Fagan


No Available Comments

Make a Comment

To make a comment you must login or register.


News Citations

  1. Notch and Memory: More Trouble for γ-Secretase Inhibition?
  2. Sorrento: Scientists Stop Clubbing, Start Tweaking, γ-Secretase

Paper Citations

  1. . The metabotropic GABAB receptor directly interacts with the activating transcription factor 4. J Biol Chem. 2000 Nov 10;275(45):35185-91. PubMed.
  2. . The GABAB receptor interacts directly with the related transcription factors CREB2 and ATFx. Proc Natl Acad Sci U S A. 2000 Dec 5;97(25):13967-72. PubMed.

Further Reading

Primary Papers

  1. . GABAB receptor antagonist SGS742 improves spatial memory and reduces protein binding to the cAMP response element (CRE) in the hippocampus. Neuropharmacology. 2005 Jun;48(7):956-64. PubMed.
  2. . SGS-742 Novartis. Curr Opin Investig Drugs. 2005 Jan;6(1):108-13. PubMed.
  3. . Safety, tolerability, and changes in amyloid beta concentrations after administration of a gamma-secretase inhibitor in volunteers. Clin Neuropharmacol. 2005 May-Jun;28(3):126-32. PubMed.