Increase in activin A may counteract decline in synaptic plasticity with age

Abstract

Activin A, a member of the transforming growth factor β (TGF-β) family, is widely recognized for its neurotrophic and neuroprotective function in the developing and injured brain, respectively. Moreover, in the healthy adult brain, activin A has been shown to tune signal processing at excitatory synapses in a fashion that improves cognitive performance. Because its level in human cerebrospinal fluid rises with age, we wondered whether activin A has a role in mitigating the gradual cognitive decline that healthy individuals experience in late-life. To interrogate the role of activin A in synaptic plasticity in the aging brain, we used an established transgenic mouse line, in which expression of a dominant-negative mutant of activin receptor IB (dnActRIB) serves to disrupt activin receptor signaling in a forebrain-specific fashion. In brain slices of young adult dnActRIB mice (2-4 months old), the NMDA receptor-dependent and -independent forms of long-term potentiation (LTP) at the Schaffer collateral-CA1 pyramidal cell synapse of the hippocampus were equally impaired relative to the extent of LTP measured in the wild-type preparation. Unexpectedly, the difference between the genotypes disappeared when the two forms of LTP were re-examined in slices from middle-aged mice (13-16 months old). Since the level of activin A and endogenous ActRIB both displayed a significant elevation in middle-aged hippocampus, we reasoned that with such a rise, the dominant-negative effect of the mutant receptors could be overcome. Substantiating this idea, we found that administration of recombinant activin A was indeed capable of restoring full-blown LTP in slices from young dnActRIB mice. Our data suggest that, beginning in the middle-aged brain, endogenous activin receptor signaling appears to become strengthened in an attempt to stave off cognitive decline. If further corroborated, this concept would also hold promise for new therapeutic venues to preserve cognitive functions in the aged brain.

Keywords: LTP; activin; age; hippocampus; synaptic plasticity.

Figure 1

Basic neurotransmission and synaptic plasticity in area CA1 are unchanged in aged dnActRIB mice. Field potentials were evoked and monitored by means of extracellular electrodes placed in CA1 stratum radiatum. (A) I-O curves for fPSPs at the SC-CA1 synapse were constructed for wt (black) and mutant preparations (red). Representative voltage trace above I-O curves was recorded in slice from 15 months old wt mouse. Electrical stimulation (50 μA) of Schaffer collaterals evoked a sequence of field potentials consisting of axonal response (fiber volley, FV) and subsequent postsynaptic response (fPSP). Stimulus artifact was truncated. (B) Short-term plasticity was probed by quadruple-pulse stimulation. The increase in 2nd to 4th fPSP relative to 1st one did not differ between genotypes. Representative voltage trace above curves illustrates characteristic potentiation of fPSPs by the four stimuli (20 Hz, 60 μA) in slice from 16 months old wt mouse. (C,D) TBS-induced NMDA-LTP of SC-CA1 synapse did not differ between genotypes (D). Time course of fPSPs potentiation in response to TBS stimulation in wt slice from representative experiment is depicted in panel (C). Insets show averaged voltage traces before (a) and 36–40 min (b) after TBS.

Figure 2

Mutant activin receptor impairs VDCC-LTP in young, but not in old hippocampus. VDCC-LTP was induced by four blocks of high-frequency stimulation at 200 Hz (for 1 s), in the presence of APV (50 μM). (A) Trajectory of fPSP slope before and after the heavy stimulation required to elicit VDCC-LTP in slice from 15 months old dnActRIB mouse. Traces above depict averaged fPSPs during baseline (a) and 36–40 min after tetanus (b). Dashed line indicates the averaged value of field PSP slope before tetanus. (B,C) Impaired VDCC-LTP in mutant (red) compared to wt hippocampus (black) was observed only in slices from young mice (B), but not from old mice (C). (D) Histogram summarizes strength of VDCC-LTP at 36–40 min post tetanus in young and old specimens of both genotypes. Suppression of VDCC-LTP in the presence of L-type Ca²⁺ channel blocker nifedipine (10 μM). Statistical comparisons were performed using an unpaired, two-tailed student’s t-test at α = 0.05 (D, young adults) or a one-way ANOVA followed by Tukey’s post-hoc test (D, old adults, F = 8.995, p = 0.009). n.s., not significant; ^*p < 0.05.

Figure 3

Both activin A and activin receptor IB are increased in aged hippocampus. (A) RT-qPCR analysis reveals comparable levels of dnActRIB mRNA in hippocampi of young and old mutant mice. (B) Aged mice have lighter hippocampus, compared to young mice. (C,D) ELISA measurement shows higher levels of endogenous activin A (C) and endogenous activin receptor ActRIB (D) in aged hippocampi. Statistical comparisons were performed using an unpaired, two-tailed student’s t-test at α = 0.05. n.s., not significant, ^*p < 0.05; ^**p < 0.01; ^***p < 0.001.

Figure 4

Application of recombinant activin A restores NMDA-LTP in young dnActRIB mice. (A) Impaired TBS-induced NMDA-LTP in young dnActRIB hippocampi (red) was rescued after pre-incubation with recombinant activin A (50 ng/mL for 3–8 h, blue), now matching the potentiation seen in wt slices (black). Superimposed traces on the right illustrate extent of fPSP potentiation in each group. (B) Histogram summarizes LTP magnitudes measured 36–40 min post TBS in young and old hippocampi of either genotype. Statistical comparisons were performed using an unpaired, two-tailed student’s t-test at α = 0.05 (B, old adults) or a one-way ANOVA followed by Tukey’s post-hoc test (B, young adults, F = 10.071, p = 0.002). n.s., not significant; ^*p < 0.05.