Nicotinic Acetylcholine Receptor β2 Subunits in the Medial Prefrontal Cortex Control Attention

See allHide authors and affiliations

Science  12 Aug 2011:
Vol. 333, Issue 6044, pp. 888-891
DOI: 10.1126/science.1207079


More than one-third of all people are estimated to experience mild to severe cognitive impairment as they age. Acetylcholine (ACh) levels in the brain diminish with aging, and nicotinic ACh receptor (nAChR) stimulation is known to enhance cognitive performance. The prefrontal cortex (PFC) is involved in a range of cognitive functions and is thought to mediate attentional focus. We found that mice carrying nAChR β2-subunit deletions have impaired attention performance. Efficient lentiviral vector–mediated reexpression of functional β2-subunit–containing nAChRs in PFC neurons of the prelimbic area (PrL) completely restored the attentional deficit but did not affect impulsive and motivational behavior. Our findings show that β2-subunit expression in the PrL PFC is sufficient for endogenous nAChR-mediated cholinergic regulation of attentional performance.

Cortical acetylcholine (ACh) release from the basal forebrain is essential for proper sensory processing and cognition (13) and tunes neuronal and synaptic activity in the underlying cortical networks (4, 5). Loss of cholinergic function during aging and Alzheimer’s disease results in cognitive decline, notably a loss of memory and the ability to sustain attention (6, 7). Interfering with the cholinergic system strongly affects cognition (3, 813). Rapid changes in prefrontal cortical ACh levels at the scale of seconds are correlated with attending and detecting cues (14, 15). Various types of nicotinic ACh receptor (nAChR) subunits are expressed in the prefrontal cortex (PFC) (1618), and in particular nAChRs containing β2 subunits are thought to enhance attention (13). However, the causal relation between nAChR β2 subunits (henceforth β2*-nAChRs) expressed in the medial PFC (mPFC) and attention performance has not yet been demonstrated.

We first determined whether absence of nicotinic β2 subunits affects attentional behavior in the five-choice serial reaction time task (5-CSRTT), a well-established test setup that taxes various aspects of attentional control over performance (19). Mice lacking β2 subunits of nAChRs (β2−/−) and their wild-type littermates (WT) were trained to detect and respond to a brief light stimulus randomly presented in one of five nose poke holes to receive a food pellet (20). β2−/− mice showed normal locomotor activity in an open field test (fig. S1), normal sensorimotor gating in a prepulse inhibition test (fig. S2), and normal acquisition in the 5-CSRTT (fig. S3). After complete acquisition of the 5-CSRTT, animals were trained at the stimulus duration of 1 s (SD1) for 10 more days until they reached stable performance (fig. S4). Baseline 5-CSRTT performance was then calculated from the 6th until the 10th session at SD1 (Fig. 1, A and B). β2−/− mice exhibited significantly more omissions than their WT littermates [F(1,27) = 12.45; P < 0.01] (Fig. 1A), whereas the level of accuracy was not significantly different [F(1,27) = 2.56; not significant (NS)] (Fig. 1B). We found no effect of genotype on any other measures, such as number of initiated trials [F(1,27) = 1.99; NS], number of premature responses [F(1,27) = 0.003; NS], correct response latency [F(1,27) = 2.03; NS], or latency to collect earned food pellets [F(1,27) = 0.12; NS] (table S1), suggesting that increased omissions reflected impairments in stimulus detection processes in β2−/− mice rather than motor or motivational deficits. β2−/− mice and their WT littermates did not differ in the number of food pellets earned by responding to a single cue light nor in the maximal number of responses in a progressive ratio for earning food pellets (fig. S5). In contrast to β2−/− mice, mice lacking α7 subunits of nAChRs (α7−/−) exhibited similar levels of omission [F(1,35) = 0.10; NS] and accuracy [F(1,35) = 0.05; NS] as their WT littermates [but see supporting online material (SOM) text; Fig. 1, C and D; and table S3]. To further characterize attentional deficits, we compared performance in a variable stimulus procedure, in which stimulus durations were randomly decreased to 0.5 and 0.25 s (fig. S6 and table S2). β2−/− mice made significantly more omissions than WT mice at every stimulus duration (fig. S6A and table S2) but had similar accuracy and motivation to earn food rewards (fig. S6B and fig. S5), whereas no difference was observed between α7−/− and WT animals (fig S6, C and D, and table S4).

Fig. 1

β2-nAChR subunit is necessary for normal performance in 5-CSRTT. (A and B) Percentage omission (A) and accuracy (B) of WT (n = 15, black) and β2−/− mice (n = 14, white) during baseline training (SD1). **P < 0.01, Newman-Keuls post hoc test. (C and D) Percentage omission (C) and accuracy (D) of WT (n = 12, black) and α7−/− mice (n = 25, white) during SD1. Data in all figures are shown as mean ± SEM.

To further understand the specific role of β2-containing nAChRs in mediating the effects of endogenous ACh on cognition (21), we selectively reexpressed the β2 subunit (22) in the prelimbic area (PrL) of the mPFC of β2−/− mice. The mPFC is critically involved in attentional performance (23). We reexpressed β2 subunits in combination with enhanced green fluorescent protein (eGFP) by injection of the β2-eGFP bi-cistronic vector (24, 25) into the PrL PFC of β2−/− mice (KOVEC). As a control, we used a lentiviral vector expressing eGFP only in β2−/− mice (KOeGFP) and WT littermates (WTeGFP). Coronal sections showing the site of lentivirus injection revealed that viral reexpression was selective to the PrL of the mPFC (Fig. 2A). The efficacy of this in vivo reexpression strategy was demonstrated by confocal analysis showing that eGFP colocalized with a neuronal marker (NeuN) in KOVEC mice (Fig. 2A) and indicates efficient transduction of β2-eGFP vectors in PrL neurons. β2 subunits do not form functional nAChRs by themselves but require nAChR α subunits to coassemble into functional receptors (26). Therefore, in KOVEC mice not all eGFP-expressing neurons will have β2*-nAChRs. Only in neurons that express nAChR α subunits will lentivirus-mediated expression of β2 subunits result in functional nAChRs containing β2 subunits. We thus made whole-cell recordings from eGFP-expressing neurons in the three groups and tested their response to ACh (Fig. 2, B to D). In WTeGFP mice, locally applied ACh (1 mM) induced inward currents with slow kinetics, characteristic of β2*-nAChRs (Fig. 2D). These currents were strongly reduced by the antagonist of β2*-nAChRs, dihydro-β-erythroidine [DHβE 1 μM; Student’s t test t(7) = –3.15; P < 0.05 (Fig. 2, D and E)]. KOeGFP neurons never showed slow inward currents in response to ACh application (Fig. 2D). Neurons in the mPFC of KOVEC mice showed slow inward currents reminiscent of functional β2*-nAChR responses [n = 9 of 17 EGFP-positive neurons (Fig. 2D)]. These currents were strongly reduced by DHβE [t(6) = –5.02; P < 0.01 (Fig. 2, D and E)], showing the successful reexpression of functional β2*-nAChRs in KOVEC mice.

Fig. 2

Lentiviral restoration of functional β2*-nAChRs in the mPFC. (A) Coronal section (1.9 mm from bregma) showing the injection site in the prelimbic mPFC (left) and confocal images of acute coronal sections showing neuronal eGFP (green) expression (red, NeuN) in KOVEC mice and merged image (right). (B) Experimental setup. (C) Patched eGFP-positive neuron. (D) Current traces recorded from WTeGFP (n = 9, black), KOeGFP (n = 9, gray), and KOVEC (n = 9, red) neurons. ACh was locally applied (1 mM, 100 ms) in control (left), in the presence of β2-containing nAChRs antagonist, DHβE 1 μM (middle), or after 30 min washout (right). (E) Summary of ACh-induced inward currents for WTeGFP (black) and KOVEC (red). nAChR current amplitudes of WTeGFP and KOVEC neurons were not statistically different.

We addressed the question of whether β2*-nAChRs specifically in the PrL mPFC would be sufficient for optimal attentional performance. We therefore tested whether impaired performance of β2−/− mice was rescued by targeted reexpression of the β2 subunit in the PrL mPFC. Preliminary analysis before viral expression showed comparable findings between the independent batches of mice, with a significant increase in omissions in β2−/− mice (fig. S7). One week after virus introduction, WTeGFP, KOeGFP, and KOVEC mice were retrained in the 5-CSRTT procedure by using SD1 for 14 days before the effects of lentiviral intervention were assessed.

At the end of these 14 days, WTeGFP and KOeGFP animals performed at the same levels as they showed before virus injection, but KOVEC mice performed significantly better than before injection (Fig. 3). The percentage of omission of the three groups of mice was differentially affected by lentivector injection [group effect F(2,30) = 10.73 and P < 0.001; injection time effect F(1,30) = 6.12, P < 0.05; group × injection time interaction F(2,30) = 9.29; P < 0.001] (Fig. 3A). Although KOeGFP mice made more omissions than WTeGFP at each time point (WTeGFP versus KOeGFP, P < 0.01), both groups exhibited the same percentage of omissions before and after virus injections (NS, eGFP) and hence were not affected by eGFP expression. Reexpression of β2 subunits in the mPFC (KOVEC) significantly decreased the percentage of omissions (KOVEC before versus KOVEC after, P < 0.001). Moreover, the rescue in KOVEC mice was complete, and these mice reached the same number of omissions as WTeGFP mice (WTeGFP versus KOVEC, NS) and made significantly fewer omissions than KOeGFP mice (KOeGFP versus KOVEC, P < 0.05). This rescue effect was selective for omissions because β2 reexpression had no significant effect on accuracy [group effect F(2,30) = 1.92, NS; injection effect F(1,30) = 2.42, NS; group × injection interaction F(2,30) = 2.36, NS] (Fig. 3B) or any other measures (table S5). This rescue effect was also observed during a variable stimulus procedure (fig. S9), as well as during a variable intertrial interval procedure in which the stimulus presentations were temporally unpredictable (fig. S10), further supporting the conclusion that β2-subunit restoration in the PrL is sufficient for proper attention performance. A similar rescue effect of β2 reexpression in KOVEC mice was observed in an independent group of animals (fig. S11). After these behavioral experiments, the mice were killed, and neuronal expression of eGFP and functional β2*-nAChRs in the PrL was confirmed. β2-subunit reexpression in the anterior cingulate had no effect on omission or accuracy (Fig. 3, C and D), in line with the finding that cholinergic projections to the anterior cingulate cortex are not involved in 5-CSRTT performance (27).

Fig. 3

Targeted reexpression of β2-nACR subunits in PrL mPFC restores performance. (A and B) Percentage omission (A) and accuracy (B) (SD1) before and after viral injection for WTeGFP (n = 11), KOeGFP (n = 11), and KOVEC mice (n = 11). **P < 0.01, compared with WTeGFP; $P < 0.05, compared with KOeGFP; +++P < 0.001, before and after virus injection, Newman-Keuls post hoc test. (C and D) Reexpression of β2-nACR subunits in the anterior cingulate cortex did not restore attention performance. Percentage omission (C) and accuracy (D) (SD1) before and after viral injection for WTeGFP (n = 11) and KOVEC mice (n = 4). *P < 0.05, compared with WTeGFP.

Our findings show that expression of β2*-nAChRs is necessary for optimal attentional performance in mice and that restoring expression of β2*-nAChRs in the mPFC PrL area is sufficient for optimal performance. Nicotinic AChRs containing β2 subunits are located on cell bodies of neurons as well as on thalamocortical afferents in the PrL PFC (16, 17). The latter have been suggested to be involved in attention and processing of sensory stimuli (17). The present study reveals that restoration of β2*-nAChR receptors, specifically in the PrL area of the mPFC, is sufficient to restore the attentional deficit of β2−/− mice to WT levels. Attentional control therefore appears to be mediated by endogenous ACh acting on β2*-nAChR receptors expressed by neurons located within the PrL mPFC, although a role for β2*-nAChRs on thalamic projections cannot be entirely excluded on the basis of the present results. Nevertheless, the nAChR system in the PrL mPFC is a principal factor in attentional control. Consistent with this, rapid changes of ACh levels in mPFC are correlated with cue attending and detection (14), an effect mainly due to mPFC β2*-nAChRs stimulation (28). Our findings have implications relevant for understanding the neurobiology of attention and suggest agonists or positive allosteric modulators at these mPFC β2*-nAChRs within the PrL PFC as potential targets for the development of more effective treatments for cognitive impairments.

Supporting Online Material

Materials and Methods

SOM Text

Figs. S1 to S11

Tables S1 to S5

Reference (29)

References and Notes

  1. Materials and methods are available as supporting material on Science Online.
  2. Acknowledgments: We thank J. Staal and H. Lodder. U.M. and A.B.S. received funding from the EU Seventh Framework Programme “Neurocypres” (HEALTH-F2-2007-202088); H.D.M., from the Netherlands Organization for Scientific Research (917.76.360 and 912.06.148), Neurobasic PharmaPhenomics, and the VU University board.

Stay Connected to Science

Navigate This Article