Nicotinic acetylcholine receptors (nAChRs) mediate and modulate synaptic transmission throughout the brain, and contribute to learning, memory, and behavior. Dysregulation of α7-type nAChRs in neuropsychiatric as well as immunological and oncological diseases makes them attractive targets for pharmaceutical development. Recently, we identified NACHO as an essential chaperone for α7 nAChRs. Leveraging the robust recombinant expression of α7 nAChRs with NACHO, we utilized genome-wide cDNA library screening and discovered that several anti-apoptotic Bcl-2 family proteins further upregulate receptor assembly and cell surface expression. These effects are mediated by an intracellular motif on α7 that resembles the BH3 binding domain of pro-apoptotic Bcl-2 proteins, and can be blocked by BH3 mimetic Bcl-2 inhibitors. Overexpression of Bcl-2 member Mcl-1 in neurons enhanced surface expression of endogenous α7 nAChRs, while a combination of chemotherapeutic Bcl2-inhibitors suppressed neuronal α7 receptor assembly. These results demonstrate that Bcl-2 proteins link α7 nAChR assembly to cell survival pathways.
Nicotinic acetylcholine receptors (nAChRs) are pentameric ligand-gated ion channels that mediate synaptic transmission at the neuromuscular junction and cholinergic synapses in the brain and periphery1. The neuronal nAChR family comprises α2-10 and β2-4 subunits, which give rise to numerous receptor subtypes through selective heteromerization2,3. However, the α7 subunit is atypical in that it forms homomeric complexes4 with five equivalent agonist-binding sites in the extracellular domain5. Native α7 receptors are expressed on both pre- and postsynaptic membranes, where they serve as modulators of circuit activity throughout the brain3,6. Though α7 nAChRs desensitize on a sub-millisecond time scale (e.g. ref. 7), their relatively high calcium (Ca2+) permeability8 has physiological importance. Indeed, deletion or selective activation of α7 receptors shows that they facilitate learning and memory, and also contribute to anxiety9,10. Likewise, deficits in α7 nAChR function are implicated in Alzheimer’s disease, schizophrenia, and pain, making the augmentation of its activity an attractive therapeutic strategy11,12.
Recombinant expression of functional α7 nAChRs is poor in most mammalian cell lines13,14,15, though we recently identified the transmembrane protein NACHO (TMEM35A) as critical for efficient expression of recombinant and native α7 nAChRs16. Furthermore, the activity of several other neuronal nAChR subtypes is also enhanced by NACHO17. Biochemical studies show that NACHO mediates α7 pentamer formation, as NACHO is required for receptor labeling by α-bungarotoxin (α-Bgt)16, which binds at the interface between assembled α7 subunits4,18. Interestingly, resistance to inhibitors of cholinesterase-3 (Ric-3) protein, which is required for nAChR function in Caenorhabditis elegans19, synergizes with NACHO to promote receptor assembly and function16.
Here, we screened a genome-wide cDNA library to determine if additional proteins further augment α7 nAChR expression in the presence of NACHO. Unexpectedly, several members from the B-cell lymphoma 2 (Bcl-2) protein family were amongst the top hits. Bcl-2 proteins are widely known for regulating apoptosis, as the binding of antiapoptotic Bcl-2 proteins to proapoptotic Bcl-2 proteins prevents signaling cascades that initiate programmed cell death20. Because overexpression of antiapoptotic Bcl-2 proteins promotes unchecked cell proliferation, these molecules have proven effective targets for anticancer drugs21. Interestingly, the activation of α7 nAChRs is also associated with prosurvival signaling pathways (e.g. ref. 22) and generally thought to protect against neuron loss in neurodegenerative disorders23. Though prosurvival Bcl-2 members are abundantly expressed in the nervous system24, whether they directly interact with α7 subunits or those of other neuronal nAChRs to facilitate cell survival has remained poorly studied.
We now find that α7 nAChRs possess a motif on the intracellular loop between transmembrane (TM) segments TM3 and TM4 that is highly conserved with the Bcl-2 homology (BH) 3 domain of proapoptotic Bcl-2 proteins. Mutation of this BH3-like motif disrupts Bcl-2-mediated α7 nAChR upregulation, though the effects of NACHO and Ric-3 remain, suggesting Bcl-2 proteins exploit unique structural pathways to enhance assembly. Moreover, manipulation of Bcl-2 family member myeloid cell leukemia-1 (Mcl-1) in neurons regulates α7 nAChR assembly, implying a functional role for this pathway in the biogenesis of native α7 receptors.
α7 nAChR surface expression is enhanced by Bcl-2 proteins
We cotransfected HEK293T cells with individual plasmids encoding the α7 subunit, NACHO, and one other gene product from a genome-wide collection of over 17,000 cDNA clones25. High-content imaging of transfected cells was then used to identify cDNA clones encoding proteins that facilitate α7 nAChR expression. Based on augmentation of cell surface α-Bgt labeling, the first, second, and fourth most effective clones encoded antiapoptotic Bcl-2 proteins, namely Mcl-1, Bcl-2-like 2 (Bcl-W), and Bcl-2-like 1 (Bcl-XL) (Table 1). For each of these clones, surface labeling of α7 receptors by α-Bgt was at least 70% of the maximal levels observed when NACHO and Ric-3 were coexpressed with α7 subunits (see ref. 17), though no other Bcl-2 proteins present in the library appeared as hits. To further evaluate the effect of Bcl-2 members on α7 nAChRs, we performed additional staining and electrophysiology experiments on transiently transfected cells. As Mcl-1 and Bcl-XL are much more abundant in brain tissue, where functional α7 receptors occur, compared to Bcl-2 itself (Genotype Tissue Expression project; https://gtexportal.org), we focused on these two proteins for most of our studies.
When Mcl-1, Bcl-XL, and Ric-3 cDNAs were cotransfected with α7 alone, total α-Bgt staining was barely, if at all detectable (Fig. 1a). However, coexpression of α7 subunits with NACHO yielded robust α-Bgt labeling in permeabilized and nonpermeabilized cells, as shown previously16, which was further enhanced by each Bcl-2 protein (Fig. 1a). When α7, NACHO, and Bcl-2 member cDNAs were cotransfected at a 1:3:5 ratio, upregulation of surface staining was particularly evident, increasing 4.2 ± 0.3 fold (mean ± SEM; n = 5) or 2.9 ± 0.2 fold (n = 5) when Mcl-1 or Bcl-XL were respectively coexpressed with NACHO (Fig. 1b; Supplementary Fig. 1a), and Bcl-W or Bcl-2 itself produced comparable effects (Supplementary Fig. 1b). By comparison, Ric-3 produced a 4.6 ± 0.2 fold (n = 5) increase in α-Bgt surface staining when coexpressed with NACHO (Fig. 1b), while Bcl-2 protein coexpression with NACHO and Ric-3 further enhanced staining (Supplementary Fig. 1c). Because Bcl-2 proteins elicited comparable increases in total versus surface α-Bgt staining (Supplementary Table 1), these data suggest intracellular assembly of α7 subunits was enhanced, rather than merely membrane trafficking. In agreement, membrane binding of the nAChR agonist [3H]epibatidine, which requires assembly of two subunits to form the orthosteric site, also increased to similar extents when Mcl-1, Bcl-XL, and Ric-3 were cotransfected with α7 and NACHO cDNA transcripts (Fig. 1c; Supplementary Fig. 1d). Moreover, Bcl-2 protein-dependent increases in total α-Bgt staining persisted in the presence of the proteasome inhibitor MG-132, and did not occur alongside comparable increases in α7 subunit expression (Supplementary Fig. 1e, f; Supplementary Table 1). This indicates that Bcl-2 member effects on α7 nAChR assembly do not stem from a suppression of ubiquitination and/or degradation of unassembled subunits.
Given that detectable electrophysiological responses have been obtained from mammalian cells when Ric-3 is coexpressed with α7 subunits (e.g. ref. 16), despite minimal α-Bgt staining, we wondered whether the expression of Bcl-2 proteins could be sufficient to produce α7-mediated currents. When α7 and Mcl-1 cDNAs were cotransfected at a 1:5 ratio, whole-cell ACh-evoked currents were routinely observed, with peak amplitudes averaging 163 ± 46 pA (n = 17), as opposed to just 1 of 16 cells transfected with α7 alone (Fig. 1d, e). Larger α7 peak currents were recorded in the presence of Ric-3 (1.6 ± 0.3 nA, n = 15), or especially NACHO (8.1 ± 1.8 nA, n = 9). Cotransfecting Bcl-2 member cDNAs with NACHO further increased α7 peak currents (Supplementary Fig. 2). Taken together, these findings reaffirm that Bcl-2 proteins and NACHO can work in synergy to enhance functional expression of α7 nAChRs.
Bcl-2 inhibitors disrupt Bcl-2-mediated α7 upregulation
Interactions between antiapoptotic Bcl-2 proteins and their proapoptotic counterparts occur through a binding groove formed by BH domains 1–3 of the former group26. Chemotherapeutic inhibitors, including the apoptosis-inducing molecule ABT-73727, occupy this groove28, which liberates the proapoptotic BH3 proteins (Supplementary Fig. 3a, b). More recently, the compounds S-63845 (S63 29) and A-1155463 (A11 30) have been reported to selectively inhibit Mcl-1 and Bcl-XL, respectively, at nanomolar concentrations. Incubation for 24 h with 100 nM S63 reduced Mcl-1 upregulation of surface α-Bgt staining from 4.9 ± 0.2 fold (n = 5) to 2.3 ± 0.1 fold (n = 5) above that of α7 + NACHO alone (Fig. 2a, b). Likewise, incubation for 24 h with 100 nM A11 abolished Bcl-XL, but not Mcl-1 or Ric-3-mediated enhancement of surface α-Bgt staining (Fig. 2a, b), resulting in staining intensity that decreased from 3.6 ± 0.2 fold (n = 5) to 0.92 ± 0.04 fold (n = 5) above that of α7 + NACHO alone. Electrophysiological data also showed that A11 strongly suppressed α7 nAChR currents obtained in the presence of Bcl-XL (Fig. 2c, d). In the micromolar range, ABT-737 countered the effect of Bcl-2 proteins on α7 nAChR surface expression (Supplementary Fig. 3c, d).
We next asked whether the overexpression of proapoptotic BH3-only proteins would also block Bcl-2 protein-mediated α7 nAChR upregulation. The cotransfection of either p53-mediated upregulator of apoptosis (Puma) or Noxa cDNA atop α7 and NACHO prevented Mcl-1 and Bcl-XL from enhancing surface α-Bgt staining (Fig. 2e). Specifically, the coexpression of Mcl-1 or Bcl-XL with Puma resulted in staining intensities that were 72 ± 3% (n = 5) and 74 ± 3% (n = 5) of that obtained without them, whereas Ric-3 increased staining 4.5 ± 0.2 fold (n = 5; Fig. 2f). Furthermore, Puma expression largely abolished ACh-evoked currents when Mcl-1, but not Ric-3 was cotransfected with α7 cDNA (Fig. 2g, h). Noxa also blocked Mcl-1 and Bcl-XL enhancement of α7 nAChR surface expression, as staining intensities were 73 ± 4% (n = 5) and 104 ± 8% (n = 5) of control values, though Ric-3 increased staining 3.3 ± 0.1 fold (n = 5; Fig. 2e, f). Overall, these data suggest the binding groove of antiapoptotic Bcl-2 members, typically engaged by proapoptotic BH3 motifs, is important for upregulation of α7 nAChR assembly.
α7−Bcl-2 interactions depend on a BH3-like binding motif
To assess the role for the Bcl-2 binding groove on α7 nAChR assembly, we explored whether that region is sufficient for Mcl-1-mediated upregulation, and how some segment of the α7 protein could act as a ligand there. The topology of antiapoptotic Bcl-2 proteins is largely cytosolic, and a C-terminal TM domain anchors them to mitochondrial and endoplasmic reticulum (ER) membranes (Fig. 3a). Deletion of the TM region of Mcl-1 disrupted its effects on α7 subunit assembly. However, replacement of the TM segment with that of the ER resident protein cytochrome B5 (CytB5) restored the upregulation of α7 pentamer formation, producing comparable effects to wild-type Mcl-1 (Fig. 3b–d). Taken together, these data demonstrate that a membrane tether is required to bring Bcl-2 proteins into proximity with α7 subunits and that their interaction likely occurs in the ER (see Discussion).
To identify regions on the α7 subunit that might interact with Bcl-2 proteins we focused on the intracellular loop between TM3 and TM4, which comprises the vast majority of the cytoplasmic portion of the protein. Interestingly, a helical motif prior to TM4 contains an amino acid sequence with striking similarity to the BH3 domain of proapoptotic Bcl-2 proteins (Fig. 4a). Hydrophobic residues are in fact conserved in α7 at three key positions known to mediate BH3 affinity for the Bcl-2 binding groove (e.g. ref. 31) (Fig. 4b). We therefore tested various deletions and mutations in this vicinity.
Deletion of the entire α7 TM3−TM4 intracellular loop prevents subunit assembly, yet removal of most of the loop still permits assembly-dependent α-Bgt binding32. One such mutant (del 347) yielded α-Bgt labeled receptors (with NACHO present) whose expression was still enhanced by Ric-3 (2.7 ± 0.2 fold, n = 5), whereas enhancement by Mcl-1 and Bcl-XL was markedly reduced to 1.0 ± 0.1 fold (n = 5) and 1.2 ± 0.03 fold (n = 5), respectively (Fig. 4c, d). More targeted removal of the helical region surrounding the BH3-like motif (del preM4) produced similarly attenuated Bcl-2 protein upregulation of total α-Bgt staining, as did point mutations of hydrophobic amino acids at positions 433, 436, and 440 to alanine residues (Fig. 4c, d). The effect of these mutations was also consistent in experiments measuring α-Bgt staining of nonpermeabilized cells (Supplementary Fig. 4a, b). All these pre-M4 helix manipulations also greatly increased surface and total α-Bgt staining in the control condition, where only α7 subunits and NACHO were coexpressed. Such findings mirror earlier reports of mutations in this region augmenting surface α-Bgt binding in α7-expressing oocytes33. The possibility that greater basal receptor expression simply precludes further upregulation by accessory proteins is countered by the L433A mutation, which yielded the highest baseline α-Bgt labeling of all mutant receptors examined yet experienced a 1.9 ± 0.1 fold (n = 5) increase in total staining with Ric-3 cotransfected (Fig. 4d). A similar effect was seen in electrophysiological data, where α7 L433A yielded consistent current responses without any chaperone protein coexpressed (Supplementary Fig. 4c, d). As a result, the most likely explanation for the reduced effect of Bcl-2 proteins on mutant receptors is a selective disruption of their interaction with α7 subunits. Though pre-M4 helix point mutations reduced Ric-3 enhancement of α7 nAChR surface expression, the chaperone still consistently produced two-fold or greater increases in total α-Bgt labeling, more than any residual effect of either Mcl-1 or Bcl-XL (Fig. 4).
To verify that the limited effect of Bcl-2 proteins on mutant α7 nAChR surface expression reflects changes in functional receptor populations, we studied the electrophysiological behavior of several mutants. Given that agonist-evoked, Ca2+-mediated fluorescence imaging plate reader (FLIPR) responses from the del 347 and del preM4 mutants were absent or negligible, relative to those produced by wild-type α7 nAChRs (Supplementary Fig. 4e), we concluded that severe perturbations of the pre-M4 helix are detrimental for channel activity. Accordingly, we focused on the I436A point mutation and performed functional recordings in the absence of NACHO, though transfected cells were incubated at 30 °C to enhance receptor surface expression (e.g. ref. 34). Consistent with α-Bgt staining, coexpression of Mcl-1 dramatically enhanced wild-type α7 peak currents over ten-fold from 34 ± 15 pA (n = 22) to 565 ± 256 pA (n = 13), whereas I436A peak currents were only modestly and nonsignificantly increased from 392 ± 97 pA (n = 17) to 709 ± 382 pA (n = 9; Fig. 4e, f).
Regulation of α4β2 nAChRs by Bcl-2 proteins
Another interesting facet of the α7 I436A mutation is that the equivalent position in most neuronal nAChR subunits contains an alanine residue (Supplementary Fig. 5a). As a result, one would expect any potential enhancement of receptor assembly by Bcl-2 proteins to be lessened by the reduced hydrophobicity of this position. To test this idea, we examined α4β2 receptors since they constitute the most abundant nicotinic subtype in the brain3. Staining of nonpermeabilized HEK293T cells cotransfected with α4 and a C-terminal, HA-tagged β2 construct revealed the presence of surface α4β2 receptors, because neither subunit expresses on the cell membrane when transfected alone (e.g. ref. 35). Cotransfection of Bcl-XL, along with α4, β2-HA, and NACHO modestly increased surface HA staining by 1.2 ± 0.2 fold (n = 4), whereas Mcl-1 yielded no apparent increase (0.9 ± 0.1 fold, n = 4; Supplementary Fig. 5b–d). Whole-cell peak current responses to ACh were also increased from 327 ± 67 pA (n = 26) to 693 ± 242 pA (n = 18) when α4β2 was coexpressed with Bcl-XL, though Ric-3 coexpression yielded greater current amplitudes (1178 ± 239 pA, n = 22) and Mcl-1, in contrast, actually reduced functional expression (115 ± 31 pA, n = 18, Supplementary Fig. 5e, f).
If the modest effects of Bcl-XL on α4β2 nAChRs were due to interactions with the pre-M4 helix, we reasoned that they could be enhanced by mutating the alanine residues at positions 567 (α4) and 427 (β2) to isoleucine, as found in the α7 subunit. Likewise, deletion of the pre-M4 helix should prevent any upregulation. Expressed on their own, α4 A567I/β2 A427I mutant receptors exhibited reduced surface staining compared to wild-type receptors, though the effect of Bcl-2 protein coexpression was increased in comparison to wild-type α4β2 (Supplementary Fig. 5b–d). The pre-M4 deletion meanwhile yielded much greater basal α4β2 surface expression, while remaining insensitive to Bcl-2 proteins, a similar effect to that observed for the equivalent deletion in α7 subunits (Supplementary Fig. 5b–d).
Mcl-1 can regulate neuronal α7 nAChR assembly
We next explored whether endogenously expressed α7 nAChRs can be regulated by Bcl-2 proteins, similar to recombinant receptors. Accordingly, we transduced cultured hippocampal neurons with lentivirus encoding Mcl-1, and assayed α7 nAChR surface expression using fluorescent α-Bgt staining and Ca2+-mediated FLIPR responses. Mcl-1-specific staining intensity essentially doubled among neurons at multiplicity of infection (MOI) 100, versus nontransduced controls (Supplementary Fig. 6a), indicating that widespread recombinant expression had occurred. Given the relatively low expression of α7 nAChRs in cultured neurons, we performed experiments around DIV 20 when protein levels are maximal (see ref. 36). Surface α-Bgt staining increased roughly twofold in neurons transduced at MOI 30 or 100 (Fig. 5a, b). As a positive control, incubation of neurons in 100 μM nicotine over several days produced robust upregulation of the α-Bgt signal, as described previously for cultured neurons (Fig. 5a, b) (e.g. ref. 37). In contrast, neither surface staining of the AMPA-type glutamate receptor subunit GluA1 nor the mean number of nuclei significantly changed in Mcl-1-transduced neurons (Fig. 5a, b; Supplementary Fig. 6b). This demonstrates that increases in α-Bgt staining are not likely to reflect a general enhancement of protein biogenesis or cell health in the presence of an antiapoptotic signaling factor.
Mcl-1 transduction also brought about a twofold increase in the nicotine-evoked FLIPR signal recorded from cultured neurons, comparable to the effect of preincubation in the same agonist (Supplementary Fig. 6c, d). These FLIPR responses were dependent on the presence of the α7-specific-positive allosteric modulator (PAM) PNU-12059615 and blocked by inclusion of the competitive antagonist methyllycaconitine (MLA) (Supplementary Fig. 6c), demonstrating that they were in fact mediated by α7 nAChR activity.
Several antiapoptotic Bcl-2 members are abundantly expressed in the brain38,39 and they may regulate the surface expression of neuronal α7 nAChRs. We addressed this possibility with the A11 and S63 compounds, which inhibit BH3 domain binding to Bcl-XL and Mcl-1, respectively. Because these compounds are toxic at high concentrations, we co-applied each inhibitor at 10 nM, a concentration that mitigated neuronal toxicity (Supplementary Fig. 6b), but still reduced Bcl-2-mediated enhancement of α7 nAChR assembly in transfected HEK293T cells (Supplementary Table 1). Used at these concentrations, together with nicotine to augment α7 receptor assembly, the BH3 mimetic compounds significantly reduced upregulation of surface α-Bgt staining intensity from 262 ± 54% to 156 ± 43% (n = 4) of control levels. Meanwhile surface GluA1 staining of nicotine-treated neurons was not significantly changed by Bcl-2 inhibitors (n = 4) (Fig. 5c, d).
In addition to neurons, functional α7 nAChRs have also been reported in astrocytes (e.g. refs. 40,41). We therefore tested whether these receptor populations may be similarly regulated by Bcl-2 proteins. Incubation with the Bcl-2 inhibitors A11, S63, and ABT-737 reduced the nicotine-evoked FLIPR signal from cultured astrocytes, versus untreated cells (Supplementary Fig. 6e, f), suggesting Bcl-2 proteins can regulate α7 nAChR expression in multiple cell types where they are endogenously coexpressed.
This study advances our understanding of nAChR biology in several fundamental ways. First, we discover antiapoptotic Bcl-2 proteins can upregulate the assembly and surface expression of α7 nAChRs. In the presence of the essential α7 chaperone NACHO, several Bcl-2 members consistently enhanced α-Bgt labeling of assembled receptors to an extent comparable with Ric-3. Yet even without NACHO, coexpression of Bcl-2 proteins enabled reliable current responses to be detected from α7 subunits that would otherwise exhibit no functional activity. Second, we demonstrate that α7 nAChRs possess a BH3-like motif necessary for their regulation by antiapoptotic Bcl-2 proteins. This motif is situated within a helical segment of the α7 TM3-TM4 intracellular loop, and mutating this site eliminates Bcl-2 member-mediated, but not Ric-3-mediated receptor upregulation. Finally, we provide evidence that neuronal α7 nAChR activity is influenced by the presence of Bcl-2 proteins. In this regard, overexpression of Mcl-1 in cultured neurons produced a twofold increase in surface α-Bgt labeling and Ca2+-mediated FLIPR responses. Overall, our data provide the first evidence for ion channel biogenesis to be regulated through molecular interactions with Bcl-2 proteins.
Although the α7 M3-M4 intracellular loop comprises nearly one-third of the mature protein, deleting most of this domain does not abolish channel function32. This suggests the intracellular region might serve as a substrate for other regulatory proteins. Interestingly, the amphipathic pre-M4 helix has been speculated as an interaction site for Ric-3, as mutations in that segment were found to reduce Ric-3-mediated upregulation of α-Bgt binding42. Our data, however, indicate Ric-3 can still facilitate assembly with the same residues absent in larger deletions, meaning that its interaction with α7 subunits does not depend solely on the pre-M4 region. Another interesting facet of the pre-M4 helix is that point mutations along the motif can enhance α7 nAChR expression in oocytes33. We confirmed this effect in mammalian cells. Particularly interesting is pre-M4 residue 436, which is alanine in many nAChR subunits like α2-4, β2, and β4, capable of assembly in HEK293T cells, but is isoleucine in α7. Remarkably, the α7 I436A mutant assembles much more readily in HEK293T cells, while losing most upregulation by Bcl-2 proteins. Because alanine at that position can be tolerated, it might seem counterintuitive for α7 subunits to retain a residue that hinders their assembly, though regulation by Bcl-2 protein chaperones may explain this idiosyncrasy. A partial snapshot of the pre-M4 helix is resolved in the Torpedo muscle nAChR43, where the helices of each subunit become increasingly close as distance from the membrane increases. Because residues from one pre-M4 helix abut those of the adjacent subunit, it is conceivable that steric clashes within this site hinder assembly under normal circumstances. It has also been postulated that α7 subunits possess an ER retention motif (RFR) within residues 446–448 42, as the RXR motif is known to limit forward trafficking in other membrane proteins44. However, the pre-M4 deletion encompassing this area produced similar enhancements of α4β2 and α7 nAChR surface expression, despite the motif not being conserved outside the α7 subunit. This suggests that segments adjacent to the RFR sequence have a much greater influence on receptor trafficking. Accordingly, a plausible explanation for Bcl-2 family-mediated upregulation is not via masking a retention signal, but rather through binding pre-M4 helices, holding apart the intracellular domains of the five subunits to facilitate assembly in an analogous manner as the pre-M4 deletion mutant.
The Bcl-2 protein family contains both antiapoptotic and proapoptotic/BH3 only members, which interact with each other through conserved BH1-4 domains20. Generally, the antiapoptotic proteins possess a hydrophobic groove formed by their BH1-3 segments that accommodates the helical BH3 ligand of proapoptotic binding partners26. Outside of the core Bcl-2 family, a growing number of unrelated, BH3 motif-containing proteins have been identified. While these noncanonical BH3-containing proteins are regulated by Bcl-2 proteins, they are not directly involved in apoptosis induction45. For example, the enzyme tissue transglutaminase46, autophagy mediator Beclin-147, and transcription regulator SUFU48 all possess BH3-like sequences through which Bcl-2 interactions regulate their activity. Bcl-2 and related proteins also interact with ryanodine and inositol trisphosphate (IP3) receptor channels and inhibit their calcium release49. This regulation is independent of the BH domain binding groove and does not involve receptor assembly50. Indeed, Bcl-2 family proteins have not previously been found to mediate ion channel assembly. Ryanodine and IP3 receptors reside in the ER, and Bcl-2 proteins are present in ER as well as mitochondria51. Interestingly, nAChRs undergo protein folding in the ER52, and we find anchoring in the ER membrane is critical for Mcl-1-mediated upregulation of α7 subunit assembly (Fig. 3). Our experiments using Bcl-2 inhibitors to suppress α7 receptor expression on both neurons and astrocytes are also consistent with the idea that some Bcl-2 proteins localize to the ER in native systems.
Antiapoptotic Bcl-2 proteins control neuronal viability in a variety of different contexts53,54,55. Meanwhile, cholinergic signaling, particularly through α7 nAChRs, also promotes the survival and synaptic integration of adult-born neurons22, leading to synergistic effects that maintain certain cell populations. For instance, the survival of tissue around ischemic brain lesions is notably enhanced by neuregulin-1 56,57, a highly variable (by alternative splicing) signaling protein, which is upregulated in injured cerebral tissue58. The treatment of neurons and microglia with exogenous neuregulin-1 increases α7 nAChR protein expression59,60, though the mechanism of this effect is unclear. Because neuregulin-1 also increases transcript levels of Bcl-2 following ischemia61, its effect on cell survival and α7 nAChR upregulation might stem from molecular interactions between Bcl-2 proteins and nicotinic receptors. Despite being beneficial for injured tissue, the antiapoptotic functions of Bcl-2 proteins can instead be problematic in cancer, where cell proliferation is unchecked by apoptosis. Nicotine and other nAChR agonists found in cigarette smoke are known to enhance Bcl-2 and Mcl-1 phosphorylation in lung cancer cell lines62,63,64, as well as reduce ubiquitin-dependent Bcl-2 degradation65, resulting in enhanced chemotherapeutic resistance. Furthermore, nicotine promotes tumor growth and metastasis in mouse models of lung cancer66. It is noteworthy that α7 nAChR activation can mediate Bcl-2 phosphorylation in certain instances above (i.e. ref. 63), because nicotine also increases α7 subunit expression in lung cancer cells67, which could provide a positive feedback mechanism. Enhanced Bcl-2 member expression could facilitate α7 subunit assembly and trafficking, leading to even more nicotine-mediated augmentation of Bcl-2 activity and the preservation of cells in an antiapoptotic state. Ultimately, more work will be needed to demonstrate a direct interaction between Bcl-2 proteins and α7 subunits during the physiological regulation of apoptosis, though our experiments demonstrate α7 nAChR expression is enhanced several fold when antiapoptotic Bcl-2 members are abundant.
Genes, molecular biology, and cell culture
The genes studied here are all human forms: CHRNA7, variant 1 (NM_000746.5), CHRNA4, variant 1 (NM_000744.6), CHRNB2 (NM_000748.2); TMEM35A, or NACHO (NM_021637.2); Ric-3, variant 1 (NM_024557.5); BCL2, variant alpha (NM_000633.2); BCL2L1, or Bcl-XL (NM_138578.2); BCL2L2, or Bcl-W (NM_004050.4); MCL1, variant 1 (NM_021960.4); BBC3, or Puma, variant 4 (NM_014417.4); PMAIP1, or Noxa (NM_021127.2); CYB5A, variant 1 (NM_148923.3). Mutant proteins containing substitutions, insertions, and deletions were generated by site-directed mutagenesis with two complementary primer reactions run in parallel, and then confirmed by sequencing. Deletion constructs used in this study, with deleted residues in parentheses are: α7 del 347 (348–454), α7 del preM4 (432–450), α4 del preM4 (563–581), β2 del preM4 (423–441), Mcl-1 or M-ΔTM (328−). The M-CytB5 fusion construct was generated using primers with ends complementary to the cytosolic-transmembrane interface of Mcl-1, and a large central insert containing the CytB5 TM segment followed by a stop codon. The forward primer sequence was as follows: 5′-CAT GTA GAG GAC CTA GAA GGT GGC AGC TGG TGG ACC AAC TGG GTG ATT CCG GCG ATT AGC GCG GTG GCG GTG GCG CTG ATG TAT CGC CTG TAT ATG GCG GAA GAT TAG ATC AGG AAT GTG CTG CTG GCT TTT GC-3′. Other primers used to generate mutations can be found in a supplementary data file. The α7-HA and β2-HA constructs contained a PSGA linker and HA tag immediately following wild-type C-terminal residues A502 and K502, respectively. Puma and Noxa expression vectors were ordered from Origene, and encoded proteins were tagged by a C-terminal Myc-DDK tag. Residue numbering of all receptors includes the signal peptide. The Mcl-1-encoding Lentiviral vector contained the EF1α promoter for transgene expression, and viral particles were packaged by VectorBuilder. MOI was calculated based on a reported titer of 2.17 × 106 U/mL.
HEK293T cells used for recombinant expression assays were cultured in DMEM medium supplemented with 10% FBS and 1 mM sodium pyruvate. Transient transfections were performed using FuGENE® HD or FuGENE® 6 transfection reagents (Promega Corporation). Unless otherwise stated, assays were performed after incubating cells for 48–72 h at 37 °C following transfection.
Primary hippocampal neurons were prepared from E18 rat hippocampi (BrainBits). Dissociation was performed using digestion in 10 U/mL papain (Worthington) for 20 min, followed by trituration with a 10 mL glass pipette. The cell suspension was plated at 10,000 cells per well (384-well plates) and maintained in NbActiv4® media (BrainBits). Cultures were transduced with Mcl-1 encoding Lentivirus at 6–8 days in vitro (DIV) and surface staining or FLIPR assays were performed at 18–21 DIV, following 4–6 days incubation in nicotine for pretreated cultures. Astrocytes used for FLIPR assays were plated at 15,000 cells per well (384-well plates), after being isolated from mouse brains and cultured according to published methods68.
Broad cDNA library screen
The cDNA library used for screening was purchased from the Broad Institute and contains 17,255 cDNA clones. Each cDNA was previously subcloned into the pLX-317 vector, and encodes a V5 tag at the C-terminus of the translated protein. The library was screened using 384-well plates, where each well was transfected with 60 ng of DNA, comprised of multiple expression vectors encoding α7 nAChR, NACHO, and a single Broad gene product at a 1:3:4 ratio. Surface α-Bgt labeling was assessed 48 h post transfection, when cells were incubated for 45 min at 37 °C with a fluorescent α-Bgt conjugate (AlexaFluor 647, Invitrogen™), prior to fixation in 4% paraformaldehyde (PFA) and imaging. A cDNA encoding a candidate α7 regulatory protein was classified as a hit if the increase in surface α-Bgt fluorescence intensity (averaged per cell) was greater than 31.0% of that induced by Ric-3 coexpression, used as a positive control. This threshold corresponded to three standard deviations from the mean change in α-Bgt intensity (−1.65%), resulting in a hit rate of 0.6%.
HEK293T cells were initially plated on uncoated plastic six-well plates, and 24 h prior to recording were treated for 5 min with CellStripper™ dissociation reagent (Corning) before being re-plated on 12 mm glass coverslips. During recording, cells were submerged in external solution composed of HyClone™ HEPES-buffered saline (GE Life Sciences) comprising 149 mM NaCl, 4 mM KCl, 10 mM HEPES, and 5 mM Glucose at pH 7.4 and 300 mOsm osmolality, supplemented with 2 mM CaCl2 and 1 mM MgCl2. Internal recording solution contained 145 mM CsCl, 2.5 mM NaCl, 10 mM HEPES, 4 mM MgATP, and 1.0 mM EGTA, with pH adjusted to 7.4 using CsOH. Osmolality was adjusted to 300 mOsm/kg using sucrose and measured with the Vapro osmometer (Wescor Inc.). Whole-cell recordings were made using the lifted-cell patch-clamp technique, with constant negative pressure applied into the recording pipette chamber to facilitate cell lifting (see ref. 69). Recording pipettes were composed of borosilicate glass (World Precision Instruments) shaped to achieve 3–5 MΩ resistance using the PC-10 micropipette puller (Narishige). Agonist-evoked peak responses were measured by rapidly applying 3 mM acetylcholine (ACh) to eGFP-expressing cells, using the MXPZT piezoelectric actuator system (Siskiyou), at a membrane holding potential of −60 mV, unless otherwise indicated. Series resistances (3–15 MΩ) were routinely compensated by at least 80%. All recordings were performed using an Axopatch 200B amplifier (Molecular Devices). Signals were sampled at 25 kHz and low-pass filtered at 10 kHz. Data were acquired and analyzed using pClamp10 software (Molecular Devices). Average peak current responses include data from GFP-expressing, nonresponsive cells, unless noted.
Immunocytochemistry and image analysis
Cells were either plated on 384-well BioCoat (Corning) or Cell Carrier Ultra clear-bottom microplates (PerkinElmer) that were precoated with poly-d-lysine. For HEK293T staining, fluorescent α-Bgt conjugates (2 μg/mL, AlexaFluor 488, Invitrogen™) or primary antibodies were incubated in culture media for 45 min at 37 °C prior to fixation and permeabilization to label surface proteins. All subsequent steps cells were performed in PBS at room temperature. In order, cells were fixed with 4% PFA for 60 min, permeabilized using 0.3% Triton X-100 (Sigma) for 15 min, and blocked in 10% normal goat serum for 30 min. For neuron staining, all steps were carried out with cells kept in HyClone™ HEPES-buffered saline (GE Life Sciences), described above, at room temperature. First, cells were fixed with 4% PFA for 60 min. Subsequently, fluorescent α-Bgt conjugates (1 μg/mL) or primary antibodies were incubated with cells for 60 min to enable surface staining, before simultaneous permeabilization and blocking in 0.2% Triton X-100 and 10% normal goat serum for 60 min. To label intracellular components, primary and secondary antibody incubation occurred sequentially for 1–2 h each. Prior to imaging, nuclei were stained with DAPI or NucBlue® reagent (Invitrogen™).
For quantitative imaging of stained cells, images were acquired using the Opera Phenix™ or Operetta™ screening system (PerkinElmer) with a 20× water objective. In every experiment five fields were acquired per well with Harmony™ high-content imaging software (PerkinElmer). Further analysis was done with the Columbus™ data storage and analysis system (PerkinElmer). Nuclei from viable cells were identified from the DAPI channel as having an area >30 μm2 for HEK293T cells or an area >10 μm2 for neurons. Cell boundaries were defined based on nuclear staining, and mean intensities were calculated from all regions of interest. Area of the cytoplasm was calculated based on the cytoplasm region of interest. Reported fluorescence intensities are averages of individually transfected or treated wells, where the mean intensity (across all identified cells) has been calculated, after subtracting the background signal from α-Bgt, HA, or antibody labeling of untransfected (HEK293T) or epibatidine cotreated (neurons) cells.
Primary antibodies used include Bcl-2 (Rb, 1:300, CST D17C4), Bcl-XL (Rb, 1:300, CST 54H6), Bcl-W (Rb, 1:300, CST 31H4), Mcl-1 (Rb, 1:300, CST D2W9E), HA (M, 1:500, Invitrogen™ 2-2.2.14) conjugated to DyLight 488 or DyLight 650, GluA1 N-terminal (M, 1:250, MilliporeSigma RH95), where M = mouse, Rb = Rabbit, and CST indicates Cell Signaling Technology. Secondary antibodies used include goat anti-mouse AlexaFluor 555 and goat anti-rabbit DyLight 650 (Invitrogen™) at 1:500 or 1:1000 dilution. In some cases, higher primary antibody concentrations were used, but kept constant within each experiment.
Epibatidine binding assays
Transfected HEK293T cells were harvested and resuspended in 50 mM cold TrisHCl buffer (pH 7.4). Cell membrane preparations were made by centrifugation at 33,000 × g for 20 min at 4 °C. Membrane pellets were homogenized for 30 s using the T-25 Ultra-Turrax homogenizer (Ika). Protein concentrations were determined using the Pierce™ BCA Protein Assay Kit (Thermo Scientific) and normalized across samples prior to the assay. Binding assays were performed in 96-well plates, where samples were incubated with 10 nM (for α7) [3H]epibatidine for 3 h at room temperature. Nonspecific binding was determined by coincubation with 10 μM unlabeled epibatidine. Binding assays were terminated by filtration through polyethylenimine-treated Unifilter GF/B 96-well plates (PerkinElmer), which were immediately washed with 500 mL assay buffer per plate. Filter plates were desiccated at 65 °C for 20 min and incubated with 50 μL MicroScint-0 scintillant cocktail (PerkinElmer). Bound [3H]epibatidine was quantified using the TopCount NXT scintillation counter (PerkinElmer).
FLIPR assay for Ca2+ influx
High-throughput FLIPR assays were performed using 384-well BioCoat (Corning) plates. Assay buffer consisted of the HyClone™ HEPES-buffered saline (GE Life Sciences) described above, though prior to recordings, cells were loaded with Calcium 5 dye (Molecular Devices) for 1 h at room temperature. Plates were washed to remove excess dye, and then placed in the FLIPR Tetra (Molecular Devices) chamber. Recordings were captured using ScreenWorks 4.0™ software (Molecular Devices). Obtaining robust α7-mediated FLIPR responses required the presence of a selective PAM to attenuate desensitization, either 50 μM NS-1738 70 for HEK293T cells or 5 μM PNU-120596 15 for neurons. Neuronal FLIPR responses were also obtained in the presence of tetrodotoxin, or TTX (500 nM), to inhibit the firing of action potentials.
Results are expressed as mean ± SEM. Analyses of sample means were performed using two-sample (assuming unequal variance) t tests and significance was determined based on two-tailed p values, unless otherwise indicated. The significance level α = 0.05 was used, though for multiple comparisons the Bonferroni correction was applied, such that the significance threshold became α/n, where n is the number of comparisons.
Further information on research design is available in the Nature Research Reporting Summary linked to this article.
The data that support the findings of this study are available from the authors upon request. Please see author contributions for specific data sets.
Albuquerque, E. X., Pereira, E. F., Alkondon, M. & Rogers, S. W. Mammalian nicotinic acetylcholine receptors: from structure to function. Physiol. Rev. 89, 73–120 (2009).
Le Novere, N., Corringer, P. J. & Changeux, J. P. The diversity of subunit composition in nAChRs: evolutionary origins, physiologic and pharmacologic consequences. J. Neurobiol. 53, 447–456 (2002).
Gotti, C., Zoli, M. & Clementi, F. Brain nicotinic acetylcholine receptors: native subtypes and their relevance. Trends Pharm. Sci. 27, 482–491 (2006).
Couturier, S. et al. A neuronal nicotinic acetylcholine receptor subunit (alpha 7) is developmentally regulated and forms a homo-oligomeric channel blocked by alpha-BTX. Neuron 5, 847–856 (1990).
Li, S. X. et al. Ligand-binding domain of an alpha7-nicotinic receptor chimera and its complex with agonist. Nat. Neurosci. 14, 1253–1259 (2011).
Miwa, J. M., Freedman, R. & Lester, H. A. Neural systems governed by nicotinic acetylcholine receptors: emerging hypotheses. Neuron 70, 20–33 (2011).
Bouzat, C., Bartos, M., Corradi, J. & Sine, S. M. The interface between extracellular and transmembrane domains of homomeric Cys-loop receptors governs open-channel lifetime and rate of desensitization. J. Neurosci. 28, 7808–7819 (2008).
Seguela, P., Wadiche, J., Dineley-Miller, K., Dani, J. A. & Patrick, J. W. Molecular cloning, functional properties, and distribution of rat brain alpha 7: a nicotinic cation channel highly permeable to calcium. J. Neurosci. 13, 596–604 (1993).
Picciotto, M. R. et al. Neuronal nicotinic acetylcholine receptor subunit knockout mice: physiological and behavioral phenotypes and possible clinical implications. Pharm. Ther. 92, 89–108 (2001).
Pandya, A. A. & Yakel, J. L. Effects of neuronal nicotinic acetylcholine receptor allosteric modulators in animal behavior studies. Biochem Pharm. 86, 1054–1062 (2013).
Dineley, K. T., Pandya, A. A. & Yakel, J. L. Nicotinic ACh receptors as therapeutic targets in CNS disorders. Trends Pharm. Sci. 36, 96–108 (2015).
Bertrand, D., Lee, C. H., Flood, D., Marger, F. & Donnelly-Roberts, D. Therapeutic potential of alpha7 nicotinic acetylcholine receptors. Pharm. Rev. 67, 1025–1073 (2015).
Cooper, S. T. & Millar, N. S. Host cell-specific folding and assembly of the neuronal nicotinic acetylcholine receptor alpha7 subunit. J. Neurochem. 68, 2140–2151 (1997).
Kassner, P. D. & Berg, D. K. Differences in the fate of neuronal acetylcholine receptor protein expressed in neurons and stably transfected cells. J. Neurobiol. 33, 968–982 (1997).
Hurst, R. S. et al. A novel positive allosteric modulator of the alpha7 neuronal nicotinic acetylcholine receptor: in vitro and in vivo characterization. J. Neurosci. 25, 4396–4405 (2005).
Gu, S. et al. Brain alpha7 nicotinic acetylcholine receptor assembly requires NACHO. Neuron 89, 948–955 (2016).
Matta, J. A. et al. NACHO mediates nicotinic acetylcholine receptor function throughout the brain. Cell Rep. 19, 688–696 (2017).
Huang, S. et al. Complex between alpha-bungarotoxin and an alpha7 nicotinic receptor ligand-binding domain chimaera. Biochem. J. 454, 303–310 (2013).
Halevi, S. et al. The C. elegans ric-3 gene is required for maturation of nicotinic acetylcholine receptors. EMBO J. 21, 1012–1020 (2002).
Youle, R. J. & Strasser, A. The BCL-2 protein family: opposing activities that mediate cell death. Nat. Rev. Mol. Cell Biol. 9, 47–59 (2008).
Cory, S., Roberts, A. W., Colman, P. M. & Adams, J. M. Targeting BCL-2-like proteins to kill cancer cells. Trends Cancer 2, 443–460 (2016).
Campbell, N. R., Fernandes, C. C., Halff, A. W. & Berg, D. K. Endogenous signaling through alpha7-containing nicotinic receptors promotes maturation and integration of adult-born neurons in the hippocampus. J. Neurosci. 30, 8734–8744 (2010).
Belluardo, N., Mudo, G., Blum, M. & Fuxe, K. Central nicotinic receptors, neurotrophic factors and neuroprotection. Behav. Brain Res. 113, 21–34 (2000).
Merry, D. E. & Korsmeyer, S. J. Bcl-2 gene family in the nervous system. Annu. Rev. Neurosci. 20, 245–267 (1997).
Yang, X. et al. A public genome-scale lentiviral expression library of human ORFs. Nat. Methods 8, 659–661 (2011).
Moldoveanu, T., Follis, A. V., Kriwacki, R. W. & Green, D. R. Many players in BCL-2 family affairs. Trends Biochem. Sci. 39, 101–111 (2014).
Oltersdorf, T. et al. An inhibitor of Bcl-2 family proteins induces regression of solid tumours. Nature 435, 677–681 (2005).
Lee, E. F. et al. Crystal structure of ABT-737 complexed with Bcl-xL: implications for selectivity of antagonists of the Bcl-2 family. Cell Death Differ. 14, 1711–1713 (2007).
Kotschy, A. et al. The MCL1 inhibitor S63845 is tolerable and effective in diverse cancer models. Nature 538, 477–482 (2016).
Tao, Z. F. et al. Discovery of a potent and selective BCL-XL inhibitor with in vivo activity. ACS Med. Chem. Lett. 5, 1088–1093 (2014).
Sattler, M. et al. Structure of Bcl-xL-Bak peptide complex: recognition between regulators of apoptosis. Science 275, 983–986 (1997).
Valor, L. M. et al. Role of the large cytoplasmic loop of the alpha 7 neuronal nicotinic acetylcholine receptor subunit in receptor expression and function. Biochemistry 41, 7931–7938 (2002).
Castelan, F. et al. Cytoplasmic regions adjacent to the M3 and M4 transmembrane segments influence expression and function of alpha7 nicotinic acetylcholine receptors. A study with single amino acid mutants. J. Neurochem. 100, 406–415 (2007).
Cooper, S. T., Harkness, P. C., Baker, E. R. & Millar, N. S. Up-regulation of cell-surface alpha4beta2 neuronal nicotinic receptors by lower temperature and expression of chimeric subunits. J. Biol. Chem. 274, 27145–27152 (1999).
Ren, X. Q. et al. Structural determinants of alpha4beta2 nicotinic acetylcholine receptor trafficking. J. Neurosci. 25, 6676–6686 (2005).
Arnaiz-Cot, J. J. et al. Allosteric modulation of alpha 7 nicotinic receptors selectively depolarizes hippocampal interneurons, enhancing spontaneous GABAergic transmission. Eur. J. Neurosci. 27, 1097–1110 (2008).
Kawai, H. & Berg, D. K. Nicotinic acetylcholine receptors containing alpha 7 subunits on rat cortical neurons do not undergo long-lasting inactivation even when up-regulated by chronic nicotine exposure. J. Neurochem. 78, 1367–1378 (2001).
Hamner, S., Skoglosa, Y. & Lindholm, D. Differential expression of bcl-w and bcl-x messenger RNA in the developing and adult rat nervous system. Neuroscience 91, 673–684 (1999).
Mori, M. et al. Expression of apoptosis inhibitor protein Mcl1 linked to neuroprotection in CNS neurons. Cell Death Differ. 11, 1223–1233 (2004).
Sharma, G. & Vijayaraghavan, S. Nicotinic cholinergic signaling in hippocampal astrocytes involves calcium-induced calcium release from intracellular stores. Proc. Natl Acad. Sci. USA 98, 4148–4153 (2001).
Shen, J. X. & Yakel, J. L. Functional alpha7 nicotinic ACh receptors on astrocytes in rat hippocampal CA1 slices. J. Mol. Neurosci. 48, 14–21 (2012).
Castillo, M. et al. Dual role of the RIC-3 protein in trafficking of serotonin and nicotinic acetylcholine receptors. J. Biol. Chem. 280, 27062–27068 (2005).
Unwin, N. Refined structure of the nicotinic acetylcholine receptor at 4A resolution. J. Mol. Biol. 346, 967–989 (2005).
Michelsen, K., Yuan, H. & Schwappach, B. Hide and run. Arginine-based endoplasmic-reticulum-sorting motifs in the assembly of heteromultimeric membrane proteins. EMBO Rep. 6, 717–722 (2005).
Aouacheria, A., Combet, C., Tompa, P. & Hardwick, J. M. Redefining the BH3 death domain as a ‘short linear motif’. Trends Biochem. Sci. 40, 736–748 (2015).
Rodolfo, C. et al. Tissue transglutaminase is a multifunctional BH3-only protein. J. Biol. Chem. 279, 54783–54792 (2004).
Maiuri, M. C. et al. Functional and physical interaction between Bcl-X(L) and a BH3-like domain in Beclin-1. EMBO J. 26, 2527–2539 (2007).
Wu, X. et al. Extra-mitochondrial prosurvival BCL-2 proteins regulate gene transcription by inhibiting the SUFU tumour suppressor. Nat. Cell Biol. 19, 1226–1236 (2017).
Vervliet, T. et al. Bcl-2 binds to and inhibits ryanodine receptors. J. Cell Sci. 127, 2782–2792 (2014).
Vervliet, T. et al. Regulation of the ryanodine receptor by anti-apoptotic Bcl-2 is independent of its BH3-domain-binding properties. Biochem. Biophys. Res. Commun. 463, 174–179 (2015).
Krajewski, S. et al. Investigation of the subcellular distribution of the bcl-2 oncoprotein: residence in the nuclear envelope, endoplasmic reticulum, and outer mitochondrial membranes. Cancer Res. 53, 4701–4714 (1993).
Smith, M. M., Lindstrom, J. & Merlie, J. P. Formation of the alpha-bungarotoxin binding site and assembly of the nicotinic acetylcholine receptor subunits occur in the endoplasmic reticulum. J. Biol. Chem. 262, 4367–4376 (1987).
Arbour, N. et al. Mcl-1 is a key regulator of apoptosis during CNS development and after DNA damage. J. Neurosci. 28, 6068–6078 (2008).
Courchesne, S. L., Karch, C., Pazyra-Murphy, M. F. & Segal, R. A. Sensory neuropathy attributable to loss of Bcl-w. J. Neurosci. 31, 1624–1634 (2011).
Nakamura, A. et al. Bcl-xL is essential for the survival and function of dfferentiated neurons in the cortex that control complex behaviors. J. Neurosci. 36, 5448–5461 (2016).
Kaneko, S. et al. Nicotine protects cultured cortical neurons against glutamate-induced cytotoxicity via alpha7-neuronal receptors and neuronal CNS receptors. Brain Res. 765, 135–140 (1997).
Shyu, W. C. et al. Neuregulin-1 reduces ischemia-induced brain damage in rats. Neurobiol. Aging 25, 935–944 (2004).
Tokita, Y. et al. Regulation of neuregulin expression in the injured rat brain and cultured astrocytes. J. Neurosci. 21, 1257–1264 (2001).
Liu, Y., Ford, B., Mann, M. A. & Fischbach, G. D. Neuregulins increase alpha7 nicotinic acetylcholine receptors and enhance excitatory synaptic transmission in GABAergic interneurons of the hippocampus. J. Neurosci. 21, 5660–5669 (2001).
Mencel, M., Nash, M. & Jacobson, C. Neuregulin upregulates microglial alpha7 nicotinic acetylcholine receptor expression in immortalized cell lines: implications for regulating neuroinflammation. PLoS ONE 8, e70338 (2013).
Guo, W. P., Fu, X. G., Jiang, S. M. & Wu, J. Z. Neuregulin-1 regulates the expression of Akt, Bcl-2, and Bad signaling after focal cerebral ischemia in rats. Biochem. Cell Biol. 88, 649–654 (2010).
Mai, H., May, W. S., Gao, F., Jin, Z. & Deng, X. A functional role for nicotine in Bcl2 phosphorylation and suppression of apoptosis. J. Biol. Chem. 278, 1886–1891 (2003).
Jin, Z., Gao, F., Flagg, T. & Deng, X. Tobacco-specific nitrosamine 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone promotes functional cooperation of Bcl2 and c-Myc through phosphorylation in regulating cell survival and proliferation. J. Biol. Chem. 279, 40209–40219 (2004).
Zhao, J., Xin, M., Wang, T., Zhang, Y. & Deng, X. Nicotine enhances the antiapoptotic function of Mcl-1 through phosphorylation. Mol. Cancer Res. 7, 1954–1961 (2009).
Nishioka, T. et al. Nicotine increases the resistance of lung cancer cells to cisplatin through enhancing Bcl-2 stability. Br. J. Cancer 110, 1785–1792 (2014).
Davis, R. et al. Nicotine promotes tumor growth and metastasis in mouse models of lung cancer. PLoS ONE 4, e7524 (2009).
Brown, K. C. et al. Nicotine induces the up-regulation of the alpha7-nicotinic receptor (alpha7-nAChR) in human squamous cell lung cancer cells via the Sp1/GATA protein pathway. J. Biol. Chem. 288, 33049–33059 (2013).
He, Y., Taylor, N. & Bhattacharya, A. Isolation and culture of astrocytes from postnatal and adult mouse brains. Methods Mol. Biol. 1938, 37–47 (2019).
Glasgow, N. G. & Johnson, J. W. Whole-cell patch-clamp analysis of recombinant NMDA receptor pharmacology using brief glutamate applications. Methods Mol. Biol. 1183, 23–41 (2014).
Timmermann, D. B. et al. An allosteric modulator of the alpha7 nicotinic acetylcholine receptor possessing cognition-enhancing properties in vivo. J. Pharm. Exp. Ther. 323, 294–307 (2007).
The authors thank Yingbo He for the preparation of astrocytes.
The authors declare no competing interests except that authors are all full-time employees of Janssen Pharmaceutical Companies of Johnson and Johnson.
Peer review information: Nature Communications thanks Xingming Deng, Ronald Lukas and other anonymous reviewer(s) for their contribution to the peer review of this work. Peer reviewer reports are available.
Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.