Current Literature In Basic Science
E(R)-GADD! Exploring Mutations in Generalized Epilepsy
Trafficking-Deficient Mutant GABRG2 Subunit Amount May Modify Epilepsy Phenotype. Kang JQ, Shen W, Macdonald RL. Ann Neurol 2013;74:547–559.
OBJECTIVE: Genetic epilepsies and many other human genetic diseases display phenotypic heterogeneity, often for unknown reasons. Disease severity associated with nonsense mutations is dependent partially on mutation gene location and resulting efficiency of nonsense-mediated mRNA decay (NMD) to eliminate potentially toxic proteins. Nonsense mutations in the last exon do not activate NMD, thus producing truncated proteins. We compared the protein metabolism and the impact on channel biogenesis, function, and cellular homeostasis of truncated γ2 subunits produced by GABRG2 nonsense mutations associated with epilepsy of different severities and by a nonsense mutation in the last exon unassociated with epilepsy. METHODS: γ-Aminobutyric acid type A receptor subunits were coexpressed in non-neuronal cells and neurons. NMD was studied using minigenes that support NMD. Protein degradation rates were determined using (35) S radiolabeling pulse chase. Channel function was determined by whole cell recordings, and subunits trafficking and cellular toxicity were determined using flow cytometry, immunoblotting, and immunohistochemistry. RESULTS: Although all GABRG2 nonsense mutations resulted in loss of γ2 subunit surface expression, the truncated subunits had different degradation rates and stabilities, suppression of wild-type subunit biogenesis and function, amounts of conjugation with polyubiquitin, and endoplasmic reticulum stress levels. INTERPRETATION: We compared molecular phenotypes of GABRG2 nonsense mutations. The findings suggest that despite the common loss of mutant allele function, each mutation produced different intracellular levels of trafficking-deficient subunits. The concentration-dependent suppression of wild-type channel function and cellular disturbance resulting from differences in mutant subunit metabolism may contribute to associated epilepsy severities and by implication to phenotypic heterogeneity in many inherited human diseases.
Commentary Trainees frequently ask how mutations in a single gene can lead to different phenotypes. Most of us answer the question (with the requisite hand waving) by restating what we all learned in high school biology: Nonsense mutations resulting in stop codons produce a truncated gene product, and missense mutations result in altered protein function. Either type of mutation can result in a lack of normal protein function, thereby causing epilepsy. The real answer is probably more subtle and elegant. The more difficult question to answer is how different nonsense mutations in the same gene lead to different epilepsy syndromes. Many nonsense mutations lead to a process involving multimeric protein complexes, known as nonsense-mediated mRNA decay (1). However, this is not always the case, particularly when mutations occur in exons near the end of the gene, as shown by Kang and coworkers in prior work (2). In their latest publication, Kang and coworkers further unravel some of the complexity underlying this deceptively simple concept by examining mutations in the GABAA receptor γ2 subunit (GABAA receptors typically consist Epilepsy Currents, Vol. 14, No. 4 (July/August) 2014 pp. 203–205 © American Epilepsy Society
of two α1, two β2 and one γ2 subunits) (3). Mutations in the γ2 subunit have been linked to syndromes across the spectrum of the epilepsies, from benign childhood absence epilepsy to the very serious Dravet syndrome (4). Regardless of the location of the nonsense mutation within the coding region (i.e., early or late exons, which ultimately code for the protein), the expression of γ2 subunits on the surface of cultured cells is dramatically decreased, compared to wild-type γ2 subunits. This raises two questions: Where did the partially translated products go? Do these truncated products adversely affect normal cellular function? To answer the first question, the authors used separate γ2 “minigenes” (consisting of intron 8 and exon 9, the last exon) with three separate point mutations in exon 9 (epilepsy-linked Q390X, W429X and epilepsy-unrelated W461X; numbering in this study reflected the immature peptide, which includes an extra 39 amino acids) to measure mRNA and protein levels in cultured cells. mRNA levels were similar for all three constructs, but intracellular protein levels of Q390X were elevated above wildtype, whereas levels of W429X were decreased and levels of W461X were virtually undetectable. The rate of synthesis of W429X and Q390X were the same as wildtype, but protein degradation rates followed a pattern that was similar to protein levels (Q390X > wild type > W429X). When co-expressed with α1 and β2 subunits in cultured cells
203
Mutations in Generalized Epilepsy
(to more closely approximate the “native” receptor complex, rather than just an isolated protein), the mutant subunits were more sensitive to glycosidase digestion, consistent with their retention in the endoplasmic reticulum (because more mature glycosylation patterns are seen when proteins are transported to the Golgi apparatus). Q390X and W429X subunits both showed increased conjugation to polyubiquitin in cultured rat cortical neuron somata (which may lead to ubiquitin-proteasome degradation and synaptic dysfunction [5]). To summarize, abnormal γ2 gene products are located in the endoplasmic reticulum and may be more susceptible to degradation – both of which may lead to abnormal neuronal firing patterns.
What is the functional impact of these findings? When α1, β2, and truncated γ2 subunits (Q390X and W429X) were expressed in cultured cells, there was a decrease in GABAA receptor currents. Aside from lack of function, could the truncated subunits be doing anything else harmful? In fact, expression of two of the truncated subunits (Q390X and W429X) led to reduced total levels of α1 subunits, which could lead to abnormal neuronal firing. Further, there was a dose-dependent decrease in α1 subunit expression with increasing relative amounts of γ2 subunits. In a more clinically relevant model, total levels of α1 subunit expression were decreased in Q390X mice engineered with the mutation but not GABRG2 knock-out mice, suggesting the loss of gene product
TABLE 1. Summary of Comparisons of γ2 Truncated Subunits Q390X (Q351X*)
W429X (W390X)
W461X (W422X)
FS/Dravet syndrome†
GEFS+
None reported
γ2 mRNA
–
–
–
γ2 protein (steady-state)
↑↑
↓
↓↓
γ2 protein degradation
↓
–/↑
n.s.
γ2 subunit mature glycosylation‡
↓↓
↓↓
n.s.
γ2 polyubiquitin conjugation (neuronal somata)
↑↑
↑
↓
γ2 subunit surface expression§
↓↓
↓↓
↓↓
GADD153 expression
↑↑
↑
↑
α1 subunit protein (steady-state)§
↓
↓
–
α1 subunit mature glycosylation†
↓
↓
–
α1 subunit protein degradation†
↑
↑
n.s.
α1 subunit surface expression§
↓
↓
–
Oligomerization (γ2/α1)
↑
–
↓
Peak current amplitudes
↓↓
↓
n.s.
Zinc sensitivity
↓↓
↓↓
n.s.
Parameter Epilepsy phenotype
Abbreviations: ↓, decreased; ↑, increased; –, unchanged; FS, febrile seizures; GEFS+, genetic epilepsy with febrile seizures plus; n.s., not shown. *Numbering in parentheses reflects the mature peptide. † The original pedigree had family members with different types of epilepsy, suggesting additional factors may have influenced the phenotype of the one patient with Dravet syndrome (10). ‡ Levels of W429X>Q390X § Levels of W461X>W429X>Q390X.
204
Mutations in Generalized Epilepsy
alone was not the only deleterious factor. One potential mechanism is organelle dysfunction. Cultured cells expressing Q390X and W429X (but not W461X) showed increased expression of the growth arrest- and DNA damage-inducible gene 153 (GADD153), one marker of endoplasmic reticulum stress and the unfolded protein response (UPR), which in turn can lead to cellular dysfunction and diversion of intracellular resources from their typical constitutive processes (7). Low levels of cell death were noted in Q390X-transfected cells but not with the other constructs. So, where does this leave our understanding of the different mutations and their corresponding phenotypes? The focus of this work was cultured HEK293T cells, cultured rat cortical neurons, and L929 cells. These data represent an important first step in defining which questions can be asked regarding intact animals, in the presence of the full circuitry of functional neurons, glia, and other cells. Are there pathological compensatory effects in other neurotransmitter receptor subunits (either GABAA, glutamate, or other channels) in response to mutations in the final exons of GABRG2? Is there evidence of cell death as a result of ER stress (including the UPR); if so, are certain cell populations (i.e., interneurons) more vulnerable than others? GADD153 has not been implicated directly in neuronal dysfunction in epilepsy but is this yet another avenue to explore? As the authors note, the amino acids missing from the Q390X mutant but present in the W429X mutant include charged residues, palmitoylation sites, tyrosine phosphorylation sites, and a GABARAP (GABA receptor binding protein) binding site. What contribution is made by each of these factors to the ultimate phenotype? GABARAP also plays a role in the later stages of macroautophagy, suggesting that there may be additional pathways of cellular dysfunction to investigate using the truncated γ2 subunits (8). Finally, findings similar to those noted by Kang and coworkers were recently reported with mutations in the second exon (Q40X), raising the question of how specific many of these cellular changes are for mutations in the final exon (9). In summary, different nonsense mutations in γ2 subunits have very different profiles of biochemical and functional abnormalities (Table 1). There are many similarities in functional outcomes (e.g., effect on α1 subunit expression, peak current amplitudes, and zinc sensitivity), although the relative levels of each may reflect subtle differences in protein stability and trafficking. The gradient of changes seen in some assays also correlates with the location of the mutation, which at some
level also may reflect their structural characteristics in neuronal membranes. Despite the advances in our understanding of the genetics of epilepsy, it is clear that many questions of cell biological nature remain to be clarified. by Adam L. Hartman, MD References 1. Leeds P, Peltz SW, Jacobson A, Culbertson MR. The product of the yeast UPF1 gene is required for rapid turnover of mRNAs containing a premature translational termination codon. Genes Dev 1991;5:2303– 2314. 2. Kang JQ, Shen W, Macdonald RL. The GABRG2 mutation, Q351X, associated with generalized epilepsy with febrile seizures plus, has both loss of function and dominant-negative suppression. J Neurosci 2009;29:2845–2856. 3. Kang JQ, Shen W, Macdonald RL. Trafficking-deficient mutant GABRG2 subunit amount may modify epilepsy phenotype. Ann Neurol 2013;74:547–559. 4. Macdonald RL, Kang JQ, Gallagher MJ. Mutations in GABAA receptor subunits associated with genetic epilepsies. J Physiol 2010;588:1861– 1869. 5. Miller RJ. How does ubiquitin regulate synapses? Let me count the ways. J Physiol 2014;592:555–556. 6. Draguhn A, Verdorn TA, Ewert M, Seeburg PH, Sakmann B. Functional and molecular distinction between recombinant rat GABAA receptor subtypes by Zn2+. Neuron 1990;5:781–788. 7. Costa RO, Lacor PN, Ferreira IL, Resende R, Auberson YP, Klein WL, Oliveira CR, Rego AC, Pereira CM. Endoplasmic reticulum stress occurs downstream of GluN2B subunit of N-methyl-d-aspartate receptor in mature hippocampal cultures treated with amyloid-beta oligomers. Aging Cell 2012;11:823–833. 8. Weidberg H, Shvets E, Shpilka T, Shimron F, Shinder V and Elazar Z. LC3 and GATE-16/GABARAP subfamilies are both essential yet act differently in autophagosome biogenesis. EMBO J 2010;29:1792–1802. 9. Ishii A, Kanaumi T, Sohda M, Misumi Y, Zhang B, Kakinuma N, Haga Y, Watanabe K, Takeda S, Okada M, Ueno S, Kaneko S, Takashima S, Hirose S. Association of nonsense mutation in GABRG2 with abnormal trafficking of GABAA receptors in severe epilepsy. Epilepsy Res 2014;108:420–432. 10. Harkin LA, Bowser DN, Dibbens LM, Singh R, Phillips F, Wallace RH, Richards MC, Williams DA, Mulley JC, Berkovic SF, Scheffer IE, Petrou S. Truncation of the GABA(A)-receptor gamma2 subunit in a family with generalized epilepsy with febrile seizures plus. Am J Hum Genet 2002;70:530–536.
205
E(R)-GADD! Exploring Mutations in Generalized Epilepsy
Trafficking-Deficient Mutant GABRG2 Subunit Amount May Modify Epilepsy Phenotype. Kang JQ, Shen W, Macdonald RL. Ann Neurol 2013;74:547–559.
OBJECTIVE: Genetic epilepsies and many other human genetic diseases display phenotypic heterogeneity, often for unknown reasons. Disease severity associated with nonsense mutations is dependent partially on mutation gene location and resulting efficiency of nonsense-mediated mRNA decay (NMD) to eliminate potentially toxic proteins. Nonsense mutations in the last exon do not activate NMD, thus producing truncated proteins. We compared the protein metabolism and the impact on channel biogenesis, function, and cellular homeostasis of truncated γ2 subunits produced by GABRG2 nonsense mutations associated with epilepsy of different severities and by a nonsense mutation in the last exon unassociated with epilepsy. METHODS: γ-Aminobutyric acid type A receptor subunits were coexpressed in non-neuronal cells and neurons. NMD was studied using minigenes that support NMD. Protein degradation rates were determined using (35) S radiolabeling pulse chase. Channel function was determined by whole cell recordings, and subunits trafficking and cellular toxicity were determined using flow cytometry, immunoblotting, and immunohistochemistry. RESULTS: Although all GABRG2 nonsense mutations resulted in loss of γ2 subunit surface expression, the truncated subunits had different degradation rates and stabilities, suppression of wild-type subunit biogenesis and function, amounts of conjugation with polyubiquitin, and endoplasmic reticulum stress levels. INTERPRETATION: We compared molecular phenotypes of GABRG2 nonsense mutations. The findings suggest that despite the common loss of mutant allele function, each mutation produced different intracellular levels of trafficking-deficient subunits. The concentration-dependent suppression of wild-type channel function and cellular disturbance resulting from differences in mutant subunit metabolism may contribute to associated epilepsy severities and by implication to phenotypic heterogeneity in many inherited human diseases.
Commentary Trainees frequently ask how mutations in a single gene can lead to different phenotypes. Most of us answer the question (with the requisite hand waving) by restating what we all learned in high school biology: Nonsense mutations resulting in stop codons produce a truncated gene product, and missense mutations result in altered protein function. Either type of mutation can result in a lack of normal protein function, thereby causing epilepsy. The real answer is probably more subtle and elegant. The more difficult question to answer is how different nonsense mutations in the same gene lead to different epilepsy syndromes. Many nonsense mutations lead to a process involving multimeric protein complexes, known as nonsense-mediated mRNA decay (1). However, this is not always the case, particularly when mutations occur in exons near the end of the gene, as shown by Kang and coworkers in prior work (2). In their latest publication, Kang and coworkers further unravel some of the complexity underlying this deceptively simple concept by examining mutations in the GABAA receptor γ2 subunit (GABAA receptors typically consist Epilepsy Currents, Vol. 14, No. 4 (July/August) 2014 pp. 203–205 © American Epilepsy Society
of two α1, two β2 and one γ2 subunits) (3). Mutations in the γ2 subunit have been linked to syndromes across the spectrum of the epilepsies, from benign childhood absence epilepsy to the very serious Dravet syndrome (4). Regardless of the location of the nonsense mutation within the coding region (i.e., early or late exons, which ultimately code for the protein), the expression of γ2 subunits on the surface of cultured cells is dramatically decreased, compared to wild-type γ2 subunits. This raises two questions: Where did the partially translated products go? Do these truncated products adversely affect normal cellular function? To answer the first question, the authors used separate γ2 “minigenes” (consisting of intron 8 and exon 9, the last exon) with three separate point mutations in exon 9 (epilepsy-linked Q390X, W429X and epilepsy-unrelated W461X; numbering in this study reflected the immature peptide, which includes an extra 39 amino acids) to measure mRNA and protein levels in cultured cells. mRNA levels were similar for all three constructs, but intracellular protein levels of Q390X were elevated above wildtype, whereas levels of W429X were decreased and levels of W461X were virtually undetectable. The rate of synthesis of W429X and Q390X were the same as wildtype, but protein degradation rates followed a pattern that was similar to protein levels (Q390X > wild type > W429X). When co-expressed with α1 and β2 subunits in cultured cells
203
Mutations in Generalized Epilepsy
(to more closely approximate the “native” receptor complex, rather than just an isolated protein), the mutant subunits were more sensitive to glycosidase digestion, consistent with their retention in the endoplasmic reticulum (because more mature glycosylation patterns are seen when proteins are transported to the Golgi apparatus). Q390X and W429X subunits both showed increased conjugation to polyubiquitin in cultured rat cortical neuron somata (which may lead to ubiquitin-proteasome degradation and synaptic dysfunction [5]). To summarize, abnormal γ2 gene products are located in the endoplasmic reticulum and may be more susceptible to degradation – both of which may lead to abnormal neuronal firing patterns.
What is the functional impact of these findings? When α1, β2, and truncated γ2 subunits (Q390X and W429X) were expressed in cultured cells, there was a decrease in GABAA receptor currents. Aside from lack of function, could the truncated subunits be doing anything else harmful? In fact, expression of two of the truncated subunits (Q390X and W429X) led to reduced total levels of α1 subunits, which could lead to abnormal neuronal firing. Further, there was a dose-dependent decrease in α1 subunit expression with increasing relative amounts of γ2 subunits. In a more clinically relevant model, total levels of α1 subunit expression were decreased in Q390X mice engineered with the mutation but not GABRG2 knock-out mice, suggesting the loss of gene product
TABLE 1. Summary of Comparisons of γ2 Truncated Subunits Q390X (Q351X*)
W429X (W390X)
W461X (W422X)
FS/Dravet syndrome†
GEFS+
None reported
γ2 mRNA
–
–
–
γ2 protein (steady-state)
↑↑
↓
↓↓
γ2 protein degradation
↓
–/↑
n.s.
γ2 subunit mature glycosylation‡
↓↓
↓↓
n.s.
γ2 polyubiquitin conjugation (neuronal somata)
↑↑
↑
↓
γ2 subunit surface expression§
↓↓
↓↓
↓↓
GADD153 expression
↑↑
↑
↑
α1 subunit protein (steady-state)§
↓
↓
–
α1 subunit mature glycosylation†
↓
↓
–
α1 subunit protein degradation†
↑
↑
n.s.
α1 subunit surface expression§
↓
↓
–
Oligomerization (γ2/α1)
↑
–
↓
Peak current amplitudes
↓↓
↓
n.s.
Zinc sensitivity
↓↓
↓↓
n.s.
Parameter Epilepsy phenotype
Abbreviations: ↓, decreased; ↑, increased; –, unchanged; FS, febrile seizures; GEFS+, genetic epilepsy with febrile seizures plus; n.s., not shown. *Numbering in parentheses reflects the mature peptide. † The original pedigree had family members with different types of epilepsy, suggesting additional factors may have influenced the phenotype of the one patient with Dravet syndrome (10). ‡ Levels of W429X>Q390X § Levels of W461X>W429X>Q390X.
204
Mutations in Generalized Epilepsy
alone was not the only deleterious factor. One potential mechanism is organelle dysfunction. Cultured cells expressing Q390X and W429X (but not W461X) showed increased expression of the growth arrest- and DNA damage-inducible gene 153 (GADD153), one marker of endoplasmic reticulum stress and the unfolded protein response (UPR), which in turn can lead to cellular dysfunction and diversion of intracellular resources from their typical constitutive processes (7). Low levels of cell death were noted in Q390X-transfected cells but not with the other constructs. So, where does this leave our understanding of the different mutations and their corresponding phenotypes? The focus of this work was cultured HEK293T cells, cultured rat cortical neurons, and L929 cells. These data represent an important first step in defining which questions can be asked regarding intact animals, in the presence of the full circuitry of functional neurons, glia, and other cells. Are there pathological compensatory effects in other neurotransmitter receptor subunits (either GABAA, glutamate, or other channels) in response to mutations in the final exons of GABRG2? Is there evidence of cell death as a result of ER stress (including the UPR); if so, are certain cell populations (i.e., interneurons) more vulnerable than others? GADD153 has not been implicated directly in neuronal dysfunction in epilepsy but is this yet another avenue to explore? As the authors note, the amino acids missing from the Q390X mutant but present in the W429X mutant include charged residues, palmitoylation sites, tyrosine phosphorylation sites, and a GABARAP (GABA receptor binding protein) binding site. What contribution is made by each of these factors to the ultimate phenotype? GABARAP also plays a role in the later stages of macroautophagy, suggesting that there may be additional pathways of cellular dysfunction to investigate using the truncated γ2 subunits (8). Finally, findings similar to those noted by Kang and coworkers were recently reported with mutations in the second exon (Q40X), raising the question of how specific many of these cellular changes are for mutations in the final exon (9). In summary, different nonsense mutations in γ2 subunits have very different profiles of biochemical and functional abnormalities (Table 1). There are many similarities in functional outcomes (e.g., effect on α1 subunit expression, peak current amplitudes, and zinc sensitivity), although the relative levels of each may reflect subtle differences in protein stability and trafficking. The gradient of changes seen in some assays also correlates with the location of the mutation, which at some
level also may reflect their structural characteristics in neuronal membranes. Despite the advances in our understanding of the genetics of epilepsy, it is clear that many questions of cell biological nature remain to be clarified. by Adam L. Hartman, MD References 1. Leeds P, Peltz SW, Jacobson A, Culbertson MR. The product of the yeast UPF1 gene is required for rapid turnover of mRNAs containing a premature translational termination codon. Genes Dev 1991;5:2303– 2314. 2. Kang JQ, Shen W, Macdonald RL. The GABRG2 mutation, Q351X, associated with generalized epilepsy with febrile seizures plus, has both loss of function and dominant-negative suppression. J Neurosci 2009;29:2845–2856. 3. Kang JQ, Shen W, Macdonald RL. Trafficking-deficient mutant GABRG2 subunit amount may modify epilepsy phenotype. Ann Neurol 2013;74:547–559. 4. Macdonald RL, Kang JQ, Gallagher MJ. Mutations in GABAA receptor subunits associated with genetic epilepsies. J Physiol 2010;588:1861– 1869. 5. Miller RJ. How does ubiquitin regulate synapses? Let me count the ways. J Physiol 2014;592:555–556. 6. Draguhn A, Verdorn TA, Ewert M, Seeburg PH, Sakmann B. Functional and molecular distinction between recombinant rat GABAA receptor subtypes by Zn2+. Neuron 1990;5:781–788. 7. Costa RO, Lacor PN, Ferreira IL, Resende R, Auberson YP, Klein WL, Oliveira CR, Rego AC, Pereira CM. Endoplasmic reticulum stress occurs downstream of GluN2B subunit of N-methyl-d-aspartate receptor in mature hippocampal cultures treated with amyloid-beta oligomers. Aging Cell 2012;11:823–833. 8. Weidberg H, Shvets E, Shpilka T, Shimron F, Shinder V and Elazar Z. LC3 and GATE-16/GABARAP subfamilies are both essential yet act differently in autophagosome biogenesis. EMBO J 2010;29:1792–1802. 9. Ishii A, Kanaumi T, Sohda M, Misumi Y, Zhang B, Kakinuma N, Haga Y, Watanabe K, Takeda S, Okada M, Ueno S, Kaneko S, Takashima S, Hirose S. Association of nonsense mutation in GABRG2 with abnormal trafficking of GABAA receptors in severe epilepsy. Epilepsy Res 2014;108:420–432. 10. Harkin LA, Bowser DN, Dibbens LM, Singh R, Phillips F, Wallace RH, Richards MC, Williams DA, Mulley JC, Berkovic SF, Scheffer IE, Petrou S. Truncation of the GABA(A)-receptor gamma2 subunit in a family with generalized epilepsy with febrile seizures plus. Am J Hum Genet 2002;70:530–536.
205
