Active glutaminase C self-assembles into a supratetrameric oligomer that can be disrupted by an allosteric inhibitor - PubMed (original) (raw)

. 2013 Sep 27;288(39):28009-20.

doi: 10.1074/jbc.M113.501346. Epub 2013 Aug 8.

Alexandre Cassago, Kaliandra de Almeida Gonçalves, Marília Meira Dias, Douglas Adamoski, Carolline Fernanda Rodrigues Ascenção, Rodrigo Vargas Honorato, Juliana Ferreira de Oliveira, Igor Monteze Ferreira, Camila Fornezari, Jefferson Bettini, Paulo Sérgio Lopes Oliveira, Adriana Franco Paes Leme, Rodrigo Villares Portugal, Andre Luis Berteli Ambrosio, Sandra Martha Gomes Dias

Affiliations

Active glutaminase C self-assembles into a supratetrameric oligomer that can be disrupted by an allosteric inhibitor

Amanda Petrina Scotá Ferreira et al. J Biol Chem. 2013.

Abstract

The phosphate-dependent transition between enzymatically inert dimers into catalytically capable tetramers has long been the accepted mechanism for the glutaminase activation. Here, we demonstrate that activated glutaminase C (GAC) self-assembles into a helical, fiber-like double-stranded oligomer and propose a molecular model consisting of seven tetramer copies per turn per strand interacting via the N-terminal domains. The loop (321)LRFNKL(326) is projected as the major regulating element for self-assembly and enzyme activation. Furthermore, the previously identified in vivo lysine acetylation (Lys(311) in humans, Lys(316) in mouse) is here proposed as an important down-regulator of superoligomer assembly and protein activation. Bis-2-(5-phenylacetamido-1,3,4-thiadiazol-2-yl)ethyl sulfide, a known glutaminase inhibitor, completely disrupted the higher order oligomer, explaining its allosteric mechanism of inhibition via tetramer stabilization. A direct correlation between the tendency to self-assemble and the activity levels of the three mammalian glutaminase isozymes was established, with GAC being the most active enzyme while forming the longest structures. Lastly, the ectopic expression of a fiber-prone superactive GAC mutant in MDA-MB 231 cancer cells provided considerable proliferative advantages to transformed cells. These findings yield unique implications for the development of GAC-oriented therapeutics targeting tumor metabolism.

Keywords: Cancer; Enzyme Inhibitors; Enzyme Mechanisms; Glutaminase; Metabolism; Warburg Effect.

PubMed Disclaimer

Figures

FIGURE 1.

FIGURE 1.

GAC polymerization is essential for enzymatic activation. A, size exclusion chromatography analysis of serial dilutions for wild-type and mutated GAC in the presence or absence of 20 m

m

phosphate. The green region delimits the expected Stokes radii between GAC dimers (D) and tetramers (T), calculated based on the crystal structure (3ss3) dimensions. The light purple region, delimited by V in the GAC.K325A graph, indicates the void volume of the gel-filtration column used. The asterisk indicates a previously published result for wild-type GAC (5). B, the software DigitalMicrograph (Gatan) was used to estimate the two orthogonal size distributions (short and long dimensions) of the wild-type and point mutant glutaminase particles, formed under the diverse conditions described in the main text (absence or presence of 20 m

m

phosphate and cross-linked versus non-cross-linked particles). C, supratetrameric organization of GAC upon the addition of 20 m

m

phosphate, a well known activator of GLS1 glutaminases. D, box plot and scatter representation of the two orthogonal dimensions for GAC wild-type and point mutants, as taken from the TEM micrographs. E, chemically cross-linked GAC superoligomers. Some of the observed filaments resemble lateral association between two simple filaments. F, GAC.K325A presents a much higher catalytic efficiency, already in the absence of phosphate, when compared with wild-type GAC and the inactive GAC.R322A. G, the high enzymatic efficiency of GAC.K325A correlates well with its tendency to self-assemble into rod-like polymers, regardless of the presence of the activator inorganic phosphate (left and middle panels) and DSS. Conversely, the catalytically inactive GAC.R322A protein remains in its tetrameric form, even in the presence of 20 m

m

phosphate (right panel). The scale bars represent 100 nm.

FIGURE 2.

FIGURE 2.

Glutaminase isozymes versus self-assembly tendency. TEM analysis of the three glutaminase isozymes, cross-linked after incubation with 20 m

m

phosphate. The average length of the superstructure correlates positively with the previously published glutaminase activity levels (5). GAC is the most active isozyme and forms the longest polymers (top left panel), followed by KGA (top right panel), with lower activity and shorter structures, and lastly LGA, which is insensitive to phosphate and accordingly does not form filaments (bottom left panel). Unless otherwise specified, the scale bars represent 100 nm.

FIGURE 3.

FIGURE 3.

BTPES inhibits GAC superstructure formation. A, dose-response profile of BPTES inhibition on GAC, assayed in the presence of 20 m

m

phosphate, showing two complementary effects on the apparent turnover rates and the Michaelis constant of the enzyme. B, BPTES traps the gating loop (black ribbons) in a rigid open conformation relative to the catalytic pocket of GAC (delimited by a green surface). C, stereographic view of a Fourier 2_F_obs − _F_calc map (at 1 σ) confirming the proposed conformation in the refined crystallographic model. D, TEM analysis of the effects of BPTES on the formation of GAC superstructure. The filament formation is hindered regardless of whether BPTES is added to the protein solution prior to (middle panel) or after (bottom panel) incubation with 20 m

m

phosphate. E and F, conversely, GAC.K325A is catalytically insensitive to BPTES treatment (E), and this inhibitor is also incapable of disrupting the non-cross-linked GAC.K325A filaments (F).

FIGURE 4.

FIGURE 4.

A model of GAC superoligomer assembly. A, the removal of high frequencies (strongly associated with noise) from the TEM micrographs highlights the right-handed double-strand nature of the superstructure, allowing the determination of its geometric features. B, cross-linked MS/MS spectra were manually validated for b and y ion series of the α and β chains of cross-linked peptides (between residues Lys181 and Lys207, Lys202 and Lys578, and Lys403 and Lys512. The ions are indicated by arrows with corresponding m/z value. C, relative position of DSS-linked lysine pairs in the tetramer, as identified by MS. The boundaries of the active sites are delimited by solid surfaces. †, Lys578 is not modeled in the crystal structure. The proposed single strand growth direction is across the longest axis of the tetramer, via an end to end interaction between pairs of N-terminal domains. Alternating orange and yellow colors facilitate the identification of the tetramers along the single strand. D, one lysine residue from each cross-linked pair was individually substituted by a glutamate and assayed for enzymatic activity (left panel) and superstructure formation (right panel). The N-terminal region mutants (GAC.K202E and GAC.K207E) enhanced both effects. On the other hand, GAC.K512E, within the glutaminase domain, generated a nonfunctional protein, similarly to the acetylation mimetic GAC.K316Q. The green region delimits the expected Stokes radii between GAC dimers (D) and tetramers (T), calculated based on the crystal structure dimensions. The asterisk indicates a previously published result for wild-type GAC (5). E, the double strand was manually modeled (panel i), using two entwined copies of the single strand from C and following the geometric restrictions of A. In panel ii, the calculated Fourier _F_calc map (with amplitudes and phases from the double strand model)—limited to 35 Å maximum resolution—is two-dimensionally projected, and the end result is shown in panel iii. All features are comparable to those observed in panel iv, concerning the presence of alternating high density narrow regions (indicated by filled triangles) and low density broad, disc-like regions (indicated by open triangles).

FIGURE 5.

FIGURE 5.

GAC superstructure in cell models. A and B, endogenous and V5-tagged ectopic proteins expressed to similar levels. C–E, a stable MDA-MB 231 clone selected after GAC.K325A-V5 transfection presented larger and more heterogeneous cell area (C), proliferated more (D), and consumed more glutamine from the culture media (E), all compared with cells bearing the V5-tagged wild-type protein or a mock plasmid. F, relative mRNA levels of ASCT2 and SN2 glutamine transporters, as defined by quantitative PCR using rRNA 18 S as a housekeeping gene, showing that the GAC.K325A-V5 cells do not overexpress these transporters. G, glutaminase activity from whole cell lysate, in the presence of 20 m

m

phosphate, showing consistently higher turnover rates for the GAC.K325A-V5 samples, against the physiological glutamine levels in tumors. H, left panel, step gradient SDS-PAGE (3–15%) followed by immunoblotting (anti-V5) of UV-induced cross-linked intracellular protein with incorporated photo-reactive amino acids, showing the tendency of the GAC.K325A-V5 to form higher molecular weight superstructures within the cells. The UV-induced cross-linking was performed in living, intact cells in culture. Right panel, densitometry was performed in conditions of nonsaturated signal, using ImageJ, to evidence the differential cross-linking of bigger species for GAC.K325A-V5. I, similar to what was observed for the recombinant protein, cells expressing the fiber-prone hyperactive GAC.K325 mutant (GAC.K325A-V5) were less sensitive to BPTES treatment, still proliferating more (left panel) and consuming more glutamine (right panel) than BPTES-treated counterparts. J, the knockdown of endogenous GAC favored the enhancing of the phenotypic differences observed above, better highlighting the outcome from GAC.K325A-V5 expression.

References

    1. Gaglio D., Metallo C. M., Gameiro P. A., Hiller K., Danna L. S., Balestrieri C., Alberghina L., Stephanopoulos G., Chiaradonna F. (2011) Oncogenic K-Ras decouples glucose and glutamine metabolism to support cancer cell growth. Mol. Syst. Biol. 7, 523. - PMC - PubMed
    1. Le A., Lane A. N., Hamaker M., Bose S., Gouw A., Barbi J., Tsukamoto T., Rojas C. J., Slusher B. S., Zhang H., Zimmerman L. J., Liebler D. C., Slebos R. J., Lorkiewicz P. K., Higashi R. M., Fan T. W., Dang C. V. (2012) Glucose-independent glutamine metabolism via TCA cycling for proliferation and survival in B cells. Cell Metab. 15, 110–121 - PMC - PubMed
    1. Gao P., Tchernyshyov I., Chang T. C., Lee Y. S., Kita K., Ochi T., Zeller K. I., De Marzo A. M., Van Eyk J. E., Mendell J. T., Dang C. V. (2009) c-Myc suppression of miR-23a/b enhances mitochondrial glutaminase expression and glutamine metabolism. Nature 458, 762–765 - PMC - PubMed
    1. Wang J. B., Erickson J. W., Fuji R., Ramachandran S., Gao P., Dinavahi R., Wilson K. F., Ambrosio A. L., Dias S. M., Dang C. V., Cerione R. A. (2010) Targeting mitochondrial glutaminase activity inhibits oncogenic transformation. Cancer Cell 18, 207–219 - PMC - PubMed
    1. Cassago A., Ferreira A. P., Ferreira I. M., Fornezari C., Gomes E. R., Greene K. S., Pereira H. M., Garratt R. C., Dias S. M., Ambrosio A. L. (2012) Mitochondrial localization and structure-based phosphate activation mechanism of glutaminase C with implications for cancer metabolism. Proc. Natl. Acad. Sci. U.S.A. 109, 1092–1097 - PMC - PubMed

Publication types

MeSH terms

Substances

LinkOut - more resources