Bl~hunica et Biopl~'sica Acta, 11351191J2}73-78 © 1992ElsevierScience Publisher, B.V.All fights rcsee,,ed0tb7~1889/92/$1)5.0D
73
A soluble factor and GTPTS are required for Dictyostelium discoideum guanylate cyclase activity C o n c h i t a C . G . M . S e h u l k e s , C u r D . S c h o e n t, J o s C . A r e n t s a n d R o d V a n D r i e l E.C. Slvtl.r lr~litlae fur Bivtlzcmicul Re~e~trctl. Unit er~ity uf Amsterdam, Amsterdam (NetheiJand~
(Received 14Ocloher 1991) tRevis=dmanuseripl rcxeivcd a) Januaq, 1992)
Key~alrds: Guanylateq~clase;GTP (DIcO'ostelium Amoeba of Dic*yosrelilon disc¢~idezvn show a rapid, transient cGMP synthcsis in response to chemotactie stimulation. Using Mg-"+-GTP as a suhslrate, guanylate cyclase (E.C. 4.6.1.2.) activity is found exclusivelyiu the parliculat,: fraction of Dictyostdium cells. Hc*c we she,w that the activity is dependent on the presence of the non-hydrolysable CTP-anaiogue GTP-t S, which itself is only a poor substratc for the enzyme under the prevailing conditions. Evidence is presente.d that a transient exposure of the enzyme to GTP'vS is sufficient to constit;Jfivclyactivate the enzyme. GTP3,S-dcpcndent aet!vily is found to requite a factor that can be separated from the enzyme by v.~shing the paniculate fraction with law salt buffer. Addition of the soluble cell fraction to these washed membranes restores ellzylllCactivity.
Introduction Dictyostel~m~ discoideum cells live as solitary amoebae as long as sufficient nutrients are available and start aggregating into multicellular structures wl:en food is exhausted. Ceil aggregation is mediated by the chemoattraetant cAMP which is secreted in pulses by starving DicO'ostdium cells. Binding of cAMP to highly specific cell surface receptors induces activation of adenyiate cyclase and guanylate cyclasc (for review see Ref. 1)- Newly synthesized cAMP is secreted, thereby relaying the extraeellular ehemotaetie signal. C~clic GMP in Dictyostelium dL~coideum is a second messenger, like in many other higher eukaryolle cell Woes [2-7]. Cyclic GMP is thoughl to mediate cellular ehemotaxis. The eGMP response in cAMP-stimulated ceils shows rapid and transient characteristics [8,9]. Within seconds after receptor stimulation, guan:~!ate
i Present address: Research Institute fur Plant Pmtccliun. Blanchhaven 12, 6709PD Wageningen,Netherlands. Abbreviations: cGMP. a~anosine 3'.5"-monophosphate; cAMP. adenosine 3".SLmonoph~phate;GTP?*S.guanosine 5'qO43-thio)lriphosphate.
Correspondence: R, Van Dnd, E.C. Slat~rInstitute for Ui~hemical Re.arch, Univc~ityof .aansterdam,PlantageMuidergracht12.1018 I v AmSteraam,Netherlands.
cyclase is activated which results in a n increase of intracellular eGMP concentration. Due to inactivation of the enzyme as a result of an adaptation mechanism and to a specific cGMP phosphodiesterase activity, the eGMP concentration decreases rapidly to the prestimulation level. Tile entire pruce~ takes only 30 s, showing a cGMP concentration peak at al~out 10 s. Cbetootaxis and concomitant changes in cytoskeleton proteins such as myosin phosphorylation and polymerizatiol:, show a similar time-course [10,11]. Mutants that lack the cGMP-speeifie phosphodiesterase show a prolonged cGMP response [/2], a different myosin response [11,13], and altered chetootaetie behaviour [14]. This suggests that cGMP is a second messenger involved in the transduction of the chetootactie signal in the cell. We have investigated the regulation of particulate guanylata cyciase of Dictyostelium. In cell hotoogenates and electropertoeahilized cells guanylate cyclase activity requires tile presence of GTP~,S, a non-hydrolyzable GTP-analoguc ( g e l 15, and Sehnen, C.D., Zegers, M., Arents, J.C. and Van Driel, R., unpublished results). Here we present evidence that a protein factor that is required for guanylate cyelase activity, can be removed by washing the particulate fraction with low salt buffer. Addition of the soMble cell fraction to washed membranes restores GTP'/S-dependent enzyme activity. A preliminary report on this work has appeared in Ref. 16
74 Materials and Methods Guanosine 3",5'-monophosphate {eGMP), 2'-deox3'adenosine 3',5'-monophosphate (dcAMP), GTP and G T P y S were obtained from Boehringer (Mannheim, Germany); [2",8'-3H]cGMP (910.2 G B q / m m o l ) was purchased from Amersham (Amersham, UK). Bovine serum albumin (BSA) ~as from Sigma fSt. Louis, MO. USA) and activated charcoal (Norit A) from Serva (Heidelberg, Germany). Polycarbonate fillers (5 /xm pore-siz-~) were from NacleF.are Corporation (Pleasanton, CA).
Odmre conditiom- and cell lysis Dico'ostelium discoideum cells, strain AX2 were grown in HL-5 medium containing maltose instead of glucose [17]. Cells were halvested at a density of approx. 5 • 10" cells/mL washed free from nutrients and resuspended in phosphate buffer 115 mM KHzPO+/ NazHPO4+ pH 6.2), to ]{17colls/ml. Starving cells were then shaken in suspension for 5 h at 2.2~C to achieve aggregation competence. Subsequently, cells ~rerc harvested, washed and resuspended at a density of 10~ cells/ml +n phosphate buffer. Washed cells were kept on ice under continuovs aeration for up to 2 h. Before breakage, cells were centrifuged and resuspended in homogenization buffer that contained 20 mM HepesNaOH (pH 7~~), 2.2 mM MgSO4, 1 mM EGTA. The homogenization buffer contained 0.1 ram G T P y S . unless indit~ated otherwise. Cells were broken by forcing the suspension through a polyearbonare filter with a pore-size of 5 a m at 0°C [17,18]. This method resulted in breakage of more than 98% of the cells, as estimated by phase contrast microscopy. The hnmogenate was used immediately.
Cell fr~ctionatia~z P,+rticolo;,, fractiotl. The partieulmL fraction was isolated by cenmfuging a cell homagcnate for 1 rain at 10000 × g and discarding the supematant, ]'he pellet was resuspended in homc~cnization buffer to the ori~nal membrane concentaation. Washed partit~dtue fiaetion. To prepare a washed particulate fraction, a cell homogenate was diluted seven times in low salt buffer (10 mM Hebes-NaOH. pH %5) and centrifuged at 1({1.1{10× g for 1 s i n . The supernatant was di~arded and the pellet was resuspended in homogcn~ation buffer to the original membrane concentration. In reconstitution experiments the pellet was resupended in the i sin+ lO000×g supernatant of a cell homogenate, a cell fiacfioa that contains all soluble cell components. ttomogenate. -t he .:x~zym¢ wdr+ ,~[al)~ f~" at [cast .5 rain+ Cell fractionation could bc carried out without considerable loss of activity. T o make the actkity in a horoogcnatc directly .eomparablt: to activity in pafficu-
late fraelions, guanylate cyclase activity in former preparaiion~ was assayed after centrifugation and resuspending the pellet in its o~vn superaatant.
Guanylare ¢5"claseaclit'i~ Guanytatc cyclase activity was measured at 20°C as desen'bed previously [19]. In this method cGMP synthesis is slatted by addition of an e.qual volume of a starting nl~ture to the cell fractions. The ~arting mix+ tare was the same as homogenization buffer except that l mM GTP and 10 mM dithiott~itol (an inhibitor of DicO'oswlium phosphodiesterase [20,21]) had been added. During the reaction 0.1 raM GTP~,S was present unless stated otherwise. Optimal as.~ay conditions were chosen on the basis of what has been published by Janssens et al. [[5,22]. Cyclic G M P synthesis was linear in time for at least 2 min. After the reaction was staffed, 30 # l samples were taken at varions time points and quenched with 30 # l of 2 M HCli34. The samples were stored at -2{PC or a-~.sayed iramediately.
Cyclic GMP assay Cyclic G M P concentrations were measured in a radio-immunoassay using a highly specific a~ti-cGMP rabbit antiserum, prepared according to 8mlner et al. [23]. The affinity of the antiserum was several orders of magnitude higher for c(;MP than for cA,~,IP (1~3-fold lower affinity.)+ dcAMP (10S-fold lower affinity) and GDP, guanosim:, AMP, AI*P and G M P (10~-fold lower aifinity). Perchloric acid extracts were neutralized by adding a mixture of KHCO 3 and K O H (1.5 and 3.5 M+ respectively) and alter shaking for i s i n , centrifuged for 15 rain at 10000 × g. After neu:ralization samples were kept on ice. 25 ~i oi supe+natam were transferred to a well of an empty microtiterplate. To each well Ill of [2".8'-~H]cGMP .',ere added ( ? ! ~ q ) m per we~) and 5 0 / t l of antHody mixture. This mixture contained phosphate buffer (200 raM KHzPO4/Na2HPO~, 10 mM EDTA. 10 mM EGTA, pH 7.0) and bovine scrllm in a ratio of 5:1 (v/v). To this mixture anti-cGMP antiserum was added to a (,oncentration of 1% (v/v). At this concentration of antiserum 0-250 nM cGMp could he measured accurately. Samples were incubated for 2-20 h at 4+
73
A soluble factor and GTPTS are required for Dictyostelium discoideum guanylate cyclase activity C o n c h i t a C . G . M . S e h u l k e s , C u r D . S c h o e n t, J o s C . A r e n t s a n d R o d V a n D r i e l E.C. Slvtl.r lr~litlae fur Bivtlzcmicul Re~e~trctl. Unit er~ity uf Amsterdam, Amsterdam (NetheiJand~
(Received 14Ocloher 1991) tRevis=dmanuseripl rcxeivcd a) Januaq, 1992)
Key~alrds: Guanylateq~clase;GTP (DIcO'ostelium Amoeba of Dic*yosrelilon disc¢~idezvn show a rapid, transient cGMP synthcsis in response to chemotactie stimulation. Using Mg-"+-GTP as a suhslrate, guanylate cyclase (E.C. 4.6.1.2.) activity is found exclusivelyiu the parliculat,: fraction of Dictyostdium cells. Hc*c we she,w that the activity is dependent on the presence of the non-hydrolysable CTP-anaiogue GTP-t S, which itself is only a poor substratc for the enzyme under the prevailing conditions. Evidence is presente.d that a transient exposure of the enzyme to GTP'vS is sufficient to constit;Jfivclyactivate the enzyme. GTP3,S-dcpcndent aet!vily is found to requite a factor that can be separated from the enzyme by v.~shing the paniculate fraction with law salt buffer. Addition of the soluble cell fraction to these washed membranes restores ellzylllCactivity.
Introduction Dictyostel~m~ discoideum cells live as solitary amoebae as long as sufficient nutrients are available and start aggregating into multicellular structures wl:en food is exhausted. Ceil aggregation is mediated by the chemoattraetant cAMP which is secreted in pulses by starving DicO'ostdium cells. Binding of cAMP to highly specific cell surface receptors induces activation of adenyiate cyclase and guanylate cyclasc (for review see Ref. 1)- Newly synthesized cAMP is secreted, thereby relaying the extraeellular ehemotaetie signal. C~clic GMP in Dictyostelium dL~coideum is a second messenger, like in many other higher eukaryolle cell Woes [2-7]. Cyclic GMP is thoughl to mediate cellular ehemotaxis. The eGMP response in cAMP-stimulated ceils shows rapid and transient characteristics [8,9]. Within seconds after receptor stimulation, guan:~!ate
i Present address: Research Institute fur Plant Pmtccliun. Blanchhaven 12, 6709PD Wageningen,Netherlands. Abbreviations: cGMP. a~anosine 3'.5"-monophosphate; cAMP. adenosine 3".SLmonoph~phate;GTP?*S.guanosine 5'qO43-thio)lriphosphate.
Correspondence: R, Van Dnd, E.C. Slat~rInstitute for Ui~hemical Re.arch, Univc~ityof .aansterdam,PlantageMuidergracht12.1018 I v AmSteraam,Netherlands.
cyclase is activated which results in a n increase of intracellular eGMP concentration. Due to inactivation of the enzyme as a result of an adaptation mechanism and to a specific cGMP phosphodiesterase activity, the eGMP concentration decreases rapidly to the prestimulation level. Tile entire pruce~ takes only 30 s, showing a cGMP concentration peak at al~out 10 s. Cbetootaxis and concomitant changes in cytoskeleton proteins such as myosin phosphorylation and polymerizatiol:, show a similar time-course [10,11]. Mutants that lack the cGMP-speeifie phosphodiesterase show a prolonged cGMP response [/2], a different myosin response [11,13], and altered chetootaetie behaviour [14]. This suggests that cGMP is a second messenger involved in the transduction of the chetootactie signal in the cell. We have investigated the regulation of particulate guanylata cyciase of Dictyostelium. In cell hotoogenates and electropertoeahilized cells guanylate cyclase activity requires tile presence of GTP~,S, a non-hydrolyzable GTP-analoguc ( g e l 15, and Sehnen, C.D., Zegers, M., Arents, J.C. and Van Driel, R., unpublished results). Here we present evidence that a protein factor that is required for guanylate cyelase activity, can be removed by washing the particulate fraction with low salt buffer. Addition of the soMble cell fraction to washed membranes restores GTP'/S-dependent enzyme activity. A preliminary report on this work has appeared in Ref. 16
74 Materials and Methods Guanosine 3",5'-monophosphate {eGMP), 2'-deox3'adenosine 3',5'-monophosphate (dcAMP), GTP and G T P y S were obtained from Boehringer (Mannheim, Germany); [2",8'-3H]cGMP (910.2 G B q / m m o l ) was purchased from Amersham (Amersham, UK). Bovine serum albumin (BSA) ~as from Sigma fSt. Louis, MO. USA) and activated charcoal (Norit A) from Serva (Heidelberg, Germany). Polycarbonate fillers (5 /xm pore-siz-~) were from NacleF.are Corporation (Pleasanton, CA).
Odmre conditiom- and cell lysis Dico'ostelium discoideum cells, strain AX2 were grown in HL-5 medium containing maltose instead of glucose [17]. Cells were halvested at a density of approx. 5 • 10" cells/mL washed free from nutrients and resuspended in phosphate buffer 115 mM KHzPO+/ NazHPO4+ pH 6.2), to ]{17colls/ml. Starving cells were then shaken in suspension for 5 h at 2.2~C to achieve aggregation competence. Subsequently, cells ~rerc harvested, washed and resuspended at a density of 10~ cells/ml +n phosphate buffer. Washed cells were kept on ice under continuovs aeration for up to 2 h. Before breakage, cells were centrifuged and resuspended in homogenization buffer that contained 20 mM HepesNaOH (pH 7~~), 2.2 mM MgSO4, 1 mM EGTA. The homogenization buffer contained 0.1 ram G T P y S . unless indit~ated otherwise. Cells were broken by forcing the suspension through a polyearbonare filter with a pore-size of 5 a m at 0°C [17,18]. This method resulted in breakage of more than 98% of the cells, as estimated by phase contrast microscopy. The hnmogenate was used immediately.
Cell fr~ctionatia~z P,+rticolo;,, fractiotl. The partieulmL fraction was isolated by cenmfuging a cell homagcnate for 1 rain at 10000 × g and discarding the supematant, ]'he pellet was resuspended in homc~cnization buffer to the ori~nal membrane concentaation. Washed partit~dtue fiaetion. To prepare a washed particulate fraction, a cell homogenate was diluted seven times in low salt buffer (10 mM Hebes-NaOH. pH %5) and centrifuged at 1({1.1{10× g for 1 s i n . The supernatant was di~arded and the pellet was resuspended in homogcn~ation buffer to the original membrane concentration. In reconstitution experiments the pellet was resupended in the i sin+ lO000×g supernatant of a cell homogenate, a cell fiacfioa that contains all soluble cell components. ttomogenate. -t he .:x~zym¢ wdr+ ,~[al)~ f~" at [cast .5 rain+ Cell fractionation could bc carried out without considerable loss of activity. T o make the actkity in a horoogcnatc directly .eomparablt: to activity in pafficu-
late fraelions, guanylate cyclase activity in former preparaiion~ was assayed after centrifugation and resuspending the pellet in its o~vn superaatant.
Guanylare ¢5"claseaclit'i~ Guanytatc cyclase activity was measured at 20°C as desen'bed previously [19]. In this method cGMP synthesis is slatted by addition of an e.qual volume of a starting nl~ture to the cell fractions. The ~arting mix+ tare was the same as homogenization buffer except that l mM GTP and 10 mM dithiott~itol (an inhibitor of DicO'oswlium phosphodiesterase [20,21]) had been added. During the reaction 0.1 raM GTP~,S was present unless stated otherwise. Optimal as.~ay conditions were chosen on the basis of what has been published by Janssens et al. [[5,22]. Cyclic G M P synthesis was linear in time for at least 2 min. After the reaction was staffed, 30 # l samples were taken at varions time points and quenched with 30 # l of 2 M HCli34. The samples were stored at -2{PC or a-~.sayed iramediately.
Cyclic GMP assay Cyclic G M P concentrations were measured in a radio-immunoassay using a highly specific a~ti-cGMP rabbit antiserum, prepared according to 8mlner et al. [23]. The affinity of the antiserum was several orders of magnitude higher for c(;MP than for cA,~,IP (1~3-fold lower affinity.)+ dcAMP (10S-fold lower affinity) and GDP, guanosim:, AMP, AI*P and G M P (10~-fold lower aifinity). Perchloric acid extracts were neutralized by adding a mixture of KHCO 3 and K O H (1.5 and 3.5 M+ respectively) and alter shaking for i s i n , centrifuged for 15 rain at 10000 × g. After neu:ralization samples were kept on ice. 25 ~i oi supe+natam were transferred to a well of an empty microtiterplate. To each well Ill of [2".8'-~H]cGMP .',ere added ( ? ! ~ q ) m per we~) and 5 0 / t l of antHody mixture. This mixture contained phosphate buffer (200 raM KHzPO4/Na2HPO~, 10 mM EDTA. 10 mM EGTA, pH 7.0) and bovine scrllm in a ratio of 5:1 (v/v). To this mixture anti-cGMP antiserum was added to a (,oncentration of 1% (v/v). At this concentration of antiserum 0-250 nM cGMp could he measured accurately. Samples were incubated for 2-20 h at 4+
