Altered Guanine Nucleotide Hydrolysis as Basis for Increased Adenylate Cyclase Activity after Cholera Toxin Treatment

Cholera toxin activation of the adenylate cyclase of mouse neuroblastoma cells occurs in situ with intact cells or in uitro with cell membranes. The in vitro activation process requires X41) and nucleotide triphosphates, in addition to toxin. The cholera toxin-activated adenylate cyclase, when assayed at physiological concentrations of M&I, (10 mnl) in assay systems containing ATP and a nucleotide-regenerat-ing system, has a very high activity and is not significantly activated by 2-chloroadenosine, prostaglandin E,, guanyl-5’- yl imidodiphosphate, or NaF. Under similar assay conditions, the activity of the normal enzyme is low, and maximal activity requires the above activators. The normal enzyme after activation by guanyl-5’-yl imidodiphosphate closely resembles the toxin-activated enzyme when assayed at 10 ml1 MgCl,. However, these two enzymes can be distin-guished from each other. When the M&I, concentration exceeds 20 mM, the toxin-activated enzyme, but not the guanyl-5’-yl imidodiphosphate-activated enzyme, again re- quires exogenously added guanine nucleotide triphosphates,

by centrifugation at 600 x g for 10 min at 4", followed by centrifugation of this supernatant at 37,000 x g for 30 min at 4". The resultant pellet was resuspended in 10 mM Tris/ maleate buffer, pH 7.4, and could be frozen and stored in liquid nitrogen without loss of activity for at least 6 months. The enzyme prepared from cells which were not treated with toxin will be referred to as normal enzyme. Toxin-activated enzyme will refer to the enzyme prepared from cells which were incubated with 20 *g/ml of toxin for 18 h, unless noted otherwise.
Adenylate Cyclase Assays -Production of CAMP was assayed at 30" using either [a~'"P]ATP or [&'*PlAMP-P(NH)P according to the method of Salomon et al. (30) with modifications as previously described (26). Unless otherwise specified in the text, assays with ATP were performed with 50 mM Trislmaleate buffer, pH 7.4, 1 rnM ATP, 6 mM MgCl*, 0.5 mM CAMP, 0.3 mM R020-1724 (CAMP phosphodiesterase inhibitor), and an ATP-regenerating system consisting of 20 mM creatine phosphate and 0.1 mglml of creatine phosphokinase. Assays with AMP-P(NH)P were as above but do not contain a regenerating system (unless specified in text) and contain 0.2 mM AMP-P(NH)P.
Proteins were determined by the method of Lowry et al. (31). All specific activities are reported as picomoles of CAMP produced/min/mg of protein.
All activity values, unless otherwise specified, were averages of duplicate determinations, varying by less than 15%.

Activation of Adenylate
Cyclase in Situ by Cholera Toxin -Incubation of intact NS20 with cholera toxin leads to increased "basal" adenylate cyclase (e.g. activity measured with homogenates at 10 mM Mg*+ in the absence of stimulators (Fig. 1)). The effects of the various known enzyme activators on the activity of the toxin-activated enzyme and normal enzyme have been compared (Table I). The activity of the normal adenylate cyclase is elevated by ClAdo, PGE,, GMP-P(NHlP, and NaF. None of these compounds significantly elevates the activity of the enzyme isolated from cells treated with toxin for 18 h. The activity of the toxin-activated enzyme is equal to the activity of the fully stimulated normal enzyme. The increases 3767 TIME (hours) FIG. 1. Activation of adenylate cyclase during incubation of NS20 cells with cholera toxin. NS20 cells were incubated with 20 pglml of cholera toxin in serum-free growth medium at 37" for various times. Cells were then washed free of toxin, homogenized, and a nucleusfree particulate membrane preparation prepared (see "Materials and Methods").
The activity of the adenylate cyclase in these preparations was assayed in the standard ATP assay system with 1 mM ATP with no other additions (0); 1.4 c(M PGE, (0); 100 jbM ClAdo (A); 5 rnM NaF (ml; or 100 PM GMP-P(NH)P CT). The time of the toxin incubation period is given along the ordinate. Panel B lists the ratio of the activity of the enzyme observed in the presence of one of the above compounds (V ,,l,,,uIB,op) to that observed in the absence of any of these compounds ( V "0 ,tLm"lalor . ) TABLE   I Comparison of normal and toxin-activated adenylate cyclase in ATP assay system All assays were performed in the ATP assay system (see "Materials and Methods") with 1 mM ATP. Additions were made as follows: ClAdo (100 PM); PGE, (1.4 PM); GMP-P(NH)P (100 PM); GTP (100 MM); GDP (100 a~); and NaF (5 mM1. in "basal" activity and the decrease in the activation produced by the above activators are not seen immediately after addition of toxin, but are subject to a characteristic lag of 1 h (2,8,(32)(33)(34). Both aspects of the in situ activation processes are virtually complete within 2 h after addition of toxin.
The neuroblastoma enzyme, like other adenylate cyclases (35-38), possesses a regulatory site which is specific for gua-Cholera Toxin Modification of Adenylate Cyclase nine nucleotides (28). GMP-P(NH)P, an analogue of GTP which is not readily converted to GDP (39), activates the normal enzyme at 10 mM MgCl, (Table I). In contrast, GMP-P(NH)P does not stimulate the toxin-activated enzyme at 10 mM MgCl,. GDP and GTP alone do not influence the activity of either enzyme. An increase in the activity of the toxinactivated enzyme, but not the normal enzyme, is observed when GTP is added in combination with PGE,. The same qualitative difference between the activity of the normal and toxin-activated enzyme is observed when the concentration of ATP in the assay is either 1.0 or 0.1 mM (data not shown).

Effects
of Mg"+, Mn", and CCL'+-Mg"+ appears to be required along with ATP to form the active metal. ATP substrate complex for the neuroblastoma (27) and other adenylate cyclases (40). Moreover, Mg'+ influences the rate of activation of the normal enzyme by GMP-P(NH)P (28). Mg*+ also affects the guanine nucleotide regulatory site of the toxin-activated enzyme (Fig. 2). The activity of the toxin-activated enzyme is progressively inhibited when MgCl, exceeds 20 mM. Addition of GMP-P(NH)P prevents any loss of activity at high MgCl, concentrations and actually results in higher activity than is seen in the absence of GMP-P(NH)P at lower Mg2+ concentrations. At 100 mM MgCl,, the activit,y of the toxin-activated enzyme, when assayed with GMP-P(NH)P, is between 6-and lo-fold higher than when assayed without GMP-P(NH)P.
At 100 mM MgCl,, the K,,, for this effect of GMP-P(NH)P on the toxin-activated enzyme is 2.9 * 1 PM (data not shown). Concentrations of Mg2+ in excess of 20 mM produce other striking changes in the toxin-activated enzyme (Table I). At 100 mM MgCll, the toxin-activated enzyme and the normal enzyme have comparable low "basal" activities. Both are stimulated by PGE, and NaF, as well as GMP-P(NH)P.
Addition of GTP or GDP (resulting in an equal mixture of GTP and GDP in the assay under these conditions) elevates the activity of the normal enzymes less than 2-fold and blocks activation of this enzyme by GMP-P(NH)P.
In contrast, the GTP/GDP mixture stimulates the toxin-activated enzyme almost as much as does GMP-P(NH)P, and the GTP/GDP mixture only slightly inhibits the activation of GMP-P(NH)P.
For the normal enzyme, Mn2+ can replace MgZ+ in the active metal. ATP substrate complex (27) dose-response curves for both the normal and toxin-activated adenylate cyclases are quite similar (Fig. 3). With both enzymes, a bell-shaped activity curve is seen in the absence of GMP-P(NH)P. GMP-P(NH)P prevents the decline in activity with increasing divalent cation concentration for both enzymes, although with Mn2+ the protection is not complete. The inhibition of the toxin-activated enzyme caused by high concentrations of MgCl, and MnCl, does not appear to be due to a nonspecific increase in the ionic strength of the assay mixture.
Addition of 200 mM sodium or ammonium chloride does not cause any inhibition (Table II). CaCl, will cause an inhibition of activity which is not prevented by GMP-P(NH)P (see below).
The activation of the normal neuroblastoma enzyme by GMP-P(NH)P appears to be irreversible. concentrations of Mg'+ or MI?+. Assays were performed in the ATP assay system with 1 rnM ATP with either MgCl, (Pnnels A and C) or MnClp (Pan& B and D). Normal enzyme assayed without GMP-P(NH)P (0) and with 100 FM GMP-P(NH)P (0). Toxin-activated enzyme. assayed without GMP-P(NH)P (ml and with 100 ELM GMP-P(NH)P (0). G, GMP-P(NH)P.
In each panel, maximal activity was determined from the l/V versus l/ion plots shown in the inset. All assays were performed in the ATP assay system with 1 rnM ATP. Where indicated, GMP-P(NH)P was also present in the assay mixture at 100 uM.   Toxinactivated enzyme was incubated at 30" in 10 mM Tris/maleate buffer, pH 7.4, and either 10 mM MgCl, (Panel A) or 50 mM MgCl, (PanelB) and the following: 100 PM GMP-P(NH)P (0); 100 PM GTP plus regenerating system (A); 100 PM ATP plus regenerating system (V); regenerating system (0); 100 pM AMP-(NH)P (0); or nothing (01. After incubation, enzyme was washed once in the buffer/Mg2+ solution, resuspended in same, and assayed in the ATP assay system with 10 rnM Mg2+ and 0.1 rnM ATP. Initial activities are the same as stated in legend to Fig. 6 Tris/maleate buffer, pH 7.4; and either 10 mM MgCl, (0, 0); 100 mM MgCl, (0, W); or 100 mM MgCl, plus 100 ,u~ GMP-P(NHlP (A). The activity remaining was then determined in the ATP assay system at 10 mM MgCl,; 0.1 mM ATP; and with either no 0) Or loo &&M GMP-P(NH)P (a, n ). When GMP-P(NH)P Was present in the incubation mixture, it was also present at 10 pM in the assays. Enzyme preparations were assayed directly after incubation. If after incubation, the enzyme is first washed twice before being assayed, the same results are obtained.
Where the activities are assayed in the absence of GMP-P(NH)P, the initial activity is that seen without incubation.
Where the activities are assayed with GMP-P(NH)P, the initial activity is that observed after a lo-min incubation with 10 mM Mg2+ and 100 PM GMP-P(NH)P.
Initial activities for normal and toxin-activated enzymes are 4.5 and 33.4 pmol/min/mg of protein, respectively.

TABLE IV
Comparison of normal and toxin-activated adenylate cyclase in AMP-P(NH)P assay system Assays were performed in the AMP-P(NH)P assay system (see Materials and Methods") with 0.2 mM AMP-P(NH)P. Where indicated, a nucleotide triphosphate-regenerating system (Reg. System) consisting of 20 mM creatine phosphate and 0.1 mglml of creatine phosphokinase was also added to the assay mixture. Other assay additions were as follows: ClAdo (100 WM); PGE, (1.4 PM); NaF (5 mM); GTP (200 PM); ATP (200 &; and GMP-P(NH)P (IO0 CLM  Effect of high Mg" on enzyme activity in AMP-P(NH)P assay system Assays were performed in the AMP-P(NH)P assay system with 0.2 rn~ AMP-P(NH)P at either 10 or 100 rn~ MgCl, as noted. Other assay additions were as follows: ClAdo (100 PM); PGE, (1.4 PM); NaF (5 mM); GTP (200 FM); ATP (200 PM); GMP-P(NH)P (100 FM); and Reg. System, the nucleotide-regenerating system as described in the legend to Table IV activation of the normal enzyme by GMP-P(NH)P (28). We find that 25 PM GDP, which inhibits 50% of the GMP-P(NH)P activation of the normal enzyme, also inhibits 50% of the GMP-P(NH)P activation of the toxin-activated enzyme. Furthermore, PGE, increases the concentration of GDP required to block the activation of the toxin-activated enzyme by GMP-P(NH)P (Fig. 8) (Table VI). Half-maximal activation occurs at 6.5 pg/ml of toxin and 30 PM NAD (data not shown).
We do not observe any effect of the cytoplasm on this process.
The requirement for a nucleotide is partially fulfilled by the inclusion of a nucleotide-regenerating system in the incubation mixture. Activation is maximal when ATP or GTP is added along with the regenerating system.
In the absence of a regenerating system, GDP, ADP, GTP, ATP, or AMP-P(NH)P do not support the activation process (Table VI). The time course of the activation process in the complete incubation mixture has a t,,, of 3 min (Fig. 9A). Based on the disappearance of stimulation by NaF, the activation process GDP (pW FIG. 8. GDP inhibition to activation of the toxin-activated adenylate cyclase by GMP-P(NH)P.
Assays were performed in the AMP-P(NH)P assay system with 5 rnM Mgy+, 0.1 rnM AMP-P(NH)P, 50 pM GMP-P(NH)P, and without any regenerating system present. appears to be complete after 20 min. Toxin activation will also take place during an adenylate cyclase assay if NAD is present (Fig. 9B)