Adenylyl Cyclase Activity of Cya1 from the Cyanobacterium Synechocystis sp. Strain PCC 6803 Is Inhibited by Bicarbonate

Abstract

Bicarbonate stimulates the activities of several class III adenylyl cyclases studied to date. However, we show here that bicarbonate decreased V_max and substrate affinity in Cya1, a major adenylyl cyclase in the cyanobacterium Synechocystis sp. strain PCC 6803. This indicates that manifestation of the bicarbonate responsiveness is specifically modulated in Cya1.

Cyclic AMP (cAMP), a ubiquitous second messenger molecule that affects a number of cellular functions in prokaryotes and eukaryotes, is mainly generated by adenylyl cyclase (AC). In mammalian cells, two types of ACs have been characterized: transmembrane ACs localized in the plasma membrane and regulated by heterotrimeric G proteins in response to various biochemical stimuli and the more recently described “soluble” AC (sAC) associated with discrete cellular components and regulated by intracellular signaling molecules, such as bicarbonate and Ca²⁺ (3, 10). These mammalian ACs are grouped into the nucleotidyl cyclase class III. Phylogenetic and biochemical analyses have shown that several prokaryotic class III ACs are closely related to mammalian sAC (5, 8). In fact, the activities of cyanobacterial class III ACs, CyaC of Spirulina platensis and CyaB1 of Anabaena sp. strain PCC 7120, are stimulated by bicarbonate and Ca²⁺ (1, 2). On the basis of the recently reported crystal structure of CyaC of Spirulina platensis, bicarbonate has been suggested to stimulate the activity through marked conformational change, although electron densities for bicarbonate were not defined (10).

cAMP signal transduction for bicarbonate signaling may be of particular importance for cyanobacteria, as the bicarbonate concentration is closely correlated with CO₂ available for photosynthesis. In the cyanobacterium Synechocystis sp. strain PCC 6803, cAMP has also been demonstrated to be a second messenger in the transduction of a light signal (reviewed in reference 9). In particular, transfer of the cells from the dark to blue light results in an increase in the level of cellular cAMP, which is bound to the cAMP receptor protein SYCRP1 involved in the biogenesis of pili (7, 12-14). This mechanism is thought to allow cells to adjust their motility in response to environmental changes in light conditions via cAMP levels. The blue-light-induced increase of cellular cAMP content is ascribed to Cya1, a class III AC that consists of a C-terminal AC catalytic domain and an N-terminal Forkhead-associated (FHA) domain (4, 11). A null mutation of the cya1 gene results in a decrease in cellular cAMP levels (∼4% of the wild-type level), and as a result, the mutant strain loses the capability of cell motility (11). Since the primary aim of phototactic movement must be to achieve higher photosynthetic performance and/or to avoid photodamage, it follows that light-dependent regulation of cell motility via Cya1 is also modulated in response to the availability of an inorganic carbon source.

In the present study, we investigated the effects of bicarbonate on Cya1 activity in vitro and found that Cya1 activity is negatively regulated by bicarbonate. The results indicated that Cya1 possesses the basic properties of a class III AC in terms of its responsiveness to bicarbonate but with inverse concentration dependence for the ion. The unique properties of Cya1 are discussed in relation to the physiological functions of Cya1 in this bacterium.

Cya1 activity is negatively regulated by bicarbonate.

To investigate the biochemical property of Cya1, we first tested the effects of bicarbonate on Cya1 AC activity. The purification of Cya1 and analysis of its AC activity were carried out as described previously (7). Figure 1A shows the effects of various salts on the AC activity of Cya1. AC activity was inhibited approximately 50% by 50 mM NaHCO₃. Since the AC activity of Cya1 requires Mn²⁺ (7), it is possible that inhibition by NaHCO₃ is attributable to Na⁺, which interferes with the function of Mn²⁺. However, the slight change of AC activity due to NaCl or KCl suggests that bicarbonate is responsible for the observed inhibition by NaHCO₃. As shown in Fig. 1B, the AC activity was progressively inhibited by increasing bicarbonate concentrations, reaching approximately one-third of the control activity at 70 mM NaHCO₃. Figure 2 shows the effects of bicarbonate on K_m and V_max for ATP in the AC activity of Cya1. Both values were markedly affected by NaHCO₃ at 50 mM; K_m increased approximately 15-fold (from 2.2 ± 0.3 to 33.9 ± 8.1 μM) and V_max decreased approximately 2-fold (from 65.7 ± 3.6 to 40.5 ± 3.2 pmol cAMP/min/nmol protein) compared to values observed in the absence of bicarbonate. These results suggest that bicarbonate suppresses Cya1 activity by reducing the affinity of ATP to its catalytic sites, as well as the turnover rate of ATP. It is of note in this context that all class III ACs reported so far are insensitive to or are stimulated by bicarbonate (1, 2, 6). Therefore, Cya1 is the first example of a class III AC whose activity is suppressed by bicarbonate.

FIG. 1.

Adenylyl cyclase activity of Cya1 is inhibited by bicarbonate. (A) Activity was measured in the presence of 50 mM NaHCO₃, NaCl, and/or KCl as indicated. (B) Activity was assayed in the presence of various concentrations of NaHCO₃. The reaction mixture included 200 μM ATP.

FIG. 2.

Effects of bicarbonate on kinetic properties of Cya1 adenylyl cyclase activity. Activity was assayed as a function of substrate ATP concentrations in the absence (closed circles) or presence (open circles) of 50 mM NaHCO₃.

Here, amino acid residues responsible for the inhibition of Cya1 by bicarbonate were not identified. However, it may be worthwhile to note that the conserved Gly-173 and Phe-178 in the bicarbonate-sensitive ACs are replaced by Ser and Tyr, respectively, in Cya1 as shown in Fig. 3, in which the amino acid sequence of the Cya1 catalytic domain was compared with those of several prokaryotic and eukaryotic ACs. Interestingly, these residues exist adjacent to Lys-177, which has been proposed as a functional site for conferring bicarbonate sensitivity in class III ACs, since the replacement of Lys-646 (equivalent to Lys-177 in Cya1) in CyaB1 of Anabaena sp. strain PCC 7120 by Ala resulted in the loss of bicarbonate sensitivity (1). Therefore, we may propose that Ser-173 and/or Tyr-178 is responsible for the unique bicarbonate sensitivity of Cya1. Mutational studies of these residues will provide decisive answers to this issue.

FIG. 3.

Amino acid sequence alignment of the catalytic region of Cya1 with various class III adenylyl cyclases. An amino acid sequence of Synechocystis Cya1 was obtained from the KAZUSA DNA Research Institute site at http://www.kazusa.or.jp/en/. Accession numbers for other aligned amino acid sequences are as follows: Anabaena CyaB1, BAA13998; Spirulina CyaC, BAA22997; Rattus sAC, AAD04035; Mycobacterium Rv1319c, Q10632; Mycobacterium Rv1264, Z77137; Rattus transmembrane AC (tmAC), M55075; and Mus tmAC9, CAA03415. Amino acids involved in substrate recognition (Lys-177), metal ion coordination (Asp-181), and transition state stabilization (Asn-258 and Arg-262) are indicated in bold type. As indicated in the right margin, the AC activities of Anabaena CyaB1, Spirulina CyaC, Rattus sAC, and Mycobacterium Rv1319c are stimulated by bicarbonate; however, those of Mycobacterium Rv1264 and Rattus tmAC are insensitive to bicarbonate (1, 2, 6). Bicarbonate has been proposed to mimic the carboxyl group of Asp conserved in the bicarbonate-insensitive ACs at the position of Thr-251 (italic type) (1). Alignment is based on the alignment of a previous report (1). Gaps introduced to maximize alignment are indicated by dashes.

Effects of the FHA domain on bicarbonate-dependent regulation of Cya1 activity.

It was previously suggested that the N-terminal FHA domain of Cya1 is responsible for regulating the AC activity in response to various changes in environmental conditions, such as blue-light intensity (4, 7, 12). To test whether the FHA domain is involved in the bicarbonate regulation of Cya1 AC activity, we characterized the properties of the truncated version of Cya1 (consisting of amino acids 95 to 338), which lacks the FHA domain. The truncated version of Cya1 lacking the FHA domain was expressed using an Escherichia coli overexpression system with the N terminus truncated and appended with a histidine tag (Novagen). For this application, the region corresponding to residues 95 to 338 of Cya1 was amplified by PCR using isolated pETCya1 plasmid (7) DNA as a template and the upstream and downstream oligonucleotide primers Cya1-F2 (5′-GGGGGGCATATGACGGAAGCGCAATTCTAC-3′; NdeI restriction site underlined) and Cya1-R (7), respectively. The amplified fragment was digested with NdeI and EcoRI and then cloned into the NdeI and EcoRI sites of the pET28(a) vector (Novagen). The resulting plasmid, named pETCya1C, was used to transform E. coli strain BL21(DE3) (Novagen). Further purification steps were the same as for the native histidine-tagged Cya1 as previously described (7). The purified truncated protein was observed as a single band on sodium dodecyl sulfate-polyacrylamide gels (data not shown) and catalyzed cAMP formation with a K_m of 2.7 ± 0.7 μM for ATP, a value compatible to that of native Cya1 (2.2 ± 0.3 μM), although specific activity was significantly (approximately fourfold) lower than that of the native Cya1 (Fig. 4). This indicates that the FHA domain is required to achieve the maximum specific activity but not determinative of the substrate affinity. For truncated Cya1, the presence of bicarbonate resulted in an approximately 12-fold increase of K_m (from 2.7 ± 0.7 to 32.7 ± 4.0 μM) and an approximately 2-fold decrease of V_max (from 17.6 ± 0.9 to 10.4 ± 0.7 pmol cAMP/min/nmol protein), which are similar responses to those of native Cya1, indicating that the FHA domain is not directly involved in the bicarbonate sensitivity of Cya1. Interestingly, however, the AC activity of the truncated Cya1 tended to be inhibited at high ATP concentrations (Fig. 4) with 50% inhibition at approximately 0.5 mM of ATP (data not shown). Furthermore, the ATP-dependent inhibition was not evident in the presence of bicarbonate (Fig. 4). ATP-dependent inhibition of AC activity and the relief of the inhibition by bicarbonate have been reported in mammalian sAC (6), although it is not clear at present how they are achieved. Notably, ATP-dependent inhibition was not obvious in the native Cya1 both in the presence and absence of bicarbonate as shown in Fig. 2.

FIG. 4.

Kinetic properties of adenylyl cyclase activity of N-terminal truncated Cya1. The activity of the purified N-terminal truncated version of Cya1 was assayed as a function of substrate ATP concentration in the absence (closed circles) or presence (open circles) of 50 mM NaHCO₃. Points reflecting substrate inhibition were omitted for regression analysis.

The N-terminal FHA domain, which may directly or indirectly interact with blue-light receptors (7, 12), has been proposed to mediate the enhancement of AC activity of Cya1 by blue light. As shown in Fig. 4, the AC activity of the N-terminal truncated Cya1 was significantly lower than that of the native Cya1, but little change was observed in substrate affinity and response to bicarbonate. These indicate that the properties of the substrate binding pocket are not greatly influenced by the deletion of the FHA domain. Cya1 shows high AC activity, independent of blue light in vitro (7), suggesting that the C-terminal catalytic domain is designed to achieve maximum enzymatic performance, possibly through direct protein-to-protein interaction between the N-terminal FHA domain and the C-terminal catalytic domain. On the basis of this view, a possible mechanism of the blue light enhancement of the Cya1 activity in vivo is the following: the FHA domain interacts with a putative blue-light receptor protein in the dark to block interaction with the catalytic domain, which shows low AC activity; then, blue light excitation liberates the photoreceptor from the FHA domain, which interacts with the catalytic domain to enhance the AC activity. Alternatively, the blue-light photoreceptor controls the interaction between the FHA domain and another putative modular protein.

Physiological function of bicarbonate responsiveness of Cya1.

In this study, we found that bicarbonate decreased the V_max of Cya1 with concomitant prominent decreases in the affinity for ATP, although the effects of the ion on the activity differed considerably from those of other ACs, which concomitantly showed an increase in V_max in the presence of bicarbonate with little decrease in the affinity for ATP (1, 2, 6). The unique response of Cya1 to bicarbonate may be related to the physiological function of this AC. The phototactic movement must be closely related to the acquisition of sufficient light for photosynthesis and/or avoidance of photodamage under excess light conditions. The negative control of Cya1 by bicarbonate infers the suppression of cell motility under high bicarbonate concentrations, which are favorable to photosynthesis. As shown in Fig. 2, the presence of bicarbonate led to the decrease in the affinity of ATP to the catalytic site, resulting in the enhanced bicarbonate-induced suppression of Cya1 activity at low cellular ATP levels. This type of regulation may be important to minimize ATP consumption as well as facilitate photosynthetic ATP synthesis. Under low bicarbonate conditions, however, Cya1 showed high AC activity, even at low ATP levels, to facilitate cell motility, which may be of advantage to avoid photodamage. Therefore, it is possible to assume that Cya1 has evolved from a general bicarbonate-sensitive class III AC by obtaining inverse sensitivity to bicarbonate in response to its physiological functions.

On the basis of these considerations, we propose that Cya1 has evolved to regulate the motility of Synechocystis sp. strain PCC 6803 by changing the cellular cAMP level in response to bicarbonate and blue light. Apparently, this type of regulation has developed in order to maximize the photosynthetic efficiency and minimize photodamage.