Monitoring the structure of Escherichia coli RNase P RNA in the presence of various divalent metal ions

Abstract

Lead(II)-induced cleavage can be used as a tool to probe conformational changes in RNA. In this report, we have investigated the conformation of M1 RNA, the catalytic subunit of Escherichia coli RNase P, by studying the lead(II)-induced cleavage pattern in the presence of various divalent metal ions. Our data suggest that the overall conformation of M1 RNA is very similar in the presence of Mg²⁺, Mn²⁺, Ca²⁺, Sr²⁺ and Ba²⁺, while it is changed compared to the Mg²⁺-induced conformation in the presence of other divalent metal ions, Cd²⁺ for example. We also observed that correct folding of some M1 RNA domains is promoted by Pb²⁺, while folding of other domain(s) requires the additional presence of other divalent metal ions, cobalt(III) hexamine or spermidine. Based on the suppression of Pb²⁺ cleavage at increasing concentrations of various divalent metal ions, our findings suggest that different divalent metal ions bind with different affinities to M1 RNA as well as to an RNase P hairpin–loop substrate and yeast tRNA^Phe. We suggest that this approach can be used to obtain information about the relative binding strength for different divalent metal ions to RNA in general, as well as to specific RNA divalent metal ion binding sites. Of those studied in this report, Mn²⁺ is generally among the strongest RNA binders.

INTRODUCTION

Folding of RNA is facilitated by the presence of metal ions. It has been demonstrated that several divalent metal ions can promote the folding of RNA molecules required for catalytic activity (1). Our studies focus on the mechanism of cleavage by RNase P, an endoribonuclease in which the catalytic activity resides in the RNA subunit, RNase P RNA (2). At physiological pH the reaction catalyzed by Escherichia coli RNase P RNA, M1 RNA, can proceed in the presence of Mg²⁺, Mn²⁺ and Ca²⁺, and among these, Mg²⁺ promotes cleavage most efficiently (3–10). These findings suggest that M1 RNA is folded in a similar way in the presence of each of these three divalent metal ions. Cross-linking analysis of M1 RNA and its substrate, and lead(II)-induced cleavage studies of M1 RNA, also indicate that the overall conformation of M1 RNA is not significantly altered due to replacement of Mg²⁺ with either Ca²⁺ or Mn²⁺ (6,11). To investigate whether the structure of M1 RNA does change as result of replacing Mg²⁺ with other divalent metal ions, we decided to study Pb²⁺ cleavage of M1 RNA in the presence of different concentrations of Mg²⁺, Ca²⁺ and Mn²⁺, as well as in the presence of other divalent metal ions that do not promote M1 RNA cleavage by themselves. Our data suggest that: (i) M1 RNA is similarly folded in the presence of Mg²⁺, Mn²⁺, Ca²⁺, Sr²⁺ and Ba²⁺, while it is changed compared to the Mg²⁺-induced conformation in the presence of other divalent metal ions, for example Cd²⁺; (ii) folding of some M1 RNA domains occurs in the presence of Pb²⁺ only, while folding of other domain(s) requires the additional presence of other divalent metal ions, cobalt(III) hexamine or spermidine; and (iii) different divalent metal ions bind with different affinities to M1 RNA, where Mn²⁺ is in general among the strongest RNA binders. Our findings are discussed in relation to the hypothesis that there are two categories of divalent metal ions, structural and catalytic, in the M1 RNA-catalyzed reaction.

MATERIALS AND METHODS

Preparation of RNA and labeling procedures

The M1 RNA gene construct behind the T7 promoter, used to generate M1 RNA according to Milligan et al. (12), has been described elsewhere (13). The pATSerCGSyn was chemically synthesized and purchased from Xeragon AG (Zürich, Switzerland) whereas yeast tRNA^Phe was purchased from Sigma (Stockholm, Sweden).

The different RNA molecules were labeled at the 3′-end with [³²P]pCp according to standard procedures. The labeled RNA were subsequently purified on denaturing polyacrylamide gels where the percentage of polyacrylamide used was dependent on the size of the RNA. The various RNA molecules of correct size were eluted as previously described (8).

Pb²⁺-induced cleavage of M1 RNA, pATSerCGSyn and yeast tRNA^Phe

M1 RNA. Lead(II)-induced cleavage of 3′-end-labeled M1 RNA was performed essentially as described previously (8,14). Approximately 20 000–30 000 c.p.m. of labeled M1 RNA was mixed with ∼2.5 pmol of unlabeled M1 RNA and preincubated in 50 mM Tris–HCl pH 7.5, 100 mM NH₄Cl and various concentrations of different metal ion concentrations as indicated. The RNA was heated for 2 min at 70°C, followed by incubation for 10 min at 37°C. In all cases we used freshly prepared MeCl₂, except for barium where Ba(OAc)₂ was used. Cleavage was initiated by the addition of freshly prepared Pb(OAc)₂ to a final concentration of 0.5 mM (final reaction volume 10 µl). The reaction was allowed to proceed for 6 min at 37°C (7 min in the case of Co(NH3)₆³⁺ and spermidine; Fig. 3). To terminate the reaction, 2 vol stop solution (30 mM EDTA in 10 M urea supplemented with 0.1% bromophenol blue) was added. The cleavage products were separated on 8% denaturing polyacrylamide gels and detected using a PhosphorImager (Molecular Dynamics).

Figure 3.

Lead(II)-induced cleavage patterns of M1 RNA in the presence of increasing concentrations (in mM) of the Mg(H₂O)₆²⁺ analog Co(NH₃)₆³⁺ and spermidine, as indicated. Cleavage was performed at the indicated concentrations and 37°C, as outlined in Materials and Methods and Figure 2 legend, with the exception that the time of incubation in the presence of Pb²⁺ was 7 instead of 6 min. Ctrl 1, incubation in the presence of 50 mM Tris–HCl pH 7.5, 100 mM NH₄Cl, 10 mM Co(NH₃)₆³⁺, 10 mM spermidine, 10 mM MgCl₂ and without Pb²⁺; Ctrl 2, as Ctrl 1 but incubation in the presence of 50 mM Tris–HCl pH 7.5 and 100 mM NH₄Cl (without Pb²⁺) only; Mg²⁺, Pb²⁺ cleavage in the presence of 50 mM Tris–HCl pH 7.5, 100 mM NH₄Cl and 5 mM MgCl₂. Roman numerals indicate the cleavage sites as shown in Figure 1.

pATSerCGSyn. After gel purification ∼40 000 c.p.m. of 3′-end-labeled pATSerCGSyn was mixed with 20 pmol of unlabeled pATSerCGSyn in 50 mM Tris–HCl pH 7.2, 100 mM NH₄Cl and different concentrations of divalent metal ions (0–50 mM). The mixture was heated and preincubated as described above. Lead(II)-induced cleavage was initiated by adding freshly prepared Pb(OAc)₂ to a final concentration of 2 mM and terminated as described above. The cleavage products were separated on a 22% denaturing polyacrylamide gel followed by detection using a PhosphorImager. The frequency of cleavage as a function of divalent metal ion concentration was calculated (see Fig. 5 for which of the fragments we used for the calculations) where 100% cleavage represents cleavage in the presence of Pb²⁺ alone. The percentage of cleavage was plotted against the divalent metal ion concentration and from these plots apparent K_i values were calculated. The apparent K_i value was defined as the concentration of divalent metal ion resulting in 50% inhibition of Pb²⁺ cleavage.

Figure 5.

(A) Lead(II)-induced cleavage of 3′-labeled pATSerCGSyn in the absence and presence of Pb(OAc)₂, as indicated. The arrows indicate the fragment (–3) used to calculate the apparent K_i values given in Table 1 and the RNase P cleavage site at position +1. The sizes of the fragments were determined by parallel runs of pATSerCGSyn cleaved with M1 RNA, which cleaves this substrate at the +1 position (see also 11). Pb²⁺ cleavage was performed at 37°C for 6 min as described in Materials and Methods. Lane 1, no Pb(OAc)₂ added; lane 2, 2 mM (final concentration) of Pb(OAc)₂ added (no other divalent metal ion added); lane 3, 2 mM Pb(OAc)₂ and 50 mM MgCl₂ (final concentrations) added. (B) Relative percentage of Pb²⁺ cleavage plotted as a function of increasing concentration of MgCl₂ using fragment –3, as indicated in (A). The apparent K_i value was determined as the concentration of MgCl₂ resulting in 50% inhibition. The apparent K_i values are summarized in Table 1.

Yeast tRNA^Phe. Pb²⁺ cleavage of yeast tRNA^Phe was essentially performed as described by Behlen et al. (15). Gel purified 3′-labeled yeast tRNA^Phe (15 000 c.p.m.) was mixed with 100 ng of unlabeled yeast tRNA^Phe in 50 mM Tris–HCl pH 7.2, 15 mM NH₄Cl, 1.5 mM spermidine and different concentrations (0–100 mM) of divalent metal ions (Mg²⁺, Ca²⁺, Sr²⁺, Mn²⁺ or Zn²⁺). The mixture was heated at 70°C for 1 min followed by incubation at 25°C for 5 min. Lead(II)-induced cleavage was initiated by the addition of freshly prepared Pb(OAc)₂ to a final concentration of 0.5 mM. The reaction was terminated after incubation for an additional 3 min at 25°C by the addition of stop solution (25 mM EDTA, 7 M urea and 0.1% bromophenol blue). The products were separated on a 10% denaturing polyacrylamide gel and apparent K_i values were calculated as described above for pATSerCGSyn.

RESULTS

Conformational changes of M1 RNA as a function of increasing Mg²⁺ concentration

Lead(II)-induced cleavage has previously been used to monitor structural changes in M1 RNA as a result of base substitutions, deletions or interaction with the substrate (14,16–19). In the first set of experiments we investigated the Pb²⁺ cleavage pattern of M1 RNA as a function of increasing Mg²⁺ concentration to probe the conformation of M1 RNA at different Mg²⁺ concentrations. From the data summarized in Figure 1 and shown in Figure 2, we conclude that M1 RNA undergoes conformational changes with increasing Mg²⁺ concentration and that M1 RNA is completely folded at 5–10 mM Mg²⁺, in keeping with previous findings (20–22). Also, consistent with previous reports (14) we observed that higher concentrations of Mg²⁺ suppressed the lead(II) cleavage pattern, indicating that Mg²⁺ binds to the same or overlapping sites as Pb²⁺. Irrespective of Mg²⁺, Pb²⁺ cleavage at positions IIc, III, IV and V was detected (for numbering of cleavage sites see Fig. 1). This suggests that the folding of these regions of M1 RNA, including the P15 internal loop, is promoted by Pb²⁺ alone, in keeping with our previous data (23). In contrast, Pb²⁺ cleavage at positions Ia, Ib and IIb was not observed or was significantly reduced (Ib) in the absence of Mg²⁺. Note that band IIb appears with a simultaneous reduction of cleavage at site IIb′. Also, a band (X) migrating slightly slower than IIa was only observed in the presence of Pb²⁺ and low Mg²⁺ concentrations. Hence, proper folding of this part of M1 RNA that results in efficient Pb²⁺ cleavage at these positions (Ia, Ib and IIb) depends on the presence of Mg²⁺. In conclusion, an increase in the concentration of Mg²⁺ resulted in conformational changes of M1 RNA. These findings are in agreement with the notion that the folding of M1 RNA proceeds via intermediates (20–22,24).

Figure 1.

A secondary structure model of M1 RNA according to Haas et al. (42). The Pb²⁺ cleavage sites are indicated by arrows and italic roman numerals (14,18), whereas the roman numerals indicate the Mg²⁺-induced cleavage sites according to Kazakov and Altman (3).

Figure 2.

Lead(II)-induced cleavage patterns of M1 RNA in the presence of increasing concentrations (in mM) of various divalent metal ions, as indicated. Cleavage was performed at the indicated concentrations and 37°C, as outlined in Materials and Methods. Only Pb²⁺, incubation in the presence of 0.5 mM Pb²⁺ only; Ctrl, incubation in the presence of the divalent metal ion, 10 mM (final concentration), as indicated in the absence of Pb²⁺. Pb²⁺ cleavage in the presence of Mg²⁺:Cd²⁺ was performed under the following conditions: 0.2 M Mg²⁺:Cd²⁺ stock solutions with 5:3 Mg²⁺:Cd²⁺ ratio (or 5:1) were prepared and these were diluted to give the final concentration in mM as indicated (data not shown for cleavage performed using the latter conditions since the Pb²⁺ cleavage patterns were identical irrespective of Mg²⁺:Cd²⁺ ratio). The roman numerals refer to the sites of cleavage in M1 RNA as shown in Figure 1.

Overall conformation of M1 RNA in the presence of other divalent metal ions

Cleavage by M1 RNA can proceed in the presence of other divalent metal ions (see above). Therefore, we next analyzed the conformation of M1 RNA as a function of increasing concentrations of other divalent metal ions, as indicated in Figure 2. Here we observed that the lead(II)-induced cleavage pattern in the presence of Ca²⁺, Mn²⁺, Sr²⁺ or Ba²⁺ was very similar to the pattern observed in the presence of Mg²⁺. The Pb²⁺ cleavage pattern generated in the presence of 5 mM Mn²⁺ is in keeping with our previous data (11). These data suggest that the global conformation of M1 RNA in the presence of each of these divalent metal ions is very similar to the structure of M1 RNA formed in the presence of Mg²⁺. However, a higher concentration of Sr²⁺ or Ba²⁺ relative to Mg²⁺ was required for proper folding of M1 RNA, while a lower concentration of Mn²⁺ was required. In the case of Ca²⁺, approximately equal concentration, as in the case of Mg²⁺, was required. Furthermore, we observed that lower and higher concentrations of Mn²⁺ and Sr²⁺ or Ba²⁺, respectively, relative to Mg²⁺, were needed to reduce the site-specific Pb²⁺ cleavage (Fig. 2). These data suggest that Mn²⁺ has a higher overall affinity for M1 RNA than the other divalent metal ions. Based on these findings it is possible to rank these divalent metal ions with respect to affinity to M1 RNA: Mn²⁺ > Mg²⁺ ≈ Ca²⁺ > Sr²⁺ ≈ Ba²⁺, where Mn²⁺ has the highest affinity.

We also studied lead(II)-induced cleavage in the presence of Co²⁺, Ni²⁺, Cu²⁺, Zn²⁺ and Cd²⁺. None of these divalent metal ions promote cleavage by themselves, rather, addition of any one of these metal ions is inhibitory (21; data not shown). From the data shown in Figure 2 it is clear that cleavage by Pb²⁺ in the presence of Ni²⁺, Cu²⁺ and Cd²⁺ resulted in a different pattern compared to the ‘Mg²⁺-induced’ cleavage pattern, in particular with respect to cleavage at sites Ia, Ib and IIb. In contrast, both Co²⁺ and Zn²⁺ promoted Pb²⁺ cleavage at sites Ia and Ib at low concentrations, while cleavage at site IIb was absent (Zn²⁺) or very weak (Co²⁺). These findings suggest that the conformation of the region(s) of M1 RNA involving sites Ia, Ib and IIb is different in the presence of in particular Ni²⁺, Cu²⁺ and Cd²⁺ relative to the conformation generated in the presence of Mg²⁺. We emphasize that none of these divalent metal ions (or combinations) by themselves induced detectable cleavage of M1 RNA under these conditions (Fig. 2).

From Figure 2 it is apparent that replacement of Mg²⁺ with Cd²⁺ resulted in a change in the conformation of M1 RNA. Previously it was shown that M1 RNA is catalytically active under reaction conditions where both Mg²⁺ and Cd²⁺ are present in a 5:1 ratio (9). Thus, we decided to study Pb²⁺ cleavage of M1 RNA in solutions with different ratios of Mg²⁺ and Cd²⁺ to investigate whether the structure of M1 RNA was changed compared to the structures observed in the presence of either of these divalent metal ions alone. As can be seen in Figure 2 the observed lead(II)-induced cleavage pattern was not significantly changed under conditions where the Mg²⁺:Cd²⁺ ratio was 5:1 or 5:3 compared to the pattern seen in the presence of only Mg²⁺ (data shown for Mg²⁺:Cd²⁺ ratio = 5:3 only). Taken together, this indicates that the conformation of M1 RNA was restored as a result of addition of Mg²⁺, giving a rationale to why M1 RNA is catalytically active in a solution of Mg²⁺ and Cd²⁺ but not in the presence of Cd²⁺ alone (or very low activity; 9).

M1 RNA structure in the presence of cobalt(III) hexamine and spermidine

Our data so far suggest that some, but not all, divalent metal ions can promote a ‘Mg²⁺-like’ conformation of M1 RNA as judged from the Pb²⁺ cleavage patterns. Cobalt(III) hexamine is considered to be an analog of fully solvated magnesium (25 and references therein). To investigate whether Pb²⁺ cleavage at particular sites Ia and Ib is promoted in the presence of this analog, we analyzed the Pb²⁺ cleavage pattern as a function of increasing concentration of Co(NH₃)₆³⁺. As shown in Figure 3, we observed that cleavage at sites Ia, Ib and IIb (to a lesser extent) were also promoted in the presence of this Mg(H₂O)₆²⁺ analog. At concentrations ≥1 mM the Pb²⁺ cleavage of M1 RNA was suppressed in a similar manner to that described above. These findings suggest that inner sphere metal ion coordination is not required to adopt a conformation of the region near sites Ia, Ib and IIb such that Pb²⁺ generates efficient cleavage at these positions.

It is known from previous studies that addition of spermidine to the M1 RNA cleavage reaction lowers the concentration requirements for Mg²⁺ (21,26). We therefore studied the influence of spermidine on the lead(II)-induced cleavage pattern. As observed in the case of addition of cobalt(III) hexamine, Pb²⁺ cleavage at sites Ia and Ib was promoted in the presence of spermidine, while cleavage in the IIa region and at IIb showed differences compared to Pb²⁺ cleavage in the presence of Mg²⁺. Given that cleavage by M1 RNA cannot be promoted by spermidine alone, this might suggest that a divalent metal ion-induced folding of these regions is necessary in order to adopt a catalytically active conformation of M1 RNA.

Relative RNA binding affinities for different divalent metal ions

As discussed above, Mn²⁺ ions bind more strongly to M1 RNA than Mg²⁺, while Sr²⁺ binds with a lower affinity. To investigate whether this is a general behavior of these divalent metal ions relative to RNA, we studied Pb²⁺ cleavage of two other RNA molecules, pATSerCGSyn and yeast tRNA^Phe (Fig. 4). The former is a well-characterized model substrate for RNase P (11,23) whereas lead(II)-induced cleavage and the structure of the latter is well-documented (27 and references therein). First, we demonstrated that pATSerCGSyn is indeed cleaved in the presence of Pb²⁺ (Fig. 5). As shown in Figure 5, cleavage occurred at several positions in the 5′ leader. Secondly, we added increasing concentrations of indicated divalent metal ions (Fig. 5; Table 1), which resulted in suppression of Pb²⁺-induced cleavage in a concentration-dependent manner (Fig. 5; data shown for Mg²⁺ only). Apparent K_i values for Mg²⁺, Mn²⁺, Ca²⁺, Sr²⁺, Zn²⁺ and Co(NH₃)₆³⁺ were calculated and the result showed that Mn²⁺ (and Zn²⁺) suppressed Pb²⁺ cleavage ∼3- and 5-fold more efficiently compared to Mg²⁺ and Sr²⁺, respectively (Table 1). The same tendency was observed when we studied suppression of Pb²⁺-induced cleavage of yeast tRNA^Phe, with the exception that a significantly higher apparent K_i value was observed for Zn²⁺. These findings, together with the data discussed above, clearly suggest that different divalent metal ions bind to RNA, generally with different strength, and that Mn²⁺ is among the strongest binders studied in this report. Furthermore, our Pb²⁺ cleavage studies using pATSerCGSyn suggest that different divalent metal ions can bind in the vicinity of the RNase P cleavage site. This is in keeping with previous reports where Mg²⁺ cleavage of a similar hairpin–loop RNA structure was studied (28,29). In addition, data obtained from the addition of Co(NH₃)₆³⁺ suggest that inner sphere coordination is not required to give suppression of Pb²⁺ cleavage in the 5′ leader.

Figure 4.

The secondary structures of pATSerCGSyn and yeast tRNA^Phe. The lead(II)-induced and RNase P cleavage sites are indicated by arrows. The number –3 refers to the cleavage fragment we used to calculate the apparent K_i values.

Table 1. Apparent K_i values for inhibition of Pb²⁺-induced cleavage of pATSerCGSyn and yeast tRNA^Phe.

Divalent Me2+	Apparent Ki/mM pATSerCGSyn	Apparent Ki/mM yeast tRNAPhe	Ionic radius (Å)a	Coordination numbera	Hardness absolutea	Exchange rates for H2O (s–1)a
Mg2+	9.3 ± 2.2	13 ± 1	0.72	6	47.59	105
Ca2+	6.7 ± 3.1	17 ± 1.6	1.12	8	19.52	108
Sr2+	16 ± 3.0	41 ± 0.1	1.18	6	27.30	108–109
Mn2+	3.4 ± 0.75	5.6 ± 0.4	0.67	6	9.02	107
Zn2+	2.8 ± 0.8	51 ± 0.8	0.74	6	10.88	107
Co(NH3)63+	1.0 ± 0.6	ND	0.61	6	8.90	1

The K_i values were determined as outlined in Materials and Methods and each number is an average of several independent experiments.

ND, not determined.

^aNumbers taken from Feig and Uhlenbeck (27) and Cowan (43).

DISCUSSION

Lead(II)-induced cleavage is a way to probe conformational changes in an RNA as a result of point mutations, deletions and substrate interaction (14,16–19,30–32). In addition, Pb²⁺ cleavage has been used to monitor the structural requirements for binding of Pb²⁺ in yeast tRNA^Phe that subsequently results in cleavage at a specific position (15,33). Here we demonstrated that it is possible to use this approach to monitor structural differences in E.coli RNase P RNA, M1 RNA, as a result of changes in divalent metal ion concentration and divalent metal ion replacements.

Kazakov and Altman (3) showed that Mg²⁺, Ca²⁺ and Ba²⁺ cleaved M1 RNA at approximately the same sites as Pb²⁺ [I (≈ Ia), II (= IIa), III–V for Mg²⁺ and Ca²⁺, while Ba²⁺ cleaves mostly at site I and to a small extent at the other positions; for details see Fig. 1]. This, together with our present data and data reported elsewhere (14), suggests that these metal ions bind to the same or to overlapping sites in M1 RNA, although with different affinities (this report). That various divalent metal ions bind to at least overlapping sites in solution is corroborated by our Pb²⁺ cleavage studies of an RNase P model substrate and of yeast tRNA^Phe, as well as crystal structure studies of the latter (for a review see 27). In conclusion, our findings suggest that it is possible to obtain information about the relative binding strengths for different divalent metal ions to RNA generally, as well as to specific divalent metal binding sites, by studying their effect on Pb²⁺ cleavage of RNA molecules.

Our data showed that Mn²⁺ binds with a higher affinity compared to Mg²⁺ to three different RNA molecules studied in this report. From the data reported by Streicher et al. (31) it appears that this is also the case in the T4 phage-derived td intron system. These data are in keeping with an earlier finding that Mn²⁺ binds more strongly to DNA than Mg²⁺ does (34), as well as the finding that cleavage both by E.coli and Bacillus subtilis RNase P RNA requires lower concentration of Mn²⁺ than Mg²⁺ for optimal cleavage (4,35). The ionic radii of Mg²⁺ and Mn²⁺, as well as the coordination of these two divalent metal ions, are very similar (Table 1). However, a major difference is that Mn²⁺ binds nitrogen more frequently compared to Mg²⁺, while Mg²⁺ (and Ca²⁺) prefers oxygen (36). Therefore, it is plausible that the increased affinity of Mn²⁺ for RNA is due to the fact that Mn²⁺ coordinates more frequently by inner sphere binding to the bases. In fact, Mn²⁺ coordinates directly to N7 of G20 in the crystal structure of yeast tRNA^Phe (37). The stronger binding of Mn²⁺ to RNA might consequently be a rationale to why addition of Mn²⁺ to the RNase P cleavage reaction resulted in an increased frequency of cleavage at an incorrect position (11).

By comparing the Pb²⁺ cleavage pattern generated in the presence of different divalent metal ions, we concluded that the conformation of M1 RNA is very similar in the presence of Mg²⁺, Mn²⁺, Ca²⁺, Sr²⁺ and Ba²⁺ (see above). Folding of M1 RNA in the presence of other divalent metal ions resulted in a changed overall conformation compared to the Mg²⁺ generated structure, in particular in regions near sites Ia, Ib and IIb. This is consistent with the notion that Mg²⁺, Mn²⁺ and Ca²⁺ alone promote catalysis (for references see above). In contrast, Sr²⁺ and Ba²⁺ do not result in cleavage at neutral pH and therefore these divalent metal ions are suggested to have a structural role under these conditions. This is in keeping with an earlier report that suggested that there are two classes of divalent metal ion binding sites in the M1 RNA-catalyzed reaction, catalytic and structural sites (21; see below). However, at higher pH and in the presence of ethanol, correct cleavage in the presence of either Sr²⁺ or Ba²⁺ alone has been observed (3), indicating that under specific conditions both these divalent metal ions can fulfil the role of catalytic metal ions. Furthermore, folding of the Pb²⁺ binding site that resulted in lead(II)-induced cleavage at the particular sites Ia and Ib was observed in the presence of Mg²⁺, Mn²⁺, Ca²⁺, Sr²⁺, Ba²⁺ (to a lower extent Zn²⁺ and Co²⁺), the Mg(H₂O)₆²⁺ analog Co(NH₃)₆³⁺ and spermidine. One interpretation of these findings is that the regions in the vicinity of sites Ia and Ib are mainly stabilized by positively charged ions, possibly by diffuse binding of fully hydrated metal ions (25). That efficient Pb²⁺-induced cleavage at sites Ia and Ib is dependent on a stabilized structure near site Ia is supported by our previous findings where we studied Pb²⁺ cleavage of M1 RNA variants carrying substitutions in the vicinity of site Ia (16,17).

Two classes of divalent metal ion binding sites have also been suggested for the group I intron ribozyme derived from Tetrahymena, L-21 ScaI (32,38,39). Here, only Mg²⁺ and Mn²⁺ can fulfil the catalytic role, i.e. bind to class I sites, whereas the presumed structural class II sites can be occupied by Mg²⁺ and Mn²⁺ as well as by Ca²⁺ and Sr²⁺ (38,40). Taken together, with respect to divalent metal ion binding, the two large ribozyme systems RNase P RNA and group I intron RNA show clear similarities (41) but also differences such as the difference in catalytic performance in the presence of Ca²⁺ alone.