Synthesis, purification, and properties of a semisynthetic ribonuclease S incorporating 4-fluoro-L-histidine at position 12.

Abstract The NH2-terminal pentadecapeptide of ribonuclease was synthesized with 4-fluoro-l-histidine (4-F-His) in place of the naturally occurring histidine at position 12. This peptide, [4-F-His 12]synthetic-(1-15), serves as an active site analogue of the native NH2-terminal eicosapeptide, RNase-S-(1-20), in that it forms a stable, noncovalent complex with the native ribonuclease fragment of residues 21 through 124, RNase-S-(21-124). A functional purification scheme has been used to prepare purified synthetic peptide in two steps. Purified semisynthetic ribonuclease S analogue complex was obtained by sulfoethyl-Sephadex chromatography of a mixture of crude, solid phase-derived peptide and native RNase-S-(21-124). Isolated [4-F-His 12]synthetic-(1-15) peptide was obtained from the purified complex after dissociation in 50 % acetic acid and gel permeation chromatography. The substitution of fluorine in the 4 position of the imidazole ring of histidine 12 does not disrupt the association of the synthetic peptide to native RNase-S-(21-124) as demonstrated by the binding constant measured in competition with native RNase-S-(1-20). Several other conformational properties of the analogue complex also are similar to those of the native RNase-S complex. However, the complex of [4-F-His 12]-synthetic-(1-15) and RNase-S-(21-124) is devoid of catalytic activity, a result attributed to the perturbation in the acid dissociation constant of the imidazole side chain by fluorine substitution.

A functional purification scheme has been used to prepare purified synthetic peptide in two steps. Purified semisynthetic ribonuclease S analogue complex was obtained by sulfoethyl-Sephadex chromatography of a mixture of crude, solid phase-derived peptide and native RNase-S-(21-124).
Several other conformational properties of the analogue complex also are similar to those of the native RNase-S complex.
Also, monoiodination of His 12 in REase-S-(1-20) has been rcportcd to eliminate the potential act'ivity obtained upon recombination with RXase-S-(21-124) as well as to reduce the binding constant for this association (9). Finn and Hofmann (15) have examined the importance to activity of the histidinyl residue at posit'ion 12 with various synthetic analogues.
It was initially shown that a I:cptide consisting of residues 1 to 13, as ~11 as a peptide of residues 1 to 12 with a UXH-terminal amide, produced full activity when combined with IlKas-S-(21.124).
However, a peptidc of residues 1 to 11, lacking His 12, teas unable to elicit activity.
Furthermore, substitution of a pyrazolyl group for the imidazole ringin position 12 of a 1 to 14 peptide complctcly eliminated activity, although the association of RSase-S-(21-124) \zas unaffected These studies serve to establish the critical catalytic requirement for an unperturbed imidazole ring at position 12 in ribonuclease

6296
The kinetic studies of Herries et al. (21) for a Step 24 substrate (cyclic CMP) and de1 Rosario and Hammes (cyclic UMP) (22) revealed kinetically apparent ionization constants in the range 5.4 to 7.5. Mechanisms consistent with the observed ionization constants may be written utilizing the histidine residues as general acid-general base catalysts (3, 23).
The involvement of His 12 and 119 was strongly supported by the structures of the active site that have emerged from the x-ray crystallographic work of Kartha et al. (24) on RNase-A and Wyckoff et al. (25) on RNase-S. The structure of a stable dinucleotide (UpcA) complexed with RNase-S is given in Ref. 3 and discussed in detail with respect to the implications for the catalytic mechanism.
Since the involvement of His 12 in the ribonuclease mechanism is so well established and its location in the folded structure well defined, this residue provides an ideal site for investigation of the structural and functional effect of minimal modification of catalytic side chains. As an example of this approach, we have synthetically incorporated 4-fluoro-t-histidine (27) into position 12 of the 1 to 15 sequence of RNase-S. If, as suggested, the mechanism of reaction of hydrolytic enzymes represented by RNase-A involves general acid-general base processes, then reduction of the pK, effected at His 12 by this substitution (27) should elicit altered kinetic properties.
This paper details the synthesis, describes the functional purification employed to obtain purified analogue peptide, and delineates the properties of this species in terms of its interaction with RNase-S-(21-124).

EXPERIMENTAL PROCEDURES
RNase-A (Worthington Biochemical Corp.) in phosphate buffer was dialyzed exhaustively against distilled water and lyophilized before use. RNase-S fragments were obtained by a procedure described nreviouslv (28).
Ribonullease enzymic activity assays were carried out spectrophotometrically against cCMP in 0.05 M Tris-HCl, pH 7.13, containing 55 mM NaCl (29). Assavs for activitv on UnA were performed using the prodedure described by Hammes and Walz (30). Concentrations of cCMP solutions were determined spectrophotometrically using an Et zti at 268 nm of 9.87 (29). Concentrations of peptide and protein solutions were based on recovery of amino acids after acid hydrolysis.
Amino acid compositions were obtained as described (28) unless otherwise noted. 4-F-His was prepared as reported earlier (27). N*-t-Boc-4-F-His was prepared by established procedures (28). The Na-t-Boc-4-F-His was converted to the N'm-Cbz derivative by reaction with Cbz chloride in sodium bicarbonate solution.
To test the effectiveness of the coupling of Na-t-Boc-F-His, an N"-t-Boc-prolyl Merrifield resin was prepared. The aminoacyl resin was deprotected, neutralized, and washed with chloroform, methvlene chloride. and DMF and then counled with iVa-t-Boc-4fluorohistidine in 5doj0 DMF-methylene chlor'ide overnight. After washing, a sample of the resin was hydrolyzed with a mixture of 0.5 ml of 6 N HCl and 0.5 ml propionic acid in a sealed, evacuated tube for 2 hours at 130" (31). Based on the recovery of proline and 4-F-His on the amino acid analyzer, the coupling was 89.5'%" complete after one coupling step.
Chemical synthesis of [4-F-His 12]synthetic-(l-15), the peptide corresponding to the NHr-terminal pentadecapeptide sequence of RNase-A with 4-F-His at position 12, was performed utilizing the Merrifield solid phase procedure with cleavage and deprotection 4 The accepted mechanism for phosphate diester hydrolysis involves initial formation of a cyclic 2':3'-phosphate intermediate (Step 1) followed by attack of water to produce the 3'-phosphate monoester (Step 2).
6 Uncorrected for losses during acid hydrolysis in the presence of resin. In the case of hydrolysis in dioxane-6 N HCl, control experiments with 4-F-His in the absence of resin indicate that losses under hydrolysis conditions may approach 65%. by anhydrous HF and piperidine-urea essentially as described previously (28).
After coupling Na-t-Boc-4-F-His to NH*-Met-Asp-Ser-resin a sample was hydrolyzed in 0.5 ml of dioxane plus 0.5 ml of concentrated HCl for 16 hours at 110". Based on the recovery of aspartic acid and 4-F-His, the coupling was 49y05 complete. At this point a second coupling was performed with AJ~-~-Boc-N~~-CBZ-~-F-His.
The amino acid composition of this sample showed no improvement in the apparent coupling yield.
Purified [4-F-His 12]synthetic-(l-15) was obtained by the following scheme. Twenty milligrams of crude, deblocked synthetic peptide were mixed with 40 mg of purified RNase-S-(21-124) in 3 ml of 0.1 M sodium phosphate buffer at pH 6.5. This sample was incubated at 5" for 1 hour and then applied to a sulfoethyl-sephadex column (2.3 X 140 cm) and eluted with 0.1 M sodium phosphate at pH 6.5. After the appearance of a peak of ultraviolet-absorbing material, the eluant was changed to 0.2 M sodium phosphate at the same pH. Elution was continued until all ultravioletpositive material came off the column and the eluate had the same conductivitv as 0.2 M phosphate. The elution profile is shown in Fig. 1 and discussed in "Results." The fractions containing the 4-fluorohistidinvl nentide nlus RNase-S-(21-124) (the semisvnthetic complex)"were'pooled, lyophilized, and then'redissolved in 8 ml of 0.05 M ammonium bicarbonate and applied to a Sephadex G-25 column (1.8 X 60 cm). The phosphate-free peak of protein was pooled and lyophilized, The yield at this point was roughly 10 mg of complex. The material was dissolved in 200 ~1 of 500/o acetic acid-water (v/v) and applied to a column (0.9 X 50-cm) of Sephadex G-75 in 5Ooj, acetic acid. The sample was eluted with 50% acetic acid and the fractions were assayed by ninhydrin reaction after alkaline hydrolysis.
The amino acid compositions of aliquots from peak tubes were obtained and the RNase-S-(21-124) and &F-His 12]synthetic- (I-15) were separately pooled and lyophiliecd several times to remove traces of acetic acid. The yield of purified synthetic peptide based on amino acid recovery was 0.5 mg. This was taken up in 1.0 ml of deionized water, providing the stock solution which was used for the experiments described below.
Difference spectra were obtained in the Cary 15 spectrophotometer using split cell cuvettes. Solutions of RNase-S-(21-124) and either  or [4-F-His 12]synthetic-(l-15) in 0.05 M Tris buffer, pH 7.11, were pipetted into separate sides of the blank and reference cuvettes. The base-line was recorded from 340 to 250 nm. The sample cuvette was then mixed by inversion and the difference spectrum was recorded.
Additions of peptide to the sample cuvette and to the peptide side of the reference cuvette were then made. An addition of buffer equal to g volume of the peptide solution added to the sample cell was added to the protein side of the reference cuvette. The sample cuvette was then remixed and the spectrum was recorded.
Additions were continued until no further change in absorbance was observed.
Thermal melting curves were obtained using stoppered cuvettes in a Gilford 240 spectrophotometer equipped with thermospacers and a Lauda constant temperature water bath. All peptide and protein samples were first passed through a 0.22~pm Millipore filter to remove nondissolved material.
Subtilopeptidase digestion at 5" was accomplished by incubating 250~~1 samples of RNase-S-(21-124), RNase-S, and semisynthetic complex at a concentration of 3 X lo-+ M in the cold room. Four micrograms of subtilopeptidase A (Sigma) was added to each incubation and RNase activity was determined as a function of time.

RESULTS
After deblocking of the crude synthetic 4-F-His peptide and fractionation on Sephadex G-25, no activity could be detected when mixed in 0.1 M Tris buffer, pH 7.13, with RNase-S-(21-124) in a synthetic peptide to native protein ratio of 20: 1. By inhibition of RNase-S activity, the ratio of crude [4-F-His 12]synthetic-(l-15) necessary to yield 50% inhibition was found to be 17.
Since the inhibition of RNase S activity by [4-F-His 12]synthetic-(l-15) implies binding of the synthetic fragment to RNase-S-(21-124), we used this capacity to purify the synthetic material.
The synthetic analogue peptide was incubated with native RNase-S-(21-124) as described under "Experimental Procedures." As illustrated by Fig. 1, ion exchange chromatography of this mixture is sufficient to resolve the resulting semisynthetic complex (Peak 2) containing 4-F-His at position 12 from both excess synthetic peptide with no affinity for RNase-S-(21-124) (Peak 1) and the excess RNase-S-(21-124) (Peak 3). Identification of the peaks was established in the following way. Peak 1 had an ultraviolet spectrum with only a shoulder in the 270 to 280 nm region and strong absorbance at shorter wavelengths, while Peaks 2 and 3 exhibited typical protein spectra with well defined maxima at 277 to 278 nm. As shown in the inset of Fig. 1, addition of native  to aliquots from tubes of Peaks 2 and 3 regenerates RNase-S as measured by hydrolysis of cCMP.
The greater ratio of RNase-S-(1-20) to  needed to elicit full activity in aliquots from Peak 2 (lower curve of inset) is due to the inhibition by the presence of the [4-F-His lalsynthetic-(l-15).
The identification of the components in Fig. 1 are further confirmed by amino acid compositions obtained from aliquots of I'eaks 1, 2, and 3. The ratio of tvrosine to nhenvlalanine is 3 for Peak 3 and 2 for I'eak 2.
consistent with the assignment of these peaks with RNase-S: (21-124) and semisynthetic complex, respectively. The isolated complex was also devoid of activity at this point.
The semisynthetic complex in Peak 2, after desalting, was separated into RNase-S-  and synthetic peptide on a Sephadex G-75 column. Fig. 2 illustrates the separation obtained.
Peak B contains the synthetic peptide as determined by amino acid analysis.
Peak A of Fig. 2 is RNase-S-(21-124) and Peak C is ammonia arising from the buffer used in the previous column. Table I gives the composition of Peak 1~ obtained after acid hydrolysis.
Up to a lo-fold excess of the isolated synthetic peptide was   mixed with 2 x 10da pmoles of RNase-S-(21-124) and then aliquots were assayed for activity against either cCMP or UpA. In both cases the measured activity was no greater than background levels (RNase-S-(21-124) alone). The inhibitory activity of the synthetic peptide was exploited to measure the binding constant to RNase-S-(21-124) as suggested by Kenkare and Richards (9) . Fig. 3 gives the activity 6298 (pH 7.13 Tris buffer) in cCMP hydrolysis titration curves for addition of increasing amounts of  to RNase-S-(21-124) (constant at 2 x lo+ pmoles) both alone (upper curve) and in the presence of two concentrations of [4-F-His 12]synthetic-(l-15).
Equation 1 Table II gives the results of these calculations.
The peptide to protein ratio is indicated for each curve. The amounts of RNase-S-(21-124) in the upper set of spectra is roughly twice that in the lower set of spectra.
Absorbance maxima at 280 and 287 nm are present in both cases, with isosbestic points at 275 and 293 nm. In Fig. 5 a plot of the corrected absorbance difference versus RNase-S-(l-20) concentration is given.
The values of T,, the temperature corresponding to the midpoint of the thermal transition curve measured by ultraviolet spectral change at 287 nm, were obtained for the semisynthetic complex and related native species. These data are given in Table III. In Fig. 6, plots are presented for the natural logarithm of the per cent of initial potential RNase-S activity remaining for analogue and RNase-S complexes versus the hours after addition of subtilopeptidase A. As indicated, the half-life of the decay of S-protein increases upon adding [4-F-His 12]synthetic-(l-15) to form the semisynthetic complex. Furthermore, addition of 2'CMP to this complex provides a further resistance to proteolysis. The uppermost line refers to the degradation of RNase-S. DISCUSSION Solid phase peptide synthesis, since its introduction by Merrifield (32,33), has proven extremely valuable in probing questions of the function of amino acid residues of natural polypeptides. One major drawback to this synthetic tool, however, is the heterogeneity of the products formed due to truncation and deletion errors (34), or to side reactions during synthesis and deblocking steps, Several procedures have been developed for achieving purification of biologically active peptides derived from solid phase synthesis.
Several of these techniques are functionally based, depending for their success on the specific binding of the .;;I //+-   peptide, the template suitable for achieving this binding is the naturally derived RNase-S-(21-124) fragment. Selective association provides a basis for removing from the mixture of synthetic peptides those which form stable noncovalent adducts. By virtue of its much higher molecular weight, different ionic character and shape, the complex is readily separable from smaller fragments in the mixture, Peptides which form weak complexes with the protein template will be discriminated against if the separation procedure is selective for more stable complexes.
Any change in the structure of the amino acid added at a given position in the peptide may disturb the association of the peptide with the template so that the selective purification described above will not be possible. Indeed, for the method to be successful, truncations or deletions in the native peptide sequence must result in diminished binding or altered chromatographic properties of the complex.
In the case of RNase-S, the work of Hofmann and Scoffone and their associates has defined the changes in binding which occur upon alterations in amino acid residues at several positions (see Ref. 20 for a complete discussion). For example, it was demonstrated that any large NH*terminal truncation of  leads to weaker binding to RNase-S-(21-124) (19). Of particular relevance to the present study however is the finding of Hofmann et al. (19) that some alteration in the nature of the side chain at position 12 is tolerated as regards binding to RNase-S-(21-124).
Thus, in terms of association to RNase-S-(1-20), fl-pyraaolylalanine could substitute for histidine at t,his locus. However, the substitution of serine at position 12 gave a peptide with a 'i-fold weaker binding 6299 capacity.
The substitution of 4-fluorohistidine, which retains the aromatic character of the imidazole, would be expected to yield a derivative with binding properties similar to those of the normal peptide or of the pyraaolylalanine-substituted peptide. In fact, the inhibition of activity observed when crude [4-F-His 12]synthetic-(l-15) is added to RNase-S' indicates that some component of the synthetic mixture is binding to RNase-S-(21-124) and preventing productive binding of native RNase-S-(l-20).
The synthetic material that is capable of binding to RNase-S-(21-124) was allowed to form the noncovalent complex and then purified by ion exchange chromatography on sulfoethyl-Sephadex.
The improvements in the isolation procedure are reported under "Experimental Procedures" and "Results." In this separation we have selected out those peptides which form stable noncovalent complesrs under this set of conditions (5', 0.1 to 0.2 M phosphate buffer, pH 6.5). Siuce the extent of binding of RNase-S-(1-20) to RNase-S-(21-124) decreases as the temperature is increased (36), and since phosphate ion also aids in stabilization of the complex, variation in temperature and phosphate concentration allow some flexibility in this separation. Thus one might achieve an even more stringent selection by changing these conditions to disrupt the weaker complexes.
In other experiments it has been observed that the position of elution of the complex in the ion exchange chromatography depends to some extent on the sequence of the synthetic peptide. The complex containing [4-F-His 12]synthetic-(l-15) elutes in the same position as a complex containing synthetic-&15) of normal sequence.
After removal of phosphate by gel filtration, the complex was disrupted in 50% acetic acid and the fragments were isolated by Sephadex G-75 molecular sieving (Fig. 2). The amino acid composition of the recovered synthetic fragment is given in Table I and serves to identify this species as [4-F-His 12]synthetic-(l-15).
The presence of fluorine in the imidazole ring gives 4-F-His an elution position on the amino acid analyzer very different from that of histidine, such that 4-F-His appears on the long column in about the same position as valine.
Since fluorine reduces the pK of the imidazole ring from 6 to 2.5, the marked change in behavior is not surprising.
Despite this large change in the pK of the side chain at position 12, this peptide is still capable of binding quite firmly to RNase-S-(21-124). Fig. 3 shows the data which allow a calculation of the quantitative strength of this interaction. The synthetic peptide with 4-F-His at position 12 binds less strongly than the native  by only a factor of 2 to 3 (Table II). This is in agreement with the results of Hofmann and co-workers (16) for peptides containing P-pyrazolylalanine at position 12. Although complex formation is relatively unperturbed, the resulting [4-F-His 121 adduct is devoid of catalytic activity. With both UpA, a step one substrate, and cyclic C'MY, a Step 2 substrate, the semisynthetic enzyme demonstrated no detectable levels of catalysis.
This loss of activity may arise from a perturbation in the active site geometry induced by the substitution, an effect on substrate binding properties, or an alteration in the functional system responsible for the prototropic shifts that facilitate bond making and breaking events.
Changes in the active site geometry of a nature sufficient to disrupt catalytic function cannot be defined in detail by any technique short of x-ray diffraction.
Gn the other hand, the strong binding of [4-F-His 12lsynthetic peptide to RNase-S-molecule to the "active site region" of the complex; such binding is known to stabilize the RNase conformation (38). The orientation of the substrate or inhibitor when bound to the complex might be altered such that the binding is nonproductive, but, due t,o the minimal modification of the active site by the fluorine substit,ution, this is considered unlikely as dell.
Finally, we must consider the altered properties of the catalytic system. ils mentioned above, the fluorine atom alters the acidic character of the imidazole ring, reducing the pK, to 2.5. Since the evidence implicates a basic group with a pK, of 5.4 (or 6.4) in the RNase reaction mechanism and since the x-ray structure places His 12 in a very favorable location to act as a proton acceptor, the most reasonable mechanism for ribonuclcase act'ion involves the histidine acting as a general base. The alteration of the pK, of this histidine makes the derivative a much weaker base. Thus the lack of activity of the [4-F-His 12]synthetic-(l-15) complex with native  finds its most acceptablc explanation in an alteration in the effectiveness of the catalytic function.
Lack of large amounts of purified synt'hetic peptide prevented titration to saturation in that case. The spectral change produced by a given amount of added peptide is, on a molecular basis, virtually the same in both cases. This is in agreement with the similarity in binding constant obtained from inhibition measurements.
It is worthwhile noting that this good binding stands in contrast to the result obtained by  iodinated at His 12 (9). This enzymatically inactive derivative, presumably iodinated at either the 2 or 4 position (39), was bound to RNase-S-(21-124) 3200 times less strongly than unmodified RNase-S-(1-20).
The difference between the binding of the iodohistidine and fluorohistidine derivatives could arise from the different properties of the two substituted imidazole rings-from the different positions of substit'ution (4-fluoro versus 2-iodo, if this is the predominant product), or from alterations in other amino acid side chains in the iodination procedure.
Because of the uncertainty of the position of iodination in the imidazole ring, the lack of activity in the latter cast is more difficult to attribute to alteration in the functional character of the histidine.
,4n attempt was made to use the difference spectra obtained to define the binding const,ants of [4-F-His 12]synthetic-(l-15) and  to RNase-S-(21-124). Under the conditions employed in this study, the ultraviolet absorbancr difference spect,rum increases linearly with each addition of RNasc-S-(1-20) to a constant' amount of  up to roughly 1.1 eq. After this point further addition of pcptide produces no further spectral changes. This is illustrated in Fig. 5 where the solid line through the points makes an abrupt shift at high RNase-S-(1-20) concentration.
These spectral data clearly indicate that the binding is too tight to allow calculation of an association constant. In contrast to our conclusion, differerlce spectra have been utilized previously to calculate binding constants for association of Rh'ase-S-(1-20) analogues to the Rxase-S-(21-124) (40). The results obtained by Finn (401, when compared to data obtained from activity assays, lead t)o the conclusion that the binding corn stants are roughly two orders of rnagnit'udc lower, or about 5 x 10" M, in the absence of substrate than in its prcscnce. Indeed, earlier studies by Finn and Hofmann (15) on the properties of synthetic analogues of RT\'ase give data showing that the association of synthetic-Rr\'ase-S- (1-13) to RNase-S-(21-124) increases as the substrate is varied from cUhlP to cCi\lP and is (21-124) observed argues against any gross distortion.
Also, in Fig. 7 (A and B) we give one view of the active site region around His 12 with a fluorine atorn substituted in position 4 (solid ball in ,i).
This picture was generated from the coordinates of the UpcAmRNaseS complex from Wyckoff et al. (25), within the XRXY program of Fcldmann et al. (37) on an ADAGE graphics unit interfaced to a PDP-10 computer.
The circles centered on the atom positions in A have been computer drawn with radii equal to the covalent radii of the various atoms in an attempt to portray a space-filling view of this region.
It is obvious from close inspection of this structure that a fluorine atom in position 4 of the imidazolc ring produces no stcric impediment and thus that any large changes in active site geometry in [4-F-His Ia]semisynthetic RNasc-8' are unlikely.
Changes in substrate binding would explain a decrease in activity, but, as illustrated in Fig. 6, bhe competitive inhibitor 2'.CMP stabilizes the semisynthetic complex against digestion b> subt'ilopeptidase while it has no stabilizing effect on RNase-S-(21-124) alone. This result indicates binding of the substrate-like nearly stoichiometric when KNA is the substrate. The experiments in Fig. 6 would also lend support to this concept, since the effect of adding the inhibitor to the semisynt,hetic complex is undoubtedly to shift the association equilibrium ton-& complex. In spite of the consistency of these arguments, it is clear from our data (obtained at pH 7.11) that the concentrations of KXase-S-(21.124) needed to observe difference spectra, generally about 2 x 10e5 M, are large enough with respect to the proposed association constant of the complex even in the absence of substrate (about 5 x 10" ,\I), to lead to stoichiomctric binding. Under thrse conditions the extrapolation procedure used previously (41) is not valid and one cannot obtain a binding constant for direct, compnrisorl to the value derived from activity assays. Sincse t>hc observed binding caonst,ant iI1 the prcscncr of substrate under these conditions is 5 x IO7 M it is still possible that, the peptideprotein asso&tion is incrcascd to SOIIX drgrcc over that in thtl abscncc of substrate.
IIowcvcr, the cstcllt of such an increasr cannot be quantitatcd by diffcrencc spc~tral data sucah as that rcportcd here. III any modification procrdure, it is incumbollt upon the illvcstigator to ascertain the cffcct of modification on the structural character of the molecule as well as to assay the rffcct on the functional propcrtics.
If the motlificatioll is substantial, a loss of activity might be due to a distortion of the active xitc. FTC have prcparcd a minimally modified scmisyntjhctic cneymc which has structural propcrtics comparable to those of the rlativc pcptitlcprotein complex.
The lack of activity of t,his derivative can thus bc directly related to the altcrcd functional character of the motlificd amilm acid side chain. The observations that wc have made arc in accord with the proposed role of the imitlazolc ring of histidinc 12 as a general base in the catalytic mechanism of RNasc-iZ and RNasc-S.