The catalytic core of RNase P
Christopher J.
Green*
,
Rafael
Rivera-León
and
Barbara S.
Vold
1
SRI International, 333 Ravenswood Avenue,
Menlo Park
, CA 94025-3493,
USA
and
1
SYVA Company, 900 Arastradero Road, PO Box 10058,
Palo Alto
, CA 94303,
USA
Received December 27, 1995;
Revised and Accepted March 5, 1996
ABSTRACT
A deletion mutant of the catalytic RNA component of
Escherichia coli
RNase P missing residues 87-241 retains the ability to interact with the protein component to form a
functional catalyst. The deletion of this phylogenetically conserved region
significantly increases the
K
m, indicating that the deleted structures may be important for binding to the
precursor tRNA substrate but not for the cleavage reaction. Under some reaction
conditions, this RNase P deletion mutant can become a relatively non-specific nuclease, indicating that this RNA's catalytic center may be more
exposed. The catalytic core of the RNase P is formed by less than one third of
the 377 residues of the RNase P RNA.
INTRODUCTION
RNase P is a ribonucleoprotein responsible for cleaving the 5'-leader sequence from precursors of tRNA and 4.5 S RNA. In bacteria,
this enzyme consists of an RNA subunit of ~400 bases and a protein subunit of ~120 amino acids. The RNA subunit has been shown to be capable of
catalyzing the cleavage of the substrate by itself under appropriate buffer
conditions and was the first true ribozyme to be discovered (
1
). The
K
m
for the RNase P RNA-catalyzed cleavage of tRNA precursors has been shown to be the same as
that for the RNase P holoenzyme, indicating that the RNA subunit supplies the
binding surface for this type of substrate. The protein subunit has been
demonstrated to reduce the ionic strength required for optimal activity and to
increase the turnover of the reaction by speeding release of the cleaved
product (
2
). Phylogenetic analyses have been used to develop the secondary structure model
for RNase P RNA shown in Figure
1
(
3
,
4
).
MATERIALS AND METHODS
Construction of RNA templates
The DNA templates for the RNase P RNA transcription were constructed by using
the polymerase chain reaction (PCR). The control normal RNase P RNA was
amplified by PCR from
E.coli
genomic DNA using primers with the following sequences:
CTCAAGCTTAATACGACTCACTATAGGGAGCTGACC and GGGAACTGACCGATAAGCCGGG. These two
primers match the 5'- and 3'-ends of the RNase P RNA gene, respectively. The first
primer includes the 17-base-long T7 RNA polymerase promoter and a
Hin
dIII restriction site at the 5'-end. The PCR-amplified DNA was cleaved with
Hin
dIII and then cloned into pUC18 cut with
Hin
dIII and
Sma
I (a blunt end restriction enzyme). This and all of the other PCR-generated constructs were then sequenced from both strands to make sure
that there were no PCR errors. To make a T7 polymerase template, this plasmid
was cleaved with
Kpn
I. Transcription from this template produces an essentially normal
E.coli
RNase P RNA as shown in Figure
1
. The only differences from the natural sequence are found at the extreme 5'- and 3'-ends that were changed to accommodate the T7
polymerase transcription start and the
Kpn
I site.
The transcription templates for deletion mutants RNase P RNA were also
constructed by the use of PCR of the original normal RNase P template. A
template that produced a deletion of the first 11 and the last 34 residues of
RNase P RNA ([Delta]1-11:344-377 P RNA) was made by PCR using primers with the sequence
CTCAAGCTTAATACGACTCACTATAGGGACAGTCGCCGCTTCGTC and GGGACAGTCATTCATCTAGGC. The
first of these primers included the 17-base-long T7 RNA polymerase promoter and a
Hin
dIII restriction site at the 5'-end for cloning as described above.
The template for the [Delta]87-241 P RNA was generated by three separate PCR reactions. The first
used the 5'-end primer used for the normal RNase P template and a primer with
the sequence: TTTGGCCTTGCTCCGTGCCCTATGGAGCC. This later primer was
complementary to the region that spanned the 87-241 deletion. The 3'-end primer used to produce the normal RNase P template was
used in a PCR reaction with a primer with the sequence: CGGAGCAAGGCCAAA. This
primer matches the sequence from 242 to 256 in RNase P RNA. The two purified
PCR-amplified fragments obtained in these reactions were then combined and PCR-amplified once more with the same primers used to make the normal
RNase P template. This resulted in a transcription template that could produce
the [Delta]87-241 P RNA (Fig.
1
). This PCR-generated DNA was also cloned and sequenced in pUC18.
RNA transcription
The cloned and sequenced templates described above were linearized with
Kpn
I to generate RNA by using T7 RNA polymerase transcription with the RiboMAX kit
(Promega). The resulting RNA transcripts were further purified on 5%
polyacrylamide-7 M urea gels although only one RNA of the expected size was produced
from each transcription. As a precaution against the possibility of
contamination with the normal RNase P RNA, the deleted versions RNAs were run
on separate gels. The secondary structures for the resulting RNase P RNA
deletion mutants shown in Figure
1
are only conjectural. They assume that the RNA secondary conformations will
follow the phylogenetically determined model of the full length RNase P RNA.
The protein component of RNase P
The vector used to express the protein component of RNase P has been described
in a previous paper (
20
). This vector was used to express the protein with an N-terminal leader that contains six consecutive histidine residues that can
be used to bind to a Ni-agarose column under either native or denaturing conditions. To remove all
traces of the host bacteria's RNA from this preparation, the recombinant
protein was put through an extensive purification procedure. The `His-tagged P' protein is extracted from the host by lysis in 6 M guanidine-HCl, loaded on the Ni-agarose column, washed extensively with 8 M urea, and eluted
with a low pH buffer in 8 M urea (
24
). The protein in 8 M urea is then passed through DEAE-cellulose to remove any RNA contaminant that survived the first step. The
resulting material is then loaded on a carboxymethyl-agarose in 8 M urea and eluted with a 0-300 mM NaCl gradient in 8 M urea, a method similar to purification
procedures used for the wild-type P protein (
25
). RNA should not bind to the carboxymethyl-agarose directly, and the urea should keep the RNA from binding to the
protein. The protein is then loaded on a new Ni-agarose column in 8 M urea, again washed extensively, and then refolded on
the column by using a gentle gradient (50 column volumes) to remove the urea.
The pure renatured protein is eluted with 250 mM imidazole and is kept at -20oC in 50% glycerol at concentrations of 0.5-1 mg/ml. It is very stable under these conditions for >= 12 months and shows no RNase P activity without the
addition of the catalytic RNA component.
The tRNA
His
precursor substrate for RNase P
The
32
P-labeled
B.subtilis
tRNA
His
precursor substrate used for this work was prepared by SP6 RNA polymerase
transcription of a gene cloned into a pSP64 plasmid (
21
). Normal RNase P cleavage of this substrate produces a eight base-pair aminoacyl stem (
21
).
RNase P processing
The processing of the substrate by the normal and deleted RNA component of RNase
P was attempted in three different buffers. They were: 100 mM NH
4
Cl, 100 mM MgCl
2
, 50 mM Tris-HCl pH 7.5, 5% polyethylene glycol (PEG, M.W. 8000; Sigma); 1.2 M NH
4
Cl, 250 mM MgCl
2
, 50 mM Tris-HCl pH 8.0; and 3 M NH
4
OAc, 250 mM MgCl
2
, 50 mM Tris-HCl pH 8.0, 0.01% sodium dodecyl sulfate (SDS).
Assays of the normal RNase P RNA component with the recombinant protein were
performed as described previously (
20
). The processing of the substrate by normal and deleted RNA versions of the
RNase P holoenzyme was performed in an unusually low ionic strength buffer
containing 50 mM NH
4
Cl, 10 mM MgCl
2
, 25 mM Tris-HCl pH 7.5. This buffer was optimized for the RNA deletion mutants but
was not optimal for the normal reconstituted RNase P. To increase the activity
of the deleted RNA-containing holoenzyme, PEG (M.W. 8000, Sigma) concentrations in the range
of 5-20% were used.
The kinetic constants of normal and [Delta]87-241 P RNA-containing RNase P were determined in 50 mM NH
4
Cl, 10 mM MgCl
2
, 25 mM Tris-HCl pH 7.5, 5% PEG buffer at 37oC. The substrate was the
32
P-labeled histidine tRNA precursor. The normal RNase P concentration was 5 * 10
-9
M, with substrate concentrations ranging from 0.067 * 10
-6
to 10
-6
M. The [Delta]87-241 P RNA-containing RNase P concentration was 10
-7
M, with substrate concentrations ranging from 0.067 * 10
-5
to 10
-5
M. The kinetic values for the [Delta]87-241 P RNA-containing holoenzyme are approximations because of the
practical difficulties in achieving substrate concentrations in excess of its
K
m
. Products and substrate were separated by gel electrophoresis (
21
). The buffer concentration is not optimal for normal RNase P and represents a
compromise between increased activity of the deletion mutant and loss of
specificity. Without PEG, the activity of the [Delta]87-241 P RNA-containing RNase P was too slight to obtain kinetic data.
Determination of the RNase P cleavage site
The substrate cleavage site caused by the normal and deleted RNase P RNA was
determined by reverse transcriptase sequencing of the cleavage products as
described previously (
21
).
RESULTS
Figure
2
A shows an autoradiograph of an attempt to cleave the substrate in high ionic
strength buffer without the protein component. Unlike the normal RNase P RNA,
neither the [Delta]87-241 nor the [Delta]1-11:344-377 P RNA alone had any detectable ability to
cleave the histidine tRNA precursor in high ionic strength buffers. In addition
to the buffer used in Figure
2
A , two other buffers were tried, also without success: 100 mM NH
4
Cl, 100 mM MgCl
2
, 50 mM Tris-HCl pH 7.5, 5% PEG (M.W. 8000; Sigma); and 3 M NH
4
OAc, 250 mM MgCl
2
, 50 mM Tris-HCl pH 8.0, 0.01% SDS. If the two deleted RNase P RNAs have any substrate
cleavage ability, it must be at least four orders of magnitude lower than that
of normal RNase P RNA. However, in the presence of the RNase P protein
component in a low ionic strength buffer, the [Delta]87-241 P RNA demonstrated significant and specific catalytic cleavage
of the histidine tRNA precursor (Fig.
2
B). The cleavage site was the same as that of the normal RNase P and resulted in
an 8 bp histidine-tRNA aminoacyl stem (
21
). However, even in the presence of the RNase P protein component, the [Delta]1-11:344-377 P RNA could not cleave the substrate. The buffer used
for these holoenzyme reactions contained 5% PEG, which, compared to buffers
without PEG, improved the cleavage reaction of the [Delta]87-241 P RNA-holoenzyme by at least an order of magnitude.
Although the buffer conditions for the substrate cleavage performed by the [Delta]87-241 P RNA-holoenzyme are not optimum for the normal RNase P, they can
be used to compare relative kinetic constants for the two catalytic species.
Without the presence of PEG, the activity of the [Delta]87-241 P RNA-holoenzyme is too slight to obtain kinetic data. Table
1
compares the
K
m
and
k
cat
of the [Delta]87-241 P RNA holoenzyme with those of normal RNase P under the same
buffer conditions. The
K
m
of the [Delta]87-241 P RNA holoenzyme is 100-fold higher, indicating that the substrate has much poorer
binding to this catalyst. Interestingly, the
k
cat
is approximately the same, indicating that the catalytic core of the RNA
moiety, when bound to the protein component, is not affected significantly by
the deletion of residues 87-241.
REFERENCES
1 Guerrier-Takada,C., Gardiner,K., Marsh,T., Pace,N. and Altman,S. (1983) Cell35, 849-857.
