Eur. J. Biochem. 76, 189-196 (1977)
Characterisation of RNA Fragments Obtained by Mild Nuclease Digestion of 30-S Ribosomal Subunits from Escherichia coli Jutta RINKE, Alexander ROSS, and Richard BRIMACOMBE
Max-Planck-Institut fur Molekulare Genetik, Abteilung Wittmann, Berlin-Dahlem (Received October 1, 1976)
When Escherichia coli 3 0 4 ribosomal subunits are hydrolysed under mild conditions, two ribonucleoprotein fragments of unequal size are produced. Knowledge of the RNA sequences contained in these hydrolysis products was required for the experiments described in the preceding paper, and the RNA sub-fragments have therefore been examined by oligonucleotide analysis. Two well-defined small fragments of free RNA, produced concomitantly with the ribonucleoprotein fragments, were also analysed. The larger ribonucleoprotein fragment, containing predominantly proteins S 4, S 5 , S 8, S 15, S 16 (17) and S20, contains a complex mixture of RNA sub-fragments varying from about 100 to 800 nucleotides in length. All these fragments arose from the 5’-terminal 900 nucleotides of 16-S RNA. corresponding to the well-known 12-S fragment. No long-range interactions could be detected within this RNA region in these experiments. The RNA from the smaller ribonucleoprotein fragment (containing proteins S7, S9, S10, S14 and S 19) has been described in detail previously, and consists of about 450 nucleotides near the 3‘ end of the 16-S RNA, but lacking the 3‘4erminal 150 nucleotides. The two small free RNA fragments (above) partly account for these missing 150 nucleotides; both fragments arose from section A of the 16-S RNA, but section J (the 3’-terminal 50 nucleotides) was not found. This result suggests that the 3‘ region of 16-S RNA is not involved in stable interactions with protein. It has been known for some time that Escherichiu coli 30-S ribosomal subunits can be split by mild nuclease digestion into two ribonucleoprotein fragments of unequal size [l -31. The smaller of these two fragments, which contains proteins S7, S9, SlO, S14 and S19 [3], has been characterized in detail, and we have recently published an analysis of the sequences of the RNA sub-fragments which are contained in this ribonucleoprotein particle [3]. These sequences cover a segment of about 450 nucleotides in the 3’proximal region of the 16-S RNA, but the last 150 nucleotides at the 3’ terminus of the 16-S molecule are absent. Further it was shown that a ‘tertiary’ (or ‘long-range’)interaction exists between distant regions of the RNA of the ribonucleoprotein fragment, since a stable RNA complex could be isolated containing two discontinuous regions of the RNA, which are separated by about 100 nucleotides in the intact 16-S molecule [4]. In this paper we describe a similar analysis of the RNA component of the larger of the two ribonucleoenzyme^. Ribonuclease TI (EC 3.1.4.8); ribonuclease A (EC 3.1.4.22).
protein fragments, which contains proteins S4, S5, S8, S15, S16(17) and S20, together with variable amounts of two or three other proteins [l]. The analysis was undertaken for a number of reasons. Firstly, it was necessary to know which RNA sequences are contained in this ribonucleoprotein fragment in order to be able to interpret our data on the distribution of proteins which are cross-linked by formaldehyde to this region of the RNA, described in the preceding paper [5]. Secondly, it is of interest to find out whether any tertiary interactions can be detected in this region of the RNA, as was the case in the RNA from the smaller ribonucleoprotein species [4]. Thirdly, the absence of the 3’-terminal 150 nucleotides of the 16-S RNA from the latter ribonucleoprotein fragment raises the question as to where these missing nucleotides are, in particular whether they can be found associated with the larger ribonucleoprotein fragment. The results showed that the RNA from the large ribonucleoprotein fragment consists of a complex series of sub-fragments covering a region of about 900 nucleotides at the 5’ end of the 16-S molecule, corresponding to the ‘12-S fragment’ [6]. We were unable to detect any stable tertiary interactions
190
RNA Fragments Obtained by Mild Digestion of E. coli 3 0 3 Ribosomes
within the RNA; as a result it is still not possible to say whether or not the tertiary interaction within this region (which has been clearly demonstrated in studies on complexes between isolated RNA and protein S4 [7,8]) exists in intact 30-S subunits. The missing 150 nucleotides from the 3‘ end of the 16-S RNA were not found associated with the large ribonucleoprotein fragment, but the greater part of this missing region was liberated from the 30-S subunits as discrete RNA fragments not associated with protein, during the initial hydrolysis. This provides further evidence that a large portion of the 3’ region of 16-S RNA is not tightly bound, and is rather easily accessible in the 30-S subunit [9 - 111.
Oligonucleotide Analysis Appropriate gel slices containing RNA fragments were homogenized, and the RNA extracted with phenol and dodecyl sulphate exactly as described [3]. After separation from phenol and isolation of the RNA by ethanol precipitation, the RNA was hydrolysed with ribonuclease TI and subjected to oligonucleotide analysis by the method of Sanger et LEI. [12], again exactly as described [3]. Analysis of the fingerprint data, and secondary digestions with ribonuclease A according to the method of Uchida et al. [13], were also made as before [3]. RESULTS
MATERIALS AND METHODS
Purification oj R N A Fragments
Isolation of R N A Fragments
30-S subunits were treated with ribonuclease TI under ‘standard’ conditions [3], and applied to a polyacrylamide gel, using the ‘pH-6’ system (see Materials and Methods and [4]). A typical gel pattern, in this instance from unlabelled subunits, is shown in Fig. 1. The two ribonucleoprotein fragments I1 and 111 can be seen, and in addition two sharp bands (marked A1 and A2) running close to the bromophenol blue marker. These two bands are particularly well defined in this gel system, and were observed when the hydrolysis was made either under ‘standard’ conditions (5 mM magnesium, slot 2, Fig. 1) or when the magnesium concentration was reduced to 1 mM (slot 3). In other experiments, using subunits labelled with 3H in the RNA and I4C in the protein moiety, it could be seen that bands A1 and A2, in contrast to ribonucleoprotein fragments I1 and 111, contained negligible amounts of protein (data not shown), and hence these bands are of free RNA as opposed to ribonucleoprotein. In experiments with 32P-labelledsubunits, identical gel patterns to that of Fig.1 were obtained, and the various components of this pattern, viz. ribonucleoprotein fragments I1 and 111, and RNA fragments A1 and A2, were subjected to further analysis. The ribonucleoprotein peak I, which has the same mobility as the unhydrolysed 30-S subunit (Fig. 1, slot I), and the heterogeneous RNA peak running ahead of bands A1 and A2 were not examined. Gel slices containing 32P-labelled ribonucleoprotein fragments I1 and I11 were applied to the 5 - 15 % gel gradient system (see Materials and Methods). This is a dodecyl sulphate gel system, with a thin layer of 3 % polyacrylamide containing 8 M urea on top of the gel, and typical RNA separations from both ribonucleoprotein fragments are shown in Fig.2. The RNA from ribonucleoprotein fragment 111 (Fig. 2 B) showed the same five sub-fragments as were previously observed on a 7 % gel with 7 M urea throughout [3], and oligonucleotide analysis of all five confirmed
32P-labelled 30-S ribosomal subunits were prepared from E. coli strain MRE 600, and were hydrolysed with ribonuclease TI in the presence of 5 mM magnesium and 2 M urea exactly as described previously [3]. The hydrolysis products were separated on a 5 % polyacrylamide gel, using the pH-6 buffer system described earlier [4], following which the gel was sliced and analysed for radioactivity. Gel slices containing small fragments of free RNA (see text) were submitted to a further electrophoresis on a 10 polyacrylamide gel containing 0.1 % dodecyl sulphate and 7 M urea, using the buffer system described earlier [4]. Gel slices containing the ribonucleoprotein fragments (see text) were treated in one of two ways. In the first method, the slices were loaded directly onto a 5 - 15 % polyacrylamide gradient gel containing dodecyl sulphate, using the buffer system described in the preceding paper [5], with the exceptions that Tris was used in place of triethanolamine, and formaldehyde was omitted. After polymerisation each gel slot was filled to a depth of 1 cm with a 3 % polyacrylamide gel containing 8 M urea in addition to dodecyl sulphate and buffer, and the sample slices were loaded on top of this layer. This gel system both removes protein and reveals ‘hidden breaks’ in the RNA. In the second method, the slices were loaded directly onto a 5 % polyacrylamide gel containing dodecyl sulphate as described [4];this system contains no urea and therefore does not reveal hidden breaks. After slicing and analysing for radioactivity, suitable gel fractions were either loaded directly onto a second gel or extracted with phenol as described [ 3 ] and then loaded onto a second gel, to reveal the hidden breaks in the RNA. This second gel was either the 5 - 15 ”/, gradient gel above, or alternatively a simple 7 % gel containing dodecyl sulphate with 7 M urea throughout, as described [4]. In all cases the final gel was sliced and analysed for radioactivity.
J. Rinke, A. Ross, and R. Brimacornbe
191 ~
A 1
1
2
3
-I -II
-m I
12
24
I 36
B
-
I
48
I 60
72
I 84
96
72
84
96
I
I
2
A1
- A2
12
24
36
48 60 Fraction number
108
Fig. 1. Hydrolysis patterns of 30-S subunits in the pH-6 gel system. The samples was of unlabelled 30-S subunits and the gel was stained with methylene blue [14]. Slot 1: 3 0 4 subunits, minus enzyme. Slot 2: 30-S subunits hydrolysed under ‘standard’ conditions [3]. Slot 3: as for slot 2, but with the magnesium concentration reduced from 5 mM to 1 mM. The positions of ribonucleoprotein fragments I, I1 and 111, the free RNA peaks A1 and A2, and the bromophenol blue dye marker are indicated
Fig. 2. [3 2 P ] R N Aprofiles from ribonucleoprotein ,fragments II and III, after electrophoresis on a 5- 15 %gradient gel containing dodecyl sulphate, with a 7 M urea layer. The fractions are 1.7-mm slices of the gel, and the position of a 5-S marker is shown. (A) RNA from ribonucleoprotein fragment 11. (B) RNA from ribonucleoprotein fragment 111, with sub-fragments numbered as in [3]
that the RNA sequences were the same as those reported [3]. This gel system gives a substantially better separation of the sub-fragments than the simple 7 % gel system [3]. Further, the fact that RNA from ribonucleoprotein IT1 showed the same five sub-fragments as before indicates that the thin urea layer on top of the gel is sufficient to reveal the hidden breaks in the RNA irreversibly. The RNA profile obtained from ribonucleoprotein fragment I1 was rather more complex than that of ribonucleoprotein fragment 111. RNA sub-fragments 1 and 2 (Fig.2A) could not always be separated, and there was usually a rather heterogeneous region in the neighbourhood of the 5-S RNA marker, between subfragments 8 and 10, which can be seen from Fig.2A. Gel fractions corresponding to the individual numbered peaks in Fig.2A were extracted with phenol, and the RNA was subjected to oligonucleotide analysis as described in Materials and Methods. Total RNA from ribonucleoprotein fragment I1 (not separated by electrophoresis on dodecyl sulphate gels) was also
analysed after phenol extraction from the pH-6 gel system (cf Fig. 1). The RNA bands A1 and A2 were purified by an electrophoresis step, using a 10% gel as described in Materials and Methods, and the result of such a purification is illustrated in Fig. 3. Both RNA species gave a single sharp peak, the one (A 1) slightly smaller than 5-S RNA, and the other (A2) slightly smaller than 4-S RNA. Both of these peaks were extracted with phenol and subjected to oligonucleotide analysis. Nucleotide Sequences in Ribonucleopvotein Fragment II
An example of a fingerprint of the largest RNA subfragment obtained from ribonucleoprotein fragment TI (fragment 1, Fig.2A) is shown in Fig.4, together with a schematic representation of the fingerprint which indicates the nomenclature used for describing the oligonucleotide spots (cf [13]). The fingerprint of total unfractionated RNA from ribonucleoprotein fragment I1 was very similar, the only obvious differ-
RNA Fragments Obtained by Mild Digestion of E. roll 30-S Ribosomes
192 I
I
I
I
I
I
I
48
60
72
84
5-5 4-5
1
36
96
Fraction number
Fig. 3. 3 2 P profiles of R N A frugmmts A1 and A2 on u I0 gel. The fractions are 1.7-mm slices of the gel, and the positions of 4-S and 5-S markers and the bromophenol blue dye are indicated. (--) RNA fragment A1 ; (---) RNA fragment A2 ~-~
ence in the latter being the presence of oligonucleotide 4- 14, whose position is indicated on the schematic diagram (Fig. 4). The oligonucleotide spots were cut out from the fingerprint, and their 32P radioactivity measured in order to obtain the molar ratios of the various oligonucleotides, as described [3]. Selected oligonucleotides were then subjected to a secondary digestion with ribonuclease A, using the method of Uchida et al. [13] (and CJ: [3]). Similar analyses were made on all the numbered RNA sub-fragments illustrated in Fig. 2A, and the oligonucleotide sequences were compared with those found in 16-S RNA [15]. The results, for those oligonucleotides which were well-separated and which could be uniquely located in the 16-S RNA by this method of analysis, are summarized in Table 1, expressed as the molar ratio of each oligonucleotide for each of the RNA sub-fragments. The analysis of total RNA from ribonucleoprotein fragment I1 showed that all the characteristic oligonucleotides in the 5’ region of the 16-S KNA (sections L through 0)were present, with the notable exception of the 5’-terminal oligonucleotide PA-A-A-U-U-G, which we were unable to detect. This region is about 900 nucleotides long, and corresponds to the 12-S fragment of Muto et al. [6]. No characteristic oligonucleotides from the 3‘ region of the 16-S RNA (sections 0’ through J) were found, from which it was concluded that the RNA from ribonucleoprotein fragment I1 and its RNA subfragments (Fig.2A) arises exclusively from sections L to 0. This restriction made it possible to designate various small oligonucleotides (e.g. 24d, 05e, 05a, see Table 1 and Fig. 4) as ‘characteristic’, since although these sequences also appear in the 3‘-proximal region [15], they each occur only once in sections L to 0. For this reason these oligonucleotides are included in Table 1, in addi-
tion to the larger characteristic oligonucleotides whose placing in the 16-S molecule is unique. Since the 5 - 15 % gel gradient system used reveals hidden breaks in the RNA, as discussed above, the RNA sub-fragments 1 to 10 (Fig.2A) must represent continuous unbroken RNA sequences. This is supported by the fact that in earlier experiments, using the less highly resolving 7 % gel system with 7 M urea throughout [3], comparable RNA profiles were obtained (data not shown). Thus the 5’ and 3’ termini of each RNA sub-fragment can be approximately located by the presence or absence of a characteristic oligonucleotide (Table 1). In all cases the fingerprint data were checked to ensure that the intervening noncharacteristic or non-separable oligonucleotides were qualitatively and quantitatively consistent with the nucleotide composition of the fragment as predicted from its characteristic oligonucleotide content. That is to say, each spot on the fingerprint showed a molarity consistent with the number of oligonucleotides expected to appear in that particular position, and the analyses with pancreatic ribonuclease were consistent with the products expected, in cases where two or more isomeric oligonucleotides ran together. For reasons of space, these data are not included in Table 1. The chain lengths of the RNA sub-fragments were estimated from their gel mobilities (Fig. 2A), and were consistent with the fingerprint data. For example, fragment 6 (Fig. 2A) has an identical mobility to that of fragment 1 (Fig. 2B) whose chain length is known to be 300 nucleotides [3]; the fingerprint analysis (c$ Table 1) for fragment 6 indicated a chain length of 260 - 300 nucleotides. Table 1 shows that RNA sub-fragments 1-4 are pure, whereas sub-fragments 5 - 7 are clearly contaminated with significant amounts of other RNA species, as indicated by the molar ratio values in brackets. The contaminating sequences are present in lower molar amounts, and extend over a wider area of the RNA than could be accounted for by the estimated chain length of the fragment concerned. Sub-fragment 8 appears to be clean, whereas sub-fragments 9 and 10 are mixtures of RNA species arising from the extreme ends of the KNA region L to 0. It is clear from Table 1 and Fig. 2A that a nuclease-stable region of the RNA exists between sections H’ and I (fragments 1-4), which corresponds to the ‘binding site’ of protein S4 [7,8], whereas the 3’ region of the RNA (sections C - 0) is more nuclease-sensitive and does not yield very well-defined fragments at this level of resolution. In view of the complex pattern of fragments obtained, a more detailed analysis to determine the precise 3’ and 5’ ends of the RNA species listed in Table 1 has not as yet been undertaken. In a further series of experiments, the RNA from ribonucleoprotein fragment I1 was first fractionated on a 5 % dodecyl sulphate gel (minus urea), and the
193
J . Rinke, A. Ross, and R. Brimacombe 49
6-11 5-12
34
12
0
13a 04h
0
O1
Fig. 4. Autoradiograph of a,fingerprint of RNA sub-jiagment I from rihonucleoprotein.fragmentII (Fig. 2 A ) . Direction of the first dimension is from right to left, and that of the second from top to bottom. The oligonucleotide spots are numbered (right side of diagram) according to the system of Uchida et al. [13]. The position of oligonucleotide 4- 14, which was not present in RNA sub-fragment 1 (see text) is indicated by the cross-hatched circle. The small oligonucleotides 03a (C-C-G)and 05x (C-C-m7G-C-G) should each be present, but the spots are too faint to be seen in this fingerprint
R N A peaks isolated in this system were subsequently applied to a second gel (plus urea) as described in Materials and Methods, in order to reveal hidden breaks and tertiary interactions in the R N A (cf: 141). This two-step gel system also serves as a method of purifying the R N A sub-fragments, particularly those of shorter chain length. A typical gel pattern of RNA from ribonucleoprotein fragment I1 fractionated on the 5 % gel system minus urea is shown in Fig. 5 (cf. [4,5]). The principle bands marked S1 and S3 were always present (the latter in rather variable amount), but band S2 was not always observed, and the minor faster-moving components were rather variable. Oligonucleotide analysis of RNA from band S1 showed that all sequences from sections L to 0 were present, and the fingerprint was indistinguishable
from that of the total R N A from ribonucleoprotein I1 (Table 1, Fig. 4). The R N A from band S1 was separated on the urea gel system (see Materials and Methods, and cf: Fig. 2A), and the resulting individual R N A sub-fragments were subjected to oligonucleotide analysis. R N A sub-fragments 1-4 (Table 1 and Fig. 2A) were all observed reproducibly, except that sub-fragments 1 and 2 could not always be separated, and a number of other smaller fragments were obtained, one of which corresponded to sub-fragment 6 (Table 1). Two other small fragments contained respectively sections L and 0,and confirmed the analysis of sub-fragments 9 and 10 (Table l), where these two small fragments had clearly not separated in the one-step purification system (Fig. 2A).Band S2 could not be distinguished from band S1 by oligonucleotide
RNA Fragments Obtained by Mild Digestion of E. coli 30-S Ribosomes
194
Table 1. Oligonucleotide analysis of R N A sub-frugments from ribonuclroprotein fragment I I (Fig.2 A ) The table lists the characteristic oligonucleotides analysed, numbered according to Fig. 4, in the order in which they appear in the 16-S RNA [15]. The sequence of each oligonucleotideis given [15], with the products of ribonuclease A digestion underlined. The molar quantities of each oligonucleotide are indicated for each of the sub-fragments, and also for unfractionated RNA from ribonucleoprotein 11.Figures in brackets denote contaminating sequences and a dash indicates the absence of a particular oligonucleotide.Hyphens representing the phosphate linkages between nucleoside in the oligonucleotides have been omitted to save space. Sub-fragment I1 is total unfractionated RNA from ribonucleoprotein 11. Oligonucleotide 16j is not distinguishable by ribonuclease A digestion from U-A-A-A-C-G (section D') 1 6 3 RNA section
Oligonucleotide Sequence
Molarity of oligonucleotide in RNA sub-fragment I1
L L H' H F F
24d 2-11a
AUUG
1.7 0.8 AAACE -~ 1.1 AUAACACUUG 0.7 CCAUCG 0.9 -~~ . CCCAG 0.8 UAACG 0.8 - __ ACCAG 0.8 CACCACUG 0.7 ACACG 0.7 CCAUG 0.6 UAUG 1.0 U ~ ~ A A U A C U U U G 1.1 0.8 C(CA)CG UACUUUCAG 1.3 1.2 AAAUCCCCC
C(CU)AACACAUG
16j 3-10 16b 05c 15e 05e 18b 0Sf 1 Sd 24c 6-11 05a 49 19c 18a 14h 4- 14 14d
Q M M M B I' I I I C Ci Ci 0 0
I
~
~~~
0
12
2
3
4
1.0 0.7
-
1.4 1.1
-
1.3 1.2 1.3 1.1 1.1 1.2 1.0 1.2 1.4 1.9 1.5 1.1 1.4 1.0
1.2 0.9 1.3 0.9 0.9 1.4 1.0 1.4 1.8 1.6 3.4 0.9 1.6 (0.4)
1.2 1.1
~~~
-
~~
~
~~
~
~~
-
~~
1.1 1.2 1.0 1.6 0.8 1.6 1.5 1.3 1.5 1.2 1.3 1.1
-
1.2 1.0 0.9 1.2 1.0 3.5 0.8 1.5 1.4 1.5 1.3 1.2 1.3 -
5
6
0.9 0.6 0.9 1.0 0.7 0.7 0.6 1.2
0.6
0.5 0.8 0.7 0.5 0.8 0.6
0.5 1.2 0.9
-
(0.3) (0.8) -
0.7 (0.3) (0.4) (0.1) (0.6) -
7
8
9
10
-
(0.4)
(0.1) 0.5 0.5 0.5 0.6 0.4 (0.2) (0.2) (0.2) (0.3) (0.4) (0.2)
-
1.4 0.5
0.3 0.3
-
-
-
-
-~
-
-
-
-
-
~-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
0.6
-
-
0.9 0.6 0.6
-
(0.1)
(0.1)
-
-
-
(0.1)
-
(0.4) (0.1) (0.2) 0.7 (0.3) (0.3) (0.2) (0.3)
-
-
0.3 0.4 0.2 0.3
~-
CCCCCCUG 1.3 1.6 ACUG 1.4 UUAAAACUCAAACG .~ 0.7 UACG ~
~-
0.4 1.3 (0.1)
-
-
(0.2) (0.1) (0.2) (0.3) (0.3) (0.3) (0.4)
-
-
-
-
-
-
-
~-
-
(0.2) 0.5 0.3
-
-
-
-
-
-
-
-
0.4
-
-
-~
I
I
I
I
I
I
Nucleotide Sequences in R N A Bunds A1 and A 2
51
0 -
1
I
24
I
I
I
36 48 60 Fraction number
I
72
I
84
I
Fig. 5. [ 3 2 P ] R N A profile f r o m ribonucleoprotein fragment I I , after electrophoresis on a 5 %, gel, minus urea. The fractions are 1.7-mm gel slices, and the positions of a 5-S marker and the bands of interest (SI, S2 and S3) are indicated
analysis, but band S3, when subjected to the second electrophoresis, reproducibly showed two bands only, which were identical with RNA sub-fragments 3 and 4 (Fig. 2A).
Oligonucleotide analysis of the purified RNA bands A1 and A2 (Fig. 3) showed that both these fragments arise from section A of the 1 6 3 RNA [lS], and a fingerprint of fragment A2 is shown in Fig.6. The fingerprint of fragment A1 was identical, with the exception that two extra spots were observed (14 h and 2 - lo), whose positions are indicated in Fig. 6. Band A1 also showed increased amounts of some smaller oligonucleotides. No nucleotide sequences from section J, comprising the 50 nucleotides at the extreme 3' end of the 16-S RNA [IS], were detected. The analysis of the fingerprints and secondary digestions with ribonuclease A were made as above, and the results are given in Table 2. The observed molar ratios of the oligonucleotides agreed very precisely with those expected from the published sequence [lS], with the notable exception that an extra molecule of A-A-G was reproducibly found in fragment A1 which cannot be accounted for. Fig. 7 shows the sequence of the appropriate part of section A, with the positions of fragments A1 and A2, as well as indicating the 3'
J. Rinke, A. Ross, and R. Brimacombe
195
.A . .. . .. CCUUGIUACACACG rC C i
I ~ ~ C ~ C C C G ~ ~ ~ ~ A C AG CGC GA UG U GG
I "
111" ( 3 -end )
AG G UG ACUG UG G G UUG AUAUUCG UG CAAAAG AAG UAG G UAG
Ekl "A2"
-Ei
(?'-end)
G G CUUACCACUUUd
J
UCG mUAACAAG
.. ..
1-1 Fig.1. Part of section A of the 16-S RNA [ 1 5 ] , encompassing R N A ,fragments A 1 and A2. The 3' and 5' termini of the fragments are indicated, as well as the 3' terminus of RNA from ribonucleoprotein 111 [ 3 ] . Hyphens between nucleosides have been omitted as in Table 1
Fig. 6. Autoradiograph of a fingerprint o j R N A ,fragment A2, afier purification on a 10 "i, gel (Fig.3).Direction of the first and second dimensions are as in Fig. 4, and the spots are again numbered according to Uchida et al. [13]. The positions of spots 14 h and 2-10, which were present in fragment A l , are indicated by the circles
terminus of the RNA from ribonucleoprotein fragment I11 [3]. DISCUSSION As mentioned in the introduction, the primary purpose for undertaking this oligonucleotide analysis was to make it possible to interpret the results of formaldehyde-induced RNA-protein cross-linking experiments [ 5 ] . The sequences described here, taken together with those already described for the RNA from ribonucleoprotein fragment 111 [3], provide a 'catalogue' of reproducibly obtainable RNA sub-fragments covering most of the 16-S RNA molecule, which should be useful in future cross-linking studies on the 30-S subunit. However, the analysis has been of interest to us for other reasons. It has been known for some time that the 3' end of the 16-S RNA molecule is fairly exposed in the 30-S particle, as evidenced by the ease of removal of the 3'terminal nucleotides by colicin E3 [9], and by the acces-
sibility to nuclease attack [lo] or modification by kethoxal [ll]. The finding that a large part of section A is released as free RNA under our hydrolysis conditions, rather than being attached to one or other of the two ribonucleoprotein fragments, suggests that not only is the whole 3' region (ie.the last 150 nucleotides) very accessible, but that this part of the RNA is not involved in strong interactions either with protein or with other regions of the RNA. This is in contrast to other areas of the 16-S RNA (e.g. in sections E'-K or C-D'), which are also accessible to kethoxal [ l l ] and to nuclease (as our own results show, see [4] and Table l), but which are nonetheless found in the ribonucleoprotein fragments I1 or 111. The fate of section J in our experiments is not known, but it seems likely that it is digested to small oligonucleotides. Our analysis was not made in sufficient detail to determine whether the total RNA from ribonucleoprotein fragment I1 represents a continuous sequence, or whether (as was the case in the RNA from ribonucleoprotein fragment I11 [3]) there are small excisions between the RNA sub-fragments, particularly in the rather sensitive region between sections Ci and 0 (Table 1) where the 16-S sequence itself is not yet fully determined [15]. It was clear from the analysis of the RNA complex 'Sl' (Fig.5) that the whole region from L to 0 can exist as a single complex species, but, in contrast to the RNA from ribonucleoprotein fragment 111 [3], it was not possible to demonstrate the existence of any stable tertiary interactions within the RNA from ribonucleoprotein fragment I1 since no oligonucleotides could be shown to be missing from the complex. In particular, the stable RNA species found in studies on complexes formed between isolated 16-S RNA and protein S4, in which sections Q, R, G and M were missing [7,8], could not so far be detected. This results from the fact that sections Q, R, G and M do not appear to be readily accessible to nuclease in the intact 30-S subunit, since the major RNA sub-fragments 1 to 4 (see Table 1) which were isolated from the RNA complexes S1 or S3 (Fig.5)
196
J. Rinke, A. ROSS,and R. Brimacombe: RNA Fragments Obtained by Mild Digestion of E. coli 3 0 3 Ribosomes
Table 2. Oligonucleotide sequencesfound in fragmenty A1 and A2 (Fig.3) The oligonucleotides are numbered as in Fig. 6, and the products of ribonuclease A digestion are underlined. The molar quantities of each oligonucleotide found are compared with those expected from the 16-S sequence [15], the relevant part of which is given in Fig. 7. Hyphens between nucleosides have been omitted as in Table 1. C* is a methylated cytidine Oligonucleotide
01 02a 02b 03d 06b 12 13c 14h 23 2-10" 37d 4-11 5-12 a
Sequence
Fragment A 1
G CG AG AAG CAAAAG UG UAG ACUG -~ UUG UC* ACACCAUG ~~
~~
AUAUUCG -~ ~ __ _
_ CUUAACCUUCG CUUA~~ACUUGG ~
~
Fragment A2
found
expected
found
expected
16.0 1.4 2.8 2.7 0.8 3.8 2.5 1.3 1.3 1.3 1.2 1.3 1.1
12 1
8.0 1.1 1.1 1.1 0.9 2.8 2.3
7 7 1 1
2
1 1 4 2 1 1 1
1 1
1
-
1
2 2 0
1.2
3
-
0 1 1 1
1.2 1.4 1.1
The alternative sequence C-C-A-C-A-C-C-A-U-G [15] does not appear to be present Uchida et a/. [13] report a different sequence viz. A-U-U-C-A-U-G.
encompassed the whole of this region. It is therefore not yet possible to say whether or not this tertiary interaction exists in the 30-S subunit, and experiments are in progress to try to detect this interaction among the minor components of hydrolyses such as that depicted in Fig. 5. It seems probable that the intact ribosome will prove to contain a number of such tertiary interactions. However, some of these are likely to be too unstable to be detectable by the methods described here and elsewhere [4], and in addition some will remain concealed within longer RNA sub-fragments which are relatively inaccessible to nuclease. For the future, it will therefore be necessary to develop techniques for intra-RNA cross-linking, in order to gain more information about the topography of the RNA within the subunits. REFERENCES 1. Morgan, J. & Brimacombe, R. (1973) Eur. J . Biochem. 37, 472 -480.
2. Roth, H. E. & Nierhaus, K. H. (1973) FEBS Lett. 31, 35-38. 3. Yuki, A. &Brimacornbe, R. (1975) Eur. J . Biochem. 56,23-34. 4. Rinke, J., Yuki, A. & Brimacombe, R. (1976) Eur. J . Biochem. 64,77 - 89. 5. Moller, K., Rinke, J., Ross, A., Buddle, G. & Brimacombe, R. (1977) Eur. J . Biochem. 76, 175-187. 6. Muto, A,, Ehresmann, C., Fellner, P. & Zimmermann, R. A. (1974) J . Mol. Biol. 86, 411 -432. 7. Mackie, G. & Zimmermann, R. A. (1975) J . Biol. Chem. 250, 4100-41 12. 8. Ungewickell, E., Ehresmann, C., Stiegler, P. & Garrett, R. A. (1975) Nucleic Acids Res. 2, 1867-1893. 9. Bowman, C. M., Dahlberg, J. E., Ikemura, T., Konisky, J. & Nomura, M. (1971) Proc. Natl Acad. Sci. U.S.A. 68, 964968. 10. Santer, M. & Santer, U. (1973) J . Bacteriol. 116, 1304-1313. 11. Noller, H. F. (1974) Biochemistry, 13,4694-4703. 12. Sanger, F., Brownlee, G. G. & Barrel], B. G. (1965) J . Mol. Biol. 13, 373 - 398. 13. Uchida, T., Bonen, L., Schaup, H. W., Lewis, B. J., Zablen, L. & Woese, C. R. (1974) J . Mol. Evol. 3, 63-77. 14. Dahlberg, A. E., Dingman, C. W. & Peacock, A. C. (1969) J. Mol. Biol. 41, 139-147. 15. Ehresmann, C., Stiegler, P., Mackie, G. A,, Zimmermann, R. A,, Ebel, J. P. & Fellner, P. (1975) Nucleic Acids Res. 2, 265 - 278.
J. Rinke, A. Ross, and R. Brimacombe, Max-Planck-Institut fur Molekulare Genetik, IhnestraDe 63/73, D-1000 Berlin (West) 33-Dahlem
Characterisation of RNA Fragments Obtained by Mild Nuclease Digestion of 30-S Ribosomal Subunits from Escherichia coli Jutta RINKE, Alexander ROSS, and Richard BRIMACOMBE
Max-Planck-Institut fur Molekulare Genetik, Abteilung Wittmann, Berlin-Dahlem (Received October 1, 1976)
When Escherichia coli 3 0 4 ribosomal subunits are hydrolysed under mild conditions, two ribonucleoprotein fragments of unequal size are produced. Knowledge of the RNA sequences contained in these hydrolysis products was required for the experiments described in the preceding paper, and the RNA sub-fragments have therefore been examined by oligonucleotide analysis. Two well-defined small fragments of free RNA, produced concomitantly with the ribonucleoprotein fragments, were also analysed. The larger ribonucleoprotein fragment, containing predominantly proteins S 4, S 5 , S 8, S 15, S 16 (17) and S20, contains a complex mixture of RNA sub-fragments varying from about 100 to 800 nucleotides in length. All these fragments arose from the 5’-terminal 900 nucleotides of 16-S RNA. corresponding to the well-known 12-S fragment. No long-range interactions could be detected within this RNA region in these experiments. The RNA from the smaller ribonucleoprotein fragment (containing proteins S7, S9, S10, S14 and S 19) has been described in detail previously, and consists of about 450 nucleotides near the 3‘ end of the 16-S RNA, but lacking the 3‘4erminal 150 nucleotides. The two small free RNA fragments (above) partly account for these missing 150 nucleotides; both fragments arose from section A of the 16-S RNA, but section J (the 3’-terminal 50 nucleotides) was not found. This result suggests that the 3‘ region of 16-S RNA is not involved in stable interactions with protein. It has been known for some time that Escherichiu coli 30-S ribosomal subunits can be split by mild nuclease digestion into two ribonucleoprotein fragments of unequal size [l -31. The smaller of these two fragments, which contains proteins S7, S9, SlO, S14 and S19 [3], has been characterized in detail, and we have recently published an analysis of the sequences of the RNA sub-fragments which are contained in this ribonucleoprotein particle [3]. These sequences cover a segment of about 450 nucleotides in the 3’proximal region of the 16-S RNA, but the last 150 nucleotides at the 3’ terminus of the 16-S molecule are absent. Further it was shown that a ‘tertiary’ (or ‘long-range’)interaction exists between distant regions of the RNA of the ribonucleoprotein fragment, since a stable RNA complex could be isolated containing two discontinuous regions of the RNA, which are separated by about 100 nucleotides in the intact 16-S molecule [4]. In this paper we describe a similar analysis of the RNA component of the larger of the two ribonucleoenzyme^. Ribonuclease TI (EC 3.1.4.8); ribonuclease A (EC 3.1.4.22).
protein fragments, which contains proteins S4, S5, S8, S15, S16(17) and S20, together with variable amounts of two or three other proteins [l]. The analysis was undertaken for a number of reasons. Firstly, it was necessary to know which RNA sequences are contained in this ribonucleoprotein fragment in order to be able to interpret our data on the distribution of proteins which are cross-linked by formaldehyde to this region of the RNA, described in the preceding paper [5]. Secondly, it is of interest to find out whether any tertiary interactions can be detected in this region of the RNA, as was the case in the RNA from the smaller ribonucleoprotein species [4]. Thirdly, the absence of the 3’-terminal 150 nucleotides of the 16-S RNA from the latter ribonucleoprotein fragment raises the question as to where these missing nucleotides are, in particular whether they can be found associated with the larger ribonucleoprotein fragment. The results showed that the RNA from the large ribonucleoprotein fragment consists of a complex series of sub-fragments covering a region of about 900 nucleotides at the 5’ end of the 16-S molecule, corresponding to the ‘12-S fragment’ [6]. We were unable to detect any stable tertiary interactions
190
RNA Fragments Obtained by Mild Digestion of E. coli 3 0 3 Ribosomes
within the RNA; as a result it is still not possible to say whether or not the tertiary interaction within this region (which has been clearly demonstrated in studies on complexes between isolated RNA and protein S4 [7,8]) exists in intact 30-S subunits. The missing 150 nucleotides from the 3‘ end of the 16-S RNA were not found associated with the large ribonucleoprotein fragment, but the greater part of this missing region was liberated from the 30-S subunits as discrete RNA fragments not associated with protein, during the initial hydrolysis. This provides further evidence that a large portion of the 3’ region of 16-S RNA is not tightly bound, and is rather easily accessible in the 30-S subunit [9 - 111.
Oligonucleotide Analysis Appropriate gel slices containing RNA fragments were homogenized, and the RNA extracted with phenol and dodecyl sulphate exactly as described [3]. After separation from phenol and isolation of the RNA by ethanol precipitation, the RNA was hydrolysed with ribonuclease TI and subjected to oligonucleotide analysis by the method of Sanger et LEI. [12], again exactly as described [3]. Analysis of the fingerprint data, and secondary digestions with ribonuclease A according to the method of Uchida et al. [13], were also made as before [3]. RESULTS
MATERIALS AND METHODS
Purification oj R N A Fragments
Isolation of R N A Fragments
30-S subunits were treated with ribonuclease TI under ‘standard’ conditions [3], and applied to a polyacrylamide gel, using the ‘pH-6’ system (see Materials and Methods and [4]). A typical gel pattern, in this instance from unlabelled subunits, is shown in Fig. 1. The two ribonucleoprotein fragments I1 and 111 can be seen, and in addition two sharp bands (marked A1 and A2) running close to the bromophenol blue marker. These two bands are particularly well defined in this gel system, and were observed when the hydrolysis was made either under ‘standard’ conditions (5 mM magnesium, slot 2, Fig. 1) or when the magnesium concentration was reduced to 1 mM (slot 3). In other experiments, using subunits labelled with 3H in the RNA and I4C in the protein moiety, it could be seen that bands A1 and A2, in contrast to ribonucleoprotein fragments I1 and 111, contained negligible amounts of protein (data not shown), and hence these bands are of free RNA as opposed to ribonucleoprotein. In experiments with 32P-labelledsubunits, identical gel patterns to that of Fig.1 were obtained, and the various components of this pattern, viz. ribonucleoprotein fragments I1 and 111, and RNA fragments A1 and A2, were subjected to further analysis. The ribonucleoprotein peak I, which has the same mobility as the unhydrolysed 30-S subunit (Fig. 1, slot I), and the heterogeneous RNA peak running ahead of bands A1 and A2 were not examined. Gel slices containing 32P-labelled ribonucleoprotein fragments I1 and I11 were applied to the 5 - 15 % gel gradient system (see Materials and Methods). This is a dodecyl sulphate gel system, with a thin layer of 3 % polyacrylamide containing 8 M urea on top of the gel, and typical RNA separations from both ribonucleoprotein fragments are shown in Fig.2. The RNA from ribonucleoprotein fragment 111 (Fig. 2 B) showed the same five sub-fragments as were previously observed on a 7 % gel with 7 M urea throughout [3], and oligonucleotide analysis of all five confirmed
32P-labelled 30-S ribosomal subunits were prepared from E. coli strain MRE 600, and were hydrolysed with ribonuclease TI in the presence of 5 mM magnesium and 2 M urea exactly as described previously [3]. The hydrolysis products were separated on a 5 % polyacrylamide gel, using the pH-6 buffer system described earlier [4], following which the gel was sliced and analysed for radioactivity. Gel slices containing small fragments of free RNA (see text) were submitted to a further electrophoresis on a 10 polyacrylamide gel containing 0.1 % dodecyl sulphate and 7 M urea, using the buffer system described earlier [4]. Gel slices containing the ribonucleoprotein fragments (see text) were treated in one of two ways. In the first method, the slices were loaded directly onto a 5 - 15 % polyacrylamide gradient gel containing dodecyl sulphate, using the buffer system described in the preceding paper [5], with the exceptions that Tris was used in place of triethanolamine, and formaldehyde was omitted. After polymerisation each gel slot was filled to a depth of 1 cm with a 3 % polyacrylamide gel containing 8 M urea in addition to dodecyl sulphate and buffer, and the sample slices were loaded on top of this layer. This gel system both removes protein and reveals ‘hidden breaks’ in the RNA. In the second method, the slices were loaded directly onto a 5 % polyacrylamide gel containing dodecyl sulphate as described [4];this system contains no urea and therefore does not reveal hidden breaks. After slicing and analysing for radioactivity, suitable gel fractions were either loaded directly onto a second gel or extracted with phenol as described [ 3 ] and then loaded onto a second gel, to reveal the hidden breaks in the RNA. This second gel was either the 5 - 15 ”/, gradient gel above, or alternatively a simple 7 % gel containing dodecyl sulphate with 7 M urea throughout, as described [4]. In all cases the final gel was sliced and analysed for radioactivity.
J. Rinke, A. Ross, and R. Brimacornbe
191 ~
A 1
1
2
3
-I -II
-m I
12
24
I 36
B
-
I
48
I 60
72
I 84
96
72
84
96
I
I
2
A1
- A2
12
24
36
48 60 Fraction number
108
Fig. 1. Hydrolysis patterns of 30-S subunits in the pH-6 gel system. The samples was of unlabelled 30-S subunits and the gel was stained with methylene blue [14]. Slot 1: 3 0 4 subunits, minus enzyme. Slot 2: 30-S subunits hydrolysed under ‘standard’ conditions [3]. Slot 3: as for slot 2, but with the magnesium concentration reduced from 5 mM to 1 mM. The positions of ribonucleoprotein fragments I, I1 and 111, the free RNA peaks A1 and A2, and the bromophenol blue dye marker are indicated
Fig. 2. [3 2 P ] R N Aprofiles from ribonucleoprotein ,fragments II and III, after electrophoresis on a 5- 15 %gradient gel containing dodecyl sulphate, with a 7 M urea layer. The fractions are 1.7-mm slices of the gel, and the position of a 5-S marker is shown. (A) RNA from ribonucleoprotein fragment 11. (B) RNA from ribonucleoprotein fragment 111, with sub-fragments numbered as in [3]
that the RNA sequences were the same as those reported [3]. This gel system gives a substantially better separation of the sub-fragments than the simple 7 % gel system [3]. Further, the fact that RNA from ribonucleoprotein IT1 showed the same five sub-fragments as before indicates that the thin urea layer on top of the gel is sufficient to reveal the hidden breaks in the RNA irreversibly. The RNA profile obtained from ribonucleoprotein fragment I1 was rather more complex than that of ribonucleoprotein fragment 111. RNA sub-fragments 1 and 2 (Fig.2A) could not always be separated, and there was usually a rather heterogeneous region in the neighbourhood of the 5-S RNA marker, between subfragments 8 and 10, which can be seen from Fig.2A. Gel fractions corresponding to the individual numbered peaks in Fig.2A were extracted with phenol, and the RNA was subjected to oligonucleotide analysis as described in Materials and Methods. Total RNA from ribonucleoprotein fragment I1 (not separated by electrophoresis on dodecyl sulphate gels) was also
analysed after phenol extraction from the pH-6 gel system (cf Fig. 1). The RNA bands A1 and A2 were purified by an electrophoresis step, using a 10% gel as described in Materials and Methods, and the result of such a purification is illustrated in Fig. 3. Both RNA species gave a single sharp peak, the one (A 1) slightly smaller than 5-S RNA, and the other (A2) slightly smaller than 4-S RNA. Both of these peaks were extracted with phenol and subjected to oligonucleotide analysis. Nucleotide Sequences in Ribonucleopvotein Fragment II
An example of a fingerprint of the largest RNA subfragment obtained from ribonucleoprotein fragment TI (fragment 1, Fig.2A) is shown in Fig.4, together with a schematic representation of the fingerprint which indicates the nomenclature used for describing the oligonucleotide spots (cf [13]). The fingerprint of total unfractionated RNA from ribonucleoprotein fragment I1 was very similar, the only obvious differ-
RNA Fragments Obtained by Mild Digestion of E. roll 30-S Ribosomes
192 I
I
I
I
I
I
I
48
60
72
84
5-5 4-5
1
36
96
Fraction number
Fig. 3. 3 2 P profiles of R N A frugmmts A1 and A2 on u I0 gel. The fractions are 1.7-mm slices of the gel, and the positions of 4-S and 5-S markers and the bromophenol blue dye are indicated. (--) RNA fragment A1 ; (---) RNA fragment A2 ~-~
ence in the latter being the presence of oligonucleotide 4- 14, whose position is indicated on the schematic diagram (Fig. 4). The oligonucleotide spots were cut out from the fingerprint, and their 32P radioactivity measured in order to obtain the molar ratios of the various oligonucleotides, as described [3]. Selected oligonucleotides were then subjected to a secondary digestion with ribonuclease A, using the method of Uchida et al. [13] (and CJ: [3]). Similar analyses were made on all the numbered RNA sub-fragments illustrated in Fig. 2A, and the oligonucleotide sequences were compared with those found in 16-S RNA [15]. The results, for those oligonucleotides which were well-separated and which could be uniquely located in the 16-S RNA by this method of analysis, are summarized in Table 1, expressed as the molar ratio of each oligonucleotide for each of the RNA sub-fragments. The analysis of total RNA from ribonucleoprotein fragment I1 showed that all the characteristic oligonucleotides in the 5’ region of the 16-S KNA (sections L through 0)were present, with the notable exception of the 5’-terminal oligonucleotide PA-A-A-U-U-G, which we were unable to detect. This region is about 900 nucleotides long, and corresponds to the 12-S fragment of Muto et al. [6]. No characteristic oligonucleotides from the 3‘ region of the 16-S RNA (sections 0’ through J) were found, from which it was concluded that the RNA from ribonucleoprotein fragment I1 and its RNA subfragments (Fig.2A) arises exclusively from sections L to 0. This restriction made it possible to designate various small oligonucleotides (e.g. 24d, 05e, 05a, see Table 1 and Fig. 4) as ‘characteristic’, since although these sequences also appear in the 3‘-proximal region [15], they each occur only once in sections L to 0. For this reason these oligonucleotides are included in Table 1, in addi-
tion to the larger characteristic oligonucleotides whose placing in the 16-S molecule is unique. Since the 5 - 15 % gel gradient system used reveals hidden breaks in the RNA, as discussed above, the RNA sub-fragments 1 to 10 (Fig.2A) must represent continuous unbroken RNA sequences. This is supported by the fact that in earlier experiments, using the less highly resolving 7 % gel system with 7 M urea throughout [3], comparable RNA profiles were obtained (data not shown). Thus the 5’ and 3’ termini of each RNA sub-fragment can be approximately located by the presence or absence of a characteristic oligonucleotide (Table 1). In all cases the fingerprint data were checked to ensure that the intervening noncharacteristic or non-separable oligonucleotides were qualitatively and quantitatively consistent with the nucleotide composition of the fragment as predicted from its characteristic oligonucleotide content. That is to say, each spot on the fingerprint showed a molarity consistent with the number of oligonucleotides expected to appear in that particular position, and the analyses with pancreatic ribonuclease were consistent with the products expected, in cases where two or more isomeric oligonucleotides ran together. For reasons of space, these data are not included in Table 1. The chain lengths of the RNA sub-fragments were estimated from their gel mobilities (Fig. 2A), and were consistent with the fingerprint data. For example, fragment 6 (Fig. 2A) has an identical mobility to that of fragment 1 (Fig. 2B) whose chain length is known to be 300 nucleotides [3]; the fingerprint analysis (c$ Table 1) for fragment 6 indicated a chain length of 260 - 300 nucleotides. Table 1 shows that RNA sub-fragments 1-4 are pure, whereas sub-fragments 5 - 7 are clearly contaminated with significant amounts of other RNA species, as indicated by the molar ratio values in brackets. The contaminating sequences are present in lower molar amounts, and extend over a wider area of the RNA than could be accounted for by the estimated chain length of the fragment concerned. Sub-fragment 8 appears to be clean, whereas sub-fragments 9 and 10 are mixtures of RNA species arising from the extreme ends of the KNA region L to 0. It is clear from Table 1 and Fig. 2A that a nuclease-stable region of the RNA exists between sections H’ and I (fragments 1-4), which corresponds to the ‘binding site’ of protein S4 [7,8], whereas the 3’ region of the RNA (sections C - 0) is more nuclease-sensitive and does not yield very well-defined fragments at this level of resolution. In view of the complex pattern of fragments obtained, a more detailed analysis to determine the precise 3’ and 5’ ends of the RNA species listed in Table 1 has not as yet been undertaken. In a further series of experiments, the RNA from ribonucleoprotein fragment I1 was first fractionated on a 5 % dodecyl sulphate gel (minus urea), and the
193
J . Rinke, A. Ross, and R. Brimacombe 49
6-11 5-12
34
12
0
13a 04h
0
O1
Fig. 4. Autoradiograph of a,fingerprint of RNA sub-jiagment I from rihonucleoprotein.fragmentII (Fig. 2 A ) . Direction of the first dimension is from right to left, and that of the second from top to bottom. The oligonucleotide spots are numbered (right side of diagram) according to the system of Uchida et al. [13]. The position of oligonucleotide 4- 14, which was not present in RNA sub-fragment 1 (see text) is indicated by the cross-hatched circle. The small oligonucleotides 03a (C-C-G)and 05x (C-C-m7G-C-G) should each be present, but the spots are too faint to be seen in this fingerprint
R N A peaks isolated in this system were subsequently applied to a second gel (plus urea) as described in Materials and Methods, in order to reveal hidden breaks and tertiary interactions in the R N A (cf: 141). This two-step gel system also serves as a method of purifying the R N A sub-fragments, particularly those of shorter chain length. A typical gel pattern of RNA from ribonucleoprotein fragment I1 fractionated on the 5 % gel system minus urea is shown in Fig. 5 (cf. [4,5]). The principle bands marked S1 and S3 were always present (the latter in rather variable amount), but band S2 was not always observed, and the minor faster-moving components were rather variable. Oligonucleotide analysis of RNA from band S1 showed that all sequences from sections L to 0 were present, and the fingerprint was indistinguishable
from that of the total R N A from ribonucleoprotein I1 (Table 1, Fig. 4). The R N A from band S1 was separated on the urea gel system (see Materials and Methods, and cf: Fig. 2A), and the resulting individual R N A sub-fragments were subjected to oligonucleotide analysis. R N A sub-fragments 1-4 (Table 1 and Fig. 2A) were all observed reproducibly, except that sub-fragments 1 and 2 could not always be separated, and a number of other smaller fragments were obtained, one of which corresponded to sub-fragment 6 (Table 1). Two other small fragments contained respectively sections L and 0,and confirmed the analysis of sub-fragments 9 and 10 (Table l), where these two small fragments had clearly not separated in the one-step purification system (Fig. 2A).Band S2 could not be distinguished from band S1 by oligonucleotide
RNA Fragments Obtained by Mild Digestion of E. coli 30-S Ribosomes
194
Table 1. Oligonucleotide analysis of R N A sub-frugments from ribonuclroprotein fragment I I (Fig.2 A ) The table lists the characteristic oligonucleotides analysed, numbered according to Fig. 4, in the order in which they appear in the 16-S RNA [15]. The sequence of each oligonucleotideis given [15], with the products of ribonuclease A digestion underlined. The molar quantities of each oligonucleotide are indicated for each of the sub-fragments, and also for unfractionated RNA from ribonucleoprotein 11.Figures in brackets denote contaminating sequences and a dash indicates the absence of a particular oligonucleotide.Hyphens representing the phosphate linkages between nucleoside in the oligonucleotides have been omitted to save space. Sub-fragment I1 is total unfractionated RNA from ribonucleoprotein 11. Oligonucleotide 16j is not distinguishable by ribonuclease A digestion from U-A-A-A-C-G (section D') 1 6 3 RNA section
Oligonucleotide Sequence
Molarity of oligonucleotide in RNA sub-fragment I1
L L H' H F F
24d 2-11a
AUUG
1.7 0.8 AAACE -~ 1.1 AUAACACUUG 0.7 CCAUCG 0.9 -~~ . CCCAG 0.8 UAACG 0.8 - __ ACCAG 0.8 CACCACUG 0.7 ACACG 0.7 CCAUG 0.6 UAUG 1.0 U ~ ~ A A U A C U U U G 1.1 0.8 C(CA)CG UACUUUCAG 1.3 1.2 AAAUCCCCC
C(CU)AACACAUG
16j 3-10 16b 05c 15e 05e 18b 0Sf 1 Sd 24c 6-11 05a 49 19c 18a 14h 4- 14 14d
Q M M M B I' I I I C Ci Ci 0 0
I
~
~~~
0
12
2
3
4
1.0 0.7
-
1.4 1.1
-
1.3 1.2 1.3 1.1 1.1 1.2 1.0 1.2 1.4 1.9 1.5 1.1 1.4 1.0
1.2 0.9 1.3 0.9 0.9 1.4 1.0 1.4 1.8 1.6 3.4 0.9 1.6 (0.4)
1.2 1.1
~~~
-
~~
~
~~
~
~~
-
~~
1.1 1.2 1.0 1.6 0.8 1.6 1.5 1.3 1.5 1.2 1.3 1.1
-
1.2 1.0 0.9 1.2 1.0 3.5 0.8 1.5 1.4 1.5 1.3 1.2 1.3 -
5
6
0.9 0.6 0.9 1.0 0.7 0.7 0.6 1.2
0.6
0.5 0.8 0.7 0.5 0.8 0.6
0.5 1.2 0.9
-
(0.3) (0.8) -
0.7 (0.3) (0.4) (0.1) (0.6) -
7
8
9
10
-
(0.4)
(0.1) 0.5 0.5 0.5 0.6 0.4 (0.2) (0.2) (0.2) (0.3) (0.4) (0.2)
-
1.4 0.5
0.3 0.3
-
-
-
-
-~
-
-
-
-
-
~-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
0.6
-
-
0.9 0.6 0.6
-
(0.1)
(0.1)
-
-
-
(0.1)
-
(0.4) (0.1) (0.2) 0.7 (0.3) (0.3) (0.2) (0.3)
-
-
0.3 0.4 0.2 0.3
~-
CCCCCCUG 1.3 1.6 ACUG 1.4 UUAAAACUCAAACG .~ 0.7 UACG ~
~-
0.4 1.3 (0.1)
-
-
(0.2) (0.1) (0.2) (0.3) (0.3) (0.3) (0.4)
-
-
-
-
-
-
-
~-
-
(0.2) 0.5 0.3
-
-
-
-
-
-
-
-
0.4
-
-
-~
I
I
I
I
I
I
Nucleotide Sequences in R N A Bunds A1 and A 2
51
0 -
1
I
24
I
I
I
36 48 60 Fraction number
I
72
I
84
I
Fig. 5. [ 3 2 P ] R N A profile f r o m ribonucleoprotein fragment I I , after electrophoresis on a 5 %, gel, minus urea. The fractions are 1.7-mm gel slices, and the positions of a 5-S marker and the bands of interest (SI, S2 and S3) are indicated
analysis, but band S3, when subjected to the second electrophoresis, reproducibly showed two bands only, which were identical with RNA sub-fragments 3 and 4 (Fig. 2A).
Oligonucleotide analysis of the purified RNA bands A1 and A2 (Fig. 3) showed that both these fragments arise from section A of the 1 6 3 RNA [lS], and a fingerprint of fragment A2 is shown in Fig.6. The fingerprint of fragment A1 was identical, with the exception that two extra spots were observed (14 h and 2 - lo), whose positions are indicated in Fig. 6. Band A1 also showed increased amounts of some smaller oligonucleotides. No nucleotide sequences from section J, comprising the 50 nucleotides at the extreme 3' end of the 16-S RNA [IS], were detected. The analysis of the fingerprints and secondary digestions with ribonuclease A were made as above, and the results are given in Table 2. The observed molar ratios of the oligonucleotides agreed very precisely with those expected from the published sequence [lS], with the notable exception that an extra molecule of A-A-G was reproducibly found in fragment A1 which cannot be accounted for. Fig. 7 shows the sequence of the appropriate part of section A, with the positions of fragments A1 and A2, as well as indicating the 3'
J. Rinke, A. Ross, and R. Brimacombe
195
.A . .. . .. CCUUGIUACACACG rC C i
I ~ ~ C ~ C C C G ~ ~ ~ ~ A C AG CGC GA UG U GG
I "
111" ( 3 -end )
AG G UG ACUG UG G G UUG AUAUUCG UG CAAAAG AAG UAG G UAG
Ekl "A2"
-Ei
(?'-end)
G G CUUACCACUUUd
J
UCG mUAACAAG
.. ..
1-1 Fig.1. Part of section A of the 16-S RNA [ 1 5 ] , encompassing R N A ,fragments A 1 and A2. The 3' and 5' termini of the fragments are indicated, as well as the 3' terminus of RNA from ribonucleoprotein 111 [ 3 ] . Hyphens between nucleosides have been omitted as in Table 1
Fig. 6. Autoradiograph of a fingerprint o j R N A ,fragment A2, afier purification on a 10 "i, gel (Fig.3).Direction of the first and second dimensions are as in Fig. 4, and the spots are again numbered according to Uchida et al. [13]. The positions of spots 14 h and 2-10, which were present in fragment A l , are indicated by the circles
terminus of the RNA from ribonucleoprotein fragment I11 [3]. DISCUSSION As mentioned in the introduction, the primary purpose for undertaking this oligonucleotide analysis was to make it possible to interpret the results of formaldehyde-induced RNA-protein cross-linking experiments [ 5 ] . The sequences described here, taken together with those already described for the RNA from ribonucleoprotein fragment 111 [3], provide a 'catalogue' of reproducibly obtainable RNA sub-fragments covering most of the 16-S RNA molecule, which should be useful in future cross-linking studies on the 30-S subunit. However, the analysis has been of interest to us for other reasons. It has been known for some time that the 3' end of the 16-S RNA molecule is fairly exposed in the 30-S particle, as evidenced by the ease of removal of the 3'terminal nucleotides by colicin E3 [9], and by the acces-
sibility to nuclease attack [lo] or modification by kethoxal [ll]. The finding that a large part of section A is released as free RNA under our hydrolysis conditions, rather than being attached to one or other of the two ribonucleoprotein fragments, suggests that not only is the whole 3' region (ie.the last 150 nucleotides) very accessible, but that this part of the RNA is not involved in strong interactions either with protein or with other regions of the RNA. This is in contrast to other areas of the 16-S RNA (e.g. in sections E'-K or C-D'), which are also accessible to kethoxal [ l l ] and to nuclease (as our own results show, see [4] and Table l), but which are nonetheless found in the ribonucleoprotein fragments I1 or 111. The fate of section J in our experiments is not known, but it seems likely that it is digested to small oligonucleotides. Our analysis was not made in sufficient detail to determine whether the total RNA from ribonucleoprotein fragment I1 represents a continuous sequence, or whether (as was the case in the RNA from ribonucleoprotein fragment I11 [3]) there are small excisions between the RNA sub-fragments, particularly in the rather sensitive region between sections Ci and 0 (Table 1) where the 16-S sequence itself is not yet fully determined [15]. It was clear from the analysis of the RNA complex 'Sl' (Fig.5) that the whole region from L to 0 can exist as a single complex species, but, in contrast to the RNA from ribonucleoprotein fragment 111 [3], it was not possible to demonstrate the existence of any stable tertiary interactions within the RNA from ribonucleoprotein fragment I1 since no oligonucleotides could be shown to be missing from the complex. In particular, the stable RNA species found in studies on complexes formed between isolated 16-S RNA and protein S4, in which sections Q, R, G and M were missing [7,8], could not so far be detected. This results from the fact that sections Q, R, G and M do not appear to be readily accessible to nuclease in the intact 30-S subunit, since the major RNA sub-fragments 1 to 4 (see Table 1) which were isolated from the RNA complexes S1 or S3 (Fig.5)
196
J. Rinke, A. ROSS,and R. Brimacombe: RNA Fragments Obtained by Mild Digestion of E. coli 3 0 3 Ribosomes
Table 2. Oligonucleotide sequencesfound in fragmenty A1 and A2 (Fig.3) The oligonucleotides are numbered as in Fig. 6, and the products of ribonuclease A digestion are underlined. The molar quantities of each oligonucleotide found are compared with those expected from the 16-S sequence [15], the relevant part of which is given in Fig. 7. Hyphens between nucleosides have been omitted as in Table 1. C* is a methylated cytidine Oligonucleotide
01 02a 02b 03d 06b 12 13c 14h 23 2-10" 37d 4-11 5-12 a
Sequence
Fragment A 1
G CG AG AAG CAAAAG UG UAG ACUG -~ UUG UC* ACACCAUG ~~
~~
AUAUUCG -~ ~ __ _
_ CUUAACCUUCG CUUA~~ACUUGG ~
~
Fragment A2
found
expected
found
expected
16.0 1.4 2.8 2.7 0.8 3.8 2.5 1.3 1.3 1.3 1.2 1.3 1.1
12 1
8.0 1.1 1.1 1.1 0.9 2.8 2.3
7 7 1 1
2
1 1 4 2 1 1 1
1 1
1
-
1
2 2 0
1.2
3
-
0 1 1 1
1.2 1.4 1.1
The alternative sequence C-C-A-C-A-C-C-A-U-G [15] does not appear to be present Uchida et a/. [13] report a different sequence viz. A-U-U-C-A-U-G.
encompassed the whole of this region. It is therefore not yet possible to say whether or not this tertiary interaction exists in the 30-S subunit, and experiments are in progress to try to detect this interaction among the minor components of hydrolyses such as that depicted in Fig. 5. It seems probable that the intact ribosome will prove to contain a number of such tertiary interactions. However, some of these are likely to be too unstable to be detectable by the methods described here and elsewhere [4], and in addition some will remain concealed within longer RNA sub-fragments which are relatively inaccessible to nuclease. For the future, it will therefore be necessary to develop techniques for intra-RNA cross-linking, in order to gain more information about the topography of the RNA within the subunits. REFERENCES 1. Morgan, J. & Brimacombe, R. (1973) Eur. J . Biochem. 37, 472 -480.
2. Roth, H. E. & Nierhaus, K. H. (1973) FEBS Lett. 31, 35-38. 3. Yuki, A. &Brimacornbe, R. (1975) Eur. J . Biochem. 56,23-34. 4. Rinke, J., Yuki, A. & Brimacombe, R. (1976) Eur. J . Biochem. 64,77 - 89. 5. Moller, K., Rinke, J., Ross, A., Buddle, G. & Brimacombe, R. (1977) Eur. J . Biochem. 76, 175-187. 6. Muto, A,, Ehresmann, C., Fellner, P. & Zimmermann, R. A. (1974) J . Mol. Biol. 86, 411 -432. 7. Mackie, G. & Zimmermann, R. A. (1975) J . Biol. Chem. 250, 4100-41 12. 8. Ungewickell, E., Ehresmann, C., Stiegler, P. & Garrett, R. A. (1975) Nucleic Acids Res. 2, 1867-1893. 9. Bowman, C. M., Dahlberg, J. E., Ikemura, T., Konisky, J. & Nomura, M. (1971) Proc. Natl Acad. Sci. U.S.A. 68, 964968. 10. Santer, M. & Santer, U. (1973) J . Bacteriol. 116, 1304-1313. 11. Noller, H. F. (1974) Biochemistry, 13,4694-4703. 12. Sanger, F., Brownlee, G. G. & Barrel], B. G. (1965) J . Mol. Biol. 13, 373 - 398. 13. Uchida, T., Bonen, L., Schaup, H. W., Lewis, B. J., Zablen, L. & Woese, C. R. (1974) J . Mol. Evol. 3, 63-77. 14. Dahlberg, A. E., Dingman, C. W. & Peacock, A. C. (1969) J. Mol. Biol. 41, 139-147. 15. Ehresmann, C., Stiegler, P., Mackie, G. A,, Zimmermann, R. A,, Ebel, J. P. & Fellner, P. (1975) Nucleic Acids Res. 2, 265 - 278.
J. Rinke, A. Ross, and R. Brimacombe, Max-Planck-Institut fur Molekulare Genetik, IhnestraDe 63/73, D-1000 Berlin (West) 33-Dahlem
