Organization of ribosomal protein genes of Escherichia coli as analyzed by polar insertion mutations.

Several mutants of lambdaspc1 and lambdafus3 have been isolated carrying DNA insertion elements that were selected for their ability to reduce the expression of the spc gene. The sizes and locations of the insertions on the phage genomes were determined by heteroduplex analysis. They were found to be located at different positions in the Spc transcription unit. The effect of these insertions on the expression of the ribosomal protein genes carried by these phages in ultraviolet light-irradiated bacteria was investigated. The insertions at intermediate positions in the transcription unit reduced the expression of some of the genes in the unit but not others. Assuming that the genes whose expressions were reduced are distal to the insertion, it was possible to determine the relative position of most of the genes in the unit. The results indicate the order of genes in the Spc transcription unit is: promoter, L14, L24, L5, S14, S8, L6, L18, (S5, L15, L30).

Several mutants of Xspcl and Xfw. 3 have been isolated carrying DNA insertion elements that were selected for their ability to reduce the expression of the spc gene. The sizes and locations of the insertions on the phage genomes were determined by heteroduplex analysis. They were found to be located at different positions in the Spc transcription unit. The effect of these insertions on the expression of the ribosomal protein genes carried by these phages in ultraviolet light-irradiated bacteria was investigated. The insertions at intermediate positions in the transcription unit reduced the expression of some of the genes in the unit but not others. Assuming that the genes whose expressions were reduced are distal to the insertion, it was possible to determine the relative position of most of the genes in the unit. The results indicate the order of genes in the Spc transcription unit is: promoter, L14, L24, Lfi, S14, S8, LG. L18. (S5, Ll5, L30).
We have isolated several transducing phages carrying bacterial DNA from the str-spc region of the Escherichia coli chromosome (1,2). The phage with the largest substitution of bacterial DNA, Afus2 (and hfus3), carries genes for 27 ribosomal proteins (r-proteins) (2,3), EF-G (4), EF-TU (4), and the cy subunit of RNA polymerasc (5). Another phage is ~spcl, which carries part of the bacterial DNA af hfue2 (2,6), and consequently it carries a subset of the genes on hfus2 (2,3). One of our goals is to determine the order of the r-protein' genes on the Afus2 genome and therefore on the E. coli chromosome.
The r-protein genes on hfus2 appear to be organized into at least four transcription units (2,4,5,7). One of these units is the Spc transcription unit, which appears to contain genes for 10 r-proteins (cf. Fig.  1 Other Methods -They are described in the preceding paper (2).

RESULTS
The physical structures of the genomes of Aspcl and Xfus3 are given in the upper part of Fig. 1. Afus2 and Xfus3 are the same except Xfus2 carries a sW allele and AfusS carries a str" allele (2). All the r-protein genes on Afus2 appear to be transcribed leftward on the phage genome (7). We have previously suggested that the r-protein genes on Afus2 are organized into at least four transcription units (2,4,5,7; and an accompanying paper, 8). The approximate positions of these units are given in Fig. 1. Aspcl carries only the cy and Spc transcription units (see introduction). The experiments reported here are primarily directed at determining the relative order of the 10 genes in the Spc transcription unit with the use of insertion mutants.
Heteroduplex Analysis of Spr Insertion Mutants -Several mutants of Asprl and Afus3 have been isolated as outlined under "Materials and Methods" on which the spc gene has been at least partially inactivated by the insertion of an IS element. The sizes of the insertions and their locations on the phage genomes have been determined by heteroduplex analysis. Aspcl and Afus2 (or AfusS) are homologous from their left end to the right bacterial-A DNA junction (called the "junction" in this paper) in Asprl (see Fig. 1) (6). They also have a A-DNA homology with the length of 5%-h units at their right ends. Thus, heteroduplcx molecules of Aspcl and hfus2 have a long double strand tail at one end and a short double strand tail at the other end separated by a large bubble (cf. Fig. 2).
Spc insertion mutants of hspcl were heteroduplexed with AfusS and Spc insertion mutants of Afus3 were heteroduplexed with Aspcl. In all cases the insertion loop was found in the long double strand region (see Fig. 2, A and B). The sizes of the insertion loops and the distances between the junction and the insertion loop (D in Fig. 2B) are given in Table I. The locations of many of the insertions are also given schematically in Fig. 1 along with the gene order that was deduced from the analysis of these mutants (see below).
Most of the insertions were the size of ISI, approximately 800 base pairs, or the size of IS2, approximately 1300 basepairs. None of the insertions have been identified except for the IS1 in  and the IS2 in hspcl-116 (14).
We found the insertions were scattered throughout the region between the junction and 4.3 kilobases to the left of the junction.
This indicates that each of the insertions is located within the bacterial DNA that is common to both Aspcl and Afus3. There seems to be a larger number of insertions near the junction in hspcl, which is the approximate location of the presumed promoter for the Spc transcription unit (2,7).
The effect of some of the insertions on the expression of the r-protein genes in UV-irradiated bacteria has been examined. Table II presents the results of experiments in which the host bacteria were prelabeled with 114C]leucine, irradiated, infected with a purified phage, and the proteins synthesized after phage infection were labeled with I:'H]leucine. The ratio of '1H/'4C for most of the r-proteins was determined.
The data was normalized to the expression of one of the genes on the phage that does not seem to be affected by the insertions (see legend to Table II). The normalized expression of each gene of the mutant phage is then expressed in Table II relative to the expression of that gene on the parent phage.
The purpose of the internal normalization is to correct for differences in the multiplicity of infection from one phage preparation to another. For derivatives of Afus3, the gene for 57 was used for this normalization.
It is located on the 11% EcoRI fragment at the right end of Afuss (see Fig. 1) (3,8). It appears to be in a transcription unit with the str gene and genes for EF-G and EF-Tu, which we call the Str transcription unit (2,7). None of the insertions in the Spc transcription unit appear to affect the expression of any of the r-protein genes to the right of the Spc unit on Afus3, including the genes in the Str unit. Thus, the use of the S7 gene for the internal standardization is justified. Aspcl does not carry any of the r-protein genes of Afuss that are located to the right of the Spc transcription unit. Thus, for derivatives of Aspcl we have used the S4 gene for normalization.
This gene appears to be in the cy transcription unit, which is located distal to the Spc transcription unit (see Fig. 1) (5). Our earlier experiments had indicated that the genes in this unit were only weakly affected by the insertions in the Spc unit (7). The experiments reported below with the Spc insertion mutants of Afuss indicate that the genes in the (Y unit are partially affected by these insertions, but not to the same extent as the genes in the Spc unit (see "Discussion").
Nevertheless, it is still possible to use the S4 gene for the standardization in comparing the various derivatives of Aspc 1. Previously we had found that the IS2 insertion in Aspcl-116 reduced the expression of the genes for S5 (spc), S8, S14, L5, L6, L14, L15, L18, L24, and L30, i.e. all of the genes in the Spc transcription unit. Similar results were seen for the experiment reported in Table II. The insertions in hspcl-126, and Aspcl-133, which are located at approximately the same position as the insertion in Aspcl-116, also had the same qualitative effect. However, the insertion in ~~~~1-133 seemed to be less polar.
In contrast to these results, the insertions located farther from the junction reduced the levels of expression of only some of the genes in the Spc transcription unit. For example, in Aspcl-140, in which the insertion is located farther to the  (3,6) and the transcription map of the r-protein genes on Afus3 (2,4,5,7,8)   left than it is in hspcl-116, the expression of genes for L24, L5, S14, and possibly the genes for L14, or LX, or both, were not affected. The expression of other genes was reduced at least partially. For hspcl-139, in which the insertion is located even farther to the left, the expression of the gene for S8, in addition to the genes for L24, L5, 514, and possibly L14 (or L15, or both), was not reduced. The simplest interpretation of gives the average of the results for S17, L16, L29, S3, S19, L22, L2, L23, L4, L3, SlO and 512.
None of these genes nor the gene for S7 is present on Aspcl, but they are present on .kfus3 (cf. 3).
by guest on March 23, 2020 http://www.jbc.org/ Downloaded from these observations is that the insertions only reduced the expression of genes located distal to the insertion, relative to the promoter. Since the direction of transcription is leftward (in Fig. 11, these genes would be located to the left of the insertion.
The genes whose expressions were not reduced would presumably be located between the insertion and the promoter, i.e. to the right of the insertion. Thus, analysis of the effect of insertions located at different positions in the transcription unit should tell us the relative order of the genes, much as would the analysis of a collection of nonsense mutations.
The genes whose expression seemed to be reduced by the insertion are underlined in Table II. For several of the mutants there is one protein whose synthesis is reduced considerably more than the others, reduced to what appears to be zero. Such is the case for L5 in Aspcl-128 and SS in Aspcl-140. It is possible that the insertion is actually located within the structural gene for these proteins. (In vitro experiments using DNA from Aspcl-140 showed that the synthesis of S8, but not other proteins, was abolished by this insertion, confirming this possibility (3). For further discussion on the in vitro experiments using this and other insertions, see an accompanying paper (31.) From the results of Table II and the physical location of the insertions, the gene order appears to be: promoter, L24, L5, S14, SS, L6, L18, (S5, L30).
The position of the genes for L14 and L15 in the unit is not clear from the results given in TaMe II because the twodimensional gel system used to separate the r-proteins for these experiments did not give a clean separation of these two proteins. However, it was possible to separate them from each other and from the other r-proteins whose genes are carried by Xspcl by electrophoresis on a one-dimensional polyacrylamide-urea gel at pH 4.5. This can be seen in the experimental results shown in Fig. 3. In this experiment, the proteins synthesized after hspcl infection of UV-irradiated bacteria were labeled with 13Hlleucine, extracts of these cells were electrophoresed on the one-dimensional polyacrylamideurea gel, and the r-proteins synthesized from Aspcl were detected by fluorography.
The bands have been identified by comparison with the mobility of purified reference r-proteins and, for some of the proteins, by demonstrating that they could be precipitated with antisera to that particular r-protein. It can be seen that proteins S5, L6, L14, LX, and L24 were synthesized in infected cells and that these five proteins can be separated in the gel system used, except that separation of L6 and S5 is not complete. (We note that the S5 made from Aspcl in the UV-irradiated bacteria appeared to have a slightly faster mobility in these gels than the reference S5. The significance of this is not known.) Fig. 4 gives the results of an experiment in which the rproteins synthesized in UV-irradiated bacteria after infection with several different insertion mutants of Aspcl were analyzed by electrophoresis on a one-dimensional urea slab gel. Several of the bands are identified on the left (see also the figure legend). The distance between the insertion and the junction for the various mutants used in this experiment (see Table I) increases as one goes from left to right across the gel. The last sample is for ~~~~1-123, which has the insertion farthest from the junction of all the Aspcl mutants.
The following observations can be made from the experiment shown in Fig. 4. The synthesis of S5, LX, and L30 was greatly reduced for all the mutants tested, indicating that the genes for these proteins are toward the end of the transcription unit. In addition to these proteins, the synthesis of L5, L14, were solubilized in urea and electrophoresed on a 10% polyacrylamide-urea gel at pH 4.5. (The details of the technique are given in the accompanying paper (2; under "Materials and Methods" and legend to Fig. 41.) A portion of the gel (Lanes 2 to 7) was processed for fluorography and an 18-day exposure is shown. Samples were: 2, original lysate; 3, anti-CL3 + LGNreated; 4, anti-S5-treated; 5, anti-L14-treated; 6, anti-L15-treated, and 7, anti-L24-treated. Fluorograms with a shorter exposure time (2 days; not shown) clearly showed that the. major bands in Lanes 3, 6, and 7 correspond to the positions of the reference proteins, L6, L15, and (the lower band of) L24, respectively. Reference proteins were run in the other lanes and a photograph of the stained se1 (not nrocessed for fluoroaranhv) is shown for these lanes. The samples were: 1, TP70; 8, TP%; 6, i3 + L6; 10, S5; 11, L14; 12, TP70; 13, L15; 14, L24; and 15, TP70. It should be noted that L24 when purified showed a single band in the same gel system corresponding to the lower one of the two stained bands shown. Protein in the upper bands was apparently produced during storage and is probably a derivative of L24. TP70 is a mixture of r-proteins extracted from 70 S ribosomes. Processing for fluorography reduces the size of the gel, and the photograph of the fluorogram was enlarged to correspond to that of the stained gel.
L24, and 514 can be seen to be reduced in the hspcl-116 and Aspcl-133 samples (Lanes 4 and 5). If one examines the samples from left to right it is clear that the synthesis of L14 begins with the sample in Lane 6, the synthesis of L24 begins in Lane 7, and the synthesis of L5 and S14 begins in Lane 8. These observations indicate the gene for L14 is closer to the promoter than the gene for L24, which in turn is closer to the promoter than the genes for L5 and S14. Since all the other genes appear to be distal to the L24 gene (cf . Table II as well as Fig. 41, it appears that the L14 gene is the first gene in the Spc transcription unit. In general, the results of the experiment in Fig. 4 are consistent with the results of the experiments in Table II. Note that the synthesis of (Y and S13 is relatively unaffected by the insertions, indicating they are in a different transcription unit. Nothing can be said about the synthesis of S4, S8, L6, Sll, L17, and L8 from the experiments in Fig. 4 since they are not separated from other r-proteins synthesized after Aspcl infection. Taking the data shown in Table II and Fig. 4 together, we conclude that the order of genes in the Spc transcription unit is: promoter, L14, L24, L5, S14, S8, L6, LB, (S5, L15, L30). As described under "Discussion," other experiments have shown that S5 is before L15 and L30, but the order of L15 and L30 has not been determined.  Fig. 3).

DISCUSSION
The results of our analysis of 16 Spc insertion mutants are summarized in the lower part of Fig. 1. The size and position of each gene is indicated along with the location of each insertion (cf. the legend to Fig. 1). The positions of the genes relative to the insertions were determined from the data in Table II. For example, the L5 gene appears to be to the right of the 1211 insertion, and the S14 gene appears to be to the right of the 140 insertion, but "cut" by the 1211 insertion. Since the 514 gene seems to be to the left of the L5 gene, we conclude that the S14 gene is approximately between the 1220 and 140 insertions as indicated in Fig. 1. The positions of the other genes in the transcription unit relative to the insertions were determined in a similar fashion. In general, the results shown in Table II and Fig. 4, as well as the relative positions of the insertions are consistent with the simple conclusion that the insertions only reduced the expression of genes located distal to the insertion. As mentioned under "Results," the present data on the polarity allows mapping of genes as follows: promoter, L14, L24, L5, S14, S8, L6, L18, (S5, L15, L30). Independent experiments using a DNA-dependent in vitro protein-synthesizing system have given more direct evidence on the order of genes in this region as follows (see accompanying paper, Ref. 3): L14, (L24, L5), S14, S8, L6, S5, (L15, L30). L18 was located between S14 and S5, but its position relative to S8 and L6 was not determined.
We also note that the order S5, (L15, L30) obtained from the in vitro experiments is consistent with our failure to isolate any insertion mutants which inactivate S5 without inactivating L15, or L30, or both. Thus, the agreement between the present mapping results and the in vitro mapping results is excellent and supports the validity of the present method based on the polarity caused by various insertions.
The only exceptional insertion we have found is 1210 (see Table I and Fig. 1). This insertion is located the farthest from the junction of all the insertions, and is in the 3% EcoRI fragment which is to the left of the 10% EcoRI fragment (cfi Fig. 11." If the physical arrangement of genes in the Spc unit is like that shown in Fig. 1, as we believe, then the 1210 insertion is distal to the spc gene. In our preliminary experiments, this insertion did not seem to decrease significantly the expression of any of the genes in the Spc unit. Yet this insertion mutant was isolated as the one affecting the expression of the spcs gene on hspcl. It is possible that only a small reduction in the expression of a spc' allele makes it recessive to a spc' allele. Since IS2 elements carry a promoter (13), one could imagine that if the direction of transcription from the insertion was opposite to the direction of transcription from the unit in which it is inserted, then the "collision" of the two polymerases could result in a decrease in the expression of promoter proximal genes.
The four insertions in ~fi.73 that were investigated seemed to decrease the expression of the r-protein genes in the (Y unit to about 50% (45 to 90%). The expression of the genes in the Spc unit that appear to be affected was reduced to 15 to 30% in most of the cases. This suggests the genes in the (Y unit are part of the Spc transcription unit to some degree. Nevertheless, it appears these genes can be expressed from another promoter that is distal to the Spc unit, i.e. the (Y promoter (5,8). Several models could accommodate this complication.
For example, the genes in the (Y unit may be expressed approximately 50% of the time from the a! promoter and 50% of the time from the Spc promoter. Or the (Y promoter may be a "secondary" promoter which is used in the Spc insertion mutants.
The effect of the Spc insertions on the expression of the genes in the (Y unit is not apparent in Table II for the hspcl mutants because all the data for these mutants was normalized to the expression of S4, one of the genes in the LY unit. If the expression of S4 is actually about 50% of its expression from the parent phage, as it is for the hfus3-Spc insertion mutants, then the data in Table II for all the Xspcl mutants should be divided by 2. This would help explain how some of the genes that are promoter proximal to the insertion appear to be expressed at twice their normal rate, e.g. S14 and L5 in  In principle, we could have mapped the genes in the Spc unit with a collection of nonsense mutants. However, the use of insertions