| Fusion plasmid | Host | After 1 h Induced | After 2 h Induced | |||
|---|---|---|---|---|---|---|
| Inducing treatment | Units GalK | Non- induced | Units GalK | Non- induced | ||
| PSAgalK | AB1157 | - | 90 | 84 | ||
| AB1157 | mitC | 182 | 2.0 | 252 | 3.0 | |
| AB1157 | - | 109 | ||||
| AB2463 | - | 120 | 1.1 | |||
| AB1157 | - | 80 | ||||
| DM1187 | - | 162 | 2.0 | |||
| DM511 | 32 ℃ | 100 | ||||
| DM511 | 42 ℃ | 210 | 2.1 | |||
| pSgalK | AB1157 | - | 70 | 78 | ||
| AB1157 | mitC | 70 | 1.0 | 102 | 1.3 | |
| PSD46gaIK | AB1157 | - | 86 | 112 | ||
| AB1157 | mitC | 216 | 2.5 | 266 | 2.4 | |
| PSD19galK | AB1157 | - | 78 | 104 | ||
| AB1157 | mitC | 78 | 1.0 | 120 | 1.2 | |
| PSD5galK | AB1157 | - | 80 | 80 | ||
| AB1157 | mitC | 90 | 1.1 | 84 | 1.1 | |
Strains AB2463(recA), DM1187(spr) and DM511(tsl) are deriva- tives of strain AB1157(rec+lex+)9. Growth conditions and measurement of galactokinase levels were as described previously6. Mitomycin C (mitC) was added to the cultures to a final concentration of 5 µg ml-1. Units galK are normalized for cell density. (The galactokinase levels measured with plasmid pSpsgalK are comparable with that promoted by pSgalK, indicating that deletion of the HaeIII-HindIII fragment in the case of pSgalK (Fig. 1) does not influence the expression of galK.)
involvement of the SOS box in the induction of the ssb gene (Table 1, Fig. 3).
For other damage-inducible genes, the regulation by LexA can be explained by the location of the SOS box in the promoter region of the corresponding gene. Examination of the DNA sequence around the LexA binding site reveals the presence of a putative promoter (‘-35’: position 141-146; ‘-10’: position 117-122; Fig. 3). Evidence for this promoter was obtained by the identification of an inducible transcript starting about 10 base pairs (bp) downstream of the LexA binding site, by S1 mapping (Fig. 4). In addition to the inducible one, a shorter transcript can be detected (Fig. 4) which is not inducible. We are at present determining the precise location of this promoter. We conclude that the ssb gene is transcribed from both a non-inducible and a damage-inducible promoter, the latter being regulated by the same repressor binding site as the uvrA promoter.
The presence of a non-inducible promoter for ssb transcrip- tion can be explained from the central role of the SSB protein in DNA replication8. On the other hand, the inducibility of the ssb gene might reflect its presumed role in the regulation of the SOS response. It has been proposed that SSB is particularly involved in the availability of signal molecules necessary for activation of RecA protein9. Therefore, the amount of SSB protein is probably critical for the cellular response to DNA damage. This is also demonstrated by the finding that presence of the ssb gene on a multicopy plasmid sensitizes wild-type cells to UV10. Furthermore, it has been reported that both in ssb mutants and in cells that overproduce the SSB protein, RecA induction is inhibited11. As RecA also binds to single-strand DNA, both proteins might compete for single-strand regions. The increase in SSB protein after SOS induction might counter- act the activation of RecA, subsequently leading to the turn-off of the SOS response.
The finding that the LexA binding site located between two genes can be functional in both directions is novel. Recently, Reed et al.12 have described a TnpR repressor, regulating two divergently transcribed genes at a single site in between these genes, and have shown that the TnpR repressor acts on both TnpA and TnpR transcription. Also in this case, the promoters of both genes are overlapping.
Our data seem to contradict the recent results of Salles and co-workers13 who found no evidence for induction of SSB after treatment of E. coli cells with DNA damaging agents. Their method is based on the immunological measurement of the amount of SSB protein present in the cell before and after treatment. The reason for this discrepancy is unknown and requires further study. One possible explanation might be that the amount of SSB protein detectable by the IRMA method is dependent on the number of SSB binding sites, which increases strongly after treatment of the cells with DNA damaging agents.
We thank Drs C. A. van Sluis and J. Brouwer for critical reading of this manuscript, E. van den Berg for help and sugges- tions and Mrs N. van Hoek for typing the manuscript. This work was supported by the Euratom contract BIO-E-408-81-NL(G).
Received 28 March; accepted 21 July 1983.
1. Glassberg, J., Meyer, R. & Kornberg, A. J. Bact. 140, 14-19 (1979).
2. Sancar, A., Williams, K. R., Chase, J. W. & Rupp, W. D. Proc. natn. Acad. Sci. U.S.A. 78, 4274-4278 (1981).
3. Kenyon, C. J. & Walker, G. C. Proc. natn. Acad. Sci. U.S.A. 77, 2819-2823 (1980).
4. Lieberman, H. B. & Witkin, E. M. Molec. gen, Genet. 183, 348-355 (1981).
5. Whittier, R. F. & Chase, J. W. Molec, gen. Genet. 183, 341-347 (1981).
6. Backendorf, C., Brandsma, J. A., Kartasova, T. & Van de Putte, P. Nucleic Acids Res. (in the press).
7. Sancar, A., Sancar, G. B., Rupp, W. D., Little, J. W. & Mount, D. W. Nature 298, 96-98 (1982).
8. Wickner, S. H. A. Rev. Biochem. 47, 1163-1191 (1978).
9. Little, J. W. & Mount, D. W. Cell 29, 11-22 (1982).
10. Brandsma, J. A., Stoorvogel, J., Van Sluis, C. A. & Van de Putte, P. Gene 18, 77-85 (1982).
11. Meyer, R. R., Voegele, D. W., Ruben, S. M., Rein, D. C. & Trela, J. M. Mutat. Res. 94, 299-313 (1982).
12. Reed, R. R., Shibuya, G. I. & Steitz, J. A. Nature 300, 381-383 (1982).
13. Salles, B., Paoletti, C. & Villani, G. Molec. gen. Genet. 189, 175-177 (1983).
14. Bolivar, F. et al. Gene 2, 95-113 (1977).
15. Rosenberg, M. & Court, D. A. Rev. Genet. 13, 319-353 (1979).
16. Berk, A .- J. & Sharp, P. A. Cell 12, 721-732 (1977).
Amplified DNA with limited homology to myc cellular oncogene is shared by human neuroblastoma cell lines and a neuroblastoma tumour
Manfred Schwab*, Kari Alitalo*, Karl-Heinz Klempnauer*, Harold E. Varmus*, J. Michael Bishop*, Fred Gilbertt, Garrett Brodeur}, Milton Goldstein & Jeffrey Trent§
* Department of Microbiology and Immunology, University of California, San Francisco, California 94143, USA
t Department of Pediatrics, Mount Sinai School of Medicine, New York, New York 10029, USA
# Departments of Pediatrics and Anatomy, School of Medicine, Washington University, St Louis, Missouri 63110, USA
§ Department of Internal Medicine and the Cancer Center, University of Arizona College of Medicine, Tucson, Arizona 85724, USA
Amplified cellular genes in mammalian cells frequently manifest themselves as double minute chromosomes (DMs) and homogeneously staining regions of chromosomes (HSRs)1,2. With few exceptions both karyotypic abnormalities appear to be confined to tumour cells3.4. All vertebrates possess a set of cellular genes homologous to the transforming genes of RNA tumour viruses, and there is circumstantial evidence that these cellular oncogenes are involved in tumorigenesis5. We have recently shown that DMs and HSRs in cells of the mouse adrenocortical tumour Y16 and an HSR in the human colon carcinoma COLO3207 contain amplified copies of the cellular oncogenes c-Ki-ras and c-myc, respectively. Both DMs and HSRs are found with remarkable frequency in cells of human neuroblastomas4,8-15. We show here that a DNA domain detect- able by partial homology to the myc oncogene is amplified up to 140-fold in cell lines derived from different human neuroblas- tomas and in a neuroblastoma tumour, but not in other tumour
cells showing cytological evidence for gene amplification. By in situ hybridization we found that HSRs are the chromosomal sites of the amplified DNA. The frequency with which this amplification appears in cells from neuroblastomas and its apparent specificity raise the possibility that one or more of the genes contained within the amplified domain contribute to tumorigenesis.
We performed initial studies with the neuroblastoma cell lines NMB8,10 and Kelly (F. G., unpublished), each of which contains HSRs, to look for evidence for amplification of known oncogenes. DNA cleaved with restriction endonuclease EcoRI was analysed by agarose gel electrophoresis and Southern blot- ting in conditions of reduced stringency for sequences homologous to 14 retroviral transforming genes (src, yes, myb, myc, erb, rel, fps, mos, fos, abl, Ha-ras, Ki-ras, fes and sis). DNA from both cell lines yielded a 2.0-kilobase pair (kbp) EcoRI fragment that hybridized with the probe for v-myc (Fig. 1). An analogous fragment was barely visible in DNA from normal human fibroblasts and cannot be seen in Fig. 1.
The major c-myc locus in human DNA lies within a 13.5-kbp EcoRI fragment16. Because human myc has diverged appreci- ably from the analogous avian gene, we could readily detect the 13.5-kbp fragment with v-myc probe only in DNA in which the locus is amplified (COLO320 DNA7 in Fig. 1). In addition to the 13.5-kbp and the 2.0-kbp fragments, a number of frag- ments is detectable with v-myc probe in all human cell lines. This result has also been reported by others16. It shows that the human genome contains a series of sequences related to the myc oncogene.
The 2.0-kbp EcoRI fragment derived from amplified DNA in neuroblastoma cells is not contained within the major locus of human c-myc 16. We explored the nature of the c-myc-related sequences in the amplified region by using probes for the 5’ and 3’ domains of v-myc (Fig. 1). The 2.0-kbp EcoRI fragment reacted primarily with the probe for the 5’ domain. The probe for the 3’ domain reacted well with the amplified c-myc locus in the DNA of COLO320 cells, but very poorly with the 2.0-kbp EcoRI fragment and not with any other fragment from neuro- blastoma DNA other than the 13.5-kbp fragment. We conclude that the 2.0-kbp EcoRI fragment contains a sequence homologous primarily to the 5’ domain of the myc oncogene.
For more detailed studies we cloned the myc-related sequence from neuroblastoma line Kelly into bacteriophage AgtWES, and subcloned into pBR322 an approximately 1.0- kbp EcoRI-BamHI fragment (Nb-1) that contained the myc- related sequence. Other investigators have recently cloned from human cells three sequences related to the 5’ end of myc16. It is likely that we have isolated an additional sequence, for the published restriction endonuclease maps are different from the one we have obtained.
We have further analysed the homology of Nb-1 to human c-myc. The c-myc locus used is an apparently normal allele cloned previously from COLO320 cells (M.S., unpublished results); the restriction endonuclease map of the cloned DNA was found to be identical to that published by other inves- tigators16. Southern blotting analyses showed that the region of homology is confined to a 1.4-kbp SacI fragment of c-myc (Fig. 2A) containing the exon that gave rise to the 5’ domain of v-myc. This result is not unexpected, as we had previously established (Fig. 1) that a probe specific for the 5’ domain of v-myc gives a strong signal on Southern blots containing genomic DNA of neuroblastomas. Additional analyses revealed that the homologous region is localized within a 0.35-kbp Xhol-PstI fragment of Nb-1 and maps to a 0.4-kbp PstI frag- ment of c-myc (Fig. 2A).
We have sequenced the 0.35-kbp Xhol-PstI fragment of Nb-1 and compared the nucleotide sequence with that of the 0.4-kbp Pst-I fragment of c-myc17. Inspection of the nucleotide sequences reveals that Nb-1 contains two blocks 74 and 59 base pairs long showing 78% homology to corresponding regions of c-myc, and separated by a stretch of 116 nucleotides that bears no apparent relationship to c-myc (Fig. 2B). The
NMB
Kelly
COLO320
COLO320
HSF
NMB
NMB
kbp
13.5
2.0-
v-myc
5’
3’
| Designation of cells | Tumour | Ref. | HSR on chromosome | DMs | amplification Fold |
|---|---|---|---|---|---|
| Kelly | Nb | F.G .* | 13p, markers | - | 100-120 |
| NGP | Nb | 10 | 4p16, 12q13 | - | 120-140 |
| NLF | Nb | G.B .* | 9q13 | - | 20-25 |
| CHP-134 | Nb | 8 | 6q, 7p | - | 20-25 |
| CHP-126 | Nb | 8,9 | 5q33 | + | 100-120 |
| IMR-32 | Nb | 13 | 1p34 | - | 15-20 |
| NMB | Nb | 8-10 | 13p11 | + | 100-120 |
| MCN-1 | Nb | 12 | - | + | 5-8 |
| AR (tumour) | Nb | This study | ? | ? | 80-100 |
| SK-N-SH | Nb | 24 | - | - | 0 |
| HA-A | Mel | J.T .* | 7p | - | 0 |
| HA-L | Mel | J.T .* | 7p | - | 0 |
| Y79 | Ret | 14 | 1p34 | - | 0 |
| COLO320 | Carc | 25 | Xt | - | 0 |
Nb, neuroblastoma; Mel, melanoma; Ret, retinoblastoma; Carc, neuroendocrine cells from a carcinoma of the colon. +, - Indicates the presence or absence of HSRs or DMs. The degree of amplification was calculated by blotting different amounts of DNA and by varying the exposure times of the autoradiographs.
* Unpublished results of F.G., G.B. and J.T., respectively.
+ C. C. Lin, personal communication.
A
5’3’
Sacl
Pstl
Pst!
Saci
Clal
c-myc L
Nb-1
EcoRI
Xhol
Pstl BamH! 1
I
1.0 kbp
1
B Xho I
CTCGAGTTTGACTCGCTACAGCCCTGCTTCTACCCGGACGAAGATGACTTCTACTTCGGC 1
1145
CGCCCAGCGAGGATATCTGGAAGAAATTCGAGCTGCTGCCCACC
61 GGCCCCGACTCGACCCCCCCGGGGGAGGACATCTGGAAGAAGTTTGAGCTGCTGCCCACG
CCGCCCCTGTCCCCTAGCCGCCGCTCCGGG
121 CCCCCGCTGTCGCCCAGCCGTGGCTTCGCGGAGCACAGCTCCGAGCCCCCGAGCTGGGTC
181 ACGGAGATGCTGCTTGAGAACGAGCTGTGGGGCAGCCCSGCCGAGGAGGACGCGTTCGGC
1339
CAAAAACATCATCATCCAGGACTGTATGTGGAGO
241 CTGGGGGGACTGGGTGGCCTCACCCCCAACCCGGTCATCCTCCAGGACTGCATGTGGAGC
GGCTTCTCGGCCGCCGCCAAGCTCG
PstI 301 GGCTTCTCCGCCCGCGAGAAGCTGGAGCGCGCCGTGAGCGAGAAGCTGCAG
Fig. 2 Analysis of homology between the myc-related sequence amplified in neuroblastoma cells and complete human c-myc. A, Mapping of the myc-related sequence to c-myc. The map of c-myc (upper map) was established by using a recombinant bacteriophage containing the c-myc locus isolated from COLO32025 cells (unpublished results of M.S.). Large rectangles symbolize major exons of human c-myc according to ref. 17. The myc-related sequence was isolated from neuroblastoma line Kelly. Total genomic DNA was partially digested with the restriction endonu- clease EcoRI and ligated into bacteriophage AgtWES arms. From ~104 recombinants a recombinant phage was purified that carries the 2.0-kbp myc-related sequence. The region related to myc was further mapped in bacteriophage and genomic DNA to a 1.0-kbp EcoRI-BamHI fragment (Nb-1; lower map), which was subcloned into pBR322 (pNb-1). The region of homology in Nb-1 is confined to a 0.35-kbp Xhol-PstI fragment. The precise regions of homology between c-myc and Nb-1, determined by nucleotide sequencing, are indicated by solid black rectangles in both diagrams. B, Comparison of the nucleotide sequences of the myc-related amplified DNA with the homologous region in c-myc. Asterisks indicate homology between the myc-related sequence amplified in neuroblastoma cells and human c-myc ; numbers refer to the nucleotide positions in c-myc according to ref. 17. Nb-1 was digested with restriction endonucleases XhoI and PstI, cloned into bacteriophages M13mp8 and M13mp9, and the nucleotide sequence spanning the Xhol-PstI restriction endonuclease sites was determined by the dideoxynucleotide chain termination method27. The nucleotide sequence of the major human c-myc locus was taken from ref. 17.
two blocks of homology with c-myc represent the only apparent explanation for the hybridization between Nb-1 and the myc probe.
A single open reading frame spans the entire sequence illus- trated in Fig. 2B. In the regions that are homologous to c-myc, the open reading frame encodes amino acid sequences that can be aligned with c-myc (data not shown). Moreover, the diverged sequence that separates the two regions of homology is virtually the same length in both Nb-1 and c-myc. It seems possible that c-myc and the domain represented by Nb-1 diverged from the same distant ancestor. We do not yet know whether the DNA of Nb-1 is derived from a functional gene, nor whether any portion of the DNA amplified in neuroblastoma cells is expressed in either normal or malignant tissues. Studies on the expression of the amplified DNA are now in progress.
We have tested nine neuroblastoma lines for amplification of the region that yields the 2.0-kbp EcoRI fragment with the myc-related sequence, using Nb-1 as the probe (Fig. 3, Table 1). In addition, we have tested a metastatic neuroblastoma (AR) obtained from a 13-month-old female patient before chemotherapy. All cell lines except one (SK-N-SH) contain DMs and/or HSRs (Table 1); the karyotype of the tumour AR has not been established. Amplification was found in all the
Kelly
CHP-134
CHP-126
IMR-32
NMB
MCN-1
SK-N-SH
COLO320 HSF
NGP
NLF
a.
b.
c.
AR
HA-A
HA-L
Y79
kbp
2.0-
cell lines showing cytological evidence of gene amplification. The line SK-N-SH, which shows neither DMs nor an HSR, does not contain an amplified myc-related sequence. The degree of amplification varied considerably between the different neuroblastoma lines, by 120-140-fold in NGP to 5-8-fold in MCN-1 (Table 1). The DNA from the tumour AR showed approximately 80-100-fold amplification. Several HSR- bearing cell lines established from other types of tumours, such as melanoma, retinoblastoma and colon carcinoma (Fig. 3), showed no amplification of the myc-related sequence: the signal obtained in Southern blots was invariably faint and not different from that displayed by DNA from skin fibroblasts. The major c-myc locus is not detected by the Nb-1 probe in the conditions of hybridization used, even when amplified as in COLO320 cells. Similar results were obtained when we analysed DNA digested with restriction endonucleases HindIII and BamHI (data not shown). Our results show that multiple cell lines derived from neuroblastomas in several laboratories and also tumour tissue from a neuroblastoma share a specific amplified sequence whose abundance is not increased in other tumour cells with karyotypic evidence of gene amplification.
We sought the chromosomal location of amplified DNA in neuroblastoma cells by using hybridization in situ with 3H- labelled DNA of Nb-1. Three cell lines have been tested-NGP, NLF and NMB (Table 1)-in all of which autoradiography revealed clustering of grains along HSRs. Figure 4 illustrates representative data for one of the two HSRs (4p) in line NGP. Details of our findings with other cell lines will be reported elsewhere (manuscript in preparation). The number of autoradiographic grains was roughly proportional to the extent of amplification in the cell lines (summarized in Table 1): the strongest signal was obtained with line NGP, weaker signals with lines NMB and NLF. We conclude that the HSRs in neuroblastoma cell lines carry amplified DNA that includes a domain distantly related to c-myc.
B
4p HSR
The HSRs in the various neuroblastoma cell lines occupy different chromosomal locations (see Table 1). Because it is likely that the DNA represented by Nb-1 comprises one or very few copies in normal cells, we suggest that amplification of the DNA is followed by translocation to any of numerous chromosomal sites. The issue can be resolved conclusively by mapping the DNA of Nb-1 on the chromosomes of normal cells (work in progress).
Other investigators have recently shown by DNA-mediated gene transfer using NIH/3T3 cells as recipients that the neuro- blastoma line SK-N-SH carries a transforming gene18 related to Ha- and Ki-ras (N-ras)19. It is possible that at least two distinct genetic events involving different oncogenes play a part in the aetiology of neuroblastoma. One event would be the presently uncharacterized change that activates an oncogene related to the ras genes. The other event would be the amplification of a presently unidentified oncogene, providing additional template for transcription and increased levels of the gene product. The two changes have not yet been detected in the same cell line or tumour.
Karyotypic evidence of gene amplification has been found in a variety of human tumours4.10,11.20-22, and recent evidence indicates that cellular oncogenes are included in at least some of these amplifications6,7,16,23. Here we have demonstrated that numerous cell lines derived from neuroblastomas share an amplified domain of DNA and that the same domain has been amplified in a neuroblastoma isolated directly from a patient prior to chemotherapy. Our results suggest that amplification of an as yet unidentified cellular gene may be instrumental in the aetiology of human neuroblastoma.
These studies were supported by grants from the NCI and the ACS (J.M.B., H.E.V., J.T.). M.S. is a Heisenberg fellow, K.H.K. is a recipient of a training fellowship from the Deutsche Forschungsgemeinschaft, and K.A. is a fellow of the NIH Fogarty International Center. J.T. is a scholar of the Leukemia Society of America. We thank J. Beckstead, M. R. Harrison, J. Kushner and V. Levin for assistance in obtaining and identify- ing the neuroblastoma tumour.
Received 23 May; accepted 29 June 1983.
1. Schimke, R. T., Kaufman, R. J., Alt, F. W. & Kellems, R. F. Science 202, 1051-1055 (1978).
2. Cowell, J. K. An. Rev. Genet. 16, 21-59 (1982).
3. Barker, P. E. Cancer Genet. Cytogenet. 5, 81-94 (1982).
4. Sandberg, A. A., Sakurai, M. & Holdsworth, R. N. Cancer 29, 1671-1679 (1972).
5. Bishop, J. M. An. Rev. Biochem. (in the press).
6. Schwab, M., Alitalo, K., Varmus, H. E., Bishop, J. M. & George, D. Nature 303, 497-501 (1983).
7. Alitalo, K., Schwab, M., Lin, C. C., Varmus, H. E. & Bishop, J. M. Proc. natn. Acad. Sci. U.S.A. 80, 1707-1711 (1983).
8. Balaban-Malenbaum, G. & Gilbert, F. Science 198, 739-741 (1977).
9. Gilbert, F., Balaban, G., Moorhead, P., Bianchi, D. & Schlesinger, H. Cancer Genet. Cytogenet. 7, 33-42 (1982).
10. Brodeur, G. M., Sikhon, G. S. & Goldstein, M. N. Cancer 40, 2256-2263 (1977).
11. Brodeur, G. M. et al. Cancer Res. 41, 4678-4686 (1981).
12. Casper, J. T. et al. Cancer Genet. Cytogenet. (in the press).
13. Tumilowicz, J. J., Nichols, W. W., Cholon, J. J. & Greene, A. E. Cancer Res. 30, 2110-2118 (1970).
14. Gilbert, F. et al. J. natn. Cancer Inst. 67, 301-306 (1981).
15. Levan, A., Manolov, G. & Clifford, P. J. natn. Cancer Inst. 41, 1377-1387 (1968).
16. Dalla-Favera, R. et al. Proc. natn. Acad. Sci. U.S.A. 79, 6497-6501 (1982).
17. Colby, W., Chen, E. Y., Smith, D. W. & Levinson, A. Nature 301, 722-725 (1983).
18. Perucho, M. et al. Cell 27, 467-476 (1981).
19. Shimizu, K. et al. Proc. natn. Acad. Sci. U.S.A. 80, 2112-2116 (1983).
20. Lubs, H. A., Salmon, J. H. & Flenigan, S. Cancer 19, 591-599 (1966).
21. Mark, J. & Granberg, I. Acta neuropath. 16, 194-204 (1970).
22. Mark, J. Acta ophthal. 48, 124-135 (1970).
23. Collins, S. & Groudine, M. Nature 298, 679-681 (1982).
24. Biedler, J. L., Helson, L. & Spengler, B. A. Cancer Res. 33, 2643-2652 (1973).
25. Quinn, L. A., Moore, G. E., Morgan, R. T. & Woods, L. K. Cancer Res. 39, 4914-4924 (1979).
26. Vennstrom, B., Moscovici, C., Goodman, H. M. & Bishop, J. M. J. Virol. 39, 625-631 (1981).
27. Sanger, F., Nicklen, S. & Coulson, A. R. Proc. natn. Acad. Sci. U.S.A. 74, 5463-5467 (1977).
28. Trent, J. M., Olson, S. & Lawn, R. Proc. natn. Acad. Sci. U.S.A. 79, 7809-7813 (1982).
Requirement for an upstream element for optimal transcription of a bacterial tRNA gene
Angus I. Lamond & Andrew A. Travers
Laboratory of Molecular Biology, Medical Research Council Centre, Hills Road, Cambridge CB2 2QH, UK
Bacterial promoters are the sites at the 5’ end of each gene that bind RNA polymerase and direct the initiation of transcription. The functional elements of Escherichia coli promoters1,2 are two highly conserved sequences, each about six nucleotides long, usually centred at sites -10 and -35, +1 being the initiating nucleotide. We have been interested in the structure of promoters of genes that are subject to stringent control, that is whose expression is reduced in conditions of amino acid shortage, such as rRNA and tRNA genes. We have therefore mapped the sequences involved in promoting in vivo transcrip- tion of a bacterial tRNATyr (tyrT) gene by fusing the tyrT promoter region to a galactokinase (galK) gene, and using in vivo expression of galactokinase activity to measure promoter strength. We show here that efficient expression from the tyrT promoter requires specific sequences upstream of the canonical promoter elements, and we suggest that these sequences con- stitute an extended promoter structure.
The wild-type tyrT promoter was inserted into the galK expression vector pKM-1 (K. M. McKenney and M. Rosenberg, personal communication) as a 300 base pair (bp) EcoRI-Aval fragment, which carries 248 bp of tyrT sequence upstream of the transcription initiation site, to give pTyr2 (Fig. 1a). Deletion mutations of the tyrT promoter were generated by cutting pTyr2 at the BstEII site and then digesting with nuclease Bal31. This produced a set of deletions which all share a common 5’ end (the EcoRI site on pKM-1) and which have 3’ ends mapping from -76 to -33 (Fig. 1b). An additional deletion was made at -98 by excising the EcoRI-BstEII fragment from pTyr2.
The pKM-1 vector carries a terminator (AfR1) inserted between the promoter cloning sites and the galK gene. We found it impossible to clone the wild-type tyrT promoter into analogous vectors lacking a terminator, as has been found in this system with other strong promoters such as leuT3 and rrnB (M. Cashel, personal communication). Presumably this effect is caused by readthrough transcription from the cloned pro- moter inhibiting plasmid replication. As the presence of a strong promoter can also influence the copy number of plasmid vec- tors4.5, we analysed the relative levels of plasmid present in each of the tyrT-galK fusions in the same growth conditions