WO1995004830A1

WO1995004830A1 - Mycoplasma expression system

Info

Publication number: WO1995004830A1
Application number: PCT/US1993/007407
Authority: WO
Inventors: F. Chris Minion; Kevin L. Knudtson
Original assignee: Biotechnology Research And Development Corporation; Iowa State University Research Foundation, Inc.
Priority date: 1993-08-06
Filing date: 1993-08-06
Publication date: 1995-02-16
Also published as: EP0715654A1; AU4804293A

Abstract

A system for identifying mycoplasma regulatory sequences with protein fusion constructs has been developed. This system has identified mycoplasma regulatory sequences that can be used in expression vectors. The expression vectors employing these mycoplasma regulatory sequences permit the expression of foreign DNA sequences in mycoplasma hosts, such a Acholeplasma.

Description

MYCOPLASMA EXPRESSION SYSTEM

Background of the Invention The class Mollicutes encompasses a group of organisms collectively known as "mycoplasmas," many of which are important human and agricultural pathogens. Despite this pathogenicity, little is known about the genetics of mycoplasmas. These organisms possess the smallest genome thought necessary for autonomous existence. Razin, Microbiol . Rev. 49: 419-55 (1985). Due to their simplicity, most mycoplasma species require complex media for growth because they lack many biosynthetic pathways. In view of such limitations, traditional genetic studies employing auxotrophic mutants have not been possible with these organisms.

Based on RNA homology, mycoplasmas are thought to be a product of degenerative evolution from Gram-positive organisms. Previous studies of 16S rRNA sequence homology have suggested that mycoplasmas are more closely related to Gram-positive organisms than Gram-negative organisms eisburg et al . , J. Bacteriol . 171: 6455-67 (1989). The differences in translational specificity that have been demonstrated between the Gram-negative and Gram-positive bacteria also appear to pertain to mycoplasmas as well. Hager & Rabinowitz, The Molecular Biology of the Bacilli 1-34 (Dubnau ed., Acad. Press 1985). The simplicity of mycoplasmas offers advantages in the context of expression systems. For example, mycoplasmas lack lipopolysaccharide and other toxic wall constituents, which would allow for simplified purification of recombinantly produced proteins. Significant problems have existed, however, with using mycoplasmas as a recombinant expression system. Adequate stability of cloned genes has previously not been achieved. Moreover, previous attempts at creating mycoplasma-based expression systems have employed gram-negative promoters, which was necessitated by the unavailability and limited knowledge regarding mycoplasma promoters, generally. The transcriptional apparatus of gram-negative bacteria, however, is often unable to correctly recognize mycoplasma promoter sequences. For instance, it has been shown that, although the rRNA promoter of Mycoplasma capricolum is recognized by both E. coli and M. capricolum RNA polymerase, it is not properly recognized in E. coli . Gafny et al . , Nucl . Acids Res . 16: 61-76 (1988). Thus, the recognition of the mycoplasma rRNA promoter is activated in E. coli under the stringent condition of amino acid starvation, which is opposite of the expected result. Additional problems exist with such use of gram-negative hosts. Signals may arise from transcription initiation at pseudo-promoter sites, which are caused by the high (A + T) content in the mycoplasma DNA of the fusion gene. Vollenweider et al . , Science 205: 508-11 (1979). Notarnicola et al . have shown that E. coli initiated translation at internal sites in a Mycoplasma hyorhinis lipoprotein structural gene. J. Biol . 172: 2986-95 (1990). It has become apparent, therefore, that the use of E. coli as a cloning host to study promoter sequences from organisms with a high (A + T) content, such as mycoplasmas, should be limited. The lack of genetic tools also has made the development of mycoplasma cloning systems difficult. Only two transposons, Tn916 and Tn4002, have been shown to be useful for studying mycoplasma genetics. Dybvig & Alderete, Plaεmid 20: 33-41 (1988); Dybvig & Cassell, Science 235: 1392-94 (1987); Mahairas & Minion, Plasmid 21: 43-47 (1989) . A number of broad host-range plasmids from Gram-positive bacteria have been examined as possible cloning vectors, but all have proven to be unstable. Dybvig, Plaεmid 21: 155-60 (1989). Naturally occurring mycoplasma plasmids have also been examined as possible cloning vectors, but they have not been shown to maintain and express a cloned gene. Dybvig et a!., IOM Letts . 1: 209-10 (1990); King & Dybvig, Plasmid 28: 86-91 (1992). A cloning system has been developed in spiroplasmas which uses a spiroplasma virus as a cloning vector, but the vector has a limited host range. Stamburski et al . ; J. Bacteriol . 173: 2225-30 (1991); Gene 110: 133-34 (1992). Mahairas and Minion have developed a cloning system based on integration of cloned genes into mycoplasma chromosomes via homologous recombination. Mahairas, et al . , Gene 93: 61-65 (1990); Mahairas & Minion, J. Bacteriol . 171: 1775-80 (1989) . The stability and versatility of this system make it possible to incorporate DNA sequences into the host. This system, however, was not designed to express foreign DNA because it lacks proper regulatory elements.

The lack of expression vectors suitable for use in mycoplasmas also has handicapped the study and production of mycoplasma proteins. Recombinant production of mycoplasma proteins in E. coli is hampered by the unique codon usage of the mycoplasmas. For instance, many mycoplasma genre, such as Mycoplasma, ϋreaplaεma and Spiroplasma , read the conventional UGA stop codon as coding for tryptophan. Some mycoplasmas, such as A. laidlawii , appear not to, however. Accordingly, a conventional expression host, such as E. coli , ceases translation at UGA, which in a normal mycoplasma background would of^en encode tryptophan. Thus, translation of a mycoplasma polypeptide would be prematurely terminated in an E. coli recipient host. This problem can be avoided by employing a mycoplasma recipient host with a suitable vector for expressing the gene of interest. Summary of the Invention

It is an object of the present invention to system to identify mycoplasma regulatory regions containing regulatory sequences. It is another object of the present invention to provide for the recombinant production of foreign proteins by mycoplasmas.

It is also an object of the present invention to provide mycoplasma regulatory sequences and regions containing these sequences for use in a recombinant expression system.

It still another object of the present invention to provide a plasmid comprising mycoplasma regulatory sequences and a site for inserting foreign DNA. It is yet another object of the present invention to provide a plasmid where the mycoplasma regulatory sequences control the expression of the foreign DNA.

In achieving these objects, there has been provided, in accordance with one aspect of the present invention, an expression system employing mycoplasma regulatory sequences to control the expression of foreign DNA in host cells. Suitable host cells include the members of the class Mollicuteε, such as Acholeplaεma .

In accordance with another aspect of the present invention, there is provided a method for identifying mycoplasma regulatory elements via a gene fusion construct with a reporter gene, such as lacZ.

In accordance with still another aspect of the present invention, there is provided a plasmid comprising a mycoplasma promoter sequence and foreign DNA, which is transformed into an appropriate host in order to produce the protein encoded by the foreign DNA. The plasmid may further comprise mycoplasma DNA that is normally located upstream of a mycoplasma promoter in the native environment. The foreign DNA may include mycoplasma DNA. The mycoplasma regulatory sequences may be from Acholeplaεma or other mycoplasma genre.

In accordance with yet another aspect of the present invention, the expression system can the include -* or more complete mycoplasma regulatory region or one c aore fragments thereof. Additionally, the expression vector of the present invention can include more than one mycoplasma regulatory sequence, or combinations of mycoplasma sequences or fragments thereof.

Other objects, features and advantages of the present invention will become apparent from the following description.

Brief Description of the Figures

FIGURE 1 is a chart of the major strains and plasmids employed to develop the present invention.

Genotypes/phenotypes of Acholeplaεma transformants are defined by the nomenclature of "original strain: :plasmid."

FIGURE 2 depicts the construction of trp'-lacZYA fusion plasmids 2004, 2005, 2006, 2009, 2010 and 2011. Single headed arrows indicate the direction of transcription. The double headed arrow and the hatched bar denote the 700 bp region found upstream of the M. capricolum rrn P2 promoter. "Λf. cap. " refers to M. capricolum; "Ap" refers to ampicillin resistance; "Gm" refers to gentamicin resistance; ^uori refers to origin of replication. This figure also sets forth β-galactosidase activity for E. coli CSH50 and Acholeplaεma ISM1520 strains transformed with each plasmid. FIGURE 3 depicts the construction of the transcriptional fusion vector pISM2050. The vector was constructed by ligating a 3.1 kb BaraHI DNA fragment containing the promoterless lacZ from plasmid pMC1871 into the BamHI site of plasmid pISM1003. The arrow indicates the direction of transcription of lacZ . The asterisk (*) indicates a BamEI site that was inactivated. "Ap" refers to a picillin resistance; "G " refers to gentamicin resistance; "Tc" refers to tetracycline resistance; "p- lacZ" refers to promoterless lacZ; ori refers to the origin of replication.

FIGURE 4 depicts data from β-gal assays performed with seven of the lacZ fusion constructs introduced into ISM1520. CSH50 and χ289 served as controls.

FIGURE 5 depicts the alignment and determination of putative -10 and -35 promoter regions driving lacZ in plasmid pISM2050 derivatives based on similarity a consensus E. coli promoter sequence. Bold letters indicate transcriptional start sites.

Figure 6 depicts the sequence upstream (a region containing regulatory sequences) of the lacZ gene in fusion construct pISM2050.1. Additionally, the lacZ-chromosomal DNA fusion is shown in italics (GATC) . The putative -35 and -10 promoter regions are underlined and the transcriptional start site is indicated by an (*) . The potential ribosomal binding site (rbs) and translational start site (start) are also underlined. These designations are also employed in Figures 7-12. Figure 7 depicts the sequence upstream of the lacZ gene in fusion construct pISM2050.2.

Figure 8 depicts the sequence upstream of the lacZ gene in fusion construct pISM2050.8.

Figure 9 depicts the sequence upstream of the lacZ gene in fusion construct pISM2050.18.

Figure 10 depicts the sequence upstream of the lacZ gene in fusion construct pISM2050.25.

Figure 11 depicts the sequence upstream of the lacZ gene in fusion construct pISM2050.69. Figure 12 depicts the sequence upstream of the lacZ gene in fusion construct pISM2050.70. A potential ribosomal binding site ("rbs") was not found.

Detailed Description of Preferred Embodiments

A lack of specific information concerning mycoplasma regulatory sequences and regions containing these sequences heretofore has combined with an inability to readily detect or identify mycoplasma promoters and other regulatory sequences by means of cloning into E. coli . A system now has been developed, however, that allows for the reproducible identification of regions containing regulatory sequences of mycoplasmas. In particular, it has been discovered that a protein fusion construct, wherein a putative mycoplasma regulatory region is tested for an ability to drive expression of a reporter gene, lacZ, can be employed advantageously in a mycoplasma, such as an Acholeplaεma host, thereby to confirm the existence of an actual regulatory sequence in the test region. The term "regulatory sequence" denotes any sequence that controls or affects the transcription of DNA into RNA or the translation of RNA into a protein. Exemplary of such regulatory sequences are promoters, ribosome binding sites, and translation start sites. A "regulatory region" contains regulatory sequences. That the lacZ gene could fulfill such a reporter function in mycoplasma was not predictable in light of the uncertainty that generally surrounded heterologous expression in mycoplasma. Thus, there was little or no information previously available on whether mycoplasma tRNA concentration and availability could accommodate expression of an exogenous gene, on whether stability of foreign mRNA would be sufficient for such expression, or on whether an heterologous expression product, if obtained, would possess adequate stability in the mycoplasma host.

Other uncertainties existed as well. For instance, the conventional stop codon, UGA, is used in many mycoplasmas as a tryptophan-coding codon. Inamine et al . , J. Bacteriol . 172: 504-06 (1990). It could not have been predicted whether this unique codon usage would interfere with the function of a reporter gene. In the context of the present invention, this consideration militates in favor of choosing a reporter gene that, like lacZ , lacks the UGA codon or is tolerant of the alternative UGA or other codon usage.

In view of the success of lacZ as a reporter gene in a mycoplasma, other commonly employed reporter genes might also be used in mycoplasmas. Problems may exist with employing other reporter genes in mycoplasma due, at least in part, to the alternative codon usage of mycoplasmas.

According to the present invention, an appropriate reporter gene with an upstream cloning site is useful for analysis of regulatory regions containing regulatory sequences of mycoplasmas. Use of an appropriate reporter gene can identify regulatory regions and assist in the analysis of defined mutations regulatory sequences. Accordingly, the present invention allows for the further identification and modification of mycoplasma regulatory sequences or regions containing these sequences.

Another aspect of the present invention relates to purified mycoplasma regulatory sequences or regions for use in an expression vector. The term "purified" in the context of this invention refers to a degree of purity greater than that found in nature, preferably a degree of purity that is sufficient for purposes of constructing an expression vector. The regulatory sequences or regions of the invention may be obtained by means such as isolation from natural sources or chemical synthesis. Purified fragments and derivatives of mycoplasma regulatory regions and sequences found in nature are also within the scope of this invention.

A further aspect of the present invention involves employing at least one mycoplasma regulatory sequence or region operably linked to foreign DNA in an expression vector. Because the regulatory sequence or region and foreign DNA are arranged in this manner, the regulatory sequence or region controls the expression of the foreign DNA when the vector is within a host, which preferably is a mycoplasma. Other appropriate hosts may be employed as well. The regulatory region, sequence, or fragments thereof may be combined when used in the expression system of the present invention. Cells transformed with these expression vectors are also within the scope of this invention.

In the context of this invention, the term "foreign DNA" includes any DNA that encodes the desired product. The foreign DNA can be isolated from natural sources, can be generated from RNA via reverse transcription or can be synthesized. Natural sources include any entity that possesses DNA or RNA, including mycoplasmas. For instance, the present invention includes the use of a mycoplasma host, such as Acholeplaεma , transformed with a plasmid containing a mycoplasma regulatory sequence and a DNA sequence (that is, a foreign DNA) from a virus, bacterium, animal or plant. Additionally, the foreign DNA can be from a mycoplasma, even from an Acholeplaεma. Preferably, the foreign DNA encodes a protein. The recombinant production of mycoplasma proteins will allow for the identification of mycoplasma antigens useful for creating vaccines. Additionally, the production of antigenic mycoplasma proteins permits the development antibodies directed against mycoplasma antigens. Such antibodies would have diagnostic and therapeutic uses. The foreign non-mycoplasmal proteins produced by the present invention are also useful, especially in the context of vaccine production and development of antibodies.

The following examples are intended to further describe and discuss the construction and use of present invention. The invention, however, is not limited to the express terms of these examples.

EXAMPLE l. Construction of lacZ transcriptional fusion constructs with the M. capricolum rrnA P2 promoter

The use of E. coli lacZ as a reporter gene in mycoplasmas was evaluated by examining the ability of a known mycoplasma or E. coli promoter to generate β- galactosidase (β-gal) activity from a transcriptional fusion with the trp'-lacZYA operon. Gene fusions have proven to be powerful tools for studying prokaryotic transcriptional and translational control elements. Silhavy & Beckwith, Microbiol . Rev. 49: 398-418 (1985); Silhavy et al . EXPERIMENTS WITH GENE FUSIONS, COLD SPRING HARBOR (1984); Simons, et al . , Gene 53: 85-96 (1987). This fusion employed E. coli translational start sites.

Plasmids pISM2004-2006 and pISM2009-2011 were constructed and transformed into E. coli CSH50 and Acholeplasma ISM1520. See Figure 1. Acholeplaεma strain designations ISM2004-2010 represent recombinants containing plasmid pISM2004-pSM2010, respectively. E. coli Dfl5α and χ289 were obtained from Bethesda Research Labs and R. Curtiss, respectively. Acholeplaεma ISM1499 was a laboratory isolate. Acholeplaεma ISM1520 was constructed by transformation of ISM1499 with the plasmid pISM1026 Tc^r. Plasmid pISM1026 Tc^r is a derivative of cloning vector pSP64 (Promega) , and contains the tetracycline resistance marker from TN916 and an approximately 5kb fragment of Acholeplaεma DNA (for insertion into the chromosome) . Plasmid pDIA15 is disclosed in De Reuse et al . , FEMS Micro . Lett . 37: 193-97 (1986). Plasmid pMC187l is disclosed in Casadaban et al . , Meth . Enzymol . 100:293-308 (1983). Plasmid pISM1003 is disclosed in Mahairas et al . , Gene 93: 61-65 (1990) . All other strains and plasmids were developed in conjunction with the present invention. The construction of the trp^,-lacZYA plasmids are depicted in Figure 2. First, the -35 and -10 regions of the M. capricolum rrnA P2 promoter or the -35 and -10 regions of the E. coli rrnB Pl promoter were cloned into the EcoRI and BatπHI sites of plasmid pISM1003 (Mahairas & Minion, εupra) to generate plasmids pISM2002 or pISM2003, respectively. Plasmids pISM2005 and pISM2006 were constructed by inserting the 7.1 kb BamHI fragment containing trp'-lacZYA from plasmid pDIA15 (De Reuse, et al . , FEMS Micro . Lett . 37: 193-97 (1986)) downstream of the promoters in plasmids pISM2002 and pISM2003. See Figure 2. Plasmids pISM2010 and pISM2009 are identical to plasmids pISM2005 and PISM2006, respectively, except that an additional 700 bp region upstream of the M. capricolum rrnB P2 promoter has been placed upstream of the promoters in the plasmids. Plasmid pISM2011 is identical to plasmid pISM2010 except that the M. capricolum promoter and rrnB P2 upstream sequences are in the reverse orientation. Acholeplaεma cells were grown in PPLO broth (Difco Laboratories, Detroit, Mich.) supplemented with 15% gamma globulin-free horse serum (GIBCO Laboratories, Grand Island, N.Y.), 2.5% yeast extract, 0.5% glucose, 2.5 mg/ml of Cefobid (Pfizer, Inc., New York, N.Y.), 0.02% (wt/vol) DNA (Sigma Chemical Co., St. Louis, Mo.), 0.002% (wt/vol) phenol red, and (when required) 1% Noble agar (Difco) . Selective levels of antibiotics for Acholeplaεma ISM1499 were 2 mg/ml tetracycline (Sigma) and 15 mg/ml gentamicin (Sigma) , and resistant cultures were maintained at 10 mg/ml tetracycline and 15 to 20 mg/ml gentamicin. E. coli cultures were grown in Luria broth containing 100 mg/ml ampicillin (Sigma) or 12.5 mg/ml tetracycline when appropriate.

The transformation of Acholeplaεma was performed as described in Mahairas & Minion, Plasmid 21: 43-47 (1989). X-gal was dissolved in dimethyl sulfoxide, and was spread on the agar surface 1 to 2 hours prior to plating the transformation mixture. Transformation of E. coli was achieved by the method of Hanahan, J. Mol . Biol . 166: 557- 80 (1983) or by electroporation using the BTX model 600 electroporator (Biotechnologies & Experimental Research, Inc., San Diego, Cal.), 1 mm gap, with settings of 0.81 kV, 50 uF, and 129 ohm, to generate a pulse length of 5 milliseconds.

Levels of β-gal activity of the transformants were measured and the results are set forth in Figure 2. β- galactosidase assays were performed as described by Schleif & Wensink, PRACTICAL METHODS IN MOLECULAR BIOLOGY (Springer-Verlag 1981) . First, 5 ml of an Acholeplaεma culture was washed with PBS and the pellet was re-suspended in 2 ml of Z buffer (0.1 M sodium phosphate, pH 7.0; 0.001 M magnesium sulfate; 0.1 M 2-mercaptoethanol) . Fifty microliters of 0.1% SDS was added to l ml of the cell suspension and vortexed for 15 seconds. One hundred microliters of the lysed cell suspension was added to 1 ml of Z buffer and equilibrated to 37°C. Two hundred microliters of a 4 mg/ml o-nitrophenyl-β-D- galactopyranoside (Sigma) solution was added. The reaction was stopped by adding 0.5 ml of 1 M sodium carbonate. The absorbance of the reaction was measured with a Spectronic 20 (Bausch & Lo b) at an absorbance of 420 nm. E. coli cultures were assayed in a similar manner except that 100 μl of a 1 ml culture was used and the cells were made permeable with 0.1% SDS and drops of chloroform. See Miller, EXPERIMENTS IN MOLECULAR BIOLOGY, COLD SPRING HARBOR LABORATORY, Cold Spring Harbor, N.Y. (1972). Fully induced E. coli strain χ289 was used as a positive control. The generation of high levels of β-gal in E. coli using the M. capricolum rrnA P2 promoter showed that the E. coli RNA polymerase recognized M. capricolum promoter sequences. In contrast, the E. coli rrnB Pl promoter was not recognized by the Acholeplaεma RNA polymerase as evidenced by the lack of mRNA produced. Although the -35 and -10 regions of these promoters were similar, the Acholeplaεma RNA polymerase was more stringent, which supports previous studies using in vitro transcription assays. Gafny et al . , Nucl . Acidε Reε . 16: 61-76 (1988).

The M. capricolum promoter was able to drive the expression of lacZ in both E. coli and Acholeplaεma ISM1520, but the overall levels of β-gal activity in Acholeplaεma were low in comparison to levels in E. coli . Slightly higher levels of β-gal activity were generated in Acholeplaεma with the additional 700 bp region from the M. capricolum rrnB Pl promoter region in plasmid pISM2010. This supported an earlier observation that the upstream region of the M. capricolum rrnB P2 promoter influences transcriptional levels. Josaitis et al . , Biochim . Biophys . Acta . 1050: 307-11 (1990). The M. capricolum promoter in the reverse orientation with respect to lacZ (pISM2011) was not able to generate β-gal activity in either Acholeplaεma or E. coli .

The low levels of β-gal activity in Acholeplaεma were observed using the M. capricolum promoter in a trp'-lacZ operon fusion could be accounted for, in part, by the single copy insertion into the chromosome or by poor recognition of the M. capricolum promoter by the Acholeplasma RNA polymerase. A more likely explanation, however, is that the translation initiation region of the E. coli trp gene promoter was not efficiently recognized in Acholeplasma . Slot blot analysis of RNA levels from Acholeplasma ISM1499, ISM1520, ISM2004, ISM2005, ISM2006, ISM2009, and ISM2010 were performed with either 16S rRNA, to standardize the assay, or a Iac2-specific probe. Counts per volume were detected by the PHOSPHORIMAGER (Molecular Dynamics, Sunnyvale, CA) and analyzed with the IMAGEQUANT program (Molecular Dynamics, Sunnyvale, CA) . All blots with the 16S rRNA probe showed hybridization. The values obtained with the lacZ probe in counts per minute after subtracting background and adjusting for amounts of RNA loaded are set forth in parenthesis following the identification of the transformant: ISM1499 (0), ISM1520 (0), ISM2004 (40,652), ISM2005 (543,651), ISM2006 (0), ISM2009 (32,574), ISM2010 (329,117) . The trp'-lacZ mRNA transcript levels in ISM2005 and ISM2010 were relatively high despite the low levels of β- gal being produced. Both of these transformants contain the M. capricolum rrnA P2 promoter. Surprisingly, the E. coli consensus promoter {E. coli rrnB Pl) alone in ISM2006 was not able to drive expression of lacZ in Acholeplasma despite close homology in the -10 and -35 sequences. See Gafny, et al . , supra . Although the E. coli consensus promoter alone failed to produce significant levels, the addition of the 700 bp M. capricolum rrnB Pl upstream sequence (ISM2009) did yield detectable, but low levels of the lacZ transcript. Transcript levels of ISM2004 were detectable and were probably due to read through from adjacent genes, a common problem when using lacZ fusion vectors. Silhavy & Beckwith, supra ; Simons, et al . supra .

EXAMPLE 2. The lacZ protein fusion library construction

Because only low levels of β-gal activity were generated from the transcriptional fusion system described above, a protein lacZ fusion construct was developed. Transcriptional and protein fusions are generally described in Casadaban et al . , Meth . Enzymol . 100: 293-308 (1983). The protein fusion vector pISM2050 was constructed to increase β-gal activity by utilizing both mycoplasma transcriptional and translational sequences cloned upstream of the promoterless lacZ.

The construction of plasmid pISM2050 is illustrated in Figure 3. The parent plasmid, pISM1003, was digested with BamHI and the 3.1 kb BamHI fragment of plasmid pMC1871 (Casadaban et al . , supra) containing a promoterless lacZ gene was ligated into the site. In order to generate a unique site for cloning, the downstream BamK site was removed by partial digestion with BamΑT followed by a fill- in reaction with Klenow fragment of DNA polymerase I, ligation to circularize, and transformation into E. coli . Transformants were selected on ampicillin containing media, and tne proper construct was confirmed by restriction enzyme analysis.

Plasmid pISM2050 was constructed with a gentamicin marker for selection in mycoplasmas, a promoterless lacZ reporter gene with unique upstream BatπHI and S al. restriction enzyme sites for cloning Sau3AI and blunt end fragments, respectively, and an ampicillin marker and origin of replication for amplification of DNA in E. coli prior to transformation into Acholeplasma . It should be noted, however, that when Sau3Al fragments are ligated into BamHI sites, the BamHI site is often lost. The ampicillin gene and origin of replication also provided a region of homology to insert the plasmid into the Acholeplasma recipient strain ISM1520 via homologous recombination. See Mahairas & Minion, J. Bacteriol . 171: 1775-80 (1989); Mahairas et al . , Gene 93: 61-66.

A library of cloned fragments was produced and screened for β-galactosidase activity in E. coli prior to introduction into Acholeplasma because of the large quantity of plasmid DNA (7 to 10 μg) required for transformation into mycoplasmas.

First, chromosomal DNA was prepared by washing the cells from a 500 ml culture with phosphate buffered saline (PBS, 10 mM sodium phosphate, 150 mM NaCl, pH 7.4) and re- suspending the pellet in 5 ml of NET buffer (0.1 M Tris, 0.15 M NaCl, and 0.08M EDTA, pH 7.5) containing 10 mg/ml Proteinase K. After 2 hours of incubation at 37°C, the cells were lysed with 1 ml of a detergent solution containing 1% sodium dodecyl sulfate (SDS) , 1% Nonidet P- 40, and 1% deoxycholate. After phenol extraction, the DNA was purified on a cesium chloride-ethidium bromide buoyant density gradient. DNA bands were collected, ethidium bromide extracted, and salt removed. Sambrook et al . , MOLECULAR CLONING, A LABORATORY MANUAL, COLD SPRING HARBOR LABORATORY, Cold Spring Harbor, N.Y. (1989). The chromosomal DNA was partially digested with Sau3AI and fragments were separated by a sucrose gradient. Sambrook et al . , loc. cit . One to six kilobase fragments were isolated and cloned into pISM2050 using standard recombinant DNA techniques as described below.

Acholeplaεma ISM1499 Sau3Al-digested chromosomal DNA and BamHI-digested vector pISM2050 were ligated together, and the ligation mixtures were used to transform E. coli DH5o. Approximately 20% of the E. coli transformants demonstrated the blue phenotype. ^■ Restriction enzyme analysis of plasmid preparations from both blue and white colonies showed that all of the blue colonies contained plasmids with mycoplasmal chromosomal DNA inserts while only 20-30% of the white colonies had plasmids with inserts. As a result, only plasmids from blue colonies were used for further study. Plasmid DNA from five independent colonies were pooled together and used to transform Acholeplaεma ISM1520. If a mycoplasma transformant from the plasmid pool demonstrated β-gal activity based on blue colony formation on X-gal-containing media, each plasmid DNA from that pool was then used to transform Acholeplasma ISM1520.

Interestingly, only 13 of the 140 (10.8%) recombinants that demonstrated promoter activity in E. coli also showed activity in ISM1520. Possible explanations include: (1) pooled plasmid preparations in the initial transformation of ISM1520 were under-represented or did not successfully transform the promoter-containing constructs; or (2) the A + T rich mycoplasma DNA may have generated pseudo-promoter activity in E. coli . Identification of the event responsible for this result, however, is not needed for the practice of this invention.

Acholeplasma transformants containing lacZ fusion constructs displayed varying blue intensity on X-gal- containing media, β-gal assays were performed with seven of the lacZ fusion constructs that were introduced into ISM1520. The results set forth in Figure 4 show that levels of β-gal production varied by 100-fold between the transformants containing the various lacZ fusion constructs. The varied production levels from cloned fragments indicate that gene regulation is occurring at the promoter and/or translational level in Acholeplaεma. Immunoblot analysis and the intensity of blue color on X- gal containing media confirmed this conclusion.

Levels of the lacZ fusion transcripts were also measured for the ISM1520 strains harboring the lacZ fusion plasmids (Figure 4) . First, total RNA was prepared using RNA STAT-60 isolation reagent (Tel-Test "B", Inc., Friendswood, Tex.) according to manufacturer's instructions from a 4 ml overnight culture. Messenger RNA levels were measured by slot blot analysis using a Minifold II apparatus (Schleicher & Schuell, Keene, N.H.) and following the procedure described in Sambrook, et al . MOLECULAR CLONING, A LABORATORY MANUAL, COLD SPRING HARBOR (1989) . Two micrograms of total RNA were placed into each slot and transferred to nitrocellulose (Schleicher & Schuell) and probed with a 1.4 kb fragment containing the 16S rRNA gene from ISM1499 or a 3.1 kb fragment containing the lacZ gene. The blot was exposed to x-ray film or was examined with a PHOSPHORIMAGER and analyzed with the IMAGEQUANT program. Levels of the lacZ mRNA were adjusted by normalizing to the 16S rRNA levels to account for differences in amounts of RNA that may have been loaded in each slot for each strain. The results in figure 4 correlate with the β-gal assay results in that strains demonstrating higher transcript levels generally had higher levels of β-gal activity.

Protein concentration was measured by using the Bio- Rad protein assay (Bio-Rad, Richmond, Cal.) according to the manufacturer's directions. The number of colony forming units (CFU) per milligram of protein was determined by dividing the number of CFU/ml by mg/ml from the average of three cultures each measured in triplicate. The amount of β-gal enzyme required to give an OD = 1.0 at an absorbance of 420 nm in one minute is 4.45 x 10¹² monomers. Towbin et al . , Proc. Nat 'l Acad. Sci . U.S.A. 76: 4350-54 (1979) . The levels of β-gal activity, therefore, was expressed as the number of monomers per CFU as shown below. Cultures were assayed in triplicate.

Units of β-gal activity = A₄₂₀ x 4.45 x 10¹² monomers time (min) x mg protein x CFU/mg

Immunoblot analysis was performed with several recombinant Acholeplasma strains to demonstrate the production of a β-gal fusion protein. Protein samples containing 25 μg protein from washed mycoplasmas were re- suspended in water and lysed with 0.01% (final concentration) SDS, then SDS-PAGE sample buffer (Laemmli, Nature 227: 680-85 (1970) was added, and the samples were boiled for 5 minutes and then separated on a 7.5% polyacrylamide resolving gel. Following electrophoresis for 4 hours at 25 mAmp constant current, proteins were transferred to nitrocellulose following the procedure of Towbin, et al . , Proc . Nat 'l Acad . Sci . USA 76: 4350-54 (1979). The blots were analyzed using a 1:3,000 dilution of a monoclonal antibody to β-gal (Promega) followed by goat anti-mouse antibody conjugated to alkaline phosphatase (Kirkegaard & Perry Laboratories, Inc., Gaithersburg, Md.) (1:1,000 dilution). The blot was developed by using the BCIP/NBT one component alkaline phosphatase substrate system (Kirkegaard & Perry Laboratories, Inc.). Acholeplaεma ISM1520 strains with high levels of β-gal activity reacted more strongly with the anti-β-gal monoclonal antibody, which supports the results of the β- gal assay.

EXAMPLE 3 DNA sequencing of promoter-containing DNA fragments The chromosomal DNA inserts adjacent to the lacZ gene in the pISM2050 derivatives were sequenced by using an oligonucleotide sequencing primer. This primer (5^/-GCTGGCGAAAGGGGGATG GCTGCAAGGCG-3^,) is reverse and complimentary to the lacZ gene at approximately 50 nucleotides downstream of the lacZ-chromosomal DNA fusion point. Other sequencing primers, such as the T7, T3 and Mi3 forward and reverse primers, were also employed in the sequencing of chromosomal inserts. The SEQUENASE kit (United States Biochemical) is suitable for this sequencing.

The chromosomal inserts were sequenced by first subcloning the inserts into pSP71 (Promega Corp.) or isolating a Clal fragment containing the insert, the junction site and a portion of lacZ from an agarose gel and purification with GeneClean. The fragments were removed from pISM2050 for sequence analysis because pISM2050 was derived from pKS (Stratagene) , which also contains the portion of lacZ that is recognized by the sequencing primer.

EXAMPLE 4 Mapping of Acholeplasma ISM1 99 promoters by primer extension

The transcriptional start sites for the promoters driving the expression of lacZ in the fusion plasmids pISM2050.1, pISM2050.2, pISM2050.8, pISM2050.18, pISM2050.25, pISM2050.69, and pISM2050.70 were mapped by primer extension method of Ausebel et al . , CURRENT PROTOCOLS IN MOLECULAR BIOLOGY (John Wiley & Sons 1989) . See Figure 5. Three microliters of the primer extension product were electrophoresed on an 8% sequencing gel. The primer extension products were compared to a sequencing ladder generated by using pISM2083 with the sequencing primer. The difference in the distance from the lacZ- chromosomal DNA fusion point to the base corresponding to the primer extension product was then mapped on a previously determined sequence map for each plasmid.

The transcriptional start sites and upstream regions of the lacZ fusion transcripts for seven of the ISM2050 derivatives are shown in Figure 5. The upstream regions were aligned to the putative sequences encompassing the -10 and -35 promoter regions. The regions were defined based on their similarity to the E. coli consensus promoter. Hawley and McClure, Nucl . Acidε Reε . 11: 2237-55 (1983). The spatial relationships between the transcriptional start and the -10 region as well as between the -10 region and the -35 region were similar to the E. coli consensus promoter. The sequences upstream of the lacZ gene in pISM2050.1, pISM2050.2, pISM2050.8, pISM2050.18, pISM2050.25, pISM2050.69 and pISM2050.70 are set forth in Figures 6-12, respectively. The potential ribosomal binding sites for each sequence were determined by aligning the sequence with the last 15 bp at the 3' end of the 16S rRNA gene of A. laidlawii .

A consensus promoter sequence for the seven pISM2050 fusion derivatives in the -10 and -35 regions were TATaW and TtcAtn, respectively. Upper case letters denote that at least 5 of the 7 pISM2050 derivatives contain the base. Lower case letters denote that at least 3 of the 7 pISM2050 derivatives contain the base in the designated position. "W" denotes A or T, and "n" denotes any base. See Figure 5. The average number of bases that align between the ISM1499 promoter regions and the E. coli consensus promoter regions are 4.4 bp for the -10 region and 3 bp for the -35 region. The 3 bp conservation at the -35 region is about 1 bp less than the 3.9 bp observed with the E. coli and phage promoters. See Figure 5. The observation that the - 10 region was more like the E. coli consensus promoter than the -35 region is consistent with previous reports examining the putative mycoplasma promoter regions. Christiansen, Microbiol . Sci . 4: 168-72, 292-95 (1987); MutO et al . , MOLECULAR BIOLOGY AND PATHOGENESIS at 331-48 (Maniloff et al . Eds., Amer. Soc. Micro. 1992); Inamine, et al . , Gene 73: 175-83 (1988).

The result that the Acholeplaεma promoters were similar to the E. coli consensus promoter is not unexpected because the lacZ fusion constructs were initially screened in E. coli based on their ability to give the Lac⁺ phenotype. The primer extension studies showed that the Acholeplaεma ISM1499 chromosomal sequences adjacent to lacZ in the plasmid derivatives of pISM2050 contained the promoters that were driving the expression of lacZ. The sequences upstream of the transcriptional start sites were aligned (Figure 5) . Defining a mycoplasma promoter by its similarity to the E. coli consensus promoter could potentially be misleading, however, because mycoplasma DNA, like E. coli promoters, are A + T rich. Therefore, in order to precisely identify a mycoplasma promoter in a mycoplasma regulatory region, it is important to correlate sequence data with promoter mapping studies. Exact identification of the mycoplasma promoter, however, is not needed for the practice of this invention.

Figures 6-12 set forth sequence data from several mycoplasma regulatory regions, which contain regulatory sequences. These regulatory regions, or fragments of these regions, can be used to construct an expression vector suitable for expressing foreign DNA sequences (typically genes) in mycoplasmas, such as Acholeplaεma . An expression vector within this invention could include the entirety of one of the mycoplasma regulatory regions identified in Figures 6-12, or one or more fragments thereof. Additionally, the expression vector of the present invention can include more than one mycoplasma regulatory region or sequence. Moreover, the expression system can include combinations of mycoplasma regions, sequences or fragments thereof.

EXAMPLE 5. Construction of a Mycoplasma Expression Vector

The expression vector pISM303 was constructed as follows. Primers PK101910 - 5' GACGGGATCCTTAAATACTAA 3' and PK101912 - 5' CGTAAGCTTCCTCCAACAACAAAAACCTTGA 3' were constructed to obtain the mycoplasma regulatory region contained in pISM2050.2. Primer PK101910 creates a BaroHI site and primer PK101912 creates a Hindlll site, both of which are underlined in the above sequences. in a polymerase chain reaction with plasmid pISM2050.2 (Figure 7) , the primers yielded a 260 base pair fragment that contained the putative -35 and -10 transcriptional start sites, a ribosomal binding site, an ATG translational start site and 24 codons of a mycoplasma structural gene. During the PCR reaction the upstream BatπHI site that had been lost during pISM2050.2 construction was recreated for cloning purposes. The 260 bp fragment was digested with Hindlll and BamHI and cloned into the general cloning vector pKS II(-) (Stratagene) to create plasmid pISM303. Cloning vector pKSII(-) is appropriate because it includes a multiple cloning site and an origin of replication. Other cloning vectors, however, are suitable for use in the present invention.

To test pISM303, a promoterless tetΛ gene was constructed by PCR and inserted into the unique BamHI site of pISM303. Plasmids with the proper tetM orientation confirmed by restriction analysis were transformed into E. coli or Acholeplaεma ISM1503. Acholeplasma ISM1503 was obtained by transforming strain ISM1499 with pISM1007. This integrative plasmid is a pKS derivative that encodes the gentamicin resistance marker from TN4O02. See Mahairas & Minion, J. Bacteriol . 171:1775-80 (1989). Other bacterial and mycoplasmal recipient strains are suitable for use with the present invention as well. Plasmid pISM303 integrates into the genome of Acholeplasma ISM1503 by homologous recombination between the common ampicillin resistance marker sequences. Other regions can be employed as well, as long as the expression vector and the recipient strain possess a homologous region.

E. coli and Acholeplasma transformants were identified by tetracycline resistance arising from expression of tetM controlled by the mycoplasmal regulatory sequence in pISM303, thereby further demonstrating the usefulness of the present invention.

It should be understood that the description, examples and figures set forth above are given by way of illustration and explanation only, because various changes and modifications within the spirit and scope of the invention will become apparent to those skilled in the art in view of the description, examples and figures.

Claims

What is claimed is

1. An expression vector comprising at least one mycoplasma regulatory sequence operably linked to a foreign DNA to control the expression of foreign DNA in a host cell.

2. An expression vector according to claim 1, wherein said host cell is a member of the class Mollicuteε .

3. An expression vector according to claim 2 wherein said host cell is a member of the genus Acholeplaεma .

4. An expression vector according to claim 1, further comprising a site for inserting foreign DNA.

5. An expression vector according to claim 4, wherein said mycoplasma regulatory sequence comprises an Acholeplaεma promoter.

6. An expression vector according to claim 1, wherein said mycoplasma regulatory sequence comprises a sequence found upstream of an Acholeplaεma promoter.

7. An expression vector comprising a mycoplasma regulatory region, wherein said mycoplasma regulatory region comprises an Acholeplaεma promoter and a sequence found upstream of an Acholeplaεma promoter.

8. An expression vector according to claim 1 wherein said mycoplasma regulatory sequence can be obtained from a plasmid selected from the group consisting of pISM2050.1, pISM2050.2, pISM2050.8, pISM2050.18, pISM2050.25, PISM2050.69 and pISM2050.70.

9. A transformed bacterial cell comprising an expression vector according to claim 1.

10. A transformed bacterial cell according to claim 12, wherein said cell is a strain of Acholeplasma.

11. A purified mycoplasma regulatory sequence.

12. A purified mycoplasma regulatory sequence according to claim 11, wherein said mycoplasma regulatory sequence can be obtained from a plasmid selected from the group consisting of pISM2050.1, pISM2050.2, pISM2050.8, pISM2050.18, pISM2050.25, pISM2050.69 and pISM2050.70.

13. A system for identifying mycoplasma regulatory sequences comprising a gene fusion construct, said construct comprising a reporter gene.

14. A system according to claim 13 wherein said reporter gene is lacZ.

15. A system according to claim 13, wherein said gene fusion construct is a protein fusion construct.

16. A method for identifying mycoplasma regulatory sequences, comprising the steps of: obtaining a gene fusion construct comprising a reporter gene, inserting mycoplasma DNA into said protein fusion construct, and determining whether the reporter gene is expressed in a mycoplasma host.

17. A method for identifying mycoplasma regulatory sequences according to claim 16, wherein said reporter gene is lacZ.

18. A method for identifying mycoplasma regulatory sequences according to claim 16, wherein said gene fusion construct is a protein fusion construct.