Gene, 23 (1983) 131-136

Elsevier

131

Gene 00787

Ml3 bacteriophage and pUC plasmids containing DNA inserts but still capable of fl-galactosidase ol-complementation

(Insertional inactivation; blue plaques; chloramphenicol acetyltransferase)

T.J. Close, J.L. Christmann and R.L. Rodriguez

Department of Genetics, University of California, Davis, CA 95616 (U.S.A.) Tel. (916) 752 - 3263

(Received January 3 1 st, 1983)

(Accepted February 28th, 1983)

SUMMARY

A DNA fragment encoding the transposon Tn9 chloramphenicol acetyltransferase gene (cut) was

inserted into Ml3 phage and pUC plasmid cloning vehicles. When the cat gene was inserted in the same

orientation as the lacZ gene, two new polypeptides were produced. One polypeptide possessed chlo-

ramphenicol acetyltransferase activity, while the other expressed /?-galactosidase a-donor activity. Both new

polypeptides were translated from a hybrid messenger RNA initiating from the Zac promoter. These

observations may help explain why not all inserts produce white plaques.

INTRODUCTION

Detection of recombinant DNA molecules can

be achieved using cloning systems that involve the

insertional inactivation of genetic markers en-

coded on the cloning vector. Messing and co-

workers have developed two such systems that

employ the Escherichia coli lad gene product,

P-gal, and a chromogenic compound (X-gal) that

turns blue when cleaved by P-gal. One system

utilizes the male-specific ssDNA bacteriophage

Ml3 (Messing et al., 1981), while the other is

Abbreviations: &al, P-galactosidase; bp, base pairs; CAT,

chloramphenicol acetyltransferase; glc, glucose; IFTG, isopro-

pyl-/3-D-galactoside; ssDNA, single-stranded DNA; X-gal, 5-

bromo-4-chloro-3-indolyl-/3-D-galactopyranoside; [ ] indicates

plasmid carrier state.

0378-1119/83/$03.00 0 1983 Elsevier Science Publishers B.V.

based on the pUC derivatives of the plasmid

pBR322 (Vieira and Messing, 1982). With the Ml3

system, recombinant bacteriophage can be dis-

tinguished from nonrecombinants by plaque color;

colorless (white) plaques indicate recombinant

phage, whereas blue plaques are indicative of non-

recombinant phage. With the pUC plasmids, re-

combinant and nonrecombinant plasmid mole-

cules can be identified as white and blue colonies,

respectively.

The formation of blue plaques (or colonies) by

cells containing nonrecombinant DNA molecules

is the result of cY-complementation of P-gal activ-

ity. The E. coli /?-gal protein is composed of 1021

amino acids, and is active as a tetramer. Certain

deletions in the IucZ gene near the promoter-prox-

imal end produce an inactive enzyme that can be

132

complemented by nonoverlapping deletions in the promoter-distal region. The latter class of dele- tions produces an a~n~te~n~ peptide called the a-donor (Langley et al., 1975), while the pro- moter-proximal deletions produce the ar-acceptor. Alpha complementation is achieved when the (Y- donor and a-acceptor aggregate to restore P-gal activity.

In the Ml3 and pUC cloning systems, the a- acceptor and lucl genes are encoded by the host. The Ml3 and pUC vectors use a segment of DNA that encodes only an a-donor polypeptide (Welply et al., 1981) and the adjacent lac control region. These vectors carry a synthetic oligonucl~tide within the a-donor gene that contains multiple restriction sites. Cloning DNA fragments into any of these restriction sites should abolish a-donor production, and thus prevent the blue color from developing.

A derivative of the chloramphenicol-resistance gene (cat) from Tn 9 has been recently cloned as a Sal1 restriction fragment (Close and Rodriguez, 1982). This 780-bp fragment, called the CAT cartridge, carries the ~bosome-binding site for the cat gene but not its promoter. In the course of isolating both strands of the CAT cartridge using an Ml3 vector, we discovered that only the anti- sense orientation of the CAT cartridge was repre- sented among the colorless plaques. Cloning the CAT cartridge in the sense o~entation resulted in the production of blue plaques. We present evi- dence which demonstrates that the blue color pro- duction is due to the synthesis of a new a-donor, whose first few amino acids are encoded by the 3’ end of the CAT cartridge, and the remainder from the original a-donor gene of the vector. These results may help to explain similar observations made by other investigators (Dixon et al., 1982; T. Fanning, pers. comm.), and suggest that inactiva- tion of ~-complementation is not completely effec- tive in detecting all recombinant DNA molecules.

MATERIALS AND METHODS

The E. co/i strain JMlOl (]A& pro/Ff]lacZ iacZA’1-4i pro) was obtained from J. Messing. An F- derivative of JMlOl was isolated by subcultur- ing the F+ strain three times in nutrient medium,

then checking the phenotype of individual colonies until a Pro- clone was found. M9 minimal medium with supplements and LB have been described (Miller, 1972). When called for, ampicillin (20 pg/ml) and chloramphenicol (30 p&;/ml) were in- cluded in the media. CAT assays were performed by the spectrophotometric method of Shaw (1975). &gal assays and in vitro ff-~omplementation as- says were performed as described by Miller (1972). Protein was assayed by the method of Bradford (1976) using bovine serum albumin as a standard. Transformation with plasmid DNA and transfec- tion with Ml3 RF DNA using CaCl,-treated E. co& cells was performed as described previously (Close and Rodriguez, 1982). Ml3 phage infec- tions were performed as described by Messing et

al. (1981). Restriction enzyme digestion patterns were

analyzed electrophoretically as described by Boli- var et al. (1977). T4 DNA ligase was prepared and ligations were performed according to the methods of Tait et al. (1980). Crude cell proteins were analyzed using the system of Laemmli (1971). Pro- teins analyzed immunolo~cally ~C~istmann and Dahmus, 1981) were challenged with rabbit an- tibodies to purified CAT (provided by D. Gold- farb), and reacted with ‘251-labeled goat antibod- ies to rabbit IgG (provided by M. Gitt). Radioac- tively labeled protein bands were detected by auto-

radiography.

RESULTS AND DISCUSSION

(a) Cloning both orientations of the CAT cartridge

in Ml3

SalI-digested pCM1 plasmid DNA (Close and Rodriguez, 1982) and M13mp73 RF DNA (Mes- sing et al., 1981) were mixed, ligated, and the product used to transfect E. coli JMlOl cells. White plaques were obtained at a frequency of about 5%. Of the 40 white plaques examined, 38 contained an intact CAT cartridge. To determine the orientation of the CAT cartridge, the DNAs from these recombinants were digested with the restriction endonucleases Bali and BglI. The asymmetric B&I site in the CAT cartridge is shown

133

Sal I, Bal I /Sal I

lac I’ [PO12 I CAT I I lac 2’

I- CAT cartridge l -I

I I >

ATG - I ATG-2

A ATGf(3

B Fig. 1. Open translational reading frames when the CAT cartridge and lacZ gene are in the same orientation. The two reading frames

are predicted from the nucieotide sequence (Messing et al., 1981; Close and Rodriguez, 1982). Open reading frame A (ATG-1)

includes the amino terminus of the IucZ gene encoded by the M13mp73 vector (solid bar), the 33 bp between the vector/cartridge

junction and the CAT initiation codon (ATG-2), and the entire cur gene sequence (solid line). Open reading frame B begins just prior

to the distal cartridge/vector junction (ATG-3) and continues through the remaining IacZ a-donor sequence (solid bar).

in Fig. 1. All 38 recombinants gave the same Bg/I + Ball double-digestion pattern with frag- ments of about 1430 and 670 bp. The opposite orientation was not found. This indicated that recombinants containing the CAT cartridge in the reverse or anti-sense orientation with respect to the facZ gene produced no functional a-donor peptide. One of these recombinants, pCM1001, was chosen for further study.

Examining the known nucleotide sequences of the phage vector (Messing et al., 1981) and the CAT cartridge (Alton and Vapnek, 1979; Close and Rodriguez, 1982) suggested that the forward or sense orientation of the CAT cartridge might be represented among the blue plaques. According to the nucleotide sequence, the promoter-proximal end of the a-region would be fused to the CAT polypeptide-coding sequence in the same transla- tional reading frame (polypeptide, A, Fig. 1). The remainder of the a-region would be fused to a sequence that encodes several amino acids at the distal end of the CAT cartridge, again in the same translational reading frame (polypeptide B, Fig. 1). If either of the two possible fusion polypeptides could function as an a-donor, then cloning the CAT cartridge in the forward orientation would result in a-complementation and the production of blue plaques.

A total of 24 blue plaques were analyzed for ssDNA molecules that were larger than the Ml3 vector. One such phage derivative, pCM1002, was

identified and characterized further by restriction analysis of its RF DNA and hybridization of the ssDNA to pCMlOO1 ssDNA. The intact SalI cartridge was present in pCM1002, and the ex- pected fragments of about 1680 and 430 bp were produced by B&I + Bgll double digestion. Fur- thermore, pCMlOO1 and pCM1002 ssDNA’s pro- duced a partially double-stranded hybridization complex with a slower mobility in agarose gel electrophoresis than either ssDNA alone (see Messing et al., 1981). These results indicate that the CAT cartridge in pCMl~2 is in the opposite orientation relative to pCMlOO1 (Table I). An a-donor activity was therefore produced by insert- ing the CAT cartridge in the forward orientation, as suggested by the nucleotide sequence (poly- peptide B, Fig. 1).

TABLE I

The color of plaques produced by Ml3 derivatives

Phage Orientation of CAT

cartridge insertion a

Plaque color

M13mp73 no CAT cartridge

pCMlOOl backward

pCM1002 forward

blue

white

blue

* In the forward orientation the sense strand of the CAT

cartridge is transcribed from the fat promoter, and in the

backward orientation the anti-sense strand is transcribed (see

Fig. 1 for an illustration of the forward orientation).

134

(b) Regulation of CAT and a-donor activities

For pCM1002, it was reasoned that CAT activ-

ity may be under the control of the luc promoter,

since one of the predicted proteins was a fusion

between the amino terminus of P-gal and the

entire CAT polypeptide (polypeptide A, Fig. 1).

Fig. 2. CAT polypeptides encoded by various plasmids. Lane

A, pBR328; lane B, pCM1001; lane C, pCMlOO2; lane D,

pCMlW4. Total protein extracts were resolved by molecular

weight on 12% polyacrylamide gels and anti-CAT antibody was

bound to proteins transferred to diaxotized paper as described

in MATERIALS AND METHODS.

The level of CAT activity was measured in cells

containing recombinant and nonrecombinant

phage and grown in the presence and absence of

the gratuitous P-gal inducer, IF’TG. WIG induc-

tion of cells containing pCM1002 increased the

level of CAT approx. lOOO-fold (data not shown).

In the opposite orientation (pCM1001) CAT activ-

ity was not detected under any conditions. These

results indicate that CAT expression is under the

control of the lac promoter when the promoter-less

CAT cartridge is inserted in the forward orienta-

tion within the a-donor region (pCM1002). CAT

fusion protein (polypeptide A) was examined im-

munologically, as shown in Fig. 2. The M, of

native CAT is 25 700 (Shaw et al., 1979). Based on

this M, and the additional size contributed by the

fusion, plasmid pBR328 should produce a native

CAT protein of M, 25 700, while pCM1002 would

be expected to produce an M, 27 800 CAT protein.

Electrophoretic analysis of pCM1002 cell extracts

revealed the presence of a prominent band corre-

sponding to an M, 27 800 protein and a lesser

amount of M, 25 700 protein. Native CAT synthe-

sis could be explained by the presence of the CAT

ribosome binding site within the first 30 bp of the

T4 DNA ligase T4 DNA ligase

Fig. 3. The construction of pUC derived plasmids containing

cat and lac genes. The plasmid pUC4 was used in both con-

structions. Plasmid pCM1004 was constructed by adding a Sal1

CAT cartridge. Plasmid pCM1003 was constructed through a

HincII fragment exchange with pPV501. pPV501 contains a

deletion in the tetracycline-resistance gene promoter resulting

in an Ap’Cm’ phenotype (Close and Rodriguez, 1982). The

plasmids pCM1003 and pCM1004 conferred ampicillin and

chloramphenicol resistance.

135

TABLE II

The effect of glucose and WIG on CAT and P-gal activities from pUC derivatives

All activities are in nmol/min/mg protein and are for extracts of JMIOl[F+] cells, which carry the a-acceptor gene, grown in LB

medium at 37’C to an absorbance at 420 nm of 1.5.

o-Donor

plasmid

puc4

pBR328

pCM1003

pCM1004

CAT

+ glc

-1I’TG

_

3100

16

166

+IF”TG

3100

1300

5000

-I&

-1PTG

19200

5 900

19200

p-gal activity

+glc - glc

+Il’TG -1I’TG +IFTG - IPTG +IFTG

i 1.0 8 138 835

16800 _ _

17800 _

26 700 < 1.0 8 179 1000

Symbols: minus (-) indicates none detected; -X 1.0 indicates that some activity was detected at a level too low for accurate

determination; glc, 0.2%; WIG, 1 mM.

CAT cartridge. The relative amounts of native and

fusion CAT from pCM1002 suggest that the ribo-

some binding site for the fucZ gene is more effec-

tive at initiating translation from the hybrid

mRNA than the ribosome-binding site of the cut

gene.

To rule out the possibility that the CAT fusion

polypeptide has a-donor activity, a plasmid,

pCM1003, specifying only the first (polypeptide

A) of the two possible polypeptides was con-

structed (Fig. 3). CAT and P-gal activities were

assayed under repressed and induced conditions

and compared to activities from cells containing a

plasmid, pCM1004, which encodes both poly-

peptides A and B (Table II, Fig. 2). P-gal activity

was detected only when polypeptide B was en-

coded by the resident plasmid. We conclude from

these results that only polypeptide A has CAT

activity, while polypeptide B has P-gal activity. As

with pCM1002, the level of CAT activity from

pCM1003 was increased several hundred-fold by

either IPTG or catabolite induction.

It was not possible to visualize the a-donor

proteins using antibody to P-gal (generously pro-

vided by A. Fowler) since the amino-terminus of

P-gal lacks antigenicity (Welply et al., 1981). In-

stead, the a-donor was examined using in vitro

cu-complementation to determine whether the new

a-donor from pCM1004 was under Iuc control.

Cell extracts containing a-acceptor from fully in-

duced JMlOl cells were mixed with extracts from

repressed and induced cells containing pCM1004.

As shown in Table III, both the magnitude of

IPTG or catabolite induction and the absolute

levels of P-gal activity are similar for cells contain-

ing either pCM1004 or the control, pUC4. From

these data it is clear that a-donor activity pro-

duced as polypeptide B from pCM1004 is also

under luc control.

In summary, inserting the CAT cartridge into

either Ml3 or pUC vectors does not affect the

regulation of a-donor synthesis. This indicates that

there is no transcriptional terminator in the CAT

cartridge, but there is a functional E. coli ribo-

some-binding site near its distal end. The two

fusion polypeptides, one with CAT activity and

the other with a-donor activity, are probably

translated from a polycistronic messenger RNA

beginning at the luc promoter.

Other situations may also give rise to a func-

TABLE III

IPTG induction of a-donor synthesis

a-Donor /I-gal activity in vitro a

-WI-G +II’TG

puc4 Cl.0 161

pCM1004 < 1.0 169

a For explanation of symbols and growth conditions see Table

II. a-acceptor extracts were obtained from JMlOl (F+) cells

grown with IETG and without glucose. a-Donor extracts were

from either F+ cells grown without IFTG and without glucose,

or F- cells grown with IETG and without glucose.

136

tional a-donor when DNA is inserted into the Ml3 or pUC cloning site. For example, the DNA fragment could be inserted in frame with the IacZ sequence and carry no translational stop signals. Alternatively, the inserted fragment could code for its own promoter, ~bosome-binding site and poly- peptide amino-terminus, again in phase with the remaining lad gene sequence. In this case a-donor activity would not be under iac control, but in- stead under the control of the inserted promoter. Since the first several amino acids of the a-donor are not essential for cy-complementation, it is not surprising that DNA fragments inserted into the Ml 3 or pUC plasmids can produce a functional a-donor. The observations made using the CAT cartridge may help to explain some difficulties that arise in cloning both orientations of certain DNA fragments using p-gal inactivation as a phenotype screen for recombinants.

ACKNOWLEDGEMENTS

We thank Crystal DiModica for her help in preparing this manuscript and R.C. Tait for his helpful comments and discussion. This research was supported by a grant to R.L.R. from the National Institutes of Health (GM291002). T.J.C. was supported by a predcctoral traineeship from the National Institutes of Health (5T32-GM- 07467). Lastly we would like to thank Tom Fan- ning and Gisela Heidecker-Fanning for helping us work out the Ml3 cloning system in our labora- tory.

REFERENCES

Alton, N.K. and Vapnek, D.: Nucleotide sequence analysis of

the chloramphenicol resistance transposon Tn9. Nature 282

(1979) 864-869.

Bolivar, F., Rodriguez, R.L., Betlach, M.C. and Boyer, H.W.:

Construction and characterization of new doning vehicles,

I. Ampicillin-resistance derivatives of the plasmid pMB9.

Gene 2 (1977) 75-93.

Bradford, M.M.: A rapid and sensitive method for the quanti-

tation of microgram quantities of protein utilizing the prin-

ciple of protein-dye binding. Anal. B&hem. 72 (1976)

248-254.

Christmann, J.L. and Dahmus, M.E.: Monoclonal antibody

specific for calf thymus RNA polymerases II, and II,. J.

Biol. Chem. 256 (1981) 11798-11803.

Close, T.J. and Rodriguez, R.L.: Construction and characteri-

zation of the chloramphenicol resistance gene cartridge: A

new approach to the transcriptional mapping of extra-

chromosomal elements. Gene 20 (1982) 305-3 16.

Dixon, K., Barnett, S.W., Lau, C.K. and Stacks, P.C.: Cloning

SV40 Hind111 restriction fragments into Ml3 bacteriophage.

Gene 18 (1982) 97-100.

Laemmli, U.K.: Cleavage of structural proteins during the

assembly of the head of bacteriophage T4. Nature 227

(1970) 680-685.

Langley, K.E., Villarejo, M.R., Fowler, A.V., Zamenhof, P.J.

and Zabin, 1.: Molecular basis of &galactosidase ol-comple-

mentation. Proc. Natl. Acad. Sci. USA 72 (1975) 1254- 1257.

Messing, J., Crea, R. and Seeburg, P.H.: A system for shotgun

DNA sequencing. Nuci. Acids. Res. 9 (1981) 309-321,

Miller, J.G.: Experiments in Molecular Genetics. Cold Spring

Harbor Laboratory, Cold Spring Harbor, NY, 1972.

Shaw. W.V.: Chloramphenicol acetyltransferase from chlor-

amphenicol resistant bacteria, in Colowick, S.P. and, Kaplan,

N.O. (Eds.), Methods in Enzymology, Academic Press, New

York, 1975, pp. 737-755.

Shaw, W.V., Packman, L.C., Burleigh, B.D., Dell, A., Morris,

H.F. and Hartley, B.S.: Primary structure of a chlo-

ramphe~~l acetyltransferase specified by R plasmids. Na-

ture 282 ( 1979) 870-872.

Tait, R.C., Rodriguez, R.L. and West, R.W.: The rapid purifi-

cation of T4 DNA ligase from a T4lig lysogen. J. Biol.

Chem. 255 (1980) 813-815.

Vieira, J. and Messing, J. The pUC plasmids, and Ml3mp%de-

rived system for insertion mutagenesis and sequencing with

synthetic universal primers. Gene 19 (1982) 259-268.

Welpley, J.K., Fowler, A.V. and Zabin, I.: &-galactosidase

ol-complementation: overlapping sequences. J. Biol. Chem.

256 (1981) 6804-6810.

Communicated by A.D. Riggs.