Introduction

Recombinant protein expression in Escherichia coli is a commonly used technique to produce a wide array of functional proteins from a variety of organisms. The pLac system (1,2) is a popular expression vector for the production of nontoxic proteins. While this isopropyl--D-thiogalactopyranoside (IPTG)-inducible system is capable of high-level protein expression, when toxic proteins are cloned into vectors of this class, low-level, untimely expression may stimulate a cascade of deleterious events ending in mutations that may affect target protein function, overproduction of target-directed proteases, or cell death. Therefore, tight expression control prior to target protein induction is critical for reproducible and functional expression of host-toxic proteins.

Several strategies have been described for reducing premature protein production. Some involve culture conditions: lower growth temperatures before and possibly during induction decrease the probability of a toxic event, and we have also observed that clones grown in stationary phase have an increased frequency of mutations in the toxic protein (data not shown). In addition, much attention has been focused on the development of tightly regulatable expression vectors. The pBAD (3) and T7 polymerase (4,5,6,7) systems have been popular vectors for the expression of toxic proteins. While the pBAD system, which relies on catabolite repression and positive induction, is tightly regulated, it produces a relatively small amount of recombinant protein. On the other hand, while T7 polymerase systems facilitate a higher level of expression than pBAD, they are prone to leaky expression. In the T7 system, recombinant protein expression is driven by promoter sequences recognized by T7 polymerase. The T7 polymerase gene is generally located on the bacterial chromosome under the control of a lactose-inducible bacterial promoter. Because these promoters are leaky in the absence of inducer, the transformation of T7-inducible plasmids bearing toxic protein sequences may result in selective pressure from immediate and untimely expression of the toxic protein.

We set out to design a protein expression system that combines tight regulation with high-level induction. To this end, we constructed an expression system based on the E. coli rhaTRS locus ((Figure 1)A). In the presence of L-rhamnose, RhaR activates transcription of rhaR and rhaS, resulting in an accumulation of RhaS. RhaS then acts as the L-rhamnose-dependent positive regulator of the rhaT promoter (8). The mechanism of expression and repression of the rhaT promoter has been studied extensively (8,9,10,11). This L-rhamnose-inducible promoter is also subject to catabolite repression (12), so although this promoter is capable of high-level recombinant protein expression in the presence of L-rhamnose, it is also tightly regulated in the absence of L-rhamnose by the addition of D-glucose (13,14,15,16,17,18). Given these properties, we elected to transfer the chromosomal sequence corresponding to the rhaT promoter and the rhaR and rhaS genes (for simplicity, referred to here as pRHA) to three different plasmid backgrounds and examined the suitability of this system for regulatable protein expression. The results presented here suggest that pRHA vectors represent a viable alternative E. coli expression system for the production of nontoxic proteins and an enabling technology for the functional production of otherwise toxic proteins.

Figure 1.

Materials and Methods

Bacterial Strains

The E. coli strains used in this study were MG1655 (19), TOP10 [F-mcrA ?(mrr-hsdRMS-mcrBC) F80lacZ?M15 ?lacX74 recA1 araD139 ?(araleu)7697 galU galK rpsL (Str^R) endA1 nupG], and BL21(DE3) [F-ompT hsdS_B (r_B-m_B-) gal dcm (DE3)] (all from Invitrogen, Carlsbad, CA, USA).

Construction of pRHA-67, pRHA-113, pRHA-109, and pMPX-66

A portion of the rhamnose regulon containing the rhaT promoter and rhaR and rhaS regulatory genes (pRHA) was PCR-amplified from the MG1655 chromosome ((Figure 1)A). Three separate PCR amplifications of this region were performed. The reverse primer used in each reaction was the same and was tagged with HindIII (see (Table 1) for all primer sequences). In each case, the forward primer was tagged with a restriction enzyme recognition site, a multicloning sequence, and an optimized Shine-Dalgarno sequence. However, the forward primer used in each reaction differed in the restriction enzyme site. For cloning into pUC18, KpnI was used; for pBR322 lacking the rop gene, NdeI was employed; and for pBR322 containing the rop gene, the recognition site used was XhoI. All three pRHA PCR products (appropriately sized at 2074 bp) were gel-purified and blunt-end ligated into pCR-Blunt II-TOPO® (Invitrogen).