An Advanced Monitoring Platform for Rational Design of Recombinant Processes - Measurement, Monitoring, Modelling, and Control of Bioprocesses - page 79

Table 1 Comparison of production processes with E. coli HMS174(DE3)(pET11aSOD): con-
ventional full induction versus limited induction with IPTG or lactose
Standard
process

Limited induction with
IPTG

Limited induction with
lactose

Spec. SOD [mg/g CDM] 64 179 88.1

Volumetric yield [g/L] 0.92 2.87 2.02

Increase of volumetric
yield

1 3.12 2.20

coefficient (Y X/S ), and consequently reduction of Y X/S results in an increasing
inducer/CDM ratio.
The ppGpp level increases to even higher values than in the standard process,
indicating high stress levels, and there is still an increase in PCN after induction.
This effect is attenuated compared with the standard process, but as the lacI repressor
gene is encoded on the plasmid, a variation in the number of repressor molecules
must be expected, which interferes with the concept of transcription tuning.

4.2 Stabilized Plasmid Copy Number

Increasing the plasmid copy number imposes additional stress on the host cell. The
plasmid used in our experiments, pET11a, is a ColE1-derived plasmid (with a ColE1
origin of replication). Replication control of this plasmid type based on two RNA
molecules (RNAII, the primer for replication, and RNAI, an antisense RNA) is well
investigated and described in detail [ 59 - 62 ]. Under starvation conditions, uncharged
transfer RNAs (tRNAs) can interact with these regulatory molecules, thereby
interfering with plasmid replication control [ 63 , 64 ]. The shortage of metabolic
resources and the depletion of amino acids after induction lead to increased levels of
uncharged tRNAs and generate starvation-like conditions. Therefore, we conclude
that the drastic increase in plasmid copy number is driven by interactions of
uncharged tRNAs with RNAI and RNAII, respectively. To reduce interference of
the tRNA with the replication regulatory system, homologies between tRNAs and
RNAI were decreased by changing the RNAI sequence into its complementary
sequence without inverting it [ 65 ]. The experiments with E. coli HMS174(DE3)
harboring the newly designed pEARL plasmid clearly show that replication control
can be maintained even under conditions of full induction (Fig. 6 ).
The product formation rate in this experiment is similar to that of the reference
experiment with the wild-type plasmid (Fig. 5 ), and consequently the influence of
the stabilized plasmid copy number on cell growth and stress level after induction
is not significant. This means that a copy number of approximately 40 plasmids per
cell is already too high and cellular capacities are exceeded even with the pEARL
plasmid. Consequently, reduction of the gene dosage in T7-based systems should
be the next step in process optimization.