The cAMP saga continues with an attempt to reconcile data.
I: The Escherichia coli glucose-lactose diauxie

June 6, 2018 eJournal
M. Crasnier-Mednansky, Ph.D., D.Sc.
CSO, Mednansky Institute, Inc.
Copyright 2018, M. Crasnier-Mednansky
Contact: martine [at] minst [dot] org

In 1942, Jacques Monod coined the term diauxie to describe a new phenomenon, a bacterial growth characterized by two distinct exponential phases separated by a complete cessation of growth.  The most striking feature of diauxie is 'complete cessation of growth' or diauxic lag between two exponential phases of growth.

Diauxie has been studied extensively and is the subject of much controversy, particularly the role of cAMP in the exemplary Escherichia coli glucose-lactose diauxie.  To understand the glucose-lactose diauxie, one must ask why following glucose consumption, E. coli stops its growth for a significant amount of time while lactose is available.  Inducer exclusion, which prevents lactose utilization during growth on glucose, is eliminated upon entry in the lag phase, in agreement with the current model implicating the phosphotransferase system or PTS, particularly Enzyme IIAGlc.  Thus, synthesis of β-galactosidase should be able to resume almost immediately, particularly because induction is known to occur in the absence of cell division (Stephenson M, 1933).  However it does not, as growth on lactose does not resume soon after entry in the lag.  In fact, β-galactosidase synthesis starts shortly before growth on lactose begins, as expected (Figure 1 in Epstein W, 1966).  Hence 'known fundamental limitations of biological sensors' were blamed for the diauxic lag, and its duration (Chu D, 2017).

The key signal in diauxie in controlling the lag phase, and ensuing β-galactosidase synthesis, is cAMP.

When E. coli grows on glucose, the cAMP level is relatively low.  Upon glucose exhaustion there is a sharp increase in intracellular cAMP, which is followed by downward steps (Figure 2 in Buettner MJ, 1973).  In the glucose-lactose diauxie, the same pattern occurs upon entry in the lag phase.  The cells however are not yet adapting to lactose because the sudden increase in cAMP leads to inhibition of lac operon transcription.  To clarify, when the lac operon is under induction, excess cAMP, resulting from the sharp increase in intracellular cAMP, impairs transcription.  In the glucose-lactose diauxie, non-growing cells in the lag phase start their 'slow' adaption to lactose while the cAMP level is falling.  It is to be understood cells growing on poor carbon sources (eliciting higher levels of cAMP as compared to glucose), and with IPTG (a much better inducer than lactose), are not affected by the cAMP level as lac transcription is fully active.

Of interest to note, a mechanism exists for supporting the proposal excess cAMP inhibits transcription of the lac operon.  Such mechanism relates to the second 'low affinity' CRP site (CRP-2) located at the site of the operator (O1), as CRP-cAMP binding at this site may inhibit promoter function.  Increasing the affinity of CRP for this site represses transcription of the lac operon (Irwin  N, 1987).  In addition, there is precedent, in the presence of two cAMP binding sites, CRP becomes a repressor at high concentration of cAMP (Nakano M, 2014, Comment).  Furthermore, distribution of RNA polymerase in the lac operon is skewed, with more RNA polymerase being located toward the 5’ than the 3’ end of the operon (Grainger DC, 2005).

Addition of cAMP eliminates the diauxic lag of the glucose-lactose diauxie, and lactose growth starts immediately after glucose consumption (Ullmann A, 1968).  By artificially increasing transcription of CRP-cAMP-dependent genes, transport of glucose is reduced, thus the level of cAMP is increased and inducer exclusion impaired.  Therefore, cells are adapting to lactose while growing on glucose and consequently growth on lactose occurs without a lag phase.

In the 'Lewis model' (Lewis M, 1996), CRP at CRP-1 assists the repressor in forming the repression loop (O1-O3) thereby promoting the repressed state.  Thus, CRP is facilitating the repressor function, as the cAMP concentration on glucose is most likely high enough for CRP-1 occupancy and formation of the O1-O3 repression loop.  It is plausible, the lactose repressor prevents RNA polymerase binding at Promoter-1, and CAP-1-bound CRP-cAMP prevents RNA polymerase binding at Promoter-2 (Malan TP, 1984).  As such, in the non-induced state (under inducer exclusion), RNA polymerase cannot initiate transcription, whatever the level of cAMP.  During the transition from the non-induced to the induced state, CRP-1-bound CRP-cAMP most likely prevents RNA polymerase occupancy at Promoter-2 (thereby favoring occupancy at Promoter-1) while the inducer-bound repressor no longer prevents RNA polymerase binding at Promoter-1.  In excess cAMP, CRP-cAMP binds to CRP-2 and inhibits transcription as a roadblock thus slowing down transcription.