Page 87 - Read Online

Page 10 of 15 Spus et al. Microbiome Res Rep 2023;2:33 https://dx.doi.org/10.20517/mrr.2023.20

lineage of the starter culture may vary due to changes in the environment (temperature, pH) and
propagation regime. In the present study, we investigated the impact of bacteriophage predation on the
culture’s population dynamics at the level of co-existing genetic lineages. To reach that goal, we used our
current knowledge of the structure and properties of a complex dairy starter culture Ur to design a defined
multi-strain starter culture (blend) and sequentially propagate it in milk for an extended period of time (538
generations). The blend included all eight genetic lineages of the Ur starter represented by 24 strains, which
extends our previous studies in which blends with three lineages represented by four strains were used .
[5]
We purposely selected strains belonging to L. cremoris lineage 1 & 5 - containing the “winner” (the fittest)
strains - which express a diverse susceptibility to the three phages used in the cocktail. We did not observe
detectable susceptibility (except strain 2MS47, L. cremoris lineage 6) to the three phages for the remaining
strains representing L. lactis lineages 2 & 4, L. cremoris 3, 6, 7 and Lc. mesenteroides ssp. cremoris lineage 8
[Figure 1].

+
[5]
According to our previous work , strains of the prt L. cremoris lineages 1 & 5 include the fittest strains
(highest growth rate in milk supplemented with casitone) covering a diverse sensitivity profile to predation
of lytic phages isolated from Ur. Our present results on the population dynamics in phage-challenged
blends provide evidence for the prediction that the fittest strains are selected against bacteriophage
predation pressure (in B-1 until 139 generations, in B-2 throughout the experiment).

Several possible events may occur during the propagation of the blend: (i) prophage induction; (ii) resistant
variants could emerge and gradually increase in abundance; (iii) bacteriophages without a host would be
“washed out” from the culture due to daily dilution; and (iv) evolved phages could emerge that predate on
previously resistant strain(s). All of these four events would shift the rules of population dynamics towards
PS or CD mechanisms and eventually impact the diversity of such a community. Based on the results
obtained in the current study, we will discuss the role of phage predation as the trigger causing a
catastrophic shift in the culture’s community that may finally lead to two alternative stable states.

Even though evidence exists that strains of the Ur starter contain inducible prophages , we did not observe
[28]
the susceptibility of Ur strains used to construct the blends to the supernatants of control blends
(throughout the propagation experiment) and phage-treated blends (at the end time point of propagation).
Based on this, we exclude the induction of an active phage crop as one of the major events impacting the
population dynamics in our blends. Perhaps, even though phage particles are released by the strains, no lysis
of the host occurs as described previously .
[29]

In the phage-challenged blends (B-1 and B-2), the final outcome of the propagation experiment was very
different in each of the two replicates. In the case of replicate B-1, phage predation caused an initial drop in
the abundance of L. cremoris lineage 1 & 5 strains, but recovery and a gradual increase were noted after 139
generations up to the end of the experiment. This increase can be explained by the emergence of a phage-
resistant strain(s) and/or by “wash out” of phages due to the serial dilution during sequential propagation
leading to an increased abundance of the faster-growing phage-resistant strain(s). We found lytic phages in
the supernatant of blend B-1 after 405 generations, indicating the presence of susceptible strain(s) at that
point of the experiment. Despite the presence of predators at the 405 generations time point, the abundance
of L. cremoris lineage 1 & 5 still increased until the end time point of propagation for 538 generations, where
we did not observe bacteriophage predation on the strains present in the blend anymore [Figure 4]. The
absence of bacteriophage predation at the 538 generations time point, conceivably due to “wash out”
between 405 and 538 generations time point, suggests that there was no phage host present anymore in the
culture allowing the population of the fastest multiplying originally-present phage-resistant strains of