Page 83 - Read Online
P. 83

Page 6 of 15                  Spus et al. Microbiome Res Rep 2023;2:33  https://dx.doi.org/10.20517/mrr.2023.20


               RESULTS
               Population dynamics in the multi-strain blends
               Strains of L. cremoris genetic lineage 1 & 5 used for the blend preparation were selected for their varying
               levels of bacteriophage sensitivity/resistance [Figure 1]. For strains of the other lineages used in the blend,
               with one exception of strain 2MS47 (L. cremoris, lineage 6), we did not detect susceptibility to any of the
               three Lactococcus phages (ϕTIFN1, ϕTIFN5 and ϕTIFN7) used in the cocktail. To evaluate the impact of the
               phage presence/absence, we monitored the abundance of all genetic lineages in both control (A-1, A-2) and
               phage-challenged blends (B-1, B-2) throughout a long-term propagation experiment [Figure 2]. At the start
               of the experiment, the distribution of abundance of representatives of the genetic lineages was the same in
               all four cultures. This initial composition was not stable throughout the propagation experiment; in control
               blends, we observed a gradual increase in the abundance of L. cremoris lineage 1 & 5 strains. At the end time
               point of the experiment, this gradual increase resulted in the dominance of the L. cremoris lineage 1 & 5
               strains in both control blends, with a relative abundance of 97% in L. cremoris strains A-1 and 98% in A-2.
               In the phage-challenged blends, we observed a far more complex behavior in terms of dynamics of relative
               abundances of all genetic lineages. Until 139 generations, both replicates B-1 and B-2 behaved in a similar
               way - the abundance of L. cremoris lineage 1 & 5 strains decreased to a level below 1%, which is in line with
               the fact that most (10/11) of the strains of L. cremoris lineage 1 & 5 are susceptible to at least one of the
               bacteriophages used in the cocktail. At the next sampling point (272 generations) the abundance of
               L. cremoris lineage 1 & 5 strains in B-2 was still very low, below 0.5% and L. lactis lineage 2 & 4 strains (cit )
                                                                                                         +
               were found to be the most dominant with 59% relative abundance, with L. cremoris lineage 3 strains (prt ) as
                                                                                                      +
               the second most abundant (25%) followed by L. cremoris lineage 7 strains (prt  and cit ) with 9% and
                                                                                      -
                                                                                             -
                                                  -
               L. cremoris lineage 6 strains (prt  and cit ) with 4.5% abundance. Finally, Lc. mesenteroides ssp. cremoris
                                           -
               lineage 8 strains (prt  and cit ) occupied 2.5% of the total population. However, in the other replicate B-1 at
                                 -
                                       +
               272 generations, L. cremoris lineage 1 & 5 strains increased in abundance to 13%, with the other prt
                                                                                                         +
               L. cremoris lineage 3 strains as the most dominant at this point with 44%, followed by L. lactis lineage 2 & 4
               strains with 25% and L. cremoris lineage 7 strains with 8% relative abundance. Previous experiments
                                                                                                         [5]
               showed that strain(s) of L. cremoris lineage 1 & 5 dominated the 4-strain blends during sequential
               propagation in milk, without phage predation, suggesting that these strains are the “winners” - i.e., the fittest
               strains for the imposed propagation regime. The fittest strains are potential targets for phage predation in
               accordance with the “kill-the-winner” concept . L. cremoris lineage 6 and Lc. mesenteroides ssp. cremoris
                                                       [8]
               lineage 8 strains were the least abundant at 5% abundance for either lineage. At later sampling points, the
               relative abundance of genetic lineages changed further. In B-1, we observed a clear pattern of a gradual
               increase in abundance of the proteolytic L. cremoris lineage 1 & 5 strains, reaching 57.5% and 86.5% at 405
               and 538 generations, respectively. Consequently, all other lineages decreased in abundance, with the biggest
               drop of prt  L. cremoris lineage 3 strains by 37 percentage points between 272 and 538 generations. The
                         +
               prolonged propagation experiment led to two different final outcomes (i.e., alternative states) in the phage
               treated blends. After 538 generations, blend B-1 was dominated by L. cremoris lineage 1 & 5 strains and
               resembled the control blends in composition. In the case of replicate B-2, however, genetic L. cremoris
               lineage 1 & 5 remained at extremely low levels in the community and L. cremoris lineage 3, the other prt
                                                                                                         +
               lineage in the culture, became the most abundant. These results show the enormous impact of lytic
               bacteriophages on population dynamics at the level of genetic lineages in defined multi-strain blends but
               also the conservation of one of the basic functionalities - proteolytic activity - at the community level.


               Bacteriophage resistance before, during and after propagation
               All 24 strains included in the blend design were individually challenged with the supernatant of the
               propagated blends collected at different time points of propagation to monitor the presence of lytic
   78   79   80   81   82   83   84   85   86   87   88