doi: 10.1371/journal.ppat.1012422. eCollection 2024 Aug.
Vancomycin-resistant Staphylococcus aureus (VRSA) can overcome the cost of antibiotic resistance and may threaten vancomycin's clinical durability
Affiliations
- PMID:39207957
- PMCID: PMC11361437
- DOI: 10.1371/journal.ppat.1012422
Item in Clipboard
Vancomycin-resistant Staphylococcus aureus (VRSA) can overcome the cost of antibiotic resistance and may threaten vancomycin's clinical durability
Samuel E Blechman et al. PLoS Pathog..
Display options
Format
Display options
Format
Abstract
Vancomycin has proven remarkably durable to resistance evolution by Staphylococcus aureus despite widespread treatment with vancomycin in the clinic. Only 16 cases of vancomycin-resistant S. aureus (VRSA) have been documented in the United States. It is thought that the failure of VRSA to spread is partly due to the fitness cost imposed by the vanA operon, which is the only known means of high-level resistance. Here, we show that the fitness cost of vanA-mediated resistance can be overcome through laboratory evolution of VRSA in the presence of vancomycin. Adaptation to vancomycin imposed a tradeoff such that fitness in the presence of vancomycin increased, while fitness in its absence decreased in evolved lineages. Comparing the genomes of vancomycin-exposed and vancomycin-unexposed lineages pinpointed the D-alanine:D-alanine ligase gene (ddl) as the target of loss-of-function mutations, which were associated with the observed fitness tradeoff. Vancomycin-exposed lineages exhibited vancomycin dependence and abnormal colony morphology in the absence of drug, which were associated with mutations in ddl. However, further evolution of vancomycin-exposed lineages in the absence of vancomycin enabled some evolved lineages to escape this fitness tradeoff. Many vancomycin-exposed lineages maintained resistance in the absence of vancomycin, unlike their ancestral VRSA strains. These results indicate that VRSA might be able to compensate for the fitness deficit associated with vanA-mediated resistance, which may pose a threat to the prolonged durability of vancomycin in the clinic. Our results also suggest vancomycin treatment should be immediately discontinued in patients after VRSA is identified to mitigate potential adaptations.
Copyright: © 2024 Blechman, Wright. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Conflict of interest statement
The authors have declared that no competing interests exist.
Figures
(A) Four VRSA strains were propagated in parallel on BHI and TSB agar media with vancomycin (32 μg/ml) and without. This was repeated for 50 propagation cycles at 37°C for ~70 h of growth per cycle. VAN-exposed lineages were propagated for a further 10 cycles [–60] on vancomycin-free agar. DNA sequencing of ancestral strains and cycle 50 and 60 evolved lineages was performed to pinpoint mutations specific to vancomycin, rather than the growth media. (B) Following growth, the outer edge of each colony was picked using a pipette tip and resuspended in 60% glycerol. Two μl of the resuspension solution was used to inoculate the next propagation cycle. (C) Three growth measures used as proxies for fitness (see Materials and Methods): (i) Colony expansion rate (re): colonies were scanned every 6 h during growth and the radius of each colony was measured from the images. There is the slope of a linear model fit to the points (see Materials and Methods). (ii) Optical density growth curves were captured and growth rate (rg) was calculated from the slope of the log of OD600. (iii) Lag time (t0) was calculated using the delay time (t1—see panel [ii] and Materials and Methods) and growth rate (rg), as well as estimated population size at inoculation (N0) and at delay time (N1). Note logarithmic scaled y-axis in (ii) and (iii). Fig 1A and 1B were made usingBiorender.com.
(A-F) Growth measurements of ancestral VRSA strains and their respective VSSA strain (where applicable) in BHI (A-C) and TSB (D-F) on solid media (n = 16 on BHI and n = 8 on TSB) and in liquid media (n = 96). (A andD) VRSA-4 and -10 had a similar colony expansion rate (re) to their respective VSSA strains, while VRSA-3 grew slower than VSSA-3. VRSA-3, -4, and -10 had a slowerre in the presence of vancomycin on BHI, but not TSB agar medium. VSSA-6 was not isolated and there of VRSA-6 did not differ in the presence of vancomycin. (B andE) VRSA-3, -4, and -10 had a slower growth rate (rg) than their respective VSSA strains, except for VRSA-3 in TSB liquid medium. All VRSA strains had a slowerrg in the presence of vancomycin, except for VRSA-6 in TSB liquid medium. (C andF) VRSA-3 and -4 had a longer lag time (t0) than their respective VSSA strains. All VRSA strains had a longert0 in the presence of vancomycin, except for VRSA-10 in BHI liquid medium. Error bars represent standard deviation. Differences between groups were compared by two-sided Wilcoxon rank-sum test and asterisks denote statistical significance after Holm-Bonferroni multiple testing correction (*** = p-adj. < 0.001, ** = p-adj. < 0.01, * = p-adj. < 0.05, ns = p-adj. ≥ 0.05). (G) Three of four VRSA strains completely reverted to susceptibility during propagation on vancomycin-free agar medium (und. = no resistant cells detected). Reversion occurred in BHI (n = 16) and TSB (n = 8) media across all replicates. VRSA-6 maintained resistance in the absence of vancomycin through 10 cycles in all cases. (H-I) VRSA-3, -4, and -10 underwent partial reversion to susceptibility when whole colonies were propagated on vancomycin-free agar (H) and when passaged in vancomycin-free liquid medium (I) (n = 3 in BHI and n = 3 in TSB).
(A-F) Shown are growth parameters of cycle 50 VAN-exposed lineages relative to ancestral VRSA strains (dotted horizontal lines), as measured in the presence (A-C; 32 μg/ml) and absence (D-F) of vancomycin. (A andD) Relative colony expansion rates (re evolved /re ancestor) of cycle 50 VAN-exposed lineages on solid media revealed that colonies generally expanded more slowly in the absence of vancomycin after evolution, with little change inre in the presence of vancomycin. Colony expansion rate was measured once per evolved lineage. (B andE) VAN-exposed lineages generally displayed faster growth rate (rg evolved /rg ancestor > 1) in liquid in the presence of vancomycin after evolution, but slower growth in its absence. (C andF) Difference in lag time between ancestral and VAN-exposed lineages (t0 evolved–t0 ancestor) in the presence (32 μg/ml) and absence of vancomycin showed increased lag times in the absence of vancomycin and decreased lag time in the presence. For each evolved lineage, growth rate and lag time were calculated from the mean of 24 replicates. Each group of relative values was compared to 1 (relative growth rate/colony expansion rate) or 0 (Δt0) by two-sided Wilcoxon signed-rank test and asterisks denote statistical significance after Holm-Bonferroni multiple testing correction (*** = p-adj. < 0.001, ** = p-adj. < 0.01, * = p-adj. < 0.05, ns = p-adj. ≥ 0.05). (G andH) Presence of mutations invanA operon in VAN-exposed (G) and VAN-unexposed (H) lineages revealed mutations were common among VAN-exposed lineages. There was global loss of thevanA operon for most VAN-unexposed lineages. The number of lineages with ≥ 1 mutation in the indicated gene is printed in each box.
(A-B) Position and identity of mutations alongddl and the surrounding region in VAN-exposed (A) and VAN-unexposed (B) lineages. Mutations in independent lineages are separated vertically. Shown is the number of lineages with a mutation inddl over the number of total cycle 50 lineages in each group. Numerous and varied mutations occurred within theddl gene of VAN-exposed lineages, implying loss or partial loss of Ddl function was adaptive in the presence of vancomycin. (C-H) Shown are colony expansion rate, growth rate, and lag time relative to ancestral strains (horizontal dotted lines). (C-E) In the presence of vancomycin, lineages with a mutation inddl (n = 44) had a faster growth rate than lineages without mutations inddl (n = 51), but no statistically significant difference was seen in colony expansion rate or lag time between the groups. (F-H) In the absence of vancomycin, lineages with a mutation inddl exhibited reduced colony expansion rate and growth rate, as well as longer lag times than lineages without mutations inddl. The groups were compared by two-sided Wilcoxon rank-sum test. Asterisks denote statistical significance after Holm-Bonferroni multiple testing correction (*** = p-adj. < 0.001, ** = p-adj. < 0.01, * = p-adj. < 0.05, ns = p-adj. ≥ 0.05).
(A) Many VAN-exposed lineages exhibited an abnormal colony morphology following the removal of vancomycin (at cycle 51) that was rapidly alleviated. Shown are representative VRSA-3 lineages after 66 hours of growth during cycles 50, 51, 52, and 60 (see also S6A Fig). Colony images were cropped and spliced from larger scans for clarity and to save room. Adjustments to brightness were applied uniformly to each cycle. (B) VAN-exposed lineages maintained resistance in the absence of vancomycin or reverted to susceptibility more slowly than their ancestral strains on both media (und. = no resistant cells detected). Reversion curves of the VAN-exposed lineages were compared with their respective ancestral strain by log-rank test and asterisks denote statistical significance after Holm-Bonferroni multiple testing correction (*** = p-adj. < 0.001, ** = p-adj. < 0.01, * = p-adj. < 0.05, ns = p-adj. ≥ 0.05). All VAN-exposed VRSA-6 lineages maintained resistance through propagation cycle 60. (C) Sankey diagram showing the number of lineages that overcame each fitness barrier. Many (64/95) lineages maintained resistance in the absence of vancomycin through cycle 60 (“VAN resistant”) and had greater fitness than their respective ancestral VRSA strain in the presence of vancomycin in two or more fitness proxies simultaneously (“fitness > VRSA (+van)”). Only a few (10/95) lineages also had greater fitness than their respective ancestral VSSA strain in the absence of vancomycin in two or more fitness proxies simultaneously (“fitness > VSSA (-van)”). * VRSA-6 evolved lineages were compared to ancestral VRSA-6 in the absence of vancomycin since an ancestral VSSA-6 was not isolated.
Similar articles
- Staphylococcus aureus VRSA-11B is a constitutive vancomycin-resistant mutant of vancomycin-dependent VRSA-11A.Périchon B, Courvalin P.Périchon B, et al.Antimicrob Agents Chemother. 2012 Sep;56(9):4693-6. doi: 10.1128/AAC.00454-12. Epub 2012 Jun 18.Antimicrob Agents Chemother. 2012.PMID:22710116Free PMC article.
- The role of the Staphylococcal VraTSR regulatory system on vancomycin resistance and vanA operon expression in vancomycin-resistant Staphylococcus aureus.Qureshi NK, Yin S, Boyle-Vavra S.Qureshi NK, et al.PLoS One. 2014 Jan 15;9(1):e85873. doi: 10.1371/journal.pone.0085873. eCollection 2014.PLoS One. 2014.PMID:24454941Free PMC article.
- Fitness cost of VanA-type vancomycin resistance in methicillin-resistant Staphylococcus aureus.Foucault ML, Courvalin P, Grillot-Courvalin C.Foucault ML, et al.Antimicrob Agents Chemother. 2009 Jun;53(6):2354-9. doi: 10.1128/AAC.01702-08. Epub 2009 Mar 30.Antimicrob Agents Chemother. 2009.PMID:19332680Free PMC article.
- Vancomycin Resistance inStaphylococcus aureus .McGuinness WA, Malachowa N, DeLeo FR.McGuinness WA, et al.Yale J Biol Med. 2017 Jun 23;90(2):269-281. eCollection 2017 Jun.Yale J Biol Med. 2017.PMID:28656013Free PMC article.Review.
- Vancomycin-resistant Staphylococcus aureus: a new model of antibiotic resistance.Hiramatsu K.Hiramatsu K.Lancet Infect Dis. 2001 Oct;1(3):147-55. doi: 10.1016/S1473-3099(01)00091-3.Lancet Infect Dis. 2001.PMID:11871491Review.
References
MeSH terms
Substances
Related information
Grants and funding
LinkOut - more resources
Full Text Sources
Medical