EXPERIMENTAL THERAPEUTICS

Linker-Improved Chimeric Endolysin Selectively Kills

Staphylococcus aureus In Vitro, on Reconstituted Human
Epidermis, and in a Murine Model of Skin Infection
Fritz Eichenseher,a,b Bjorn L. Herpers,c Paul Badoux,c Juan M. Leyva-Castillo,d Raif S. Geha,d Mathijs van der Zwart,e James McKellar,e
Ferd Janssen,e Bob de Rooij,e Lavanja Selvakumar,b Christian Röhrig,b Johan Frieling,e Mark Offerhaus,e Martin J. Loessner,a
Mathias Schmelchera,b

a Institute of Food, Nutrition and Health, ETH Zurich, Zurich, Switzerland

b Micreos GmbH, Wädenswil, Switzerland
c Regional Public Health Laboratory Kennemerland, Haarlem, The Netherlands
d Boston Children’s Hospital, Harvard Medical School, Boston, Massachusetts, USA
e Micreos Human Health B.V., Bilthoven, The Netherlands

ABSTRACT Staphylococcus aureus causes a broad spectrum of diseases in humans

and animals. It is frequently associated with inflammatory skin disorders such as atopic
dermatitis, where it aggravates symptoms. Treatment of S. aureus-associated skin infec-
tions with antibiotics is discouraged due to their broad-range deleterious effect on
healthy skin microbiota and their ability to promote the development of resistance.
Thus, novel S. aureus-specific antibacterial agents are desirable. We constructed two
chimeric cell wall-lytic enzymes, Staphefekt SA.100 and XZ.700, which are composed
of functional domains from the bacteriophage endolysin Ply2638 and the bacteriocin
lysostaphin. Both enzymes specifically killed S. aureus and were inactive against com-
mensal skin bacteria such as Staphylococcus epidermidis, with XZ.700 proving more
active than SA.100 in multiple in vitro activity assays. When surface-attached mixed

Downloaded from https://journals.asm.org/journal/aac on 17 November 2023 by 111.95.125.135.

staphylococcal cultures were exposed to XZ.700 in a simplified microbiome model, the
enzyme selectively removed S. aureus and retained S. epidermidis. Furthermore, XZ.700
did not induce resistance in S. aureus during repeated rounds of exposure to sublethal
concentrations. Finally, we demonstrated that XZ.700 formulated as a cream is effec-
tive at killing S. aureus on reconstituted human epidermis and that an XZ.700-contain-
ing gel significantly reduces bacterial numbers compared to an untreated control in a
mouse model of S. aureus-induced skin infection.

KEYWORDS atopic dermatitis, bacteriophage, endolysin, antibiotic resistance,

microbiome
Copyright © 2022 Eichenseher et al. This is an

S
open-access article distributed under the terms
taphylococcus aureus is an important Gram-positive opportunistic pathogen, which of the Creative Commons Attribution 4.0
causes mortality and morbidity in humans and animals worldwide. While it colonizes International license.
20 to 30% of healthy individuals, it can cause a wide spectrum of (often hospital- Address correspondence to Mathias
Schmelcher, mathias.schmelcher@hest.ethz.ch.
acquired) diseases in susceptible patients, ranging from relatively mild to life-threatening
The authors declare a conflict of interest. M.J.L.
conditions (1, 2). This includes skin infections (such as impetigo, folliculitis, and cellulitis), is an advisor for Micreos, a company producing
wound and soft tissue infections, abscesses on skin and internal organs, osteomyelitis, phage-based products. F.E., M.v.d.Z., J.M., F.J.,
B.d.R., L.S., C.R., J.F., M.O., and M.S. are
pneumonia, meningitis, endocarditis and bloodstream infections, and sepsis (2, 3). S. aur- employees of Micreos. All other authors
eus infections are often difficult to treat by conventional antibiotic therapy because the declare no conflict of interest.
bacteria can hide in protected niches within the body, such as biofilms, abscesses, and Received 30 November 2021
Returned for modification 13 January 2022
intracellular compartments, where they frequently exist in a dormant state (4–6). Accepted 16 March 2022
Moreover, S. aureus readily acquires resistance to multiple antibiotics, with methicillin-re- Published 13 April 2022
sistant S. aureus (MRSA) being the most prominent example (7).

May 2022 Volume 66 Issue 5 10.1128/aac.02273-21 1

Staphylococcus aureus-Specific Chimeric Endolysins Antimicrobial Agents and Chemotherapy

Atopic dermatitis (AD) is a chronic inflammatory skin condition that affects up to

25% of children and up to 10% of adults worldwide (8) and is associated with dry and
itchy skin, which can have a significant impact on quality of life (9). S. aureus coloniza-
tion and a reduced diversity of the skin microbiome are highly prevalent in AD
patients, with 30% to 100% of these patients being colonized with S. aureus, depend-
ing on the method used (10). S. aureus can aggravate inflammation via secretion of vir-
ulence factors such as enterotoxins, which, in turn, trigger the release of proinflamma-
tory cytokines (11). Vice versa, a weakened skin barrier as a frequent consequence of
AD facilitates S. aureus colonization. AD is commonly treated with emollients, anti-
inflammatory agents such as topical corticosteroids, immunosuppressants, or mono-
clonal antibodies, whereas antibiotic therapy aiming at the elimination of S. aureus is
indicated only in clinically infected cases, for which AD patients have an increased risk
(11–13). Treating S. aureus in AD with (broad-spectrum) antibiotics is discouraged due
to their undesired impact on the healthy skin microbiome and the associated risk of
antibiotic resistance induction (14). Instead, novel antimicrobial agents that specifically
target S. aureus but are ineffective against coagulase-negative staphylococcal (CoNS)
species such as Staphylococcus epidermidis and other beneficial commensal skin bacte-
ria would be desirable.
Endolysins are bacteriolytic enzymes encoded by bacteriophages, which degrade
from within the peptidoglycan (PG) bacterial cell wall at the end of the phage replica-
tion cycle, causing cell death and the release of progeny phages (15). In Gram-positive
organisms, the PG is exposed to the outside due to the absence of an outer membrane,
for which reason endolysins are effective against these bacteria also when added exter-
nally as recombinant proteins and have therefore been suggested as a novel class of
antibacterial agents (16). The most important advantages of endolysins as antimicro-
bials are (i) their rapid killing kinetics (17), (ii) the high level of specificity for their target
cells (18), (iii) their activity against drug-resistant bacteria, dormant cells, and biofilms
(19, 20), and (iv) their low chance of inducing bacterial resistance (21) due to their
highly conserved target sites in the PG. Staphylococcal phage endolysins have been
shown to effectively kill staphylococci, both in vitro and in animal infection models
(22). However, they are usually specific for staphylococci at the genus level (including
S. aureus and CoNS) (20), whereas endolysins active exclusively against S. aureus have

Downloaded from https://journals.asm.org/journal/aac on 17 November 2023 by 111.95.125.135.

not been described. Staphylococcal endolysins feature a modular architecture, typically
comprising two enzymatically active domains (EADs) and one SH3b-type cell wall bind-
ing domain (CBD). In most cases, the two EADs include an N-terminal cysteine-histi-
dine-dependent amidohydrolase/peptidase (CHAP) domain cleaving the D-Ala-Gly
bond of staphylococcal PG and one amidase domain targeting the MurNAc-L-Ala bond.
In a “lysis from without” setting, the CHAP domains contribute most to the lytic activity
of these enzymes, whereas the amidase domains are essentially inactive (23–25). The
endolysin of staphylococcal phage 2638A (Ply2638) is different in the way that it fea-
tures a highly active amidase domain and an N-terminal M23 endopeptidase instead of
a CHAP domain, which targets the same PG bond (D-Ala-Gly) but contributes only little
to the activity of the enzyme (20, 26).
Here, we describe the construction and functional characterization of two chimeric
peptidoglycan hydrolases (PGHs) named Staphefekt SA.100 (SA.100) and XZ.700, which
were generated to render the enzyme more specific toward S. aureus. This was accom-
plished by replacing the native low-activity M23 domain of Ply2638 by that of lysosta-
phin (LST), a bacteriocin produced by Staphylococcus simulans and directed against S.
aureus (27). The M23 endopeptidase domain of LST targets the pentaglycine bridge
(Gly-Gly activity), an interpeptide bridge unique to S. aureus PG, whereas CoNS feature
modifications in their interpeptide bridges, which render them less susceptible to the
action of LST (28, 29). SA.100 and XZ.700, while featuring a similar domain architecture,
differ in a linker region connecting their M23 and amidase domains. We first compared
both chimeric enzymes in multiple in vitro activity assays. Then, we further character-
ized the more active construct, XZ.700, investigating its specificity toward S. aureus, its

May 2022 Volume 66 Issue 5 10.1128/aac.02273-21 2

Staphylococcus aureus-Specific Chimeric Endolysins Antimicrobial Agents and Chemotherapy

FIG 1 Modular architectures and lytic activities of staphylococcal phage endolysin Ply2638, lysostaphin, and the chimeric lysin Staphefekt SA.100
constructed from both enzymes. (A) Schematic representation of native and chimeric enzymes. Functional domains are represented by colored bars, and

Downloaded from https://journals.asm.org/journal/aac on 17 November 2023 by 111.95.125.135.

respective amino acid positions are indicated. The dashed line indicates the fusion site in the chimeric enzyme. M23, M23 endopeptidase domain; Ami,
amidase domain; SH3b, SH3b cell wall binding domain. (B) Specific activities of parental and chimeric enzymes against S. aureus ATCC 12600, as
determined by turbidity reduction assays. Specific activity values were determined from the steepest slopes of individual lysis curves (decrease in optical
density over time) over a range of enzyme concentrations (12.5 to 100 nM). (C) Normalized and control-corrected lysis curves obtained with an enzyme
concentration of 100 nM. Error bars represent SEM from 2 biological replicates, including 2 technical replicates each.

potential to induce bacterial resistance, and its efficacy at killing S. aureus on reconsti-
tuted human epidermis (RHE) and in a mouse model of superficial skin infection.

RESULTS
Substitution of the M23 domain of the Ply2638 endolysin yields an active
chimeric PGH. We created a chimeric lysin by substituting the M23 D-alanyl-glycine en-
dopeptidase domain of the staphylococcal phage endolysin Ply2638 with the M23 gly-
cyl-glycine endopeptidase domain of the potent Staphylococcus aureus-specific bacter-
iocin lysostaphin (LST). This resulted in the construct M23LST_Ami2638_SH3b2638,
here named Staphefekt SA.100 (SA.100) (Fig. 1A). When we compared the parental and
chimeric enzymes in turbidity reduction assays (TRAs) against live S. aureus cells in sus-
pension, we found that SA.100 exhibits a similar specific activity to Ply2638 and is
slightly more active than LST, even though this difference was not statistically signifi-
cant (P . 0.05; ANOVA with post hoc Tukey honestly significant difference [HSD] test)
(Fig. 1B). The specific activity is derived from the steepest slopes (change in optical
density at 600 nm [DOD600]/min) of individual lysis curves (as exemplified in Fig. 1C),
measured over a range of enzyme concentrations (12.5 to 100 nM, corresponding to
0.7 to 5.6 m g/mL in the case of SA.100). Interestingly, SA.100 displayed faster lysis

May 2022 Volume 66 Issue 5 10.1128/aac.02273-21 3

Staphylococcus aureus-Specific Chimeric Endolysins Antimicrobial Agents and Chemotherapy

FIG 2 Comparison of anti-S. aureus activity of SA.100 and XZ.700. (A) Schematic representations of both constructs. Compared to SA.100, the 44-amino-
acid linker region at the N terminus of the amidase domain of Ply2638 is absent in XZ.700. (B) Representative image of an MIC assay with both enzymes.
MIC values were obtained by exposure of 2  105 CFU to 3  105 CFU of S. aureus 305 Newbould to serial dilutions of SA.100, XZ.700, or controls with
and without bacteria. The image was taken after 20 h of incubation. The MIC values of Staphefekt SA.100 and XZ.700 were 62.5 nM (3.6 m g/mL) and
15.6 nM (0.8 m g/mL), respectively, in three independent assays. (C) Thermal stability of SA.100 and XZ.700. Proteins at 200 nM concentration were
subjected to standard turbidity reduction assays against S. aureus SA113 after heat exposure for 10 min in PBS with subsequent cooling on ice. Error bars
represent standard deviations from 3 biological replicates. (D) Quantitative killing assays with both enzymes. Log reductions of S. aureus ATCC 12600
cultures after exposure to the enzymes for 120 min in PBS-T are shown. The upper limit of detection was a 4-log reduction. Error bars represent standard
deviations from at least 4 biological replicates.

kinetics than Ply2638, i.e., it caused an earlier onset of bacterial lysis, even though the

Downloaded from https://journals.asm.org/journal/aac on 17 November 2023 by 111.95.125.135.

steepest slopes of the two lysis curves were similar (Fig. 1C).
A linker-improved version of SA.100 features improved antibacterial activity.
Optimization of linker regions between individual functional domains of PGHs can sub-
stantially improve their activity (30). SA.100 contains an interdomain region consisting
of two natural linkers at the C terminus of the LST-derived M23 domain and the N ter-
minus of the Ply2638-derived amidase domain. In an attempt to further enhance the
staphylolytic activity of SA.100 and based on learnings from our extensive engineering
activities on staphylococcal PGHs in recent years (31–33), we created a linker-modified
derivative of SA.100, which was named XZ.700 and features a deletion of a 44-amino-
acid region at the N-terminal end of the Ply2638-derived protein fragment while
retaining the same domain architecture (Fig. 2A; Fig. S1 in the supplemental material).
Direct comparison of both proteins in three different in vitro activity assays revealed
that XZ.700 consistently displayed higher activity against S. aureus than SA.100 (Fig. 2B
to D). When tested in MIC assays against a selection of MRSA and methicillin-suscepti-
ble S. aureus (MSSA) strains, the average obtained MIC values for SA.100 and XZ.700
were 19.96 m g/mL (;350 nM) and 3.87 m g/mL (;75 nM), respectively, with no obvious
differences in susceptibility between MRSA and MSSA (Table 1; Fig. 2B). Of note, the
only S. epidermidis strain tested within this experiment, ATCC 12228, proved insensitive
to both enzymes (MIC . 128 m g/mL) (Table 1).
Kinetic measurements in TRAs under standard conditions (i.e., in phosphate-buffered
saline with 0.01% Tween 20 [PBS-T] at room temperature) revealed an approximately
70% increase in lytic activity of XZ.700 compared to SA.100. Thermodynamic stabilities
as measured in the same assay format were found to be similar for both enzymes. When
heated to increasing temperatures for 10 min prior to conducting TRAs under standard

May 2022 Volume 66 Issue 5 10.1128/aac.02273-21 4

Staphylococcus aureus-Specific Chimeric Endolysins Antimicrobial Agents and Chemotherapy

TABLE 1 MICs of SA.100 and XZ.700 against various staphylococcal strainsa

MIC (mg/mL) of:

Bacterial strain XZ.700 SA.100

S. aureus strains (MRSA) 2 16
130710015047
130215015001 2 16
130403015176 16 64
130603015366 2 16

S. aureus strains (MSSA)

305 (Newbould) 0.8 3.6
ATCC 29213 4 8
150825048101 4 16
150911021401 2 32
150918032701 2 8
305 (Newbould)
S. epidermidis strain
ATCC 12228 .128 .128
an = 3 for strain 305 Newbould; n = 1 for all other strains.

conditions, lytic activities of SA.100 and XZ.700 were gradually reduced, and they were
completely abolished when temperatures exceeded 63°C (Fig. 2C).
The difference in activity between SA.100 and XZ.700 was most pronounced when
they were compared in quantitative killing assays for their efficacy at reducing concen-
trations of viable S. aureus in suspension. Both enzymes reduced S. aureus numbers
(CFU/mL) at concentrations as low as 1 ng/mL, and killing efficacies increased with
increasing enzyme concentrations, with a higher rate observed for XZ.700 (Fig. 2D). At
15.6 ng/mL, XZ.700 caused a reduction in CFU/mL by .2.5 log units within 120 min,
whereas an 8-fold higher concentration of SA.100 was needed to achieve a similar
effect. These results corroborate the previous findings demonstrating a substantial
improvement in staphylolytic activity of XZ.700 compared to SA.100 and led us to char-
acterize XZ.700 in more detail.
XZ.700 is specific for Staphylococcus aureus. To provide more evidence for the

Downloaded from https://journals.asm.org/journal/aac on 17 November 2023 by 111.95.125.135.

specificity of XZ.700 for S. aureus, MICs were determined against a comprehensive collec-
tion of bacterial strains from different geographic regions, including 50 MRSA strains, 50
MSSA strains, and 20 non-S. aureus strains. The latter consisted of 17 staphylococcal spe-
cies other than S. aureus and 3 nonstaphylococcal organisms (Corynebacterium jeikeium,
Cutibacterium acnes, and Saccharomyces cerevisiae) (Table S1). For the tested S. aureus
isolates (MSSA and MRSA), MICs ranged from 0.5 to 16 m g/mL of XZ.700 (Fig. 3A). The
distribution of MICs did not significantly differ between MSSA and MRSA strains (median
MICs of 2 and 4 m g/mL, respectively; P = 0.11, independent-samples t test), and also the
mean MIC values of the MSSA and MRSA populations (3.89 6 2.88 and 4.90 6 3.88,
respectively) were not significantly different from each other (P = 0.14). None of the
tested S. aureus strains showed intrinsic resistance to XZ.700. Importantly, of the 17
staphylococcal non-S. aureus strains, 10 strains did not show any susceptibility to XZ.700
activity (MICs . 256 m g/mL; Table S1). Most of the strains that were susceptible to the
enzyme represent species closely related to S. aureus, sharing similar virulence factors
and/or causing similar infections in humans or other mammalian species (34–36). These
included three Staphylococcus pseudintermedius strains isolated from wound infections
in dogs and one Staphylococcus chromogenes strain isolated from bovine mastitis, all
showing an MIC of ,2 m g/mL. Furthermore, three Staphylococcus lugdunensis strains iso-
lated from clinical samples of different human infections showed an MIC of 8 m g/mL.
The other coagulase-negative staphylococcal species, including S. epidermidis and the
nonstaphylococcal control strains, all were not susceptible to XZ.700-mediated killing
(Table S1). The specificity of XZ.700 for S. aureus was further demonstrated in quantita-
tive killing assays, using three S. aureus and three S. epidermidis strains (Fig. 3B). While

May 2022 Volume 66 Issue 5 10.1128/aac.02273-21 5

Staphylococcus aureus-Specific Chimeric Endolysins Antimicrobial Agents and Chemotherapy

FIG 3 Specificity of XZ.700 for S. aureus. (A) MIC distribution of XZ.700 across 120 bacterial strains, including S. aureus (MSSA and MRSA) and non-S. aureus
strains. (B) Quantitative killing assays with XZ.700 and multiple S. aureus (red) and S. epidermidis (blue) strains. S. aureus 1, strain 140122015197; S. aureus 2,
strain 200309031601; S. aureus 3, strain 200224033101; S. epidermidis 1, strain 140712015198; S. epidermidis 2, strain 200414016601; S. epidermidis 3, strain
200331020501. Error bars represent standard deviations from 3 biological replicates.

XZ.700 showed concentration-dependent killing of all tested S. aureus strains, no meas-

urable activity was detected against the three S. epidermidis strains.
To further substantiate the specificity of XZ.700 for S. aureus, we grew mixed cul-
tures of S. aureus and S. epidermidis on transwell membranes (inoculated at a ratio of
1:1), mimicking a simplified skin microbiome (Fig. 4A). After 2 h of growth, the surface-
adhered cultures were treated with 100 m g/mL of XZ.700 in PBS-T or PBS-T alone as a
placebo control for 1 h, after which the liquid was removed. Following further incuba-
tion for 21 h in the absence of any treatment, bacterial cells were harvested and S. aur-
eus and S. epidermidis numbers determined. As shown in Fig. 4B, S. aureus overgrew S.
epidermidis by approximately 1.5 log units (ratio, 25:1) in the placebo control within

Downloaded from https://journals.asm.org/journal/aac on 17 November 2023 by 111.95.125.135.

24 h of incubation (P , 0.0001; unpaired t test). However, in the XZ.700-treated sam-
ples, S. aureus concentrations were found to be 2.5 log units below those of S. epider-
midis (ratio, 1:335) at the end of the experiment (P , 0.001). This suggests that XZ.700
selectively kills S. aureus in mixed surface-adhered bacterial cultures while allowing S.
epidermidis to grow.
XZ.700 does not induce resistance in S. aureus. Unlike classical antibiotics or the
bacteriocin LST, endolysins have been suggested to be refractory to resistance devel-
opment in bacteria due to their highly conserved binding and cleavage sites in the
peptidoglycan (16). XZ.700 is derived from the phage endolysin Ply2638 but also con-
tains one catalytic domain from LST. This being said, we set out to investigate whether
XZ.700 could provoke resistance in S. aureus after several rounds of repeated exposure
to sub-MICs. Mupirocin, an antibiotic widely used as a topical agent against superficial
S. aureus infections or colonization of skin and mucous membranes, was included in
these experiments for comparison. After 20 cycles of repeated exposure, no induction
of resistance was observed in either MSSA or MRSA strains against XZ.700, while MICs
of mupirocin gradually increased to 8 times the original MIC in all strains (Fig. 5).
Although the MICs for XZ.700 fluctuated by one 2-fold dilution step compared to the
starting MIC, no gradual increase of MICs was observed. At cycle 20, the difference in
MIC fold changes between XZ.700 and mupirocin across all tested strains was statisti-
cally significant (P , 0.0001; unpaired t test).
XZ.700 in solution or formulated as a cream reduces numbers of viable MRSA
on reconstituted human epidermis. The absence of resistance development and the
specificity of XZ.700 for S. aureus make this enzyme a promising novel antibacterial

May 2022 Volume 66 Issue 5 10.1128/aac.02273-21 6

Staphylococcus aureus-Specific Chimeric Endolysins Antimicrobial Agents and Chemotherapy

FIG 4 Selective killing of S. aureus by XZ.700 in mixed surface-adhered staphylococcal populations. (A) A
mixed culture of S. aureus and S. epidermidis was grown on transwell membranes (;104 CFU/cm2 of a 1:1
mixture at the time of seeding) inserted into agar-filled wells for 24 h in total. Two hours after seeding,
the adhered bacteria were exposed to XZ.700 solution or a control treatment (buffer) for 1 h before
being further incubated for 21 h in the absence of the enzyme solution/buffer. (B) Concentrations of S.
aureus and S. epidermidis on transwell membranes 21 h after treatment with buffer (placebo) or 100 m g/
mL of XZ.700. Error bars represent standard deviations from 6 independent experiments. ****, P , 0.0001;

Downloaded from https://journals.asm.org/journal/aac on 17 November 2023 by 111.95.125.135.

***, P , 0.001.

candidate agent for the treatment of S. aureus infections in which the preservation of a
healthy microbiome is desired, such as in skin infections. To evaluate the potential of
XZ.700 as an agent to decrease the S. aureus burden on colonized or infected human
skin, we tested the enzyme in solution and formulated as a cream on MRSA-colonized
reconstituted human epidermis (RHE). The RHE samples we obtained for this purpose
had been cultivated from human keratinocytes for 17 days and were fully differenti-
ated, with stratum corneum, stratum lucidum, stratum granulosum, stratum spinosum,
and stratum basale visible in hematoxylin-eosin saffron (HES)-stained cross-sections
(Fig. 6A). On the day of the experiment, the RHE samples immersed in antibiotic-free
maintenance medium were inoculated with 6  106 CFU/mL of S. aureus ATCC 33591
(MRSA; MIC of XZ.700, 0.50 6 0.27 m g/mL). Following incubation at 35°C for 4 h, the
medium was removed and the colonized RHE treated with XZ.700 in solution (32 m g/
mL; 64-fold MIC) or a cream formulation at three different concentrations (32 m g/mL,
128 m g/mL, 512 m g/mL) for 30, 60, or 120 min or left untreated as a control. Residual
viable S. aureus cells after the treatment were determined both in the apical fraction
(nonadhered bacteria) and the homogenized tissue sample (adhered bacteria). As
shown in Fig. 6B, the 60-min treatments were most effective at reducing nonadherent
S. aureus numbers for both XZ.700 solution and the different cream formulations.
While there was a trend toward higher efficacy with increasing concentrations in the
cream, these differences were not statistically significant. The most effective treatment

May 2022 Volume 66 Issue 5 10.1128/aac.02273-21 7

Staphylococcus aureus-Specific Chimeric Endolysins Antimicrobial Agents and Chemotherapy

FIG 5 Fold changes in XZ.700 and mupirocin MICs against MSSA and MRSA strains over 20 cycles of repeated
exposure to sub-MICs. Each curve represents mean values and SEMs from 2 strains tested in duplicate. Strains
used for this experiment included ATCC 29213 (MSSA) and the clinical isolates 170606037501 (MSSA),
130603015366 (MRSA), and 150212029501 (MRSA).

was the XZ.700 cream at 128 m g/mL applied for 60 min, resulting in a significant 2-log
reduction compared to the untreated control (P , 0.01). In contrast, none of the 30-
min treatments led to a significant reduction in bacterial numbers. Similar observations
were made for the adherent cells in the tissue homogenates (Fig. 6C). Here, the con-
centration-dependent effect was more pronounced, with XZ.700 cream at 128 m g/mL
and 512 m g/mL, applied for 60 min, causing the highest reduction in adhered S. aureus
numbers compared to the control (approximately 3 log units; P , 0.01). Overall, these
results demonstrate that XZ.700 is effective against S. aureus also in more complex

Downloaded from https://journals.asm.org/journal/aac on 17 November 2023 by 111.95.125.135.

environments such as reconstituted tissues and in cream formulations, as they are
commonly used for topical applications on skin.
XZ.700 gel reduces S. aureus numbers in a murine skin infection model. To eval-
uate the anti-S. aureus activity of XZ.700 also in vivo, we employed a mouse model of
superficial skin infection combined with an in vivo bioluminescence imaging system. In
a preliminary experiment, we monitored the concentration of a bioluminescent S. aur-
eus USA300 strain (MRSA) on tape-stripped skin of female and male mice over a period
of 79 h following initial inoculation with 108 CFU/animal. For this purpose, biolumines-
cence produced by viable bacteria was recorded at multiple time points during the
entire 3-day experiment, and the resulting color-scale images were overlaid with gray-
scale photographs of the immobilized mice (Fig. S2A). Furthermore, bioluminescence
was quantified and plotted over time as relative intensity (Fig. S2B). We found that the
intensity of the bioluminescence signal (corresponding to the bacterial load on the
murine skin; Fig. S3) decreased by around 42% in males within the first 8 h after infec-
tion, whereas in females, it remained unchanged or even slightly increased during the
same period. Overall, male mice showed a markedly faster clearance of bacteria than
female mice, with percent intensity values significantly different between the two
groups at time (t) of 8 h (P = 0.02). This was also reflected in the number of viable S.
aureus cells obtained from homogenized skin biopsy specimens at the end of the
experiment. While the bacterial concentration on male skin was 1.6 6 0.7  105 CFU/
cm2, it was more than 2-fold higher in female mice (3.6 6 1.5  105 CFU/cm2),
although this difference was not statistically significant (Fig. S2C). Overall, the results
obtained from this preliminary study suggest that female mice are less efficient at

May 2022 Volume 66 Issue 5 10.1128/aac.02273-21 8

Staphylococcus aureus-Specific Chimeric Endolysins Antimicrobial Agents and Chemotherapy

Downloaded from https://journals.asm.org/journal/aac on 17 November 2023 by 111.95.125.135.

FIG 6 Activity of XZ.700 in solution and cream formulation against MRSA on reconstituted human
epidermis (RHE). (A) HES-stained vertical paraffin section of RHE used in these experiments, with
stratum corneum (sc), stratum lucidum (sl), stratum granulosum (sg), stratum spinosum (ss), and
stratum basale (sb) visible. (B and C) Concentrations of viable nonadherent (B) and adherent (C) S.
aureus ATCC 33591 (MRSA) after treatment of colonized RHE with XZ.700 in solution or cream
formulation at different concentrations for 30, 60, or 120 min. Bacterial concentrations after treatment
significantly different from the respective untreated controls are marked with asterisks (*, P , 0.05; **,
P , 0.01). Error bars represent standard deviations from 3 biological replicates. The used enzyme
concentrations of 32, 128, and 512 m g/mL correspond to 0.96, 3.84, and 15.36 m g/cm2 RHE,
respectively.

clearing S. aureus infection on tape-stripped skin than male mice, for which reason we
chose female mice for the following infection and treatment study.
Similar to what has been described above, mice were infected with 108 CFU of the
bioluminescent S. aureus on tape-stripped skin, and the infected areas (around 2 cm2)

May 2022 Volume 66 Issue 5 10.1128/aac.02273-21 9

Staphylococcus aureus-Specific Chimeric Endolysins Antimicrobial Agents and Chemotherapy

FIG 7 Efficacy of XZ.700 formulated as cream or gel in a mouse model of S. aureus skin infection. (A) Representative bioluminescence images of tape-
stripped mice infected with the bioluminescent S. aureus strain USA300 LAC::lux and treated with XZ.700 formulated as cream or gel (30 m g XZ.700 per mL
cream or gel) or the respective vehicle controls (Veh). (B) Luminescence intensity measured over time in infected animals treated with XZ.700 (XZ) gel (left)
or cream (right) or the respective vehicle controls (Veh) (n = 6). (C) S. aureus CFU counts/cm2 on murine skin left untreated or undergoing various
treatments as described in panels A and B 74 h after infection (n = 6). (D) Representative images showing the appearance of murine skin 74 h after
infection with S. aureus and various treatments as described above.

were treated with XZ.700 formulated as gel or cream or corresponding vehicle controls
(gel or cream without enzyme) 6 times during a period of 70 h, with the first treatment
applied at 3 h postinfection. The amounts of cream and gel applied per animal during the
entire experiment were approximately 30 mg and 25 mg, respectively. Bioluminescence
imaging revealed that topical treatment with XZ.700 gel significantly enhanced the clear-
ance of S. aureus on tape-stripped skin compared with untreated mice or vehicle control

Downloaded from https://journals.asm.org/journal/aac on 17 November 2023 by 111.95.125.135.

animals, whereas treatment with XZ.700 cream resulted in a similar clearance as observed
in untreated or vehicle cream-treated mice (Fig. 7A and B). The luminescence intensity on
XZ.700 gel-treated animals was significantly lower than on the respective vehicle control
mice at the end of the experiment (P , 0.05; Fig. 7B). These results were confirmed by
enumeration of S. aureus on skin homogenates 74 h after the infection (Fig. 7C). While the
reduction in CFU/cm2 skin in XZ.700 cream-treated animals compared to untreated or
cream vehicle controls was not statistically significant, XZ.700 gel caused a significant
reduction compared to the respective untreated and gel vehicle controls (P , 0.01). We
observed very weak redness in almost all mice before the cutaneous S. aureus infection,
consistent with the mechanical injury induced by tape stripping. This redness was not
worsened by the infection and disappeared 48 h after infection. For some mice treated
with vehicle gel, we observed slight skin dryness starting at 36 h and remaining until
72 h after the infection, whereas this was not observed in mice treated with XZ.700 gel
(Fig. 7D).

DISCUSSION
Novel classes of antibacterial agents that are specific for the target pathogen and
refractory to resistance development are highly desirable, particularly in the light of
the increasing prevalence of antibiotic-resistant strains worldwide, which represents
one of the most urgent problems of our time (37). One of the biggest disadvantages of
using conventional antibiotics for the treatment of bacterial infections, besides their
high chance of inducing resistance, is their broad-range detrimental effect on com-
mensal bacterial populations. This particularly applies to the treatment of skin

May 2022 Volume 66 Issue 5 10.1128/aac.02273-21 10

Staphylococcus aureus-Specific Chimeric Endolysins Antimicrobial Agents and Chemotherapy

infections which are caused or aggravated by S. aureus, such as AD, where any off-tar-
get activity against beneficial microorganisms of the healthy skin microflora should be
avoided. The two chimeric PGHs described here and, in particular, XZ.700 due to its
enhanced lytic activity against S. aureus compared to SA.100, are promising candidates
for such applications. SA.100, the modular design and construction of which is
described here for the first time, has previously been used successfully in a cetomacro-
gol-based cream formulation for the treatment of chronic and recurrent S. aureus-
related dermatoses in three single-patient case studies (38). All three patients were suf-
fering from skin conditions (folliculitis or impetigo) associated with S. aureus infection
and had been treated unsuccessfully with antibiotics prior to the start of SA.100 treat-
ment. Importantly, no induction of resistance to SA.100, as determined by MIC assays
with S. aureus isolates obtained from one of the patients, was observed despite
repeated application of the endolysin (up to 2 times daily for several months) (38). In
another study, SA.100 in the same cetomacrogol cream induced a clinically relevant
and statistically significant improvement of both severity scores, as well as quality of
life, in 43 patients with atopic dermatitis (39).
Besides antibiotics, also, LST is known to be prone to resistance development. This
is because this bacteriocin cleaves within the pentaglycine bridge, which is the most
variable part of S. aureus PG, whereas all staphylococcal phage endolysins described to
date target more conserved portions of the PG (20). Various S. aureus mutant strains
featuring alterations within the pentaglycine bridge and, consequently, reduced sus-
ceptibility to LST have been described (40). This being said, our finding that XZ.700
does not induce resistance in MRSA and MSSA strains during repeated cycles of expo-
sure is highly encouraging and in agreement with previous findings on SA.100 (38). In
contrast, increases in MIC of more than 500-fold had been observed in similar experi-
ments with LST (41). The absence of resistance induction in SA.100 and XZ.700 can
likely be attributed to the combination of two different EADs (i.e., a Gly-Gly endopepti-
dase and a MurNAc-L-Ala amidase) within the same molecule. In this case, two simulta-
neous independent mutations within the PG would be required to render the strains
resistant, which is highly unlikely to occur. This is also in agreement with previous
research showing that the rate of resistance development against staphylococcal PGHs
decreases with an increasing number of different EADs within the same enzyme (41).

Downloaded from https://journals.asm.org/journal/aac on 17 November 2023 by 111.95.125.135.

Similar to XZ.700, the chimeric PGH constructs K-L and L-K described in this study con-
tain an LST-derived M23 domain besides two other EADs originating from a phage
endolysin. Despite the demonstrated functionality of the M23 domain in the context of
these chimeras (via mass spectrometry analysis of digested PG), K-L and L-K displayed
73- to 291-fold reduced fold changes in MICs compared to LST when tested in
repeated-exposure experiments. This being said, it would be interesting to investigate
whether both EADs included in XZ.700 retain their catalytic activity, e.g., via PG diges-
tion product analysis or active-site knockouts. Further research could also include test-
ing bacteria repeatedly exposed to XZ.700 for their susceptibility to LST and determin-
ing the activity of XZ.700 against known LST-resistant strains.
The replacement of the native Ply2638 M23 endopeptidase domain by the (penta-
glycine bridge-specific) M23 domain of LST in SA.100 and XZ.700 was expected to shift
the specificity of these enzymes toward S. aureus. However, the high degree of specific-
ity observed for XZ.700 (illustrated by MICs .256 m g/mL and absence of any measura-
ble activity against S. epidermidis in quantitative killing assays) (Fig. 3; Table S1) was
rather surprising given the highly conserved PG cleavage site of its Ply2638-derived
amidase domain (MurNAc-L-Ala) (20), which is also present in S. epidermidis (28). This
may argue for a substantially higher contribution of the M23 domain to the overall ac-
tivity of XZ.700 than the amidase domain. Alternatively, this effect could be explained
by the possibility that, despite its highly conserved cleavage site, also, the amidase do-
main exhibits a certain preference for S. aureus PG by recognizing and binding to a
larger portion of the PG structure, extending beyond the actual scissile bond and
including parts of the pentaglycine bridge. This specificity could be further enhanced

May 2022 Volume 66 Issue 5 10.1128/aac.02273-21 11

Staphylococcus aureus-Specific Chimeric Endolysins Antimicrobial Agents and Chemotherapy

by the presence of the SH3b binding domain in direct proximity of the amidase do-
main. The SH3b cell wall-targeting domain of LST has been demonstrated to require
an intact pentaglycine bridge (as only present in S. aureus) to exhibit its full binding
capacity (42–44), and, given the high degree of conservation of SH3b domains
throughout all known staphylococcal PGHs (45), the same has been suggested for
SH3b domains derived from staphylococcal phage endolysins (46). The observation
that the S. aureus specificity of XZ.700 does not only hold true in experiments with
individual bacterial strains but also in our mock microbiome study in the presence of S.
epidermidis (Fig. 4) is highly encouraging in light of a potential application of this
enzyme on human skin as a microbiome-friendly anti-S. aureus agent. There is a broad
consensus that S. aureus plays an important role in aggravating disease in AD patients
(10, 14, 47). While, also, S. epidermidis can cause infections in humans (48), there is cur-
rently more evidence for its protective effect within the skin microbiome (e.g., by pro-
tecting against infection by pathogenic bacteria, promoting wound repair, tuning skin
immunity, and protecting against skin tumors) than for its role as an aggravator in AD
(49, 50). Importantly, we observed a shift in our mixed staphylococcal culture toward S.
epidermidis 21 h after the end of the XZ.700 treatment, suggesting that the enzyme
might be capable of exerting a long-term positive effect on a skin microbiome.
However, it is important to acknowledge the limitations of this simplified microbiome
model, which consists of only two bacterial species and therefore does not take into
account possible effects of other skin microbiota on the microbiome during endolysin
treatment. Therefore, these experiments should be followed up by studies employing
more complex microbial compositions or even full skin microbiomes.
The comparative in vitro characterization of our two chimeric lysins revealed that
XZ.700 outperforms SA.100 in three independent activity assays, i.e., MICs, TRAs, and
quantitative killing assays (qKAs) (Fig. 2). High thermostability has previously been
reported for individual phage-derived lytic enzymes, such as the virion-associated PGH
gp36C of Pseudomonas aeruginosa phage phiKMV (51) or the Listeria phage endolysin
PlyP35, which retained considerable lytic activity even after heating to 90°C (52). While
SA.100 and XZ.700 exhibited lower thermostability, both enzymes retained approxi-
mately 90% of their activity after exposure to 42°C, which is higher than normal human
skin temperature by 5 to 9°C, suggesting sufficient stability for skin applications. Also

Downloaded from https://journals.asm.org/journal/aac on 17 November 2023 by 111.95.125.135.

under these conditions, the activity of XZ.700 surpassed that of SA.100. Taken together,
these results clearly demonstrate that XZ.700 is the more potent enzyme, even though
results from the different assays may differ on a quantitative level. Similar quantitative
differences have been reported for other lytic enzymes that have been compared in
multiple activity assays (20, 41, 53), and this observation has been explained with dif-
ferences in sensitivity between these assays and assay-specific inherent biases, each of
them favoring enzymes with certain biochemical properties such as regarding mole-
cule size, charge, or hydrophobicity (16). The increased activity of XZ.700 compared to
SA.100 can most likely be attributed to the deletion of a Ply2638-derived linker region
in XZ.700, which represents the only difference between the two enzymes. The poten-
tial impact of interdomain linkers on lytic activity of (chimeric) endolysins has been rec-
ognized in earlier studies (30, 54, 55), with the exchange or variation of such linkers
being able to either boost, decrease, or even completely abolish the activity of the re-
spective enzymes (30). The likely reason for this is that the length, structure, and flexi-
bility of the linker peptide determines the geometric orientation of the two adjacent
functional domains relative to each other and, therefore, their degree of alignment
with the respective cleavage or binding sites in the PG (30, 54).
Besides its high activity in vitro against planktonic and adhered S. aureus cells, as
demonstrated in this study, XZ.700 has recently been reported to also effectively de-
grade MRSA biofilms grown on titanium surfaces under static and dynamic conditions
for 24 or 48 h (56). This antibiofilm activity, which has also been described for several
other PGHs (19, 57, 58), is an important feature when considering the application of
such lytic enzymes as antibacterial agents, given the prominent role of biofilms in

May 2022 Volume 66 Issue 5 10.1128/aac.02273-21 12

Staphylococcus aureus-Specific Chimeric Endolysins Antimicrobial Agents and Chemotherapy

bacterial infections and their resilience to conventional treatment regimens (6, 22, 59).
One reason for this observed resilience is that bacteria within such biofilm commun-
ities frequently exist as persisters, i.e., dormant, nondividing cells, which display high
tolerance to many antibacterial agents (60). By attacking the bacterial PG, PGHs exhibit
a mode of action that is independent of the metabolic state of the cell, which is in
agreement with previous reports demonstrating activity of PGHs against persister cells
(19, 61). Besides biofilms, dormant bacteria also exist in abscesses and intracellular
compartments (5). A recent study reported activity of PGHs fused to cell-penetrating
peptides against S. aureus residing intracellularly within various cell lines and within
abscesses in a mouse model (31).
The promising in vitro results with XZ.700 obtained here and in the aforementioned
biofilm study are corroborated by our findings in the reconstituted human epidermis
model and the mouse model of superficial skin infection. Reconstructed human epithe-
lia models have been described to closely mimic in vivo human tissues in terms of mor-
phological (i.e., the presence of a multistratified epithelium), biochemical, and physio-
logical properties and therefore currently represent the most promising alternative to
animal models, ex vivo explants, and submerged cell monolayers for safety and efficacy
evaluation of topically applied agents (62, 63). Here, we demonstrated that XZ.700 can
significantly reduce nonadherent and adherent S. aureus numbers in a complex envi-
ronment mimicking human skin when formulated as a solution or a cetomacrogol-
based cream (Fig. 6). Of note, a similar cream formulation with SA.100 as the main in-
gredient had previously proven safe in human clinical trials (11, 12). Interestingly, the
60-min treatments with XZ.700 led to significant reductions in bacterial numbers in 7
out of 8 cases, whereas 120 min treatments significantly reduced the numbers in only
3 out of 8 cases. One can speculate if this could possibly indicate a relatively fast inacti-
vation of the enzyme under the applied experimental conditions or it could, rather, be
attributed to the inherent variability of the experiment. In any case, this question war-
rants further investigation in view of the intended application of XZ.700 on human
skin. Contrary to our observations in the reconstituted human epidermis model, appli-
cation of an identical XZ.700 cream did not significantly reduce S. aureus numbers
compared to the untreated and vehicle controls in our murine skin infection model

Downloaded from https://journals.asm.org/journal/aac on 17 November 2023 by 111.95.125.135.

(Fig. 7C). In this experimental setup, the gel formulation of XZ.700 proved significantly
more effective at reducing bacterial numbers on the skin than the cream. This differ-
ence can likely be explained by cream- and gel-specific matrix effects, with the gel for-
mulation allowing for a more efficient release of the enzyme from the matrix and, con-
sequently, a higher local concentration of active enzyme on the treated skin.
Interestingly, we found that, in the absence of XZ.700 treatment, male mice cleared S.
aureus from their skin significantly faster than female mice. Gender bias in bacterial
infections (including those caused by S. aureus) has been reported previously, with the
direction of the bias depending on the pathogen and the site of infection (64, 65). It
remains unclear if the gender-specific difference observed here can have an impact on
XZ.700 efficacy. While the results of our murine skin infection model are encouraging
by demonstrating efficacy of XZ.700 in a complex in vivo setting, one should keep in
mind the limitations of murine models from a translational point of view. In fact, there
are important differences in skin morphology (e.g., regarding skin thickness, the num-
ber of epidermal layers, and adhesion to underlying tissues) and wound-healing mech-
anisms between mice and humans (66). This being said, future research should focus
on evaluating the efficacy of XZ.700 also in larger animal models such as pigs, which
are more suitable to mimic the situation in human skin (67).
Overall, the results from our study corroborate the high potential of PGHs in general
as a novel class of antibacterial agents and of XZ.700 in particular for the treatment of
S. aureus-induced or -aggravated skin conditions due to its unique specificity for S. aur-
eus, its high antibacterial activity, and the absence of resistance induction against this
enzyme.

May 2022 Volume 66 Issue 5 10.1128/aac.02273-21 13

Staphylococcus aureus-Specific Chimeric Endolysins Antimicrobial Agents and Chemotherapy

MATERIALS AND METHODS

Bacterial strains and culture conditions. Staphylococcus aureus strains ATCC 12600 and SA113 (68)
were used for turbidity reduction assays (TRAs) and quantitative killing assays (qKAs) conducted in this
study. In addition, the following clinical isolates available in the strain collection of the Regional Public
Health Laboratory Kennemerland were used for comparison of killing activity against S. aureus and S. epi-
dermidis: S. aureus strains 140122015197 (blepharitis isolate), 200309031601 (skin isolate from acne
patient), and 200224033101 (skin isolate from rosacea patient) and S. epidermidis strains 140712015198
(blepharitis isolate), 200414016601 (human blood isolate), and 200331020501 (human skin isolate). S.
aureus ATTC 12600 and S. epidermidis strain 140712015198 were also used for mixed-culture transwell
experiments. Strains used for MIC assays are listed in Table 1 and Table S1 in the supplemental material.
ATCC 29213 (MSSA) and the clinical isolates 170606037501 (MSSA; isolate from prosthetic joint infec-
tion), 130603015366 (methicillin-resistant S. aureus [MRSA]; isolate from nasal colonization screening),
and 150212029501 (MRSA; isolate from furunculosis) were used for resistance induction experiments. S.
aureus strain ATCC 33591 was used for experiments with RHE and the bioluminescent MRSA strain
USA300 LAC::lux (69) for the murine skin infection model.
Staphylococcal strains were grown in tryptic soy broth (TSB) or brain heart infusion (BHI) at 37°C or
35°C. Escherichia coli strains were cultured in Luria-Bertani (LB) medium supplemented with appropriate
antibiotics for plasmid selection.
DNA techniques and cloning procedures. Standard molecular cloning techniques (70) were used
to generate plasmid constructs encoding recombinant endolysins. In the first round of cloning, the
endolysin gene ply2638 identified in Staphylococcus phage 2638a (71) was inserted into BamHI and SalI
sites of pQE-30 (Qiagen), resulting in pHPL2638. Its heterologous expression in E. coli results in an N-ter-
minally 6His-tagged version of Ply2638 (HPL2638) that can be purified by immobilized metal ion affin-
ity chromatography (IMAC). In a similar manner, a construct encoding an N-terminally His-tagged ver-
sion of mature lysostaphin (lysostaphin fragment Ala248-Lys493) (72) was generated (pHLST). A
truncated variant of Ply2638 lacking the N-terminal M23 domain, pHAmi2638_SH3b2638, was created
by inserting a ply2638 fragment corresponding to Leu138-Lys486 into SacI-SalI sites of pQE-30. A con-
struct encoding an N-terminally His-tagged version of a fusion protein subsequently named Staphefekt
SA.100 (pHM23LST_Ami2638_SH3b2638; SA100) was then obtained by inserting a fragment correspond-
ing to the M23 domain of lysostaphin into the BamHI/SacI site of pHAmi2638_SH3b2638. This construct
was initially generated to allow for direct comparison with the His-tagged parental enzymes HPL2638
and HLST. All pQE-30-based plasmid constructs were transformed into E. coli XL1-Blue MRF9 (Agilent) by
electroporation and verified by Sanger sequencing (GATC, Konstanz, Germany).
For further experiments with SA.100 (and its derivative XZ.700), non-His-tagged versions of these
enzymes were obtained from Micreos Human Health (Bilthoven, The Netherlands). Compared to SA.100,
XZ.700 features a deletion of a 44-amino-acid region at the N-terminal end of the Ply2638-derived pro-
tein fragment. Furthermore, two amino acids derived from the SacI restriction site originally used for
cloning in the pQE-30-based constructs are no longer present in these non-His-tagged versions.
Protein expression and purification. Expression of recombinant proteins in E. coli was carried out
essentially as previously described (33). In brief, bacterial cultures were grown under agitation in LB me-

Downloaded from https://journals.asm.org/journal/aac on 17 November 2023 by 111.95.125.135.

dium modified for protein expression (20) at 37°C and supplemented with suitable antibiotics for plas-
mid selection until an OD600 of 0.5 was reached. Following cooling on ice, 0.5 M isopropyl- b -D-thiogalac-
topyranoside (IPTG) was added for induction of protein expression, and incubation under agitation was
continued for 18 h at 19°C. Cells were pelleted by centrifugation and frozen at 280°C. Depending on
the presence of a His tag, pellets were thawed and resuspended in either lysis buffer for IMAC (50 mM
NaH2PO4, 300 mM NaCl, 10 mM imidazole, and 30% glycerol, pH 8, for His-tagged proteins) or wash
buffer for cation exchange chromatography (CIEX) (50 mM Na2HPO4, 50 mM NaCl, and 20% glycerol, pH
7.4, for proteins without His tag), and cells were disrupted by one passage through a Stansted Fluid
Power pressure cell homogenizer at 100 MPa. His-tagged proteins were purified by IMAC as described
before (20), using low-density nickel resin (ABT, Madrid, Spain), and eluted in IMAC elution buffer
(50 mM NaH2PO4, 300 mM NaCl, 250 mM imidazole, and 30% glycerol, pH 8). Proteins without His tags
were purified by CIEX as previously described (31), using a HiTrap Sepharose fast-flow (SP-FF) column on
an Äkta fast-performance liquid chromatography (FPLC) device (GE Healthcare, Uppsala, Sweden), and
eluted with a 1%/min gradient of CIEX elution buffer (50 mM Na2HPO4, 1 M NaCl, and 20% glycerol, pH
7.4). For in vitro experiments, proteins were dialyzed against phosphate-buffered saline with 0.01%
Tween 20 (PBS-T). Dialyzed proteins were filter sterilized (0.2 m m), protein identity and purity evaluated
by SDS-PAGE, and protein concentration measured with a NanoDrop ND-1000 spectrophotometer
(NanoDrop Technologies, Wilmington, DE, USA). When proteins were produced for in vivo experiments,
the purification procedure described above was modified to yield endotoxin-free preparations, essen-
tially as previously described (20). Endotoxin concentrations were determined by using an EndoZyme kit
(Hyglos, Regensburg, Germany) according to the manufacturer’s instructions. For experiments with
reconstituted human epidermis and the murine skin infection model, XZ.700 was formulated as cetoma-
crogol-based oil-in-water cream (38). Additionally, a methylhydroxypropylcellulose-based gel formula-
tion of XZ.700 was used for the in vivo model (15% methylhydroxypropylcellulose).
TRAs. TRAs were performed essentially as previously described (30). In brief, frozen S. aureus cells
were thawed and diluted in PBS-T (pH 7.4), and aliquots of the suspension were mixed in a 96-well plate
with endolysin dilutions in the same buffer, with final concentrations ranging from 12.5 to 200 nM, so
that the initial OD600 of the suspensions was 1.0. Buffer without enzyme served as a negative control.
The decrease in optical density over time in each well was monitored for 30 min at 10-s intervals using a
FLUOstar Omega plate reader (BMG Labtech, Ortenberg, Germany). The resulting lysis curves were

May 2022 Volume 66 Issue 5 10.1128/aac.02273-21 14

Staphylococcus aureus-Specific Chimeric Endolysins Antimicrobial Agents and Chemotherapy

normalized, corrected for the no-enzyme control, and fitted to a sigmoidal function with 5 parameters
as described before (73). The lytic activity of the endolysin under the conditions tested was calculated
from the steepest slope of the fitted curve. To determine the temperature stability of endolysins, pro-
teins at a concentration of 200 nM were heated to temperatures between 42°C and 72°C for 10 min,
cooled down on ice, and then tested in a TRA as described above, in comparison with a nonheated
control.
qKAs. To quantify the killing activity of endolysins, S. aureus was grown for 120 min in TSB to an
OD600 of approximately 0.5 and harvested by centrifugation, and the pellet was resuspended in PBS-T to
a concentration of 107 CFU/mL. Aliquots (100 m L) of the suspension were mixed in 96-well plates with
100 m L each of serially diluted endolysins in PBS-T or buffer alone as a control. Plates were incubated at
37°C for 120 min. Then, 20-m L samples were removed from each well, serially diluted, and plated on TSB
agar plates for enumeration of CFUs after overnight incubation.
MIC assay. MICs of SA.100 and XZ.700 against various staphylococcal strains were determined
essentially as described before (74). In brief, 200 m L of an overnight culture of a staphylococcal strain
was transferred to 5 mL TSB and incubated for 4 h at 35°C. Cells were spun down and resuspended in
PBS to a density of 1 McFarland. This suspension was then diluted in cation-adjusted Mueller-Hinton
broth (CAMHB) to a bacterial concentration of 105 to 106 CFU/mL. Stock solutions of endolysins were 2-
fold serially diluted in CAMHB, and 25.6 m L of each enzyme dilution was mixed with 174.4 m L of bacterial
suspension in a 96-well plate. After overnight incubation at 35°C, bacterial growth was assessed visually
per well. The MIC was the lowest concentration at which no growth of the bacterium was observed.
Mixed culture transwell experiments. To determine the species-specific killing of S. aureus by XZ.700
in a mixed culture with S. epidermidis on a solid surface, both species were grown on 12-mm transwell cell cul-
ture inserts (Corning, Glendale, AZ, USA) with a 0.4-m m pore size. To this end, growth medium (1 M9 mini-
mum salts [Sigma-Aldrich], 0.5 g/L casein amino acids, 0.5 g/L glucose, 2 mM MgSO4, 0.1 mM CaCl2, 1 mM thi-
amine, 0.05 mM nicotinamide, and 6 g/L agar) was filled into individual wells of a 12-well plate and allowed
to solidify. Frozen stocks of S. aureus and S. epidermidis were diluted in minimal medium (1 M9 salts, 0.1 mM
CaCl2, 1 mM thiamine, and 0.05 mM nicotinamide) (adapted from reference 75) to an OD600 of 0.1, and then,
50 m L of each strain was added to 9.9 mL medium. Two hundred microliters of this diluted mixed suspension
were added to transwells (corresponding to approximately 104 CFU/cm2) and centrifuged for 5 min at
2,000  g to remove the liquid. The transwells were then inserted into agar-containing wells and incubated at
35°C in a humidity-controlled incubator. After 2 h of incubation, transwells were treated with 100 m g/mL
XZ.700 in 100 m L PBS-T or buffer alone as a control (placebo). Treatment was stopped after 1 h by centrifuga-
tion to remove all liquid from the transwells, after which they were returned to the agar-containing wells and
further incubated for 21 h at 35°C. At the end of the incubation period, bacteria were harvested from transwell
membranes by filling the transwells with 500 m L PBS-T and gently pipetting up and down. This wash step
was repeated with a further 500 m L PBS-T, and suspensions were pooled. The suspension was then serially
diluted and plated onto CASO agar plates, CASO agar plates supplemented with 1 m g/mL fusidic acid, and
CASO agar plates supplemented with 350 m g/mL potassium tellurite for enumeration of total bacteria, S. epi-
dermidis only, and S. aureus only, respectively.
Resistance induction assay. Induction of resistance against XZ.700 and mupirocin was evaluated in

Downloaded from https://journals.asm.org/journal/aac on 17 November 2023 by 111.95.125.135.

two MSSA and two MRSA strains in duplicate by passing 20 consecutive cycles of exposure to subinhibi-
tory concentrations (one-half MIC) of both agents and monitoring the MICs at every cycle. Every cycle
consisted of a separate MIC determination as described above, with XZ.700 starting at a range of 32 m g/
mL to 0.5 m g/mL in the first cycle and mupirocin at a range of 1,024 nM to 64 nM. Between every cycle,
the cells in the wells at one-half the MIC (the well with the highest XZ.700 or mupirocin concentration
still allowing visible growth) were transferred to 5 mL TSB, incubated for 4 h at 35°C, and harvested by
centrifugation as described above. The pelleted cells served as inoculum for the next cycle of MICs. The
fold change of MICs of XZ.700 and mupirocin compared to the respective MICs at the beginning of the
experiment was plotted for each round of repeated exposure to visualize possible emergence of resist-
ance. All strains used for this experiment were sensitive to mupirocin, as determined by the Vitek 2 sys-
tem (bioMérieux, France) following the EUCAST standards (MICs # 2 m g/mL).
Activity of XZ.700 in an RHE model. Reconstituted human epidermis (RHE; SkinEthic) samples for
colonization with S. aureus and treatment with XZ.700 in solution and cream formulations were obtained
from Episkin (Lyon, France). Colonization and treatment experiments were performed by VitroScreen
(Milan, Italy). RHE samples (0.5 cm2) were cultivated for 17 days from normal human keratinocytes
(Episkin) on inert 0.4-m m polycarbonate filters at the air-liquid interface in a chemically defined medium
(Episkin), a procedure reproducing epidermal morphology. Each RHE batch was tested for the absence
of HIV-1 and HIV-2, hepatitis B and hepatitis C viruses, and Mycoplasma. Immediately after arrival in the
laboratory, the RHE samples were removed from the nutrient solution and rapidly transferred to 6-well
plates previously filled with antibiotic-free maintenance medium (1 mL/well; Episkin) at room tempera-
ture and incubated overnight at 35°C, 5% CO2, and relative humidity of 90%.
An overnight culture of S. aureus strain ATCC 33591 in BHI broth was diluted to an OD600 of 0.1 and
further incubated at 35°C for 2 h to reach the exponential-growth phase. Bacteria were then harvested
by centrifugation and resuspended in a 0.9% saline solution at an OD600 of 0.1. The CFU/mL of this sus-
pension was determined by serial dilution plating prior to using it for colonization of the RHE.
The maintenance medium of the RHE was exchanged, and 30 m L of S. aureus suspension was added,
resulting in a bacterial concentration of approximately 5  106 CFU/mL. Following incubation at 35°C for
4 h, all liquid was removed, and the colonized RHE was treated with 15 m L of XZ.700 solution at 32 m g/
mL or 15 m L of XZ.700 cream at 3 different concentrations (32, 128, and 512 m g/mL) for either 30 min,
1 h, or 2 h. Untreated RHE served as a control. At the end of the treatment, 600 m L of medium was used

May 2022 Volume 66 Issue 5 10.1128/aac.02273-21 15

Staphylococcus aureus-Specific Chimeric Endolysins Antimicrobial Agents and Chemotherapy

to rinse the RHE surface by repeatedly pipetting up and down and subsequently collected to recover
any nonadhered bacteria (apical fraction). The RHE tissue retaining any adhered bacteria was then col-
lected in a tube containing 1.5 mL buffer solution and zirconium oxide beads and homogenized using a
Minilys homogenizer (Bertin Corp., Rockville, MD, USA; 3 cycles of 30 s each at the lowest power).
Bacterial concentrations in both apical fraction and tissue homogenate were determined by serial dilu-
tion plating on BHI agar.
Mouse model of superficial skin infection. Six- to 8-week-old female and male BALB/c mice
(Charles River Laboratories, Wilmington, MA, USA) were housed in a pathogen-free environment on an
ad libitum diet. All procedures performed on the mice were in accordance with the Animal Care and Use
Committee of the Boston Children's Hospital.
The preparation of bacteria and infection of mice were carried out essentially as previously described
(76). On the day of infection, an overnight culture of the bioluminescent MRSA strain USA300 LAC::lux
was diluted 1:50 and incubated for another 2 h at 37°C. Bacteria were harvested and concentrated in
PBS to approximately 2  109 CFU/mL. The exact bacterial concentration was determined by serial dilu-
tion plating. Female or male mice were anesthetized with a mixture of ketamine and xylazine (60 to
100 mg/kg and 5 to 10 mg/kg body weight, respectively), and the back skin was shaved and tape
stripped with Tegaderm 6 times. After 18 h, mice were anesthetized with a mixture of ketamine and
xylazine and topically treated with 50 m L (1  108 CFU) of the prepared S. aureus suspension on the
tape-stripped skin with the help of a cotton swab.
In a preliminary study, the bacterial burden on female and male mice (n = 3 for each group) was
determined over time. For this purpose, in vivo bioluminescence imaging was performed at different
time points (8 h, 23 h, 31 h, 47 h, 56 h, 72 h, and 78 h) using a Pearl Trilogy small animal imaging system
(LI-COR, Lincoln, NE, USA) (77). In vivo bioluminescence imaging data were presented on a color scale
overlaid on a grayscale photograph of mice and quantified as total bioluminescence for a circular region
(with background subtracted) using the Image Studio software (LI-COR). To enumerate the bacterial
load on the skin at the end of the experiment, two 8-mm skin biopsy specimens were obtained. After
mechanical homogenization using a Bio-Gen PRO200 homogenizer (Biogen, Cambridge, MA, USA) at
15,000 rpm for 1 min, serial dilutions of skin homogenates were cultured on CHROMagar plates.
For the treatment study, female mice were anesthetized and infected as described above in two inde-
pendent experiments. Three hours postinfection, the infected skin was left untreated or treated with vehicle
cream or gel (negative controls) or cream or gel containing XZ.700 at 30 m g/mL with the help of a cotton
swab. The quantity of product applied onto the infected skin was calculated by the gravimetric difference
between the cotton swab saturated with cream or gel before and after the treatment. Following the first treat-
ment, further treatments were done twice per day at 9 a.m. at 5 p.m. for 3 days. The mice received 6 treat-
ments in total. Before each treatment, the bacterial load on the infected skin was evaluated by in vivo biolumi-
nescence imaging, and at the end of the experiment, S. aureus concentrations on the skin were determined,
both as described above. Altogether, we conducted two separate experiments with 3 mice per group each
(i.e., n = 6). The statistical power was 90%, with a significance level (a) of 0.05 (https://clincalc.com/).

SUPPLEMENTAL MATERIAL

Downloaded from https://journals.asm.org/journal/aac on 17 November 2023 by 111.95.125.135.

Supplemental material is available online only.
SUPPLEMENTAL FILE 1, PDF file, 1 MB.

ACKNOWLEDGMENTS
J.M.L.-C. and R.S.G. received support from the National Institutes of Health/National
Institute of Allergy and Infectious Diseases Atopic Dermatitis Research Network grant
U19AI1176.3. Furthermore, this work was supported by funds from Micreos.
Conceptualization, F.E., B.L.H., J.F., B.d.R., M.J.L., and M.S.; methodology, F.E., P.B.,
M.v.d.Z, J.M., F.J., and L.S.; formal analysis, F.E., B.L.H., J.M.L.-C., R.S.G., M.v.d.Z, B.d.R., J.F.,
F.J., and M.S.; supervision, F.E., B.L.H., B.d.R., C.R., J.F., M.O., M.J.L., and M.S; funding
acquisition, J.M.L.-C. and R.S.G.; resources, M.O. and M.J.L.; writing – original draft
preparation, F.E. and M.S.; writing – review and editing, all authors.
M.J.L. is an advisor for Micreos, a company producing phage-based products. F.E.,
M.v.d.Z, J.M., F.J., B.d.R., L.S., C.R., J.F., M.O., and M.S. are employees of Micreos. All other
authors declare no conflict of interest.

REFERENCES
1. Kluytmans J, Van Belkum A, Verbrugh H. 1997. Nasal carriage of Staphylococ- 3. Lowy FD. 1998. Staphylococcus aureus infections. N Engl J Med 339:
cus aureus: epidemiology, underlying mechanisms, and associated risks. Clin 520–532. https://doi.org/10.1056/NEJM199808203390806.
Microbiol Rev 10:505–520. https://doi.org/10.1128/CMR.10.3.505. 4. Fraunholz M, Sinha B. 2012. Intracellular Staphylococcus aureus: live-in
2. Tong SY, Davis JS, Eichenberger E, Holland TL, Fowler VG, Jr. 2015. Staphy- and let die. Front Cell Infect Microbiol 2:43. https://doi.org/10.3389/fcimb
lococcus aureus infections: epidemiology, pathophysiology, clinical mani- .2012.00043.
festations, and management. Clin Microbiol Rev 28:603–661. https://doi 5. Leimer N, Rachmuhl C, Palheiros Marques M, Bahlmann AS, Furrer A,
.org/10.1128/CMR.00134-14. Eichenseher F, Seidl K, Matt U, Loessner MJ, Schuepbach RA, Zinkernagel

May 2022 Volume 66 Issue 5 10.1128/aac.02273-21 16

Staphylococcus aureus-Specific Chimeric Endolysins Antimicrobial Agents and Chemotherapy

AS. 2016. Nonstable Staphylococcus aureus small-colony variants are 25. Becker SC, Swift S, Korobova O, Schischkova N, Kopylov P, Donovan DM,
induced by low pH and sensitized to antimicrobial therapy by phagolyso- Abaev I. 2015. Lytic activity of the staphylolytic Twort phage endolysin
somal alkalinization. J Infect Dis 213:305–313. https://doi.org/10.1093/ CHAP domain is enhanced by the SH3b cell wall binding domain. FEMS
infdis/jiv388. Microbiol Lett 362:1–8. https://doi.org/10.1093/femsle/fnu019.
6. Otto M. 2008. Staphylococcal biofilms. Curr Top Microbiol Immunol 322: 26. Abaev I, Foster-Frey J, Korobova O, Shishkova N, Kiseleva N, Kopylov P,
207–228. https://doi.org/10.1007/978-3-540-75418-3_10. Pryamchuk S, Schmelcher M, Becker SC, Donovan DM. 2013. Staphylococ-
7. Chambers HF, Deleo FR. 2009. Waves of resistance: Staphylococcus aureus cal phage 2638A endolysin is lytic for Staphylococcus aureus and harbors
in the antibiotic era. Nat Rev Microbiol 7:629–641. https://doi.org/10 an inter-lytic-domain secondary translational start site. Appl Microbiol
.1038/nrmicro2200. Biotechnol 97:3449–3456. https://doi.org/10.1007/s00253-012-4252-4.
8. Silverberg JI. 2017. Public health burden and epidemiology of atopic dermati- 27. Schindler CA, Schuhardt VT. 1964. Lysostaphin: a new bacteriolytic agent
tis. Dermatol Clin 35:283–289. https://doi.org/10.1016/j.det.2017.02.002. for the Staphylococcus. Proc Natl Acad Sci U S A 51:414–421. https://doi
9. Lewis-Jones S. 2006. Quality of life and childhood atopic dermatitis: the .org/10.1073/pnas.51.3.414.
misery of living with childhood eczema. Int J Clin Pract 60:984–992. 28. Schleifer KH, Kandler O. 1972. Peptidoglycan types of bacterial cell walls
https://doi.org/10.1111/j.1742-1241.2006.01047.x. and their taxonomic implications. Bacteriol Rev 36:407–477. https://doi
10. Geoghegan JA, Irvine AD, Foster TJ. 2018. Staphylococcus aureus and .org/10.1128/br.36.4.407-477.1972.
atopic dermatitis: a complex and evolving relationship. Trends Microbiol 29. Zygmunt WA, Browder HP, Tavormina PA. 1968. Susceptibility of coagu-
26:484–497. https://doi.org/10.1016/j.tim.2017.11.008. lase-negative staphylococci to lysostaphin and other antibiotics. Appl
11. Totte J, de Wit J, Pardo L, Schuren F, van Doorn M, Pasmans S. 2017. Tar- Microbiol 16:1168–1173. https://doi.org/10.1128/am.16.8.1168-1173.1968.
geted anti-staphylococcal therapy with endolysins in atopic dermatitis 30. Romero P, Bartual SG, Schmelcher M, Glück C, Hermoso JA, Loessner MJ.
and the effect on steroid use, disease severity and the microbiome: study 2018. Structural insights into the binding and catalytic mechanisms of
protocol for a randomized controlled trial (MAAS trial). Trials 18. https:// the Listeria monocytogenes bacteriophage glycosyl hydrolase PlyP40. Mol
doi.org/10.1186/s13063-017-2118-x. Microbiol 108:128–142. https://doi.org/10.1111/mmi.13922.
12. de Wit J, Totte JEE, van Mierlo MMF, van Veldhuizen J, van Doorn MBA, 31. Röhrig C, Huemer M, Lorgé D, Luterbacher S, Phothaworn P, Schefer C,
Schuren FHJ, Willemsen SP, Pardo LM, Pasmans S. 2019. Endolysin treat- Sobieraj AM, Zinsli LV, Mairpady Shambat S, Leimer N, Keller AP,
ment against Staphylococcus aureus in adults with atopic dermatitis: a Eichenseher F, Shen Y, Korbsrisate S, Zinkernagel AS, Loessner MJ,
randomized controlled trial. J Allergy Clin Immunol 144:860–863. https:// Schmelcher M. 2020. Targeting hidden pathogens: cell-penetrating enzy-
doi.org/10.1016/j.jaci.2019.05.020. biotics eradicate intracellular drug-resistant Staphylococcus aureus. mBio
13. Alexander H, Paller AS, Traidl-Hoffmann C, Beck LA, De Benedetto A, Dhar 11:e00209-20. https://doi.org/10.1128/mBio.00209-20.
S, Girolomoni G, Irvine AD, Spuls P, Su J, Thyssen JP, Vestergaard C, Werfel 32. Sobieraj AM, Huemer M, Zinsli LV, Meile S, Keller AP, Röhrig C, Eichenseher
T, Wollenberg A, Deleuran M, Flohr C. 2020. The role of bacterial skin F, Shen Y, Zinkernagel AS, Loessner MJ, Schmelcher M. 2020. Engineering
infections in atopic dermatitis: expert statement and review from the of long-circulating peptidoglycan hydrolases enables efficient treatment of
systemic Staphylococcus aureus infection. mBio 11:e01781-20. https://doi
International Eczema Council Skin Infection Group. Br J Dermatol 182:
.org/10.1128/mBio.01781-20.
1331–1342. https://doi.org/10.1111/bjd.18643.
33. Verbree CT, Dätwyler SM, Meile S, Eichenseher F, Donovan DM, Loessner
14. Niebuhr M, Mai U, Kapp A, Werfel T. 2008. Antibiotic treatment of cutane-
MJ, Schmelcher M. 2018. Corrected and republished from: identification
ous infections with Staphylococcus aureus in patients with atopic dermati-
of peptidoglycan hydrolase constructs with synergistic staphylolytic ac-
tis: current antimicrobial resistances and susceptibilities. Exp Dermatol
tivity in cow's milk. Appl Environ Microbiol 84:e02134-17. https://doi.org/
17:953–957. https://doi.org/10.1111/j.1600-0625.2008.00734.x.
10.1128/AEM.02134-17.
15. Young R. 1992. Bacteriophage lysis: mechanism and regulation. Microbiol
34. Maali Y, Badiou C, Martins-Simoes P, Hodille E, Bes M, Vandenesch F, Lina G,
Rev 56:430–481. https://doi.org/10.1128/mr.56.3.430-481.1992.
Diot A, Laurent F, Trouillet-Assant S. 2018. Understanding the virulence of
16. Schmelcher M, Donovan DM, Loessner MJ. 2012. Bacteriophage endoly-
Staphylococcus pseudintermedius: a major role of pore-forming toxins. Front
sins as novel antimicrobials. Future Microbiol 7:1147–1171. https://doi
Cell Infect Microbiol 8:221. https://doi.org/10.3389/fcimb.2018.00221.
.org/10.2217/fmb.12.97.

Downloaded from https://journals.asm.org/journal/aac on 17 November 2023 by 111.95.125.135.

35. Franca A, Gaio V, Lopes N, Melo LDR. 2021. Virulence factors in coagu-
17. Nelson D, Schuch R, Chahales P, Zhu S, Fischetti VA. 2006. PlyC: a multi-
lase-negative staphylococci. Pathogens 10:170. https://doi.org/10.3390/
meric bacteriophage lysin. Proc Natl Acad Sci U S A 103:10765–10770.
pathogens10020170.
https://doi.org/10.1073/pnas.0604521103. 36. Gonzalez-Martin M, Corbera JA, Suarez-Bonnet A, Tejedor-Junco MT.
18. Loessner MJ, Wendlinger G, Scherer S. 1995. Heterogeneous endolysins in Lis- 2020. Virulence factors in coagulase-positive staphylococci of veterinary
teria monocytogenes bacteriophages: a new class of enzymes and evidence interest other than Staphylococcus aureus. Vet Q 40:118–131. https://doi
for conserved holin genes within the siphoviral lysis cassettes. Mol Microbiol .org/10.1080/01652176.2020.1748253.
16:1231–1241. https://doi.org/10.1111/j.1365-2958.1995.tb02345.x. 37. Centers for Disease Control and Prevention. 2019. Antibiotic resistance
19. Gutierrez D, Ruas-Madiedo P, Martinez B, Rodriguez A, Garcia P. 2014. threats in the United States, 2019. U.S. Department of Health and Human
Effective removal of staphylococcal biofilms by the endolysin LysH5. PLoS Services, Centers for Disease Control and Prevention, Atlanta, GA.
One 9:e107307. https://doi.org/10.1371/journal.pone.0107307. 38. Totte JEE, van Doorn MB, Pasmans S. 2017. Successful treatment of
20. Schmelcher M, Shen Y, Nelson DC, Eugster MR, Eichenseher F, Hanke DC, chronic Staphylococcus aureus-related dermatoses with the topical endo-
Loessner MJ, Dong S, Pritchard DG, Lee JC, Becker SC, Foster-Frey J, lysin Staphefekt SA.100: a report of 3 cases. Case Rep Dermatol 9:19–25.
Donovan DM. 2015. Evolutionarily distinct bacteriophage endolysins fea- https://doi.org/10.1159/000473872.
turing conserved peptidoglycan cleavage sites protect mice from MRSA 39. Moreau M, Seite S, Aguilar L, Da Cruz O, Puech J, Frieling J, Demessant A.
infection. J Antimicrob Chemother 70:1453–1465. https://doi.org/10.1093/ 2021. Topical S. aureus-targeting endolysin significantly improves symp-
jac/dku552. toms and QoL in individuals with atopic dermatitis. J Drugs Dermatol 20:
21. Schuch R, Nelson D, Fischetti VA. 2002. A bacteriolytic agent that detects 1323–1328. https://doi.org/10.36849/jdd.6363.
and kills Bacillus anthracis. Nature 418:884–889. https://doi.org/10.1038/ 40. Gründling A, Missiakas DM, Schneewind O. 2006. Staphylococcus aureus
nature01026. mutants with increased lysostaphin resistance. J Bacteriol 188:6286–6297.
22. Schmelcher M, Loessner MJ. 2021. Bacteriophage endolysins—extending https://doi.org/10.1128/JB.00457-06.
their application to tissues and the bloodstream. Curr Opin Biotechnol 41. Becker SC, Roach DR, Chauhan VS, Shen Y, Foster-Frey J, Powell AM,
68:51–59. https://doi.org/10.1016/j.copbio.2020.09.012. Bauchan G, Lease RA, Mohammadi H, Harty WJ, Simmons C, Schmelcher
23. Becker SC, Dong S, Baker JR, Foster-Frey J, Pritchard DG, Donovan DM. M, Camp M, Dong S, Baker JR, Sheen TR, Doran KS, Pritchard DG, Almeida
2009. LysK CHAP endopeptidase domain is required for lysis of live staph- RA, Nelson DC, Marriott I, Lee JC, Donovan DM. 2016. Triple-acting lytic
ylococcal cells. FEMS Microbiol Lett 294:52–60. https://doi.org/10.1111/j enzyme treatment of drug-resistant and intracellular Staphylococcus aur-
.1574-6968.2009.01541.x. eus. Sci Rep 6:25063. https://doi.org/10.1038/srep25063.
24. Schmelcher M, Korobova O, Schischkova N, Kiseleva N, Kopylov P, 42. Gründling A, Schneewind O. 2006. Cross-linked peptidoglycan mediates
Pryamchuk S, Donovan DM, Abaev I. 2012. Staphylococcus haemolyticus lysostaphin binding to the cell wall envelope of Staphylococcus aureus. J Bac-
prophage PhiSH2 endolysin relies on cysteine, histidine-dependent ami- teriol 188:2463–2472. https://doi.org/10.1128/JB.188.7.2463-2472.2006.
dohydrolases/peptidases activity for lysis 'from without.' J Biotechnol 43. Mitkowski P, Jagielska E, Nowak E, Bujnicki JM, Stefaniak F, Niedziałek D,
162:289–298. https://doi.org/10.1016/j.jbiotec.2012.09.010. Bochtler M, Sabała I. 2019. Structural bases of peptidoglycan recognition

May 2022 Volume 66 Issue 5 10.1128/aac.02273-21 17

Staphylococcus aureus-Specific Chimeric Endolysins Antimicrobial Agents and Chemotherapy

by lysostaphin SH3b domain. Sci Rep 9:5965. https://doi.org/10.1038/ 61. Defraine V, Schuermans J, Grymonprez B, Govers SK, Aertsen A, Fauvart
s41598-019-42435-z. M, Michiels J, Lavigne R, Briers Y. 2016. Efficacy of artilysin Art-175 against
44. Gonzalez-Delgado LS, Walters-Morgan H, Salamaga B, Robertson AJ, resistant and persistent Acinetobacter baumannii. Antimicrob Agents Che-
Hounslow AM, Jagielska E, Sabała I, Williamson MP, Lovering AL, Mesnage mother 60:3480–3488. https://doi.org/10.1128/AAC.00285-16.
S. 2020. Two-site recognition of Staphylococcus aureus peptidoglycan by 62. Gordon S, Daneshian M, Bouwstra J, Caloni F, Constant S, Davies DE,
lysostaphin SH3b. Nat Chem Biol 16:24–30. https://doi.org/10.1038/ Dandekar G, Guzman CA, Fabian E, Haltner E, Hartung T, Hasiwa N, Hayden
s41589-019-0393-4. P, Kandarova H, Khare S, Krug HF, Kneuer C, Leist M, Lian G, Marx U,
45. Becker SC, Foster-Frey J, Stodola AJ, Anacker D, Donovan DM. 2009. Dif- Metzger M, Ott K, Prieto P, Roberts MS, Roggen EL, Tralau T, van den Braak
ferentially conserved staphylococcal SH3b_5 cell wall binding domains C, Walles H, Lehr CM. 2015. Non-animal models of epithelial barriers (skin,
confer increased staphylolytic and streptolytic activity to a streptococcal intestine and lung) in research, industrial applications and regulatory toxi-
prophage endolysin domain. Gene 443:32–41. https://doi.org/10.1016/j cology. ALTEX 32:327–378. https://doi.org/10.14573/altex.1510051.
.gene.2009.04.023. 63. Zuang V, Asturiol Bofill D, Viegas Barroso J, Belz S, Berggren E, Bernasconi
46. Haddad KH, Schmelcher M, Sabzalipoor H, Seyed HE, Moniri R. 2018. C, Bopp S, Bostroem A, Bouhifd M, Casati S, Coecke S, Cole T, Corvi R,
Recombinant endolysins as potential therapeutics against antibiotic-re-
Halder M, Holley T, Horvat T, Janusch Roi A, Kienzler A, Landesmann B,
sistant Staphylococcus aureus: current status of research and novel deliv-
Madia F, Milcamps A, Munn S, Paini A, Palosaari T, Price A, Prieto PM,
ery strategies. Clin Microbiol Rev 31:e00071-17. https://doi.org/10.1128/
Richarz A, Triebe J, Wittwehr C, Worth A, Whelan M. 2016. EURL ECVAM
CMR.00071-17.
status report on the development, validation and regulatory acceptance
47. Williams MR, Nakatsuji T, Gallo RL. 2017. Staphylococcus aureus: master
manipulator of the skin. Cell Host Microbe 22:579–581. https://doi.org/10 of alternative methods and approaches. Publications Office of the Euro-
.1016/j.chom.2017.10.015. pean Union, Luxembourg.
48. Sabate BM, Harris LG, Thompson K, Stanic B, Morgenstern M, O'Mahony L, 64. Castleman MJ, Pokhrel S, Triplett KD, Kusewitt DF, Elmore BO, Joyner JA,
Richards RG, Moriarty TF. 2017. Pathogenic mechanisms and host interac- Femling JK, Sharma G, Hathaway HJ, Prossnitz ER, Hall PR. 2018. Innate
tions in Staphylococcus epidermidis device-related infection. Front Micro- sex bias of Staphylococcus aureus skin infection is driven by alpha-hemol-
biol 8:1401. https://doi.org/10.3389/fmicb.2017.01401. ysin. J Immunol 200:657–668. https://doi.org/10.4049/jimmunol.1700810.
49. Pastar I, O'Neill K, Padula L, Head CR, Burgess JL, Chen V, Garcia D, 65. Yanke SJ, Olson ME, Davies HD, Hart DA. 2000. A CD-1 mouse model of
Stojadinovic O, Hower S, Plano GV, Thaller SR, Tomic-Canic M, Strbo N. infection with Staphylococcus aureus: influence of gender on infection
2020. Staphylococcus epidermidis boosts innate immune response by acti- with MRSA and MSSA isolates. Can J Microbiol 46:920–926.
vation of gamma delta T cells and induction of perforin-2 in human skin. 66. Zomer HD, Trentin AG. 2018. Skin wound healing in humans and mice:
Front Immunol 11:550946. https://doi.org/10.3389/fimmu.2020.550946. challenges in translational research. J Dermatol Sci 90:3–12. https://doi
50. Nakatsuji T, Cheng JY, Gallo RL. 2021. Mechanisms for control of skin .org/10.1016/j.jdermsci.2017.12.009.
immune function by the microbiome. Curr Opin Immunol 72:324–330. 67. Brackman G, Coenye T. 2016. In vitro and in vivo biofilm wound models
https://doi.org/10.1016/j.coi.2021.09.001. and their application. Adv Exp Med Biol 897:15–32. https://doi.org/10
51. Lavigne R, Briers Y, Hertveldt K, Robben J, Volckaert G. 2004. Identification .1007/5584_2015_5002.
and characterization of a highly thermostable bacteriophage lysozyme. 68. Iordanescu S, Surdeanu M. 1976. Two restriction and modification sys-
Cell Mol Life Sci 61:2753–2759. https://doi.org/10.1007/s00018-004-4301-y. tems in Staphylococcus aureus NCTC8325. J Gen Microbiol 96:277–281.
52. Schmelcher M, Waldherr F, Loessner MJ. 2012. Listeria bacteriophage pepti- https://doi.org/10.1099/00221287-96-2-277.
doglycan hydrolases feature high thermoresistance and reveal increased 69. Thurlow LR, Hanke ML, Fritz T, Angle A, Aldrich A, Williams SH,
activity after divalent metal cation substitution. Appl Microbiol Biotechnol Engebretsen IL, Bayles KW, Horswill AR, Kielian T. 2011. Staphylococcus
93:633–643. https://doi.org/10.1007/s00253-011-3372-6. aureus biofilms prevent macrophage phagocytosis and attenuate inflam-
53. Kusuma C, Kokai-Kun J. 2005. Comparison of four methods for determin- mation in vivo. J Immunol 186:6585–6596. https://doi.org/10.4049/
ing lysostaphin susceptibility of various strains of Staphylococcus aureus. jimmunol.1002794.
Antimicrob Agents Chemother 49:3256–3263. https://doi.org/10.1128/ 70. Sambrook J, Fritsch EF, Maniatis T. 1989. Molecular cloning: a laboratory
AAC.49.8.3256-3263.2005.

Downloaded from https://journals.asm.org/journal/aac on 17 November 2023 by 111.95.125.135.

manual, vol 2. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY.
54. Hermoso JA, Monterroso B, Albert A, Galan B, Ahrazem O, Garcia P, Martinez- 71. Kwan T, Liu J, DuBow M, Gros P, Pelletier J. 2005. The complete genomes
Ripoll M, Garcia JL, Menendez M. 2003. Structural basis for selective recogni- and proteomes of 27 Staphylococcus aureus bacteriophages. Proc Natl
tion of pneumococcal cell wall by modular endolysin from phage Cp-1. Acad Sci U S A 102:5174–5179. https://doi.org/10.1073/pnas.0501140102.
Structure 11:1239–1249. https://doi.org/10.1016/j.str.2003.09.005. 72. Thumm G, Götz F. 1997. Studies on prolysostaphin processing and char-
55. Schmelcher M, Tchang VS, Loessner MJ. 2011. Domain shuffling and mod- acterization of the lysostaphin immunity factor (Lif) of Staphylococcus
ule engineering of Listeria phage endolysins for enhanced lytic activity simulans biovar staphylolyticus. Mol Microbiol 23:1251–1265. https://doi
and binding affinity. Microb Biotechnol 4:651–662. https://doi.org/10
.org/10.1046/j.1365-2958.1997.2911657.x.
.1111/j.1751-7915.2011.00263.x.
73. Korndörfer IP, Danzer J, Schmelcher M, Zimmer M, Skerra A, Loessner MJ.
56. Kuiper JWP, Hogervorst JMA, Herpers BL, Bakker AD, Klein-Nulend J, Nolte PA,
2006. The crystal structure of the bacteriophage PSA endolysin reveals a
Krom BP. 2021. The novel endolysin XZ.700 effectively treats MRSA biofilms
unique fold responsible for specific recognition of Listeria cell walls. J Mol
in two biofilm models without showing toxicity on human bone cells in vitro.
Biol 364:678–689. https://doi.org/10.1016/j.jmb.2006.08.069.
Biofouling 37:184–193. https://doi.org/10.1080/08927014.2021.1887151.
74. Wiegand I, Hilpert K, Hancock RE. 2008. Agar and broth dilution methods to
57. Sass P, Bierbaum G. 2007. Lytic activity of recombinant bacteriophage
phi11 and phi12 endolysins on whole cells and biofilms of Staphylococcus determine the minimal inhibitory concentration (MIC) of antimicrobial sub-
aureus. Appl Environ Microbiol 73:347–352. https://doi.org/10.1128/AEM stances. Nat Protoc 3:163–175. https://doi.org/10.1038/nprot.2007.521.
.01616-06. 75. Reed P, Atilano ML, Alves R, Hoiczyk E, Sher X, Reichmann NT, Pereira PM,
58. Olsen NMC, Thiran E, Hasler T, Vanzieleghem T, Belibasakis GN, Mahillon Roemer T, Filipe SR, Pereira-Leal JB, Ligoxygakis P, Pinho MG. 2015. Staph-
J, Loessner MJ, Schmelcher M. 2018. Synergistic removal of static and ylococcus aureus survives with a minimal peptidoglycan synthesis
dynamic Staphylococcus aureus biofilms by combined treatment with a machine but sacrifices virulence and antibiotic resistance. PLoS Pathog
bacteriophage endolysin and a polysaccharide depolymerase. Viruses 10: 11:e1004891. https://doi.org/10.1371/journal.ppat.1004891.
438. https://doi.org/10.3390/v10080438. 76. Malhotra N, Yoon J, Leyva-Castillo JM, Galand C, Archer N, Miller LS, Geha
59. Gutierrez D, Fernandez L, Rodriguez A, Garcia P. 2018. Are phage lytic pro- RS. 2016. IL-22 derived from gammadelta T cells restricts Staphylococcus
teins the secret weapon to kill Staphylococcus aureus? mBio 9:e01923-17. aureus infection of mechanically injured skin. J Allergy Clin Immunol 138:
https://doi.org/10.1128/mBio.01923-17. 1098–1107.e3. https://doi.org/10.1016/j.jaci.2016.07.001.
60. Kint CI, Verstraeten N, Fauvart M, Michiels J. 2012. New-found fundamen- 77. Leyva-Castillo JM, McGurk A, Geha MDR. 2020. Allergic skin inflammation
tals of bacterial persistence. Trends Microbiol 20:577–585. https://doi and S. aureus skin colonization are mutually reinforcing. Clin Immunol
.org/10.1016/j.tim.2012.08.009. 218:108511. https://doi.org/10.1016/j.clim.2020.108511.

May 2022 Volume 66 Issue 5 10.1128/aac.02273-21 18