Efficient and Selective Biosynthesis of a Precursor-Directed FK506 Analogue: Paving the Way for Click Chemistry

Abstract

The medically important immunosuppressant FK506 is a structurally complex macrolactone biosynthesized by a combined polyketide synthase and a nonribosomal peptide synthetase enzyme complex. Its acyltransferase domain 4 (AT4) selects an unusual extender unit, resulting in an allyl moiety on carbon 21 of the macrolactone backbone. Based on the AT4 domain, chemobiosynthetic processes have been developed that enable the introduction of diverse moieties at the carbon 21 position. However, the novel moieties that were introduced into the polyketide backbone are chemically inert. Reported here is a novel and efficient chemobiosynthetic approach that ensures high titer of an FK506 analogue containing a propargyl moiety. The novel FK506 analogue displays lower immunosuppression activity than FK506 with significantly reduced cytotoxicity. More importantly, the propargyl moiety contains a terminal alkyl group, which makes click chemistry reactions possible; this approach may potentially be translated to other medically important drugs of polyketide origin.

Full text

Ecient and Selective Biosynthesis of a Precursor-Directed FK506
Analogue: Paving the Way for Click Chemistry
Dusan Goranovic, Branko Jenko, Barbara Ramsak, Ajda Podgorsek Berke, Leon Bedrac, Jaka Horvat,
Martin Sala, Damjan Makuc, Guilhermina M. Carriche, Luana Silva, Aleksandra Lopez Krol,
Alen Psenicnik, María Beatriz Durán Alonso, Martina Avbelj, Stojan Stavber, Janez Plavec,
Tim Sparwasser, Rolf Muller, Gregor Kosec, Stefan Fujs, and Hrvoje Petkovic*
Cite This: J. Nat. Prod. 2025, 88, 619−630
Read Online
ACCESS Metrics & More Article Recommendations *
sı Supporting Information
ABSTRACT: The medically important immunosuppressant FK506 is
a structurally complex macrolactone biosynthesized by a combined
polyketide synthase and a nonribosomal peptide synthetase enzyme
complex. Its acyltransferase domain 4 (AT4) selects an unusual
extender unit, resulting in an allyl moiety on carbon 21 of the
macrolactone backbone. Based on the AT4 domain, chemo-
biosynthetic processes have been developed that enable the
introduction of diverse moieties at the carbon 21 position. However,
the novel moieties that were introduced into the polyketide backbone
are chemically inert. Reported here is a novel and ecient
chemobiosynthetic approach that ensures high titer of an FK506
analogue containing a propargyl moiety. The novel FK506 analogue
displays lower immunosuppression activity than FK506 with
significantly reduced cytotoxicity. More importantly, the propargyl moiety contains a terminal alkyl group, which makes click
chemistry reactions possible; this approach may potentially be translated to other medically important drugs of polyketide origin.
Polyketides (PKs) are a large group of biogenetically
related compounds with a diverse spectrum of activities;
they are biosynthesized by the large group of closely related
polyketide synthase (PKS) enzymes that display an enormous
structural diversity.
1
A particularly important PK group is
constituted by structurally closely related macrolactones, such
as rapamycin and the FKBP12-binding compounds FK506 and
FK520. These metabolites exhibit a broad spectrum of
pharmacological activities including immunomodulation, anti-
cancer, and neuroprotection properties.
2
Due to its powerful
and selective immunosuppression activity, FK506 has been
widely used to prevent organ rejection
3,4
and, to some extent,
in the treatment of inflammation-related conditions
5
such as
atopic dermatitis.
6
Despite their potent activity, PKs frequently require further
optimization by structural modification to improve their
pharmacokinetic and pharmacodynamic properties. This is
most often carried out by semisynthetic approaches; on the
other hand, these are often very dicult to carry out, due to
the structural complexity of natural products.
Precursor-directed chemobiosynthesis remains a valuable
and industrially important method for obtaining new analogues
of promising therapeutic molecules.
7
Chemobiosynthesis is
carried out by feeding a synthetic precursor to an engineered
strain that has been deprived of one of its naturally occurring
intermediates or that is unable to synthesize a building block
needed for the biosynthesis of the target metabolite. Instead of
the natural building block, synthetic precursors with dierent
structures are fed into the culture of an engineered strain
during the biosynthesis, thus resulting in the formation of a
structural analogue of the target natural product.
7
This
approach has been used, for example, for the large-scale
production of the antiparasitic PK doramectin, where
unnatural cyclohexane carboxylic acid is incorporated instead
of the natural starter unit.
8
A number of unnatural precursor-
derived chemobiosynthesis processes have also been developed
for the biosynthesis of dierent PKs, including FK506 and
rapamycin.
9−12
FK506 (Tacrolimus, Figure 1) is a macrocyclic polyketide
produced by a hybrid type I polyketide synthase (PKS)/
nonribosomal peptide synthetase (NRPS) system of Strepto-
myces tsukubaensis and some other Streptomyces species.
12
Received: April 5, 2024
Revised: February 1, 2025
Accepted: February 5, 2025
Published: March 10, 2025
Articlepubs.acs.org/jnp
© 2025 The Authors. Published by
American Chemical Society and
American Society of Pharmacognosy 619
https://doi.org/10.1021/acs.jnatprod.4c00394
J. Nat. Prod. 2025, 88, 619−630
This article is licensed under CC-BY 4.0
Downloaded via 187.147.180.219 on March 31, 2025 at 14:59:47 (UTC).
See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

Typically, the PKS complex type I is composed of modules
that contain domains with dierent activities, including
acyltransferase (AT), ketosynthase (KS) and acyl-carrier
protein (ACP), which are all obligatory domains present in
every functional extender module.
13
In the PKS type I
complex, such as that found in the FK506 biosynthetic gene
cluster (BGC) in Streptomyces tsukubaensis, acyltransferase
(AT) domains are responsible for the selection and
incorporation of simple monomeric building blocks. Extender
AT domains usually exhibit a strict specificity toward a single
α-carboxyacyl-CoA building block. Interestingly, however, AT
domains associated with crotonyl-CoA carboxylase/reductase
(CCR)-generated extender units can show relaxed specificity,
and thus frequently give rise to PK structure diversification.
14
This is also the case for the AT4 domain of the FK506 PKS,
which most often incorporates the unusual extender unit
allylmalonyl-CoA (Figure 1).
15,16
The selective introduction of novel extender units would
bring a much larger degree of PK diversification, and thus, the
relaxed specificity of the acyltransferase (AT) domain in a
typical PKS module would significantly increase the potential
for novel chemobiosynthetic processes. With a few exceptions,
AT domains remain stringent catalytic selectivity, predom-
inantly favoring malonyl-CoA and methylmalonyl-CoA, while
exhibiting lower preference for substrates such as ethylmalonyl-
CoA and methoxymalonyl-CoA. Rarely do AT domains display
selectivity for unusual extender units, including FK506
(allylmalonyl-CoA
16
) and stambomycin (hexanoyl-CoA
17
).
A feature of the AT4 domain from FkbB of FK506 PKS is its
selectivity to accept unusual allylmalonyl-CoA, propylmalonyl-
CoA, and ethylmalonyl-CoA extender units (Figure 1), but not
usual methylmalonyl-CoA and malonyl-CoA extender units.
This is the reason why the AT4 domain from the FK506 PKS/
NRPS complex, which shows relaxed specificity for unusual
acyl-CoA extender units, represents a very interesting model
system that has been the focus of a number of studies in the
past.
15,16
Following the identification of the origin of the unusual
extender unit allylmalonyl-CoA at position C21 of FK506,
16
chemobiosynthesis procedures
15,18
were developed by applying
N-acetylcysteamine ester (SNAC) precursors, which are
accepted by AT modules due to their similarity with native
coenzyme-A-activated extender units.
19,20
However, most of
the unnatural extender units (precursors) incorporated to
position C21 of FK506 have resulted in the introduction of
structural moieties which are very dicult to chemically
modify. In addition, the ecacy of these chemobiosynthesis
processes is often very low, hence generating very low titers of
the target products (often, in nanomolar quantities), which
make this kind of bioprocess technically and economically
unfavorable and it thus cannot be readily translated to the
industrial scale.
In this work, we have used the engineered strain of S.
tsukubaensis ΔallR
18
to develop a chemobiosynthesis process
for the ecient and highly selective incorporation of the
unnatural extender unit propargylmalonyl-SNAC. We have
also developed a method for the ecient synthesis of the
precursor propargylmalonyl-SNAC and have optimized an
industrially ecient chemobiosynthesis bioprocess for the
production of a novel FK506 analogue that contains a terminal
alkyne functional group at its C21 position. This technology is
potentially applicable to any other enzymatic system of type I
PKS, opening a new way for the generation of novel polyketide
derivatives containing alkyne functional groups and hence
paving the way for “click chemistry” approaches to the
development of novel PK natural products.
■RESULTS AND DISCUSSION
Use of the FK506-Producing ΔallR Strain, Where the
Production of the Allylmalonyl-CoA Extender Unit Is
Disrupted. We developed a chemobiosynthetic process where
we fed an unnatural α-carboxyacyl-CoA extender unit analogue
containing a terminal alkyl moiety propargylmalonyl-SNAC
[(S,S-bis(2-acetamidoethyl) 2-(prop-2-yn-1-yl) propanebis-
(thioate)] to cultures of a S. tsukubaensis ΔallR mutant strain
that has an inactivated allR gene. Since this gene encodes the
crotonyl-CoA carboxylase/reductase, a key enzyme involved in
the formation of ethylmalonyl-SCoA and allylmalonyl-SCoA,
18
these two natural extender units are not biosynthesised in this
strain. Therefore, the exclusive biosynthesis of a target FK506
analogue modified at the C21 position can only be achieved in
the S. tsukubaensis ΔallR strain when an unnatural extender
unit, fed during the chemo-biosynthetic process, is selected by
the AT4 domain.
A few C21-modified FK506 analogues were generated
through biosynthetic engineering or feeding procedures in
the past.
15
However, the titer achieved for these metabolites
was most often below 1 mg/L, which is not sucient for
thorough semisynthetic work, preclinical evaluations and any
potential economical transfer to the industrial scale.
11
Figure 1. Structure of FK506 and the propargyl-FK506 analogue, showing the atom numbering used in our study.
Journal of Natural Products pubs.acs.org/jnp Article
https://doi.org/10.1021/acs.jnatprod.4c00394
J. Nat. Prod. 2025, 88, 619−630
620

Therefore, in the scope of this work, and to ensure sucient
titer of the target product propargyl-FK506, we developed an
ecient precursor-directed chemobiosynthetic process with
the S. tsukubaensis ΔallR strain containing an in-frame deletion
of allR
18
(Figure 2), which produces neither FK506 nor
FK520. To achieve this, it was necessary to develop an ecient
procedure for the synthesis of the precursor of the unnatural
extender unit propargylmalonyl−S-N-acetylcysteamine (prop-
argylmalonyl-SNAC).
Synthesis of Propargylmalonyl-SNAC [S,S-Bis(2-
acetamidoethyl)2-(prop-2-yn-1-yl)propanebis(thioate)].
Click chemistry is a powerful tool in the area of synthetic
chemistry; it by-passes the need for laborious and often very
complex synthetic steps, followed by purification processes,
that are characteristic of conventional chemical synthesis.
21,22
First defined by Nobel laureate Sharpless and associates in
2001, click chemistry consists of a few stereospecific, modular
reactions with a high thermodynamic driving force that occur
in simple reaction conditions and preferably in an aqueous
environment. “Click reactions” often enable high yields of lead
compound. For example, among click chemistry reactions,
copper(I)-catalyzed azide alkyne cycloaddition is the most
used and versatile “click” reaction, which can be used to
derivatize the propargyl moiety at the C21 position of
FK506.
21,23−25
The alkyl moiety introduced into the PK
scaold would not only ensure a simple and straightforward
reactive moiety at the desired position in the PK backbone, but
would also simplify the entire semisynthetic procedure, by
avoiding the protection and deprotection steps that are due to
undesired o-target reactions.
26
We initiated the synthesis of the target extender unit
propargylmalonyl-SNAC from dimethyl 2-(prop-2-yn-1-yl)-
malonate and then hydrolyzed it to a malonic acid derivative
by aqueous NaOH (Step1, Figure 3,Figure S1). We further
transformed 2-(prop-2-yn-1-yl) malonic acid to the corre-
sponding malonyl chloride (Step 2, Figure 3,Figure S2) that
ultimately gave rise to the target compound S,S-bis(2-
acetamidoethyl) 2-(prop-2-yn-1-yl)propanebis(thioate), by
using N-acetylcisteamin (Step 3, Figure 3,Figure S3). The
detailed three-step procedure for the synthesis of propargyl-
malonyl-SNAC (S,S-bis(2-acetamidoethyl)2-(prop-2-yn-1-yl)-
propanebis(thioate)) was carried out as described in
Supporting Information (Step 3, Figure 3,Figure S4−S6).
Figure 2. Construction of strain S. tsukubaensis ΔallR. A) the deletion of allR in the FK506 biosynthetic gene cluster disrupts the production of
natural allylmalonyl-CoA in FK506, as described by Kosec et al.
18
The allA,allK,allR, and allD (A-K-R-D) gene products, part of the FK506
biosynthetic gene cluster, are involved in the biosynthesis of the allylmalomyl-CoA extender unit, as presented in panel B. B) Proposed biosynthesis
pathway of unusual extender unit allylmalonyl used by the AT4 domain of FK506 PKS synthase.
15,16
Propyl-SCoA is proposed as the starter unit
and malonyl-SCoA as the extender unit, selected by the KS and AT domains of AllA, respectively, resulting in the C5 carbon unit after
decarboxylative condensation carried out by the KS domain. It seems that no other putative candidate reductase and dehydratase genes involved in
the formation of the double bond are present in the FK506 gene cluster. Other ubiquitous enzymes such as fatty acids synthase (FAS) which are
encoded in other regions of the chromosome in S. tsukubaensis likely carry out these reactions. CCR-homologue 2-pentenoyl-CoA carboxylase/
reductase (allR) catalyzes the reductive carboxylation of α,β-unsaturated 5 carbon 2-pentenoyl-ACP substrate followed by the final
dehydrogenation of acyl-ACP dehydrogenase by AllD, resulting in allylmalonyl-ACP.
Figure 3. Schematic presentation of the three-step synthesis of
propargylmalonyl-SNAC [(S,S-bis(2-acetamidoethyl) 2-(prop-2-yn-1-
yl)propanebis(thioate)]. The chemical structures of the intermediates
and the final target product propargylmalonyl-SNAC [(S,S-bis(2-
acetamidoethyl)2-(prop-2-yn-1-yl)propanebis(thioate)] were con-
firmed by 1H NMR analysis. Finally, the structure of the final
product propargylmalonyl-SNAC was reconfirmed by 13C NMR
spectra (Figure S4) and MS analysis (Figure S5). The high purity of
propargylmalonyl-SNAC was confirmed by HPLC analysis (Figure
S6).
Journal of Natural Products pubs.acs.org/jnp Article
https://doi.org/10.1021/acs.jnatprod.4c00394
J. Nat. Prod. 2025, 88, 619−630
621

N-Acetylcysteamine (SNAC) thioesters are often used as
test surrogates for acyl carrier protein (ACP)-tethered
intermediates. In this chemobiosynthetic process, the synthesis
of the unnatural extender unit may carry additional costs.
SNAC-thioester of propargylmalonyl may not be the most
economical version of this activated malonate extender unit. A
number of alternative thioesters could instead be generated
that replaced SNAC thioesters and became more econom-
ical.
27
Chemobiosynthesis Process Optimization by Allyl-
malonyl-SNAC Feeding of S. tsukubaensis Wild Type
and S. tsukubaensis ΔallR Strains. Relatively low titers of
FK506, from a few milligrams, up to 50 mg/L, are typically
achieved with the S. tsukubaensis NRRL18448 wild-type
strain.
28
Consequently, even lower titers of an FK506 analogue
are achieved when applying a chemobiosynthetic procedure to
the engineered strain S. tsukubaensis ΔallR, as shown by Kosec
et al. (2012).
18
To improve the ecacy of the chemo-
biosynthetic process with propargyl-SNAC, we initially carried
out medium optimization work for the production of FK506
by testing dierent carbon sources. As previously reported,
16,18
FK506 yields were approximately 30−70 mg/L (Experimental
Section) when using the S. tsukubaensis NRRL18488 wild-type
strain in PG3 medium on a shaker scale. Initially, we used
dextrin (90 g/L) as the main carbon source in the PG3
production medium. Dextrin was thereafter replaced with
several alternative carbon sources, always maintaining the same
total carbon concentrations, since the use of alternative starch
sources was known to exert a significant impact on FK506
titer.
28
Thus, starch and soluble starch from various suppliers
were tested, and their eect on FK506 production by the S.
tsukubaensis wild-type strain was evaluated (Figure 4).
We observed a significantly higher titer of FK506 in the
production medium that contained soluble starch. Corn starch
without amylase pretreatment resulted in FK506 titers that
were comparable to those of the PG3 control medium that
contained potato dextrin. However, a significant increase in
yields resulted from the addition of amylase to PG3 medium
containing corn starch, which resulted in a 2-fold increase in
FK506 titers, indicating that the starch properties and the
degree of hydrolysis had a crucial impact on the final FK506
yields. Potato and tapioca soluble starches were tested as
alternatives to avoid the need for amylase addition. Use of
soluble starches led to a significant increase in FK506
production, with FK506 titers reaching approximately 200
mg/L at the shake flask-level using the S. tsukubaensis
NRRL18488 wild-type strain (Figure 4).
The increase in the FK506 yield was likely due to the
capability of the producing strain for faster starch assimilation
while catabolic repression was avoided by glucose as a result of
starch hydrolysis. The PG3 production medium, where dextrin
was replaced by soluble starch, was designated as PG3_SS.
Optimization of the Chemobiosynthetic Process by
Feeding Allylmalonyl-SNAC to the S. tsukubaensis ΔallR
Strain. In the next step, we evaluated dierent feeding regimes
and the quantity of an unnatural extender unit (precursor) to
be fed during the process. For this purpose, we used
allylmalonyl-SNAC, considering that allylmalonyl-CoA is a
native extender unit selected by the AT4 domain.
18
In our earlier work,
18
we carried out a chemosynthetic
procedure on the FK506-nonproducing S. tsukubaensis ΔallR
strain, by transferring 33% (v/v) of the production culture after
3 days of cultivation to fresh production medium that already
contained allylmalonyl-SNAC. The culture was then cultivated
for an additional 6 days.
16,18
Although we applied the same
conditions (i.e., feeding allylmalonyl-SNAC), we obtained
significantly lower FK506 titers. Compared to the wild-type
strain S. tsukubaensis NRRL18448 that achieved titers of 30−
70 mg/L, the ΔallR strain reached up to 15 mg/L.
18
As
reported in our previous work,
18
the relatively low titer of the
target product was in part likely attributed to toxicity of the
unnatural SNAC-thioester of the allylmalonyl extender unit,
since the addition of a SNAC-ester to the wild-type strain
already resulted in a reduction in the FK506 yield of
approximately 50%.
18
Optimization of the feeding procedure was carried out by
adding a 10% solution of allylmalonyl SNAC thioester (w/v),
prepared in DMSO, directly to the cultivation broth of the S.
tsukubaensis ΔallR mutant, to dierent final concentrations and
at various time points during the cultivation procedure (Table
1).
Total concentrations of allylmalonyl-SNAC that were added
to the culture ranged from 0.5 to 3.5 g/L. The S. tsukubaensis
wild-type strain was used as the control strain to evaluate the
potentially negative eects of the SNAC-thioester on biomass
formation. Initial testing of dierent feeding regimes was
carried out in Falcon tubes, containing 5 mL of PG3_SS
production medium. Cultivation was carried out at 28 °C for 7
days (Table 1,Figure 5).
Interestingly, the addition at a low total concentration of
allylmalonyl-SNAC to wild-type strain S. tsukubaensis
NRRL18448 showed a positive eect on FK506 yields.
When feeding allylmalonyl-SNAC to the engineered strain S.
tsukubaensis ΔallR, the highest FK506 titer achieved reached
up to approximately 50%, compared to the control wild type
strain S. tsukubaensis NRRL18448 (Table 1 and Figure 5,
regimes 1−14). This increase in FK506 titer was observed,
while the total amount of added allylmalonyl-SNAC did not
exceed 1 g/L. Clearly, the increase in FK506 titer in the WT
strain S. tsukubaensis NRRL18448 indicates that a natural
supply of the allylmalonyl-CoA extender unit is a limiting
factor in the biosynthesis of FK506. The feeding regimes where
the total amount of SNAC-thioester added to the culture
exceeded 1.5 g/L resulted in a decreased FK506 titer when
using wild-type strain S. tsukubaensis NRRL18448, indicating
that higher concentrations of the unnatural extender unit
allylmalonyl-SNAC dissolved in DMSO are toxic to the culture
(Table 1 and Figure 5, regimes 10−14).
Figure 4. Evaluation of dierent starch sources as the main carbon
source present in PG3 medium for the production of FK506.
Journal of Natural Products pubs.acs.org/jnp Article
https://doi.org/10.1021/acs.jnatprod.4c00394
J. Nat. Prod. 2025, 88, 619−630
622

The eect of dierent feeding regimes on the final FK506
titer was particularly pronounced with the S. tsukubaensis ΔallR
strain. A comparative correlation of the FK506 titers achieved
with S. tsukubaensis ΔallR and the wild type strains (Figure 6),
simultaneously cultivated in identical cultivation media and the
same allylmalonyl-SNAC feeding procedure, was regarded as
an indicator of the ecacy of the allylmalonyl-SNAC thioester
feeding procedure.
The highest titer of FK506 achieved when applying the
chemobiosynthetic procedure, feeding allylmalonyl-SNAC
thioester to the S. tsukubaensis ΔallR strain, was approximately
100 mg/L. This titer was obtained when a maximum total
amount of 3.0 g/L allylmalonyl-SNAC was added to the
culture broth at 7 time points during cultivation (Table 1,
regime #10). At each time point, the allylmalonyl-SNAC ester
was fed to the culture broth to a final concentration of 0.5 g/L,
starting at the time of inoculation and then repeating the
feeding procedure every 24 h during the six-day cultivation
procedure. Thus, the total concentration of allylmalonyl-SNAC
ester was 3.0 g/L.
The most ecient chemobiosynthetic procedure at the 5 mL
scale was subsequently tested in shake flasks, and the FK506
yield was determined after 6−9 days of cultivation, in order to
determine the optimal cultivation time. Allylmalonyl-SNAC
thioester was added to the production medium at concen-
trations of 0.5−1.5 g/L, at the time of inoculation and
thereafter continuing with the same feeding rate every 24 h
during 7 days of cultivation. Thus, from 4 g/L to a maximum
of 12 g/L of total allylmalonyl-SNAC thioester had been fed to
the culture by the time the last feeding had been carried out
(Figure 6).
Table 1. Optimization of the Chemobiosynthetic Procedure by Applying Dierent Regimes of Allylmalonyl-SNAC Feeding
a
Time of
allylmalonyl-SNAC
addition to the
culture
Addition of allylmalonyl-
SNAC in production
medium [g/L]
Total concentration of
allylmalonyl-SNAC in
production medium [g/L]
FK506 yield [mg/
L] S. tsukubaensis
NRRL18448
% FK506
compared to
control
FK506 yield
[mg/L] S.
tsukubaensis
ΔallR
Mutasynthesis
procedure
eciency
1 Control 0.00 0.00 184.19 100% 6.7 0%
2 At inoculation 0.50 0.50 219.77 119% 14.03 6%
3 At inoculation 0.75 0.75 206.21 112% 16.85 8%
4 At inoculation 1.00 1.00 194.96 106% 29.69 15%
5 Day 3 0.50 0.50 208.64 113% 16.21 8%
6 Day 3 0.75 0.75 215.15 117% 26.77 12%
7 Day 3 1.00 1.00 206.47 112% 35.73 17%
8 At inoculation +
day 3 0.50 1.00 231.88 126% 28.02 12%
9 At inoculation +
day 3 0.75 1.50 193.75 105% 32.23 17%
10 At inoculation +
days 1−60.50 3.50 179.87 98% 103.01 57%
11 Days 1−6 0.50 3.00 146.42 79% 74.75 51%
12 Days 2−6 0.50 2.50 176.87 96% 62.86 36%
13 Days 3−6 0.50 2.00 164.72 89% 38.80 24%
14 Days 3−6 0.75 3.00 150.14 82% 53.22 35%
a
The ecacy of the chemobiosynthetic procedure (last column) is presented as a ratio of the FK506 titer achieved by this procedure compared to
the FK506 titer achieved with the parent strain S. tsukubaensis NRRL18448 (Figure 5).
Figure 5. Optimization of the chemobiosynthetic procedure; evaluation of dierent feeding regimes of allylmalonyl-SNAC on FK506 titers using
wild type and the S. tsukubaensis ΔallR mutant strain at the 5 mL scale. Feeding regimes under 1−14 are described in Table 1.
Journal of Natural Products pubs.acs.org/jnp Article
https://doi.org/10.1021/acs.jnatprod.4c00394
J. Nat. Prod. 2025, 88, 619−630
623

The results from the chemobiosynthetic procedure in shake
flasks were in good correlation with the results obtained at the
5 mL scale (Figure 6B). When allylmalonyl-SNAC feeding was
carried out to S. tsukubaensis ΔallR cultures to a final
concentration of 0.5 g/L at each time-point, approximately
90 mg/L FK506 was achieved after 7−8 days of cultivation. At
the lowest total concentration of 0.5 g/L allylmalonyl-SNAC-
thioester feeding, we observed a minor negative eect on
FK506 titer in the wild-type strain (up to 15% reduction).
However, the FK506 titer obtained with the WT strain S.
tsukubaensis NRRL18448 at the highest total feeding
concentrations of 1.0 and 1.5 g/L was significantly reduced
to approximately 50−65% and 80−85%, respectively. Partial
cell volume (PCV) was measured to determine any potential
eect of SNAC-thioester feeding on the formation of biomass
following 9 days of cultivation. PCV measurements (Table S1)
indicated that the addition of allylmalonyl-SNAC had an
inhibitory eect on biomass formation and pH value, which
resulted in a reduction in FK506 biosynthesis by the wild-type
strain. Based on the data gathered with the S. tsukubaensis
ΔallR strain, where allylmalonyl-SNAC feeding was carried
out, we can conclude that feeding 0.5 g/L SNAC-thioester at
each time-point was the most ecient and economical
approach. Moreover, although the S. tsukubaensis ΔallR mutant
strain was used, the optimized feeding procedure achieved a
relatively high titer of around 100 mg/L of FK506, compared
to the native strain. Importantly, we have demonstrated that
even without the auxiliary gene allR (Figure 2) being involved
in the synthesis and provision of native allylmalonyl-CoA, the
corresponding AT4 domain was able to eciently select for the
allylmalonyl-SNAC precursor.
Figure 6. Chemobiosynthetic process in 250 mL flasks. Addition of allylmalonyl-SNAC was carried out at inoculation and every 24 h thereafter
until the 6th day of cultivation. Cultivation was then carried out for an additional 3 days in the absence of any additional feeding and to a total of 9
days; A) wild-type S. tsukubaensis NRRL18448 control; B) ΔallR mutant strain.
Journal of Natural Products pubs.acs.org/jnp Article
https://doi.org/10.1021/acs.jnatprod.4c00394
J. Nat. Prod. 2025, 88, 619−630
624

Development of a Chemobiosynthetic Process for
the Production of C21 Propargyl-FK506. The chemo-
biosynthetic process for the production of propargyl-FK506
was carried out by feeding the propargylmalonyl-SNAC
extender unit (10% w/v in DMSO) to the S. tsukubaensis
ΔallR strain. We applied the chemobiosynthetic procedure
according to the optimized procedure described above, where
0.5 g/L propargylmalonyl-SNAC was fed to the production
medium every 24 h up to 7 days (Figure 7). As presented in
Figure 7, when feeding the unnatural extender unit
propargylmalonyl-SNAC, the chemobiosynthetic process was
slightly less ecient compared to the process with
allylmalonyl-SNAC. The likely reason is that the AT4 domain
has lower anity for propargylmalonyl-SNAC than for
allylmalonyl-CoA. Nevertheless, a final yield of around 70
mg/L of propargylmalonyl-FK506 was achieved, thus easily
ensuring sucient amounts of the target compound. The
HPLC analysis of crude material and of the purified propargyl-
FK506 analogue is presented in Figures S7 and S8,
respectively. LC-MS analysis of the final product isolated by
preparative HPLC is presented in Figure S9.
AT-swap in PKS systems has been a successfully employed
technology for two decades. For example, Del Vecchio et al.
13
have demonstrated that the AT4 domain from the
erythromycin PKS module 4 can be successfully replaced
with the AT2 domain from the rapamycin PKS module 2 to
alter its specificity from methylmalonyl-CoA to malonyl-CoA.
It is not yet fully understood how the FK506 PKS complex,
in collaboration with an auxiliar biosynthetic complex, transfers
an allylmalonyl-CoA extender unit to the AT4 domain of the
FkbB protein of FK506 PKS. Goranovicet al.
16
and Mo et al.
15
demonstrated that allylmalonyl-CoA is synthesized by an
unusual small protein complex containing KS-AT-ACP
domains encoded by a small gene cluster (allA, allK, allR
and allD), located on the side of the FK506 biosynthetic gene
cluster (Figure 2). Experiments by Jiang et al. suggest that both
ACP and CoA can be acyl donor candidates for the transfer of
an allylmalonyl unit catalyzed by AT4 from the FK506 PKS.
29
Therefore, it is possible that any type I PKS complex
containing a heterologous AT4 domain from the FkbB module
4 of FK506 PKS could potentially accept an unnatural
propargylmalonyl extender unit and thus result in a polyketide
backbone containing a propargyl moiety in the presence of
propargylmalonyl-SNAC. Alternatively, to ensure a more
ecient transfer of the unnatural extender unit propargylma-
lonyl-SNAC, auxiliary genes for the provision of allylmalonyl-
CoA should be used in a heterologous host together with a
type I PKS complex containing a heterologous AT4 domain
from the FkbB module 4 of FK506 PKS. If successful, this
approach would enormously expand our capability to
derivatize numerous PKs of medical and industrial importance.
In this work, we have established a reproducible production
method and industrially relevant titer of the target compound,
and we therefore easily prepared sucient amounts of pure
propargyl-FK506 at the shake-flask level.
Structure Confirmation of C21 Propargyl-FK506. The
elucidation of the structure of C21 propargyl-FK506 was
achieved through one- and two-dimensional NMR experi-
ments. 1H and 13C NMR spectra of a C21 propargyl-FK506
sample showed two sets of signals in a ratio 2:1 (Figures S10−
S11 and S16); these were attributed to two conformational
isomers (rotamers) arising from the restricted rotation of the
amide bond. The starting point for the 1H and 13C NMR
assignment was a propargyl group attached to C21, which
showed characteristic chemical shifts for C/H38−C/H40.
1H−1H (Figures S12−S13) and 1H−13C correlation signals via
single and multiple bonds (Figures S14−S15) enabled the
unequivocal assignment of signals on the macrolactam ring.
The 1H−13C gHMBC spectrum (Figure S15) showed
correlation signals to ketone carbons C9 (δC= 199.40 and
200.09 ppm) and C22 (δC= 210.13 and 209.38 ppm). C8
showed characteristic chemical shifts of the amide group (δC=
167.47 and 167.70 ppm). The carbon atom C1 of the ester
group (δC170.42 and 170.50 ppm) was confirmed by multiple
bond correlation signals with both the H2 and H26 protons.
Two sets of signals were observed in the 1H and 13C NMR
spectra, which were attributed to the cis and trans rotamers
along the peptide bond. The ratio between the cis and trans
rotamers was 2:1 for C21 propargyl-FK506. 1H and 13C NMR
chemical shifts for both rotamers of C21 propargyl-FK506 are
Figure 7. Chemobiosynthetic process for the production of FK506 and propargyl-FK506 in shake flasks. Allylmalonyl-SNAC and
propargylmalonyl-SNAC (10% w/v solutions in DMSO) were added to the production medium at the time of inoculation and every 24 h
thereafter, up to 7 days.
Journal of Natural Products pubs.acs.org/jnp Article
https://doi.org/10.1021/acs.jnatprod.4c00394
J. Nat. Prod. 2025, 88, 619−630
625

presented in Table 2. The chemical structure is in full
agreement with the C21 propargyl-FK506 structure presented
in Figure 1.Cis and trans rotamers along the amide bond can
be distinguished with respect to the characteristic dierences in
chemical shifts of C2 and C6 (δC57.5 and 39.86 ppm for the
cis rotamer; δC53.36 and 45.01 ppm for the trans rotamer).
Evaluation of the Biological Activity of C21 Prop-
argyl-FK506. In the scope of this work, we isolated sucient
amounts of propargyl-FK506 to determine its structure
(Supporting Information) and evaluate its activity. In order
to compare the toxicity of FK506 and its C21 propargyl
analogue, we established primary T cell cultures (DC Balb/c,
T cells C57Bl/6) and exposed them to increasing concen-
trations of FK506 and propargyl-FK506. Rates of live cells
were evaluated by staining for CD4+and LIVE/DEAD Fixable
Aqua, following a 4-day treatment; frequencies were
normalized to nonstimulated live cells (Figure 8,Figure
S18). High cell death rates were observed when testing FK506
at dierent concentrations (0.5 1, 5, 10, and 100 nM).
However, a toxic eect of C21 propargyl-FK506 was observed
only when this analogue was applied at concentrations higher
than 10 nM.
To investigate the immunosuppressive potential of the C21
propargyl-FK506 analogue, we compared the expansion of
alloreactive eector T cells that were either exposed to this
compound or to FK506 while grown for 4 days in coculture
Table 2. NMR Spectroscopic Data (600 MHz, Pyridine-d5) for Propargyl-FK506
Major rotamer (cis)
a
Minor rotamer (trans)
a
Position δC, type δH(Jin Hz) δC, type δH(Jin Hz)
1 170.42, C −170.50, C −
2 57.5, CH 5.16, m 53.36, CH 5.59, d (5.4)
3 /
b
/
b
/
b
/
b
4 21.97, CH21.6−1.7, m, 1.7−1.8, m 22.32, CH21.5−1.6, m, 1.6−1.7, m
5 25.27, CH21.46, m, 1.62, m 25.64, CH21.38, m, 1.55, m
6 39.86, CH23.29, m, 4.75, m 45.01, CH23.6−3.7, m
8 167.47, C −167.70, C −
9 199.40, C −200.09, C −
10 99.40, C −100.13, C −
11 36.33, CH 2.60, m 35.9, CH 2.80, m
12 33.49, CH21.82, m, 2.22, m 33.32, CH21.89, m, 2.22, m
13 74.73, CH 3.68−3.74, m 75.15, CH 3.68−3.74, m
14 73.48, CH 4.24, d (9.6) 74.41, CH 4.31, dd (9.6, 2.9)
15 76.66. CH 3.90, m 78.40, CH 3.90, m
16 34.99, CH21.47, m, 1.79, m 35.57, CH21.65, m, /
b
17 26.93, CH 1.98, m 27.0−27.2, CH 2.05, m
18 49.57, CH22.0−2.1, m 48.09, CH22.01, m, /
b
19 140.59, C −141.48, C −
20 123.16, CH 5.16, m 123.38, CH 5.20, d (10.1)
21 53.20, CH 3.99, dt (10.2, 7.0) 53.09, CH 4.05, dt (10.0, 7.2)
22 210.13, C −209.38, C −
23 48.31, CH22.74, dd (14.4, 6.4), 3.17, m 46.75, CH23.00, dd (16.3, 7.6), 3.09, dd (16.3, 4.8)
24 69.81, CH 4.56, m 69.24, CH 4.65, m
25 41.85, CH 2.12, m 41.65, CH 2.25, m
26 81.34, CH 5.87, d (5.8) 79.93, CH 5.93, d (5.0)
27 132.80, C −133.29, C −
28 133.76, CH 5.52, d (8.9) 132.56, CH 5.50, d (9.0)
29 35.9, CH 2.44, m 35.9, CH 2.44, m
30 36.81, CH21.15, m, 2.17, m 36.83, CH21.15, m, 2.17, m
31 85.44, CH 3.24, m 85.50, CH 3.24, m
32 74.24, CH 3.71, m 74.26, CH 3.70, m
33 33.88, CH21.61, m, 2.11, m 33.89, CH21.61, m, 2.11, m
34 31.66, CH21.0−1.1, m, 1.5−1.7, m 31.69, CH21.0−1.1, m, 1.5−1.7, m
35 17.30, CH31.30, d (6.5) 16.7, CH31.21, d (6.8)
36 20.30, CH30.97, d (6.5) 20.9, CH30.98, d (6.3)
37 16.7, CH31.69, s 17.44, CH31.90, s
38 21.47, CH22.54, m, 2.76, m 20.9, CH22.62, m, 2.84, m
39 83.51, C −83.78, C −
40 71.24, CH 2.70, t (2.5) 71.25, CH 2.72, t (2.6)
41 11.26, CH31.23, d (6.8) 11.05, CH31.19, d (7.0)
42 13.75, CH31.84, s 14.12, CH31.90, s
43 56.55, CH33.43, s 56.27, CH33.44, s
44 57.94, CH33.46, s 58.11, CH33.47, s
45 57.56, CH33.47, s 57.53, CH33.46, s
a
Ratio between cis and trans rotamers is 2:1.
b
NMR chemical shifts could not be unequivocally assigned due to significant signal overlap.
Journal of Natural Products pubs.acs.org/jnp Article
https://doi.org/10.1021/acs.jnatprod.4c00394
J. Nat. Prod. 2025, 88, 619−630
626

with Dendritic cells (DCs) derived from BALB/c mice;
proliferation was evaluated via CellTrace Violet stain
incorporation. As expected, FK506 exhibited strong immuno-
suppressive activity and could inhibit cell proliferation at 1 nM
(Figure 9). Gating strategies are presented in Figure S20.
Immunosuppression by C21 propargyl-FK506 was detected at
5 nM (Figure 9), with greatly reduced cell numbers at cell
cycle 4 (Figure 9); yet, cell viability was sustained at this
concentration (Figure 8). Thus, it can be concluded that C21
propargyl-FK506 is less toxic to primary T cell cultures than
FK506, while exerting its immunosuppressive eect at higher
concentrations.
Interestingly, although allyl and propargyl moieties only
marginally dier in their structure (terminal triple bond instead
of a double bond), the replacement of the allyl moiety with the
propargyl-moiety did have a profound eect on the
immunosuppressive activity and toxicity of the novel
compound. Propragyl-FK506 displays significantly reduced
immunosuppressive activity while being significantly less toxic
to primary T cell cultures.
Nevertheless, this FK506 analogue still preserves some
immunosuppressive eect and could thus be used for further
diversification of the terminal alkyl group of the propargyl-
moiety at carbon C21 of the FK506 backbone. As exemplified
in the literature, FK506 analogues with reduced immunosup-
pressive activity and toxicity, and increased antifungal or nerve-
regenerating activities are of interest in the scope of drug-
development eorts.
10,30
■EXPERIMENTAL SECTION
General Experimental Procedures. NMR spectra were acquired
on an Agilent Technologies 800 MHz NMR spectrometer equipped
with a cryoprobe and operating at 800 MHz for 1H and at 200 MHz
for 13C nuclei. All NMR data were processed using Mnova software
(Mestrelab Research S.L.). Detection of FK506 and analogues were
carried by Nucleosil EC100−3 C18, reversed-phase HPLC column.
The mobile phase used for isocratic elution was composed of water,
acetonitrile, MTBE and phosphoric acid (58.29:34.4:7.29:0.02, v/v/
v/v). Isolation of propargyl-DK506 was carried out by an extraction
procedure followed by preparative HPLC purification using Knauer
preparative HPLC with a Macherey-Nagel C18 column as stationary
phase. LC-MS/MS analyses were performed on an Agilent 1100 with
a reversed phase analytical C18 column (Gemini C18 column, 5 μm,
150 mm ×2 mm i.d., Phenomenex, Torrance, CA, USA). The mass
selective detector (Quattro Micro API, Waters, Milford, MA, USA)
equipped with an electrospray ionization, and cone voltage of 20 V
and capillary voltage of 3.0 kV were used for positive ionization of the
analytes.
Cultivation of Engineered Cultures. The S. tsukubaensis NRRL
18448 strain was used for optimization of the production medium.
The FK506 nonproducing S. tsukubaensis ΔallR mutant strain with an
inactivated allR gene (crotonyl CoA carboxylase in the FK506 gene
cluster) was used for chemo-biosynthetic bioprocess experiments.
16,18
For the preparation of spore stocks, the S. tsukubaensis strains were
cultivated as a confluent lawn on ISP4 agar sporulation medium
31
for
10−14 days at 28 °C. For liquid cultures, spores of the S. tsukubaensis
strains were inoculated in VG3 seed medium
16
and incubated at 28
°C and 250 rpm for 24−48 h. This seed culture (10% v/v) was then
used to inoculate 250 mL Erlenmeyer flasks containing 50 mL of
production medium based on the PG3 production medium recipe, as
described previously.
16
Cultivation was carried out at 28 °C and 250
rpm for 6−7 days. Dierent sources of starch were used as a main
carbon source in the PG3 medium, as described in the Results and
Discussion. Partial degradation of the starch was carried out by the
addition of alpha-amylase (Glentham life sciences, GE7409) to the
PG3 medium. The eect of SNAC-thioesters on biomass formation
was evaluated by the packed cell volume (PCV). For the
determination of PCV, 10 mL of culture broth was transferred to a
15 mL Falcon tube and centrifuged at 4000 rpm for 10 min. The
Figure 8. Cytotoxicity of FK506 and C21 propargyl-FK506. After 4
days of culture, CD4+live cells were analyzed by flow cytometry
(Figure S18). Bar graphs represent the frequency of live cells
normalized to nonstimulated control samples.
Figure 9. Immunosuppressive activity of FK506 and C21 propargyl-FK506 at 0.5, 1, and 5 nM concentrations. Naive CD4+T-cells derived from
C57BL/6 mice were cocultured with allogeneic bone marrow-derived BALB/c DCs, and incubated with 0.5, 1, and 5 nM concentrations of either
tacrolimus or propargyl-FK506, in order to compare the immunosuppressive potential of these two compounds. Representative flow cytometry
results showing the rates of live CD4+T cells at proliferation cycles 1 and 4 under each of the conditions: control and treated with 3 dierent
concentrations of each of the compounds. The proliferation was assessed by flow cytometry (Figure S19). The gating strategy is presented in Figure
S20. The data shown here and Figure S19 are the combined results from three independent experiments. Graphs show mean values, and error bars
represent SDs unless otherwise specified. *P< 0.05; **P< 0.01; ***P< 0.001; ****P< 0.0001.
Journal of Natural Products pubs.acs.org/jnp Article
https://doi.org/10.1021/acs.jnatprod.4c00394
J. Nat. Prod. 2025, 88, 619−630
627

volume of the sediment, which contained the S. tsukubaensis biomass,
was then recorded.
Detection of FK506 and Propargyl-FK506. The analytical
procedures used in this work have been previously described.
16
Briefly, after 6−7 days of cultivation, the broth was mixed with an
equal volume of methanol (1:1) and the soluble fraction loaded onto
a Nucleosil EC100−3 C18, reversed-phase HPLC column. The mobile
phase used for isocratic elution was composed of water, acetonitrile
(MeCN), methyl tert-butyl ether (MTBE) and phosphoric acid
(H3PO4) (58.29:34.4:7.29:0.02, V/V/V/V). Chromatographic peaks
corresponding to FK506 and FK520 were identified using an FK506
and FK520 external standard (obtained from Lek/Sandoz) and
ChromQuest software, which was used for data analysis. Using this
method, the detection limit for FK506 and FK520 was 0.5 mg/L.
Statistical Analysis of FK506-Related Metabolites. When the
M4018 transformants were compared with each other, statistical
analysis was performed using analysis of variance (ANOVA), with
SPSS windows version 26.0 (SPSS Inc., Chicago, IL, USA). Mann−
Whitney U-tests were used for the comparisons of the FK506 titers
from the parental strain and their transformants, at the level of P<
0.05.
32
Molecular Biology Methods. In this study, the S. tsukubaensis
ΔallR strain with an inactivated allR gene was used, as described
previously.
18
Briefly, Goranovicet al.
16
identified a group of genes
encoding the biosynthesis of the extender unit (allylmalonyl-CoA)
that forms the allyl group at carbon C21 of FK506 (Figure 1). This
group of genes contains a small independent diketide synthase alllK
system involved in the biosynthesis of the allyl group. Based on this
finding, Goranovicet al.
16
proposed a biosynthetic pathway for the
provision of an unusual five-carbon extender unit allylmalonyl-CoA,
which is carried out by a novel diketide synthase complex. This small
group of genes also contains the allR gene (Figure 2), which plays an
important role in the provision of the allyl group at carbon C21 of
FK506 (Figure 1). allR gene is a homologue of the crotonyl-CoA
carboxylase/reductase that catalyzes the carboxylation and reduction
of crotonyl-ACP toward 2-pentenoyl-ACP, which is an important step
in the biosynthesis of the unusual extender unit allylmalonyl-ACP, as
described by Goranovicet al.
16
Based on this information, Kosec et
al.
18
constructed the strain with carrying an inactivated allR gene, and
developed an ecient chemobiosynthetic process. The main
advantage of this strain is that it cannot produce FK506 or any
other FK506-related product. Only after feeding synthetically
prepared precursors such as allylmalonyl-SNAC is the biosynthesis
of FK506 reestablished at a very high eciency. Therefore, this
chemobiosynthetic bioprocess ensures the exclusive production of
target FK506-related products, which represents an industrially
important advantage.
Synthesis of Allylmalonyl-SNAC and Propargylmalonyl-
SNAC. Synthesis of allylmalonyl-SNAC [S,S-bis(2-acetamidoethyl)2-
allylpropanbis(thiolate)] was carried out as described by Kosec et al.
18
The synthesis of [S,S-bis(2-acetamidoethyl)2-(prop-2-yn-1-yl)-
propanebis(thioate)], which contains a terminal alkyne (also
designated as propargylmalonyl-SNAC), was carried out for the first
time during this work, and it is therefore described in the Results and
Discussion.
Isolation of FK506 Analogues. Around 1000 mL of a
fermentation broth culture of S. tsukubaensis 2-(prop-2-yn-1-yl)
malonic acid-containing C21 propargyl-FK506 was obtained. Fresh
broth was subjected to downstream processing by the following
procedure: Whole broth was mixed with 1 L of 2-propanol and the
mixture was stirred vigorously at room temperature (RT) for 1 h. The
suspension was centrifuged at 4500 rpm for 10 min, resulting in 1850
mL of clear supernatant and 157 g of dry biomass. Both fractions were
analyzed by HPLC for propargyl-FK506 content. The supernatant
was concentrated under reduced pressure on a rotary evaporator to a
final volume of 500 mL. The resulting aqueous concentrate was
subjected to L/L extraction using equivolume amounts of toluene.
The extraction was carried out at RT and was repeated twice. The
phases were separated in a separating funnel and both phases were
analyzed by HPLC for propargyl-FK506 content. The organic extract
was concentrated to give 538 mg of an orange oily slurry. The crude
material was subjected to further purification using normal phase silica
gel column chromatography. The mobile phase consisted of
dichloromethane (DCM) and MeOH, starting with pure DCM and
increasing the MeOH share by 10% in each subsequent fraction up to
a final MeOH share of 50%. 500 mL (2 volumes) of mobile phase was
used for each subsequent elution. The eluates were analyzed for
FK506 content by HPLC and it was observed that most of the FK506
was eluted within the mobile phase with a 20% MeOH. Following the
evaporation of the solvent, 300 mg of orange-brown material was
obtained and subjected to purification by preparative HPLC.
Preparative HPLC purification was carried out using Knauer
preparative HPLC with a Macherey-Nagel C18 column (dimensions
18 mm ×5 mm, particle size: 5 μm) as stationary phase. The mobile
phase consisted of a two-solvent system: (A) 60% H2O, 33% MeCN
and 7% MTBE with 2 ‰H3PO4and (B): 40% H2O, 50% MeCN and
10% MTBE with 2 ‰H3PO4. The initial composition of the mobile
phase was 50% A - 50% B and the gradient was increased over 32 min
to 100% B. The flow rate was 12 mL/min and the load was 50 mg of
crude material. The sample was purified twice, and the fractions
containing a single peak at RT= 22.2 min were collected and
subjected to isolation. Isolation was carried out by L/L extraction
using MTBE as the organic solvent. The organic phase was dried over
anhydrous Na2SO4and concentrated to yield 70 mg of white
crystalline material. HPLC analysis of the final product, a typical
preparative HPLC chromatogram, and the LC-MS analysis are shown
below in Figures S7, S8 and S9. The final material was dried overnight
under vacuum to give a final amount of 64 mg and it was used for
two-dimensional NMR experiments directed at elucidating the
structure of the compounds.
Confirmation of the C21 Propargyl-FK506 Structure.
Samples were dissolved in deuterated pyridine (Pyridine-d5). 1D
(1H NMR, 13C NMR) and 2D (1H−1H COSY, 1H−1H TOCSY,
1H−13C HSQC, and 1H−13C HMBC) NMR spectra were acquired
on an Agilent Technologies 800 MHz NMR spectrometer equipped
with cryo-probe and operating at 800 MHz for 1H and at 200 MHz
for 13C nuclei. The temperature for the samples was set at 298 K. 1H
and 13C NMR chemical shifts were reported in parts per million and
referenced with respect to the residual solvent signals, corresponding
to TMS at δ= 0.0 ppm. All NMR data were processed using Mnova
software (Mestrelab Research S.L.).
In addition, we reconfirmed the structure of C21 propargyl-FK506
by applying LC-MS/MS analysis (Table S2), as described in detail in
Supporting Information (6. Additional structural confirmation of C21
propargyl-FK506 by LC-MS/MS analysis).
33
Evaluation of the Immunosuppressive Activity of C21
Propargyl-FK506. All mice were maintained under specific
pathogen-free conditions at the animal facility at TWINCORE
(Hannover, Germany). All animal procedures, such as organ
collection, were performed in compliance with the German animal
protection law and were approved by the Lower Saxony Committee
on the Ethics of Animal Experiments as well as the responsible state
oce (Lower Saxony State Oce of Consumer Protection and Food
Safety under permit number 33.9−42502−04−15/1851).
Mouse T Cell Cultures. CD4+CD25−T cells were enriched from
the spleen and lymph nodes of C57BL/6 wild type male mice using a
Mouse CD4+ T Cell Isolation Kit (Stem Cell Technologies, Canada)
and incubated with anti-CD25-PE antibody (eBioscience, USA). Cells
were then passed through magnetic columns. This protocol yielded an
average of 90% pure CD4+ T cells. T helper cultures were kept for 4
days in IMDM GlutaMAX medium (Life Technologies, USA)
supplemented with 10% fetal calf serum (FCS, Biochrom, UK), 500
U penicillin-streptomycin (PAA laboratories, Canada) and 50 μMβ-
mercaptoethanol (Life Technologies, USA). At day 0, 2−3×105
naive T cells were cultured per well, in the presence of plate-bound
anti-CD3ε(10 ug mL−1, clone 145−2C11; Bio X Cell, USA), anti-
CD28 (1ug/mL, clone 37.51; Bio X Cell, USA), anti-TGF-β1 (2 ng
mL-1; Peprotech, USA), rmIL-6 (5 ng mL-1; Peprotech, USA) and
rmIL-1β(50 ng mL-1; Peprotech, USA).
Journal of Natural Products pubs.acs.org/jnp Article
https://doi.org/10.1021/acs.jnatprod.4c00394
J. Nat. Prod. 2025, 88, 619−630
628

In Vitro Allogeneic T-Cell Stimulation. CD4+CD25−T cells
were enriched from C57BL/6 wild type male mice using a Mouse
CD4+ T Cell Isolation Kit (Stem Cell Technologies, Canada) and
incubated with anti-CD25-PE antibody (eBioscience, USA). Cells
were then passed through magnetic columns. This protocol yielded an
average of 90% pure CD4+ T cells. 1.5 ×105T cells labeled with
CellTrace Violet Cell Proliferation (Life Technologies, USA) were
cocultured with 5 ×104GM-CSF bone-marrow-derived allogeneic
DCs from BALB/c mice. The coculture was kept for 4 days in RPMI
1640 GlutaMAX medium (Life Technologies, USA) supplemented
with 10% heat-inactivated FCS (Biochrom, Germany), 500 U
penicillin-streptomycin (PAA laboratories, Canada) and 50 μMβ-
mercaptoethanol, in the presence or absence of the compounds
FK506 and C21 propargyl-FK506. The statistical parameters applied
can be found in the figure legends. Data were analyzed by using
GraphPad Prism 7.0 software (GraphPad Software, La Jolla, Calif).
Statistical analyses were performed as follows: two-way ANOVA
followed by Sidak multiple comparison was used to analyze
experiments with 2 variables and 3 or more groups, and one-way
ANOVA followed by Duvett comparison with a control was used for
experiments with 1 variable and 3 or more groups. The experiments
with 2 groups were analyzed with the Student ttest. In all cases, P<
0.05 was considered statistically significant.
Flow Cytometry and Antibodies. Following their isolation from
spleen and peripheral lymph nodes, lymphocyte cell suspensions were
incubated with Fc Block (clone 2.4G2) for 5 min before staining.
Clones and suppliers of mAbs and reagents were: anti-CD4 PE
(GK1.5) and anti-CD44 FITC (IM7), from eBioscience (USA); Aqua
reagent for live/dead discrimination, from BioLegend (USA); and
CellTrace Violet Cell Proliferation kit, from Life Technologies.
Cytometric data were acquired on a CyAn ADP (Beckman Coulter,
USA) and analyzed with FlowJo Software (Treestar, Ashland, OR,
USA).
■ASSOCIATED CONTENT
Data Availability Statement
The NMR data for propargyl-FK506 has been deposited in the
Natural Products Magnetic Resonance Database (NP- MRD;
www.np-mrd.org) and can be found at NP0332700 (10.
57994/1960).
*
sı Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acs.jnatprod.4c00394.
1H NMR, 13C NMR and MS spectra for all synthesized
or isolated compounds, 1H NMR, 13C NMR, 1H−1H
gCOSY NMR, and 1H−1H TOCSY NMR spectra;
1H−13C gHSQC NMR, 1H−13C gHMBC NMR, 13C
NMR of isolated compound; additional experimental
details, including information on feeding regimes of the
packed cell volume and pH value of cultures of S.
tsukubaensis strains and T cell proliferation assays (PDF)
■AUTHOR INFORMATION
Corresponding Author
Hrvoje Petkovic−University of Ljubljana, Biotechnical
Faculty, Department of Food Science and Technology, 1000
Ljubljana, Slovenia; orcid.org/0000-0003-1377-9845;
Phone: +38640488498; Email: hrvoje.petkovic@bf.uni-
lj.si
Authors
Dusan Goranovic−Acies Bio, d.o.o., 1000 Ljubljana, Slovenia
Branko Jenko −Acies Bio, d.o.o., 1000 Ljubljana, Slovenia
Barbara Ramsak −University of Ljubljana, Biotechnical
Faculty, Department of Food Science and Technology, 1000
Ljubljana, Slovenia
Ajda Podgorsek Berke −Acies Bio, d.o.o., 1000 Ljubljana,
Slovenia
Leon Bedrac−Acies Bio, d.o.o., 1000 Ljubljana, Slovenia
Jaka Horvat −Acies Bio, d.o.o., 1000 Ljubljana, Slovenia
Martin Sala −National Institute of Chemistry, 1000
Ljubljana, Slovenia; orcid.org/0000-0001-7845-860X
Damjan Makuc −National Institute of Chemistry, 1000
Ljubljana, Slovenia
Guilhermina M. Carriche −Institute of Medical Microbiology
and Hygiene and Research Center for Immunotherapy (FZI),
University Medical Center of the Johannes Gutenberg-
University, Mainz 55131, Germany; Institute of Infection
Immunology, TWINCORE, Centre for Experimental and
Clinical Infection Research, a Joint Venture Between the
Medical School Hannover (MHH) and the Helmholtz Centre
for Infection Research (HZI), Hannover 30625, Germany
Luana Silva −Institute of Medical Microbiology and Hygiene
and Research Center for Immunotherapy (FZI), University
Medical Center of the Johannes Gutenberg-University, Mainz
55131, Germany; Institute of Infection Immunology,
TWINCORE, Centre for Experimental and Clinical Infection
Research, a Joint Venture Between the Medical School
Hannover (MHH) and the Helmholtz Centre for Infection
Research (HZI), Hannover 30625, Germany; orcid.org/
0000-0001-7297-9586
Aleksandra Lopez Krol −Institute of Medical Microbiology
and Hygiene and Research Center for Immunotherapy (FZI),
University Medical Center of the Johannes Gutenberg-
University, Mainz 55131, Germany; Institute of Infection
Immunology, TWINCORE, Centre for Experimental and
Clinical Infection Research, a Joint Venture Between the
Medical School Hannover (MHH) and the Helmholtz Centre
for Infection Research (HZI), Hannover 30625, Germany
Alen Psenicnik −University of Ljubljana, Biotechnical
Faculty, Department of Food Science and Technology, 1000
Ljubljana, Slovenia
María Beatriz Durán Alonso −Department of Biochemistry
and Molecular Biology and Physiology, University of
Valladolid, 47005 Valladolid, Spain
Martina Avbelj −University of Ljubljana, Biotechnical
Faculty, Department of Food Science and Technology, 1000
Ljubljana, Slovenia
Stojan Stavber −Department of Physical and Organic
Chemistry, Jozef Stefan Institute, 1000 Ljubljana, Slovenia
Janez Plavec −National Institute of Chemistry, 1000
Ljubljana, Slovenia; EN →FIST Centre of Excellence, 1000
Ljubljana, Slovenia; Faculty of Chemistry and Chemical
Technology, University of Ljubljana, 1000 Ljubljana,
Slovenia; orcid.org/0000-0003-1570-8602
Tim Sparwasser −Institute of Medical Microbiology and
Hygiene and Research Center for Immunotherapy (FZI),
University Medical Center of the Johannes Gutenberg-
University, Mainz 55131, Germany; Institute of Infection
Immunology, TWINCORE, Centre for Experimental and
Clinical Infection Research, a Joint Venture Between the
Medical School Hannover (MHH) and the Helmholtz Centre
for Infection Research (HZI), Hannover 30625, Germany
Rolf Muller −Helmholtz Institute for Pharmaceutical
Research Saarland (HIPS), Helmholtz Centre for Infection
Research (HZI), and Department of Pharmacy, Saarland
Journal of Natural Products pubs.acs.org/jnp Article
https://doi.org/10.1021/acs.jnatprod.4c00394
J. Nat. Prod. 2025, 88, 619−630
629

University, 66123 Saarbrucken, Germany; orcid.org/
0000-0002-1042-5665
Gregor Kosec −Acies Bio, d.o.o., 1000 Ljubljana, Slovenia;
Centre of Excellence for Integrated Approaches in Chemistry
and Biology of Proteins (CIPKeBiP), SI-1000 Ljubljana,
Slovenia
Stefan Fujs −Acies Bio, d.o.o., 1000 Ljubljana, Slovenia;
Centre of Excellence for Integrated Approaches in Chemistry
and Biology of Proteins (CIPKeBiP), SI-1000 Ljubljana,
Slovenia
Complete contact information is available at:
https://pubs.acs.org/10.1021/acs.jnatprod.4c00394
Notes
The authors declare no competing financial interest.
■ACKNOWLEDGMENTS
The study was supported by the Ministry of Higher Education,
Science and Technology (Slovenian Research Agency, ARRS,
grant nos. P4-0116 and L4-5530 to H.P.; Slovenian Research
and Innovation Agency, ARIS, grant no. P1-0034 to M.S.) and
by the German Research Foundation (Project Number
318346496, SFB 1292 TP18 and 490846870, SFB355 TPA4
to T.S.). PhD study of A.P. was supported by a young
researcher grant from the Slovenian Research Agency (No.
53621). Project ID: 318346496-SFB 1292, TP18.
■REFERENCES
(1) Katz, L.; Donadio, S. Annu. Rev. Microbiol. 1993,47, 875−912.
(2) Kolos, J. M.; Voll, A. M.; Bauder, M.; Hausch, F. Front.
Pharmacol. 2018,9, 1425.
(3) Rath, T. Expert Opin. Pharmacother. 2013,14, 115−122.
(4) Azzi, J. R.; Sayegh, M. H.; Mallat, S. G. J. Immunol. 2013,191,
5785−5791.
(5) Migita, K.; Eguchi, K. FK506: Anti-Inflammatory Properties.
Current Medicinal Chemistry - Anti-Inflammatory &Anti-Allergy Agents.
Current Medicinal Chemistry - Anti-Inflammatory &Anti-Allergy Agents
2003.2260
(6) Simpson, D.; Noble, S. Drugs 2005,65, 827−858.
(7) Kennedy, J. Nat. Prod. Rep. 2008,25, 25−34.
(8) Cropp, A.; Chen, S.; Liu, H.; Zhang, W.; Reynolds, K. A. J. Ind.
Microbiol. Biotechnol. 2001,27, 368−377.
(9) Gregory, M. A.; Petkovic, H.; Lill, R. E.; Moss, S. J.; Wilkinson,
B.; Gaisser, S.; Leadlay, P. F.; Sheridan, R. M. Angew. Chem., Int. Ed.
Engl. 2005,44, 4757−4760.
(10) Jung, J. A.; Lee, H. J.; Song, M. C.; Hwangbo, A.; Beom, J. Y.;
Lee, S. J.; Park, D. J.; Oh, J. H.; Ha, S.-J.; Cheong, E.; Yoon, Y. J. J.
Nat. Prod. 2021,84, 195−203.
(11) Lechner, A.; Wilson, M. C.; Ban, Y. H.; Hwang, J.-Y.; Yoon, Y.
J.; Moore, B. S. ACS Synth. Biol. 2013,2, 379−383.
(12) Barreiro, C.; Martínez-Castro, M. Appl. Microbiol. Biotechnol.
2014,98, 497−507.
(13) Del Vecchio, F.; Petkovic, H.; Kendrew, S. G.; Low, L.;
Wilkinson, B.; Lill, R.; Cortés, J.; Rudd, B. A. M.; Staunton, J.;
Leadlay, P. F. J. Ind. Microbiol. Biotechnol. 2003,30, 489−494.
(14) Wilson, M. C.; Moore, B. S. Nat. Prod. Rep. 2012,29, 72−86.
(15) Mo, S.; Kim, D. H.; Lee, J. H.; Park, J. W.; Basnet, D. B.; Ban,
Y. H.; Yoo, Y. J.; Chen, S. W.; Park, S. R.; Choi, E. A.; Kim, E.; Jin, Y.
Y.; Lee, S. K.; Park, J. Y.; Liu, Y.; Lee, M. O.; Lee, K. S.; Kim, S. J.;
Kim, D.; Park, B. C.; Lee, S. G.; Kwon, H. J.; Suh, J. W.; Moore, B. S.;
Lim, S. K.; Yoon, Y. J. J. Am. Chem. Soc. 2011,133, 976−985.
(16) Goranovic, D.; Kosec, G.; Mrak, P.; Fujs, S.; Horvat, J.; Kuscer,
E.; Kopitar, G.; Petkovic, H. J. Biol. Chem. 2010,285, 14292−14300.
(17) Ray, L.; Valentic, T. R.; Miyazawa, T.; Withall, D. M.; Song, L.;
Milligan, J. C.; Osada, H.; Takahashi, S.; Tsai, S.-C.; Challis, G. L.
Nat. Commun. 2016,7, 13609.
(18) Kosec, G.; Goranovic, D.; Mrak, P.; Fujs, S.; Kuscer, E.; Horvat,
J.; Kopitar, G.; Petkovic, H. Metab. Eng. 2012,14, 39−46.
(19) Chuck, J. A.; McPherson, M.; Huang, H.; Jacobsen, J. R.;
Khosla, C.; Cane, D. E. Chem. Biol. 1997,4, 757−766.
(20) Frykman, S.; Leaf, T.; Carreras, C.; Licari, P. Biotechnol. Bioeng.
2001,76, 303−310.
(21) Kaur, J.; Saxena, M.; Rishi, N. Bioconjugate Chem. 2021,32,
1455−1471.
(22) Thirumurugan, P.; Matosiuk, D.; Jozwiak, K. Chem. Rev. 2013,
113, 4905−4979.
(23) Jiang, X.; Hao, X.; Jing, L.; Wu, G.; Kang, D.; Liu, X.; Zhan, P.
Expert Opin. Drug Discovery 2019,14, 779−789.
(24) Wang, X.; Huang, B.; Liu, X.; Zhan, P. Drug Discovery Today
2016,21, 118−132.
(25) Presolski, S. I.; Hong, V. P.; Finn, M. G. Curr. Protoc. Chem.
Biol. 2011,3, 153−162.
(26) Fernandes, P.; Martens, E.; Pereira, D. J. Antibiot. 2017,70,
527−533.
(27) Kosec, G.; Goranovic, D.; Horvat, J.; Fujs, S.; Jenko, B.;
Petkovic, H. Novel Polyketide Compounds and Methods of Making
Same. EP2658855 A2, 2013.
(28) Martínez-Castro, M.; Salehi-Najafabadi, Z.; Romero, F.; Pérez-
Sanchiz, R.; Fernández-Chimeno, R. I.; Martín, J. F.; Barreiro, C. Appl.
Microbiol. Biotechnol. 2013,97, 2139−2152.
(29) Jiang, H.; Wang, Y.-Y.; Guo, Y.-Y.; Shen, J.-J.; Zhang, X.-S.;
Luo, H.-D.; Ren, N.-N.; Jiang, X.-H.; Li, Y.-Q. FEBS J. 2015,282,
2527−2539.
(30) Beom, J. Y.; Jung, J. A.; Lee, K. T.; Hwangbo, A.; Song, M. C.;
Lee, Y.; Lee, S. J.; Oh, J. H.; Ha, S. J.; Nam, S. J.; Cheong, E.; Bahn, Y.
S.; Yoon, Y. J. J. Nat. Prod. 2019,82, 2078−2086.
(31) Kieser, T.; Bibb, M. J.; Buttner, M. J.; Chater, K. F.; Hopwood,
D. A. Practical Streptomyces Genetics; John Innes Foundation, 2000.
(32) IBM Corp. IBM SPSS Statistics for Windows [Internet]; IBM
Corp.: Armonk, NY, 2017.
(33) Mevizou, R.; Aouad, H.; Sauvage, F. L.; Arnion, H.; Pinault, E.;
Bernard, J. S.; Bertho, G.; Giraud, N.; Alves de Sousa, R.; Lopez-
Noriega, A.; Di Meo, F.; Campana, M.; Marquet, P. Pharmacol. Res.
2024,209, 107438−10752.
Journal of Natural Products pubs.acs.org/jnp Article
https://doi.org/10.1021/acs.jnatprod.4c00394
J. Nat. Prod. 2025, 88, 619−630
630