4-Chloroisocoumarins as Chlamydial Protease Inhibitors and Anti-Chlamydial Agents

4-Chloroisocoumarin compounds have broad inhibitory properties against serine proteases. Here, we show that selected 3-alkoxy-4-chloroisocoumarins preferentially inhibit the activity of the conserved serine protease High-temperature requirement A of Chlamydia trachomatis. The synthesis of a new series of isocoumarin-based scaffolds has been developed, and their anti-chlamydial properties were investigated. The structure of the alkoxy substituent was found to influence the potency of the compounds against High-temperature requirement A, and modifications to the C-7 position of the 3-alkoxy-4-chloroisocoumarin structure attenuate anti-chlamydial properties.


Introduction
Chlamydia trachomatis is a gram-negative bacteria and common sexually transmissible infection (STI).Chlamydia is the most-frequently diagnosed bacterial STI, with the majority of diagnoses for both men and women being from 15 to 24 years of age [1].Persistent infection can cause infertility in men and women as well as pelvic inflammatory disease (PID) and ectopic pregnancy [2].
Current treatment regimens use Azithromycin or Doxycycline.Azithromycin can and has previously been used to treat Neisseria gonorrhoeae and Mycoplasma genitalium, which are often present as a co-infection with C. trachomatis.Each of these infections may present asymptomatically, and with increasing rates of resistance to Azithromycin in Neisseria gonorrhoeae, there are concerns with overreliance in its use as a front-line anti-chlamydial therapy.Recent reports recommend the use of Ceftriaxone preferentially to Azithromycin in concurrent chlamydial infections [3].The contraindication of Doxycycline for use with pregnant women restricts its potential as a replacement front-line therapy [4].Therefore, new drugs with novel modes of action are needed.
The serine protease High-temperature requirement A (HtrA) has become an attractive anti-microbial target.The development of small molecules targeting HtrA in Eschericia coli (DegS) [5] and Helicobacter pylori [6] has demonstrated the potential of HtrA inhibitors as a valid antibacterial target in vitro.The HtrA of C. trachomatis (CtHtrA) functions as a proteolytic enzyme regulating homoeostasis, and it chaperones membrane proteins through the periplasm [7].CtHtrA likely has a critical role during stress conditions [8], and exposure of C. trachomatis to CtHtrA inhibitors during the mid-replicative phase of the development cycle is lethal [9].These inhibitors are effective as they prevent both functions of HtrA and restrict bacterial replication, preventing the production of further progeny.
The development of CtHtrA inhibitors has produced diphenylphosphonate peptidebased inhibitors such as JO146 (Figure 1).JO146 features a tripeptidomimetic backbone with a terminal N-Boc capping group and an electrophilic diphenylphosphonate headgroup that reacts with the serine protease to form a reversible covalent complex with CtHtrA (IC 50 = 12 µM) [9].JO146 exhibits anti-chlamydial properties against C. trachomatis and other chlamydial species with minimal cytotoxic properties against mammalian cell lines including HEp-2 cells, McCoy B cells and ex vivo Koala mononuclear blood cells [9].Similar structures with modifications to the N-terminal capping group retain their anti-chlamydial activity [10].However, their inhibitory activity against CtHtrA has not been investigated.
with a terminal N-Boc capping group and an electrophilic diphenylphosphonate headgroup that reacts with the serine protease to form a reversible covalent complex with CtHtrA (IC50 = 12 µM) [9].JO146 exhibits anti-chlamydial properties against C. trachomatis and other chlamydial species with minimal cytotoxic properties against mammalian cell lines including HEp-2 cells, McCoy B cells and ex vivo Koala mononuclear blood cells [9].Similar structures with modifications to the N-terminal capping group retain their antichlamydial activity [10].However, their inhibitory activity against CtHtrA has not been investigated.Successive optimization studies have investigated the effects of replacing various amino acids within the JO146 structure, and substitution on the pyrrolidine ring, with some improvement to the inhibitory activity without loss of anti-chlamydial properties [11,12].However, off-target effects have been observed through the inhibition of Human Leukocyte Elastase (HLE) [9,[11][12][13], a serine protease that has been implicated in the pathogenesis of Chronic Obstructive Pulmonary Disease [14].Thus, we hypothesized that HLE inhibitors may exhibit inhibitory activity against CtHtrA and that these chemical scaffolds may serve as a source of new lead compounds as CtHtrA inhibitors.
Isocoumarin compounds have been investigated for many years as general serine protease inhibitors, and 3,4-dichloroisocoumarin is a commercial serine protease inhibitor.In the development of HLE inhibitors, 3-alkoxy-4-chloroisocoumarins and their C-7substituted derivatives have been identified as rapid and potent inhibitors [15][16][17][18].As the C-3 and C-7 positions govern the selectivity and drive the inhibitory properties of 3alkoxy-4-chloroisocoumarins against serine proteases [15][16][17][18], we focused our investigation on uncovering the structure-activity relationships (SARs) of these positions (Figure 2) with regard to their inhibitory properties against CtHtrA.We also sought to investigate the anti-chlamydial properties of the newly synthesized 4-chloroisocoumarins.  Successive optimization studies have investigated the effects of replacing various amino acids within the JO146 structure, and substitution on the pyrrolidine ring, with some improvement to the inhibitory activity without loss of anti-chlamydial properties [11,12].However, off-target effects have been observed through the inhibition of Human Leukocyte Elastase (HLE) [9,[11][12][13], a serine protease that has been implicated in the pathogenesis of Chronic Obstructive Pulmonary Disease [14].Thus, we hypothesized that HLE inhibitors may exhibit inhibitory activity against CtHtrA and that these chemical scaffolds may serve as a source of new lead compounds as CtHtrA inhibitors.

Compound
Isocoumarin compounds have been investigated for many years as general serine protease inhibitors, and 3,4-dichloroisocoumarin is a commercial serine protease inhibitor.In the development of HLE inhibitors, 3-alkoxy-4-chloroisocoumarins and their C-7substituted derivatives have been identified as rapid and potent inhibitors [15][16][17][18].As the C-3 and C-7 positions govern the selectivity and drive the inhibitory properties of 3-alkoxy-4-chloroisocoumarins against serine proteases [15][16][17][18], we focused our investigation on uncovering the structure-activity relationships (SARs) of these positions (Figure 2) with regard to their inhibitory properties against CtHtrA.We also sought to investigate the anti-chlamydial properties of the newly synthesized 4-chloroisocoumarins.
with a terminal N-Boc capping group and an electrophilic diphenylphosphonate headgroup that reacts with the serine protease to form a reversible covalent complex with CtHtrA (IC50 = 12 µM) [9].JO146 exhibits anti-chlamydial properties against C. trachomatis and other chlamydial species with minimal cytotoxic properties against mammalian cell lines including HEp-2 cells, McCoy B cells and ex vivo Koala mononuclear blood cells [9].Similar structures with modifications to the N-terminal capping group retain their antichlamydial activity [10].However, their inhibitory activity against CtHtrA has not been investigated.Successive optimization studies have investigated the effects of replacing various amino acids within the JO146 structure, and substitution on the pyrrolidine ring, with some improvement to the inhibitory activity without loss of anti-chlamydial properties [11,12].However, off-target effects have been observed through the inhibition of Human Leukocyte Elastase (HLE) [9,[11][12][13], a serine protease that has been implicated in the pathogenesis of Chronic Obstructive Pulmonary Disease [14].Thus, we hypothesized that HLE inhibitors may exhibit inhibitory activity against CtHtrA and that these chemical scaffolds may serve as a source of new lead compounds as CtHtrA inhibitors.
Isocoumarin compounds have been investigated for many years as general serine protease inhibitors, and 3,4-dichloroisocoumarin is a commercial serine protease inhibitor.In the development of HLE inhibitors, 3-alkoxy-4-chloroisocoumarins and their C-7substituted derivatives have been identified as rapid and potent inhibitors [15][16][17][18].As the C-3 and C-7 positions govern the selectivity and drive the inhibitory properties of 3alkoxy-4-chloroisocoumarins against serine proteases [15][16][17][18], we focused our investigation on uncovering the structure-activity relationships (SARs) of these positions (Figure 2) with regard to their inhibitory properties against CtHtrA.We also sought to investigate the anti-chlamydial properties of the newly synthesized 4-chloroisocoumarins.

Design and Synthesis of 4-Chloroisocoumarin Compounds
Figure 2 shows the series of 3-alkoxy-4-chloroisocoumarin compounds synthesized in the current work.The general procedure used to prepare 2a-g is shown in Scheme 1. Homophthalic acid was converted to the corresponding monoesters (1a-g) using the appropriate alcohols with sulfuric acid.Compounds 1a-g were cyclized to the corresponding 3-alkoxy-4-chloroisocoumarin compounds (2a-g) using phosphorus pentachloride.The cyclization/chlorination step has been reported to take up to 16 h at 80-110 • C [19], but in our hands this reaction was complete after 4 h at 40
Chloro-and methoxy-analogues (6g and 6h, respectively) were synthesized using the procedure shown in Scheme 3. Compound 3 was reduced to the corresponding amine (7) using an adapted procedure [23] and diazotized to give 5-chlorohomophthalic acid [22] (8).Monoesterification with (±)-sec-butanol ( 9) and cyclization with phosphorus pentachloride gave the corresponding dichloroisocoumarin (6g).Diazotization of 7 followed by acidic thermolysis gave the corresponding phenol (10) [24], which was esterified to the corresponding dimethyl ester (11) using a Fischer esterification and then methoxylated with methyl iodide to give (12) [24].The methyl ethers were hydrolyzed under basic conditions, affording (13) [25].Compound 13 was monoesterified with (±)-sec-butanol to give (14) and then cyclized using the described procedure.We note that cyclization of the methoxy and chloro analogues under the reported conditions (using toluene as a solvent at reflux) did not afford the desired product; hence it may be possible that other unexplored C-7 substituents may be compatible with this chemistry.
Homophthalic acid was nitrated using red fuming nitric acid [22] and monoesterified to the corresponding sec-butoxy monoester.This was cyclized to give 3-sec-butoxy-7-nitro-4-chloroisocoumarin (5) and then reduced to the amine (6a) with zinc dust in acetic acid.Acylation to the corresponding amides and reductive double-alkylation gave the dimethylamino analogue (6b).Diazotization reactions were optimized to synthesize iodo (6e) and bromo (6f) analogues but did not afford the respective phenolic or chloro moieties.
Chloro-and methoxy-analogues (6g and 6h, respectively) were synthesized using the procedure shown in Scheme 3. Compound 3 was reduced to the corresponding amine (7) using an adapted procedure [23] and diazotized to give 5-chlorohomophthalic acid [22] (8).Monoesterification with (±)-sec-butanol ( 9) and cyclization with phosphorus pentachloride gave the corresponding dichloroisocoumarin (6g).Diazotization of 7 followed by acidic thermolysis gave the corresponding phenol (10) [24], which was esterified to the corresponding dimethyl ester (11) using a Fischer esterification and then methoxylated with methyl iodide to give (12) [24].The methyl ethers were hydrolyzed under basic conditions, affording (13) [25].Compound 13 was monoesterified with (±)-sec-butanol to give ( 14) and then cyclized using the described procedure.We note that cyclization of the methoxy and chloro analogues under the reported conditions (using toluene as a solvent at reflux) did not afford the desired product; hence it may be possible that other unexplored C-7 substituents may be compatible with this chemistry.

Inhibitory Properties of 4-Chloroisocoumarins against CtHtrA and HLE
The inhibitory activity of compounds 2a-g against CtHtrA was evaluated.These compounds did not exhibit substantial interference with fluorescence signals during the protease inhibition assays (see Figure S1).As these compounds are reportedly covalent inhibitors of serine proteases, a true IC 50 is unable to be calculated.Instead, the residual activity of the protease is monitored over time in response to a range of compound concentrations, and an IC 50 value is interpolated from a dose-response curve (see Figures S2-S4 and Tables S1 and S2).This IC 50 value is then compared to a positive control (JO146 was used for 2a-g, and 2g was used for 6a-h), giving the 'relative potency'.This is the established method for assessing CtHtrA inhibitory potency [9,11,12].Table 1 shows the inhibitory properties of these compounds relative to the known inhibitor JO146 (Figure 1).Extension of the R 1 carbon chain from methyl to ethyl (2a vs. 2b) gave a 1.5-fold increase in activity (relative to JO146), and potency reduced as the chain length was extended (2b vs. 2c and 2d).This effect may be due to depth restrictions within the S1 subpocket, as kinetic studies on the rates of protease inhibition in HLE were reduced by larger C-3 alkoxy substituents [17].Chain branching (iso-propyl and sec-butyl substituents) increased the activity, resulting in a 1.4-fold increase for 2e over 2b and a 2.8-fold improvement for 2g over 2c.However, the isobutyl analogue 2f exhibited a slight reduction in activity compared to 2c, which suggests that elongation or increases in steric bulk toward the end of the aliphatic chain are not well tolerated.C-7 functionalized 4-chloro-3-sec-butoxyisocoumarins 6a-h were compared against 2g rather than JO146, as 2g is structurally more comparable, and improvements to its potency were the aim of this investigation.Table 1 shows that all C-7 modifications resulted in reduced activity compared to 2g.Compounds substituted with small, polar and electrondonating groups were identified as the most potent compounds in this series (6a-c and 6h), with the exception of 6g, which had the highest activity.
Substituent lipophilicity, as determined from their substituent hydrophobicity constants (π; see Table S3) [26], was identified as a significant factor in attenuating the CtHtrA inhibitory properties at the C-7 position.In general, the compounds bearing lipophilic substituents had little to no activity (6d and 6e), while analogues with less lipophilic substituents were better CtHtrA inhibitors (6e-g, 6a vs. 6b and 6c vs. 6d).
As compounds 2a-g showed effective inhibition of CtHtrA relative to JO146, attention was turned to evaluating their inhibitory properties against HLE to ascertain selectivity between the two proteases.As shown in Table 1, 2a-g showed poor to modest inhibition of HLE in comparison to JO146, possibly due to their reportedly high molecular turnover, as 3-choro-and 3,4-dichloroisocoumarin have a molecular turnover of 15 and 3.1, respectively, against HLE [16].Compound 2a showed the poorest inhibition relative to JO146.Substitution with longer carbon chains (2c and 2d vs. 2a and 2b) produced the most active inhibitors, potentially owing to better stabilization within the S1 subpocket after hydrolysis.Similar to their inhibitory properties against CtHtrA, 2e and 2g afforded the most potent inhibition of HLE in this series.This suggests comparable subpocket homology between the two proteases.
4-Chloroisocoumarins 2a-g were found to inhibit CtHtrA preferentially over HLE.Compounds 2a and 2b showed a strong preference for the inhibition of CtHtrA, but this is likely due to their preference for hydrolysis by HLE.Of note, the increase in aliphatic chain length increased the selectivity toward HLE (2a-d).Isomerization of the butyl chain resulted in a slight preference for HLE (2d and 2f), while 2c and 2e and also 2d and 2g showed a significant increase in preference for CtHtrA.This finding indicates that substitution at the C-1 ′ position of the aliphatic chain could be a driver for selectivity between the two enzymes, with 2g showing the most promise as a CtHtrA inhibitor.

4-Chloroisocoumarin Cytotoxic Effects in HEp-2 Cells and McCoy B Cells
The cytotoxic effects of the 4-chloroisocoumarin compounds 2a-g and 6a-h (24 h) were tested against HEp-2 and McCoy B cells using the MTS cell viability assay.HEp-2 and McCoy B cells were selected, as these are routinely used for the cultivation of chlamydial species, and low cytotoxicity against host cells is required to evaluate the anti-chlamydial properties of a test compound [27].As shown in Figure 3, it was found that compounds 2b-2g had no effect on HEp-2 cell viability at concentrations up to 125 µM.The methyl-substituted compound 2a reduced HEp-2 cell viability to 50% of DMSO-treated control cells at 125 µM but did not produce significant effects at the lower concentrations of 62. 5   Substitution of the C-7 position in 2g with halogens (6e-6g) increased the cytotoxicity against HEp-2 cells at 125 µM, while only 6e retained some cytotoxicity at 62.5 µM, as shown in Figure 3. Compound 6a reduced cell viability slightly at 125 µM, while 6b was not found to significantly impact cytotoxicity.Acetamides 6c and 6d and the methoxy analogue 6h did not significantly affect cell viability at any tested concentration.From these data, it was determined that HEp-2 cells could be used as a host cell for anti-chlamydial testing at a concentration of 62.5 µM, as this was the highest concentration that could be used without having a large impact on HEp-2 cell viability across the tested con- Substitution of the C-7 position in 2g with halogens (6e-6g) increased the cytotoxicity against HEp-2 cells at 125 µM, while only 6e retained some cytotoxicity at 62.5 µM, as shown in Figure 3. Compound 6a reduced cell viability slightly at 125 µM, while 6b was not found to significantly impact cytotoxicity.Acetamides 6c and 6d and the methoxy analogue 6h did not significantly affect cell viability at any tested concentration.From these data, it was determined that HEp-2 cells could be used as a host cell for anti-chlamydial testing at a concentration of 62.5 µM, as this was the highest concentration that could be used without having a large impact on HEp-2 cell viability across the tested concentration range.Compounds were also tested against McCoy B cells (see Figure S5) but exhibited cytotoxicity at 31.25 and 62.5 µM, and thus this cell line was unable to be used for the cultivation of chlamydia to evaluate anti-chlamydial properties.

4-Chloroisocoumarin Compound Effects against C. trachomatis
The workflow for the evaluation of chlamydial growth inhibition is described in the supplementary information (Figure S6).Briefly, HEp-2 cells were seeded in a 96-well plate and grown for 24 h; they were then infected with a fixed quantity of C. trachomatis.The infection was propagated for 4 h, and then a mixture of test compound and cycloheximide was added and the plate was incubated for a further 26 h (30 h post infection).The cells were fixed with methanol and then stained with fluorescent antibodies to visualize chlamydial inclusions, and the growth was measured using inclusion counts relative to vehicle controls.
The 3-alkoxy-4-chloroisocoumarins 2a-2f did not significantly reduce the proliferation of C. trachomatis, with the exception of 2g, which reduced chlamydial growth by 49% (Figure 4).C-7-substituted derivatives of 2g were generally more active than the unsubstituted analogue 2g.Amino (6a) and dimethylamino (6b) analogues successfully showed that ≥90% of chlamydial inclusions were affected, suggesting that small polar electrondonating groups at the C-7 position facilitate inhibition of chlamydial growth against C. trachomatis.Consistent with this, the methoxy substituted analogue 6h had similar levels of activity.The acetamido (6c), iodo (6e) and chloro (6g) moieties did not significantly affect the efficacy of 2g at 62.5 µM, while bromination (6f) abolished observable activity.Trifluoromethylacetamide (6d) was found to decrease chlamydial growth with comparable efficacy to the methoxy (6h) analogue.Compounds with statistically significant anti-chlamydial activity were generally found to be poor inhibitors of CtHtrA (6a, 6b, 6d, 6e, 6g, 6h), while compounds with poor inhibitory activity against C. trachomatis proliferation exhibited good activity against CtHtrA (2a-g).Although this illustrates that C-7-substituted 4-chloroisocoumarins can attenuate the growth of C. trachomatis, it is likely that their activity is not due to the inhibition of CtHtrA.
Compounds 6a and 6b were further investigated as they exhibited the most activity against C. trachomatis at 62.5 µM.In anti-chlamydial assays, the MIC (minimum inhibitory concentration) of a compound is defined as twice the MICTP (the MIC transition point), which is the concentration required to alter >90% chlamydial morphology (i.e., a reduction Compounds with statistically significant anti-chlamydial activity were generally found to be poor inhibitors of CtHtrA (6a, 6b, 6d, 6e, 6g, 6h), while compounds with poor inhibitory activity against C. trachomatis proliferation exhibited good activity against CtHtrA (2a-g).Although this illustrates that C-7-substituted 4-chloroisocoumarins can attenuate the growth of C. trachomatis, it is likely that their activity is not due to the inhibition of CtHtrA.
Compounds 6a and 6b were further investigated as they exhibited the most activity against C. trachomatis at 62.5 µM.In anti-chlamydial assays, the MIC (minimum inhibitory concentration) of a compound is defined as twice the MIC TP (the MIC transition point), which is the concentration required to alter >90% chlamydial morphology (i.e., a reduction in size) [27].For 6a and 6b, dose-response relationships for their anti-chlamydial activity were identified (Figure 5).However, 6a did not consistently show inhibition greater than the 90% threshold at any concentration tested.Compound 6b showed greater than 90% inhibition at concentrations ≥ 31.25 µM.These results show that while both 6a and 6b exhibit dose-response relationships against the inhibition of C. trachomatis growth, only compound 6b has a calculable MIC of 62.5 µM, as per the reported definition [27].
Compounds with statistically significant anti-chlamydial activity were generally found to be poor inhibitors of CtHtrA (6a, 6b, 6d, 6e, 6g, 6h), while compounds with poor inhibitory activity against C. trachomatis proliferation exhibited good activity against CtHtrA (2a-g).Although this illustrates that C-7-substituted 4-chloroisocoumarins can attenuate the growth of C. trachomatis, it is likely that their activity is not due to the inhibition of CtHtrA.
Compounds 6a and 6b were further investigated as they exhibited the most activity against C. trachomatis at 62.5 µM.In anti-chlamydial assays, the MIC (minimum inhibitory concentration) of a compound is defined as twice the MICTP (the MIC transition point), which is the concentration required to alter >90% chlamydial morphology (i.e., a reduction in size) [27].For 6a and 6b, dose-response relationships for their anti-chlamydial activity were identified (Figure 5).However, 6a did not consistently show inhibition greater than the 90% threshold at any concentration tested.Compound 6b showed greater than 90% inhibition at concentrations ≥ 31.25 µM.These results show that while both 6a and 6b exhibit dose-response relationships against the inhibition of C. trachomatis growth, only compound 6b has a calculable MIC of 62.5 µM, as per the reported definition [27].

General Procedure for the Synthesis of Unsubstituted Homophthalate Monoesters (1a-1g)
Homophthalic acid (1.5 g, 8.3 mmol) was dissolved in the corresponding anhydrous alcohol (20 equiv.), to which sulfuric acid (98%, 1 mL) was added.The reaction mixture was heated under reflux for methyl, ethyl and isopropyl alcohols, and at 80 • C for other alcohols, until the starting material was no longer present by TLC analysis (ethyl acetate mobile phase, typically 1-1.5 h, except the methyl ester, which took 30 min).Work-up for methyl and ethyl analogues followed work-up procedure A. All other analogues followed work-up procedure B.
Work-up A: The reaction was diluted with ethyl acetate (30 mL) and washed with saturated sodium bicarbonate (3 × 50 mL).The organic phase was removed, and the aqueous phase was acidified to pH 2-3 with aqueous hydrochloric acid (1 M) and then washed with ethyl acetate (3 × 50 mL).The combined extracts were washed with brine (100 mL), dried using anhydrous magnesium sulfate and then concentrated with reduced pressure to afford solids that were crystallized from 10% ethyl acetate in hexane.
Work-up B: The reaction was diluted with chloroform (50 mL) and washed with water (2 × 100 mL) and then brine (50 mL) and dried using anhydrous magnesium sulfate.The solvent was removed with reduced pressure and dissolved in a minimal amount of ethyl acetate (typically 2-3 mL) at 60 • C and then slowly diluted with hexanes until the solution became slightly cloudy.The mixture was cooled to room temperature over 30 min and then further with an ice bath for one hour.The resulting white needles were filtered and washed with hexanes (2 × 20 mL) and dried with reduced pressure for 16 h. 1 H, 13 C NMR spectra and melting points for compounds 1a-e were consistent with those previously reported [19,30].

General Procedure for the Synthesis of 3-alkoxy-4-chloroisocoumarins (2a-2g) (General Procedure A)
This procedure was adapted from the literature [19].Homophthalate monoester (up to 4 g) was dissolved in anhydrous toluene (120 mL/g), and the flask was backfilled three times with argon gas.Phosphorus pentachloride (3.0 equiv.) was added in a single portion with stirring, and the flask was evacuated for one minute and then backfilled with argon gas and stirred at 40 • C for four hours.The reaction was cooled on an ice bath, and water (50 mL) was added and stirred while cold for five minutes.The toluene phase was separated and washed sequentially with water (50 mL), saturated aqueous sodium bicarbonate (50 mL) and brine (50 mL) and then dried using magnesium sulfate and concentrated with reduced pressure to afford a yellow or orange oil that was purified on silica gel (0-20% dichloromethane in hexanes) and then recrystallized by heating to 40 • C in hexane and then cooling to −15 • C (approximately 1 mL of hexane per 100 mg of material).
3.1.6.Synthesis of (±)-7-(dimethylamino)-3-sec-butoxy-4-chloroisocoumarin (6b) Compound 6a (0.400 g, 1.50 mmol) was added to glacial acetic acid (12 mL) in a sealed vial, and the mixture was stirred vigorously until all material dissolved.Paraformaldehyde (0.449 g, 10 equiv.) was added and stirred at room temperature with the vial sealed for one hour.Sodium cyanoborohydride (0.470 g, 5 equiv.)was added and stirred overnight.The reaction mixture was neutralized by slow addition to saturated aqueous sodium carbonate (100 mL) and then extracted with chloroform (3 × 50 mL).The combined extracts were washed with water (2 × 100 mL) and brine (100 mL) and dried using anhydrous potassium carbonate.The solvent was removed with reduced pressure and purified using gradient silica gel column chromatography with the crude mixture loaded using 10% dichloromethane in hexane and eluted with 25-50% dichloromethane in hexane in 5% increments to afford a bright yellow oil (0.154 g, 35%).
TALON ® metal affinity resin (Takara cat no.635652, 5 mL) was added to each lysate tube and washed sequentially with lysis buffer (2 × 45 mL) and stringency buffer (4 × 45 mL) and then eluted with elution buffer (2 5 mL), allowing incubations at 4 • C for 20 min per wash, and centrifuging at 2500× g and 4 • C between washes to pellet the resin.The eluates were dialyzed in Tris buffer (pH 7.0, 50 mM, 4 L) with magnesium chloride (20 mM) at 4 • C five times over 24 h, with a minimum of three hours per dialysis change.Protein expression and isolation were confirmed with SDS-PAGE Coomassie stained gel, which was examined for the band at the correct size, and a BCA assay to confirm purified protein concentration prior to use.

CtHtrA Protease Inhibition Assay
CtHtrA was added to 25 wells in a black 96-well plate (40-80 µL, 10 µL increments), to which Tris-HCl buffer (50 mM, pH 7.0 with 20 mM MgCl 2 ) was added to bring the total volume to 90 µL for 10 wells and 85 µL for the remaining 15.JO146 (10 µM, 5 µL) was added to 5 wells and JO146 (1 µM, 5 µL) was added to another 5 wells, with DMSO (5 µL) being added to the remaining 5 as a control.The plate was incubated at 37 • C for 10 min before substrate (MCA-ENLHLPLPIIFK-DNP, 10 µL, 1 mg/mL, 50% isopropanol in buffer) was added and analyzed in a TECAN Infinite Pro M200 plate reader with an excitation wavelength of 340 nm and fluorescence measured at 405 nm with four reads per well at 60 s intervals.A minimum volume of CtHtrA granting a fluorescence change of 300 min −1 was used for subsequent assay.
The determined appropriate volume of CtHtrA in buffer was added to a 96-well non-transparent black plate and diluted to 85 µL with buffer solution.Drug or control (5 µL) at a maximum of 20% DMSO/80% buffer was added to the appropriate well with DMSO controls, and the plate was incubated for 10 min to allow activation by the protein.MCA-ENLHLPLPIIFK-DNP (Mimotopes, Australia, 1 mg/mL, 10 µL) was added and immediately monitored on a TECAN M200 Infinite plate reader at 37 • C every 60 s for 30 min, using the same wavelengths as for the activity evaluation.
The rate of change of fluorescence was compared to controls to determine % inhibition of protein in Microsoft Excel TM , and these values were exported to GraphPad Prism for IC 50 calculation.The positive control for 2a-g was JO146, and (±)-2g was used as a positive control for 6a-h.

HLE Protease Inhibition Assay
Elastase was purchased from Athens Research and Technology (Cat No. 16-15-051200).Elastase (100 µg) was dissolved in Tris buffer (100 mM Tris, 20 mM CaCl 2 , pH 8.1, 3.14 mL), giving a concentration of 1.117 µM, and 25 µL was added to each well containing the respective inhibitor.The plate was incubated at 37 • C for 15 min.
N-Suc-AAA-pNA (Sigma Aldrich, Bayswater, Australia, S4760, 100 µL, 1 mg/mL, 45.1 µmol) was added to each well, and the absorbance was monitored every six minutes at 405 nm.The rate of change of absorbance was used as a comparison to controls to determine the % inhibition of protein in Microsoft Excel TM (v2402), and these values were exported to GraphPad Prism (v8.0.1) for IC 50 calculation.

MTS Cell Viability Assay
HEp-2 (human epithelial type 2, CCL-23, ATCC ® , Virginia, USA) or McCoy B (mouse fibroblast, CRL-1696, ATCC ® ) cells were grown to 80-100% confluence, and cell density was determined with a haemocytometer.Cells were then diluted to 1 × 10 5 cells/mL, and 100 µL was added to each well in a 96-well plate, with 3 wells left as controls.Cells were grown for 24 h, and the media was removed and then replaced with fresh media containing 0.1% DMSO and the drug for testing in triplicate.Drugs and cells were incubated for 24 h, and then 20 µL of MTS and PMS solution was added and then incubated for a further four hours.Absorbance was measured at 490 nm on a TECAN M200 Infinite plate reader and data were processed using Microsoft Excel TM with the following equation to determine cell viability relative to DMSO controls.
The supernatant was removed, and 100 µL of the respective chlamydia species in media (100,000 IFU/mL) was added, giving an MOI of 0.5.The plate was sealed and centrifuged at 500× g/28 • C for 30 min and then incubated for a further 3.5 h at 37 • C with 5% CO 2 .The supernatant was removed, and 100 µL of cycloheximide in DMEM (1 µg/mL) was added in addition to 100 µL of drug stock solution to each respective well with DMSO and media negative controls and also azithromycin positive control.The plate was incubated for a further 26 h.
The supernatant was removed, and wells were washed with PBS (2 × 150 µL) and then fixed in methanol for eight minutes.The methanol was removed and washed with additional PBS (2 × 150 µL) and stored at 4 • C with 100 µL PBS per well and then stained and analyzed using fluorescence microscopy, which is outlined below.
Cells in fixed plates were permeabilized with 0.5% Triton X-100 in PBS for 15 min and then blocked with 1% BSA in PBS for 45 min.Anti-HtrA polyclonal antibody (sera) (1:600) (in-house [8]), anti-mouse tubulin (Sigma-Aldrich Cat no.T-5168, 1:3000) and DAPI (125 ng/mL) in PBS were added to each well (70 µL) and then gently shaken at room temperature while shielding it from light for 45 min.The primary antibodies were removed, and the wells were washed with PBS (4 × 150 µL).Secondary antibodies (goat anti-rabbit Alexafluor 488, Thermo Fisher Cat No. A-11008 and goat anti-mouse Alexafluor 568 (Thermo Fisher Cat No. A-11004)) in PBS (70 µL, 1:600 respectively) were added and the wells were gently shaken at room temperature while shielding them from light for 45 min.The antibodies were removed, and the wells were washed with PBS (5 × 150 µL); cells were then stored with PBS (100 µL per well) at 4 • C pending analysis.
Plates were analyzed on an INcell Analyzer 2200 (Cytiva Life Sciences, Marlborough, MA, USA) using both laser autofocus (10%) and software autofocus modes, with ten random images per well covering a minimum of 80% of the well at 10x magnification.Non-morphologically affected inclusions were counted using a macro in ImageJ (v1.54f), and MIC was defined as being one two-fold dilution more concentrated than MIC TP (where MIC TP is defined as the lowest concentration to produce ≥90% morphologically abnormal inclusions) relative to solvent controls [27].

Figure 5 .
Figure 5. Dose-response relationships for 6a and 6b against C. trachomatis (30 h post infection).Values represent the average ± standard error of the mean from three independent experiments.