S. pneumoniae was grown at 37°C in a casein hydrolysate-based liquid medium (AGCH) in which ZnSO₄ was depleted and adjusted according to the experimental needs. This medium also contained 0.2% yeast extract and 0.3% sucrose (Lacks et al., 1986). Transformation was performed as previously reported (Lacks et al., 1986). Selection of transformants was made in 1 μg/ml tetracycline, 250 μg/ml kanamycin, or 2.5 μg/ml chloramphenicol when strains were transformed with plasmid pLS1, the kanamycin resistance cassette (kan), or the chloramphenicol resistance cassette (cat), respectively. To induce DNA relaxation, 0.25 μg/ml (0.25 × MIC) of novobiocin (Nov) was added to the culture. Growth was monitored by measuring OD_620nm either in an UV–visible spectrophotometer (Evolution 201, Thermo Scientific) or in a microplate reader (Infinite F200, Tecan). Both measures correlate linearly by means of the equation y = 0.2163 x + 0.1151 (y = microplate reader measure, x = spectrophotometer), with an R² of 0.98 (Ferrándiz et al., 2018).

Strain P_ZntopA (Figure 1A) containing topA (spr1141) under the control of the P_Zn promoter at the spr1865 locus, which is not esssential for growth (Martín-Galiano et al., 2014), was constructed as follows. Three DNA fragments were obtained by PCR. All primers are detailed in Table 1. The first PCR product (2,157 pb) included topA flanked by restriction sites present in primers TOPAUPSAC (SacI) and TOPADOWNSAL(SalI), which were used to amplify topA from R6 chromosome. The second product (671 pb) contained a consensus transcriptional terminator followed by the C-terminus of spr1866 amplified from the chromosome of R6P_ZnhlpΔhlp (Ferrándiz et al., 2018) with primers SPR1866R and TERSAL (SalI). The third PCR product (1,704 pb) contained P_Zn, kan, the N-terminus of spr1865, spr1864, and spr1863. It was also amplified from R6P_ZnhlpΔhlp with primers PZNRSAC (SacI) and SPR1863F. These three PCR products were digested and ligated. The ligation product was used as a template to obtain a 4,532-bp PCR product with the oligonucleotide pair SPR1866R/SPR1863F, which was used to transform R6 competent cells. Transformants were selected by plating in AGCH-agar medium without ZnSO₄ (to avoid overexpression of topA) containing 250 μg/ml kanamycin. To confirm the insertion in spr1865, amplification from the chromosome was performed with primers SPR1866R2 and SPR1863F2 flanking the inserted DNA. Primers TOPAUP2, TOPARTF, TOPARTR, and TOPADOWN were used to sequence P_ZntopA insertion.

To construct strain ΔtopAP_ZntopA (Figure 1A), topA was disrupted in the chromosome of P_ZntopA by homologous recombination as follows. Three DNA fragments were obtained by PCR amplification. Two fragments upstream and downstream topA of 1,103 and 1,149 bp were amplified with primers TOPAKOF1 and TOPAKORI (EcoRI) and TOPAKOF3 and TOPAKOR2 (HindIII), respectively, using R6 DNA as template. A third DNA fragment of 924 bp containing the cat cassette was amplified from plasmid pJS3 with primers ECOCATUP (EcoRI) and CATDOWNHIN (HindIII). The three fragments were digested and ligated, and the ligation product was used as a template to obtain a 3,176 bp PCR product (oligonucleotide pair TOPAKOF1/TOPAKOR2), which was used to transform P_ZntopA. Transformants were selected by plating in AGCH-agar medium supplemented with 150 μM ZnSO₄ (to allow topA expression) and 2.5 μg/ml chloramphenicol. To confirm the disruption, amplification from the chromosome was performed with primers TOPAKOF2 and TOPAKOR3 flanking the replaced DNA. To sequence the construct, primers TOPAUP2, TOPARTF, TOPARTR, TOPADOWN, CATMED, and CAT191 were used.

Total RNA was extracted from 10 ml of culture (OD_620nm = 0.4) with RNeasy kit (Qiagen), following the manufacturer’s instructions, with the exception that RNA was treated 3-fold with DNase I. cDNAs were synthesized from 5 μg of RNA with SuperScript™ IV Reverse Transcriptase (Thermo-Fisher) for 10 min at 55°C. A total of 2 μl of diluted cDNA (2,000-fold for 16S rDNA and one 20-fold for the rest of amplicons) were used as template in a subsequent qRT-PCR (CFX Opus 96, Bio-Rad) using 10 μl of SsoAdvanced Universal SYBR Green Supermix (Bio-Rad). Amplification was achieved as previously described (Ferrándiz et al., 2010). Primer pairs used (GYRARTF/GYRARTR, TOPARTF/TOPARTR, and 16SDNAF3/16SDNAR3) are indicated in Table 1. Three independent assays were performed for quantification. Analysis of qRT-PCR data was performed using the 2^−ΔΔCq method (Livak and Schmittgen, 2001) using an internal fragment of 16S rDNA gene as an internal control and 0 ZnSO₄ condition expression levels as the calibrator.

Whole cell lysates (~ 5 × 10⁵ cells) were obtained by centrifugation of a 10-ml culture (OD_620nm = 0.4), resuspended in 400 μl of phosphate buffered saline and sonicated for 20 min (30 s ON/30 s OFF) with a Bioruptor® Pico sonication device (Diagenode). Lysates (~ 2 × 10⁴ cells) were separated on Any kD™ Criterion™ TGX Stain-Free™ Protein Gels (Bio-Rad). They were transferred to 0.2 μm PVDF membranes with a Trans-Blot Turbo Transfer System (Bio-Rad) at 25 V, 1 A for 30 min. Membranes were blocked with 5% milk in Tris-buffered saline for 2 h and incubated with anti-TopoI (diluted 1:500), anti-GyrA (diluted 1:2,000; Ferrándiz et al., 2016), and anti-RpoB (Ferrándiz et al., 2021; diluted 1:500). Anti-rabbit IgG-Peroxidase antibody (Sigma-Aldrich) was used as the secondary antibody. SuperSignal West Pico chemiluminescent substrate (Thermo-Fisher) was used to develop the membranes. Signal was detected with a ChemiDocTM MP system (Bio-Rad). Images were analyzed using Image Lab™ software (Bio-Rad). The number of molecules of TopoI and GyrA per cell were estimated by Western blot and by counting CFUs (colony-forming units) at each growth condition. Purified GyrA (2.5 –20 ng) and TopoI (1.5 –24 ng) proteins were used to standarize the correlation between protein amount and immunostaining intensity, which was used to estimate the amount of these proteins in cell lysates. RpoB was used as an internal loading control. For that, an average of the RpoB signal of all wells was calculated, and the deviation of the RpoB signal on each well from this value was assessed. These values were used to adjust TopoI and GyrA signals. CFUs were determined by plating cell extracts on AGCH with 0.2% yeast extract and 0.3% sucrose agar plates. Molecular masses of GyrA and TopoI are 92.04 and 79.38 kDa, respectively. Determinations were performed in triplicate.

Plasmid DNA topoisomers were analyzed in neutral/neutral two-dimensional agarose gels. The first dimension was run at 1.5 V/cm in a 0.4% agarose (Seakem; FMC Bioproducts) gel in 1 × Tris-borate-EDTA (TBE) buffer for 20 h at room temperature. The second dimension was run at 7.5 V/cm in 1% agarose gel in 1 × TBE buffer for 7–9 h at 4°C. Chloroquine (Sigma) was added to the TBE buffer in both, the agarose and the running buffer. Chloroquine is a DNA intercalating agent that removes negative Sc in bacterial plasmids. Increasing concentrations of chloroquine progressively eliminate negative Sc until the plasmid is relaxed and can then introduce net positive Sc. In this way, the use of adequate concentrations of chloroquine during each dimension in the 2D analysis allows the efficient resolution of the different topoisomers (Schvartzman et al., 2013). Gels were stained with 0.5 μg/ml ethidium bromide for 1 h at room temperature. When this staining was not enough for topoisomers visualization, after electrophoresis, gels were subjected to Southern hybridization. A 240-bp PCR fragment amplified from pLS1 obtained as described (Ferrándiz et al., 2010) was used to probe on two-dimensional agarose gels transferred to nylon membrane (Inmobylon NY⁺, Millipore). Streptavidin-HRP (SouthernBiotech) was used to detect the DNA and signal was developed with the SuperSignal West Pico chemiluminescent substrate (Thermo-Fisher). Images were captured in a ChemiDoc Imaging System (Bio-Rad) and analyzed with the Image Lab software (BioRad).

Cultures of strains P_ZntopA and ΔtopAP_ZntopA were grown at mid-log growth phase in media with different amounts of ZnSO₄. Samples (from 5 × 10⁷ to 2 × 10⁸ cells) were collected, washed in 10 mM phosphate buffer (pH 7.2), and fixed in 2% of paraformaldehyde for 48 h. Fixed cells were washed, suspended in buffered salt solution (137 mM NaCl, 5.4 mM KCl, 10 mM Tris–HCl, and pH 7.6) and incubated with 5 μM Sytox™ Orange Nucleic Acid Stain (Invitrogen) for 5 min at room temperature. ProLong™ Gold Antifade Mountant (Invitrogen) was added to fixed cells, and the mixture was transferred onto poly-L-lysine coated glass slides (Sigma-Aldrich). Slides were observed using a confocal microscope STELLARIS 8–FALCON/STED (Leica Microsystems) with a HC PL APO 100×/1.40 NA × OIL immersion objective. Super resolution images were acquired by Stimulated Emission Depletion (STED) microscopy using 660_nm depletion laser. Image J software was used for image analysis. Average Sytox fluorescence intensity was measured from the nucleoids, defined as the region of each cell with intensity values between 20 and 225. Raw integrated density measurements divided by area (Mean Gray Values) were obtained for 1,307–4,562 nucleoids. GraphPad Prism 9.1 was used to represent the average intensities vs. Sc density (σ) and to perform a simple linear regression.

To ascertain the role of TopoI levels in cell viability, a strain derived from R6 carrying a deletion of topA at its native chromosomal location and an ectopic topA copy under the control of the P_Zn promoter was constructed as described in the section “Materials and methods” (Figure 1A). This strain, named ΔtopAP_ZntopA, depended on ZnSO₄ to grow (Figure 1B). It did not grow in the absence of ZnSO₄, but growth was restored when ≥75 μM ZnSO₄ was added. On the other hand, strain P_ZntopA, which contains topA at its native location and an additional ectopic topA copy under the control of P_Zn, grew normally either in the absence or in the presence of 75 μM of ZnSO₄, while higher ZnSO₄ concentrations (150 or 300 μM) compromised its growth. Accordingly, TopoI levels correlated with the concentration of ZnSO₄ in the growth medium (Figure 2).

The number of TopoI molecules per cell at each condition was estimated by using Western-blot to determine the amount of protein and by counting CFUs. Results are showed in Table 2. In strain ΔtopAP_ZntopA, a residual amount of TopoI (1.5 ng) was detected in the absence of ZnSO₄, while the number of molecules increased to 665, 848, and 1,247 in the presence of 75, 150, and 300 μM ZnSO₄, respectively (Table 2). At 300 μM ZnSO₄, the number of TopoI molecules was similar to the values estimated for R6 at any ZnSO₄ concentration. The highest number of TopoI molecules was detected in P_ZntopA; being of 2,879 with 150 μM and 3,352 with 300 μM of ZnSO₄. This high number of TopoI molecules exerted a negative effect on the growth of this strain (Figure 1B) compared with its growth at either no ZnSO₄ (1,726) or 75 μM of ZnSO₄ (2,250).

No ZnSO₄-dependent variation in the number of gyrase molecules was detected in strains R6, P_ZntopA or ΔtopAP_ZntopA (Table 2). The ratio of GyrA:TopoI molecules in R6 was of 1:0.66, 1:0.71, 1:0.60, and 1:0.65 when grown in 0, 75, 150, and 300 μM ZnSO₄, respectively. In P_ZntopA, the GyrA:TopoI ratio experienced a progressive reduction as ZnSO₄ concentration increased, until a value of 1:1.26 at 300 μM, while half the proportion was observed in R6 strain at the same ZnSO₄ concentration. This is in agreement with the progressive reduction of cell growth of R6P_ZntopA at increasing ZnSO₄ concentration compared to R6 strain (Figure 1B). GyrA:TopoI ratio was higher in ΔtopAP_ZntopA than in R6 when ZnSO₄ concentrations of 75 or 150 μM ZnSO₄ were used. The highest proportion (1:0.09, calculated with protein amount) was observed in this strain grown without ZnSO₄.

Consistent with the variation of TopoI levels in ΔtopAP_ZntopA, topA expression, measured by qRT-PCR, progressively increased as higher ZnSO₄ concentrations in the medium were used (Figure 2C). The expression level of topA in medium with 300 μM of ZnSO₄ was 1.6-fold higher than that in the presence of 75 μM ZnSO₄, this is, 2.7 vs. 1.7. This increase was consistent with the 1.8-fold increase observed by Western-blot (Figure 2A). The expression levels of gyrA and rpoB were independent of ZnSO₄ (Figure 2C).

These results show that the constructed strains allow to control the level of TopoI inside the cell by ZnSO₄ addition in a Sc-independent manner, from very low to wild type levels, in ΔtopAP_ZntopA, and to near double the wild type levels in P_ZntopA. Both strains were further used to assess the effect of TopoI levels in Sc and cell growth.

We investigated the levels of Sc in the aforementioned strains, ΔtopAP_ZntopA and P_ZntopA. Measurement of Sc density (σ) of the plasmid pLS1 present in the cells was carried out as described in the section “Materials and methods.” In the presence of residual amounts of TopoI (0 ZnSO₄), a Sc density (σ) of −0.086 was observed in strain ΔtopAP_ZntopA (Figure 3). When the number of TopoI molecules increased to 665 (75 μM ZnSO₄), 848 (150 μM ZnSO₄), or 1,247 (300 μM ZnSO₄), the density of negative Sc decreased to values of −0.060, −0.057, and −0.053, respectively, similar to σ values observed in R6 and R6P_ZntopA (Figure 3). Therefore, the variation of TopoI levels at the different ZnSO₄ concentrations tested was consistent with the Sc level.

The growth of strain ΔtopAP_ZntopA in the presence of novobiocin (Nov), which targets gyrase, was studied. To modulate the levels on TopoI expression, this strain was grown in media with either 0, 25, or 200 μM of ZnSO₄ (Figure 4A). Its parental strain (P_ZntopA) grown in medium without ZnSO₄ was used as a control. As already seen in this study, ΔtopAP_ZntopA was unable to grow in the absence of ZnSO₄. Notably, treatment with a subinhibitory Nov concentration (0.25 × MIC), restored growth in the absence of ZnSO₄, with a duplication time of 91 ± 6 min, similar to that of P_ZntopA treated with Nov (82 ± 3 min; Figure 4A). Although ΔtopAP_ZntopA in media with 25 μM of ZnSO₄ grew slower (96 ± 3 min) than P_ZntopA without ZnSO₄ (75 ± 9 min), addition of Nov reduced the duplication time to 77 ± 5 min, close to that of untreated P_ZntopA.

Growth of ΔtopAP_ZntopA in medium with the three concentrations of ZnSO₄ studied above, rendered three levels of TopoI expression as shown by Western-blot (Figure 4B): nearly no protein without ZnSO₄, a low amount of TopoI with 25 μM ZnSO₄, and an amount equivalent to that of the control strain with 200 μM ZnSO₄. Without Nov treatment (NT), in the absence of ZnSO₄, a faint band of TopoI was detected (1.2 ± 0.3 ng) in strain ΔtopAP_ZntopA, which represented about 10% the amount of TopoI observed in P_ZntopA control strain (11.6 ± 1.1 ng). In contrast, in the presence of 25 μM ZnSO₄, a low but higher amount of TopoI (2.8 ± 0.4 ng) was detected, which represented 24% the TopoI present in the control strain. Under 200 μM of ZnSO₄, the amount of TopoI (9.0 ± 2.4 ng) was close to that of the control strain (76%). Treatment of the control strain with Nov (0.25 × MIC) decreased 2.2-fold the amount of TopoI due to DNA relaxation. As expected, no change in the amount of TopoI was induced by Nov in ΔtopAP_ZntopA under any growth conditions since topA expression was under the control of P_Zn, whose activity does not dependent on the Sc level.

Measurement of Sc density of plasmid pLS1 present in ΔtopAP_ZntopA in the absence of TopoI expression was −0.080 and for ΔtopAPZntopA grown in 200 μM of ZnSO₄, was of −0.057. When ΔtopAPZntopA was grown in the absence of ZnSO₄, but the presence of Nov, DNA relaxed by 15.0% (σ values from −0.080 to −0.068), consistent with the growth restoration observed (Figure 4C). Therefore, the release of DNA Sc via addition of either ZnSO₄ or Nov allowed restoration of growth (Figure 4C).

Super-resolution confocal microscopy was used for the first time to estimate compaction of nucleoids. Samples analyzed contained different amounts of TopoI by regulated expression (Figure 5). Three strains representing control situations were studied. ΔtopAP_ZntopA grown in the absence of ZnSO₄ showed the highest σ (−0.086) and the lowest TopoI amount (1.5 ng, Table 2). P_ZntopA in the absence of ZnSO₄ represented the wild-type situation, with an equilibrium σ value of −0.060 and normal TopoI levels (12.2 ng, Table 2). R6 treated with Nov for 30 min represented the most relaxed situation (σ = −0.024). As observed in the confocal micrographs, these strains showed highly compacted (ΔtopAP_ZntopA), intermediate-compacted (P_ZntopA), and low-compacted (R6 plus Nov) nucleoids (Figure 5A). Quantification of mean gray values of nucleoids in the mentioned strains with different TopoI amounts revealed a good correlation (R² = 0.93) between these values and σ values obtained by classic analysis of plasmid topoisomers by 2D-gel electrophoresis (Figures 5B, C). Good correlations were also found between GyrA:TopoI ratio and σ (R² = 0.97; Figure 6).