Identification of small molecules that inhibit B. ovis replication in THP-1 cells

We screened 1,280 small molecules from the Prestwick Chemical Library to identify compounds that inhibited B. ovis replication in mammalian cells. Our initial aim was to identify molecules that impair the fitness of B. ovis within the host intracellular niche without exerting direct antibacterial effects in axenic culture. We performed this screen using a luciferase-expressing strain of B. ovis ATCC 25840, in which the lux operon was integrated into a neutral chromosomal site. Luminescence served as a proxy for intracellular replication within host cells. This effort yielded 49 compound hits that met our criteria for selectively inhibiting B. ovis in the intracellular niche (S1 Table). To identify potential host cellular pathways that influence B. ovis fitness in THP-1 macrophages, we prioritized compounds with well-characterized mechanisms of action, including multiple FDA-approved drugs. Notably, several hits targeted pathways involved in calcium (Ca²⁺) homeostasis or dopamine signaling (S1 Table). We selected 10 compounds for validation using the same chemical stocks from the initial high-throughput screen, evaluating their cytotoxicity as well as their effectiveness against Brucella under both intracellular and axenic growth conditions. These validation assays confirmed four initial hit compounds—nicardipine, cilnidipine, lomerizine, and bifonazole—as inhibitors of intracellular replication that were not toxic to host cells at the assessed concentration (Table 1; S1 Fig).

Download:

• PNG
  larger image
• TIFF
  original image

Table 1. Validation of selected hits from the Prestwick Chemical Library. Axenic IC₅₀, intracellular IC₅₀, and host cell CC₅₀ values were determined through dose-response assays using compound stocks from the chemical library. Hits confirmed as non-toxic intracellular B. ovis inhibitors based on these validation assays are marked with an asterisk.

https://doi.org/10.1371/journal.pgen.1011795.t001

Previous screens identified calcium channel blockers as inhibitors of B. abortus replication within the intracellular niche [15,16], and a recent case–control study concluded that the use of dihydropyridine-type Ca²⁺ channel blockers lowered the risk of active tuberculosis [24]. For tuberculosis, the effect of these compounds on disease development may relate to the role of L-type Ca²⁺ channels in supporting iron import [25], which is crucial for Mycobacterium tuberculosis fitness in host cells. Given the clinical potential of Ca²⁺ channel blockers for the treatment of intracellular infections and their emergence in our drug screens against Brucella growth, we prioritized two FDA-approved dihydropyridine-class Ca²⁺  channel blockers—nicardipine and cilnidipine—for further investigation. Both are widely used to treat hypertension and angina pectoris [26].

Differential activity of dihydropyridines against Brucella in host cells and axenic culture

We next re-evaluated the effects of nicardipine and cilnidipine in our assays using freshly prepared stock solutions to more quantitatively establish potency. Both drugs showed a similar potency for inhibiting intracellular growth of B. ovis, with comparable IC₅₀ values of 2.3 and 2.5 µM, respectively, when growth was assessed after 48 h of treatment (Fig 1A). To determine whether the observed inhibitory activity was directed at the host cell or at Brucella itself, we carried out dose-response assays with static axenic B. ovis cultures (i.e., in the absence of THP-1 cells; Fig 1B). Nicardipine had no effect on the terminal optical density at any concentration tested. Although nicardipine and cilnidipine share a common dihydropyridine scaffold (Fig 1C), B. ovis proved more sensitive to cilnidipine. In our high-throughput screen, compounds were designated axenic inhibitors only if they produced > 35% inhibition at the screening concentration of approximately 5 µM; cilnidipine fell just below this cutoff, although modest growth suppression was evident in the primary data (S1 Fig). Follow-up dose-response assays using fresh chemical stocks revealed an IC₅₀ of ~5 µM for cilnidipine and complete growth inhibition at concentrations ≥ 10 µM (Fig 1B).

Download:

• PNG
  larger image
• TIFF
  original image

Fig 1. The dihydropyridine-class Ca²⁺ channel blockers nicardipine and cilnidipine inhibit intracellular B. ovis growth and cilnidipine inhibits B. ovis axenic growth.

(A) Intracellular inhibitory activities of nicardipine and cilnidipine during B. ovis THP-1 macrophage infection. B. ovis luminescence was measured 48 h post-treatment and normalized to signal from infected untreated cells. IC₅₀ values based on curve fits are shown. (B) Axenic inhibitory activity of nicardipine and cilnidipine during B. ovis growth in liquid medium. Optical densities at 600 nm were measured after 48 h of growth and normalized to untreated cultures. (C) Chemical structures of nicardipine and cilnidipine. In A and B, data represent the mean and standard deviation of 3 biological replicates.

https://doi.org/10.1371/journal.pgen.1011795.g001

Nicardipine treatment alters the THP-1 metallome

Nicardipine and cilnidipine are well-known inhibitors of Ca²⁺ import into vascular smooth muscle and cardiac cells [26] but, to our knowledge, their effect on levels of Ca²⁺ or other metals in monocytic cell lines such as THP-1 has not been examined. To evaluate intracellular metal levels following dihydropyridine treatment, we used triple quadrupole inductively coupled plasma mass spectrometry (ICP-QQQ) to analyze a panel of elements in THP-1 macrophage-like cells (Figs 2 and S2). The cells were treated with 25 µM nicardipine, a concentration near the reported IC₅₀ for L-type Ca²⁺ channels at membrane potentials of −15 to −30 mV [27], a range typical of human macrophages and monocytic cells [28,29]. Metal concentrations were normalized to phosphorus, of which advantages for intracellular metal analysis have been shown in multiple studies [30–33] Most metal levels remained unchanged following nicardipine treatment (S2 Fig). However, we observed a ~ 20–40% increase in intracellular calcium and manganese levels in treated THP-1 cells (Fig 2A and 2B).

Download:

• PNG
  larger image
• TIFF
  original image

Fig 2. Nicardipine treatment alters the THP-1 macrophage metallome.

Elemental content of untreated THP-1 cells or cells treated with 25 µM nicardipine for 48 hours was determined using triple quadrupole inductively coupled plasma mass spectrometry (ICP-QQQ). Each element was normalized by the total phosphorus in each sample (M/P). (A) Calcium and (B) Manganese levels were significantly elevated upon nicardipine treatment. Other elements are presented in S2 Fig. Bars represent the mean ± standard deviation of 7 biological replicates measured over 2 independent experiments. The full metallome data set was analyzed using multiple unpaired t-tests and Bonferroni-Dunn method to adjust for multiple comparisons (***, adjusted P < 0.001).

https://doi.org/10.1371/journal.pgen.1011795.g002

The mechanism underlying elevated calcium remains unclear but may reflect compensatory calcium influx triggered by depletion of intracellular Ca²⁺ stores or altered activity of alternate Ca²⁺ transport pathways [34,35]. Notably, inhibition of voltage-gated Ca²⁺ channels has been shown to promote calcium influx in dendritic cells and reduce Mycobacterium tuberculosis burden in mice [36], supporting a model in which nicardipine impacts calcium signaling in immune cells. Increased manganese levels following nicardipine treatment is notable, as elevated intracellular Mn has been linked to activation of host-defense pathways against microbial infection [37,38]. Together, our results indicate that nicardipine selectively alters intracellular calcium and manganese levels in a human monocytic cell line. Although the downstream effects of these metal shifts on host cell physiology are not yet defined, they may help explain the enhanced efficacy of nicardipine against B. ovis in the intracellular niche compared to axenic conditions (Fig 1).

Mutations in transporter genes confer resistance to dihydropyridines in axenic culture

Nicardipine disrupted calcium and manganese levels in THP-1 host cells (Fig 2), which may influence B. ovis fitness within the intracellular niche. However, cilnidipine had direct antimicrobial activity against B. ovis in axenic culture as measured by a terminal density assay (Fig 1B), suggesting that the intracellular inhibition observed upon dihydropyridine treatment may be at least partly independent of host-mediated effects. We hypothesized that B. ovis susceptibility to the drug is influenced by specific bacterial genes. To identify such genes, we selected for spontaneous cilnidipine-resistant mutants. Briefly, we spread approximately 10⁸ wild-type B. ovis cells on solid medium containing 25 μM cilnidipine. After 72 hours we recovered a small number of cilnidipine-resistant mutant clones. Whole-genome sequencing of six of these clones revealed mutations at one of two loci: bepE (locus tags BOV_RS01560; BOV_0306) or emrA (BOV_RS10885; BOV_A0108) (Table 2). bepE is a pseudogene that encodes a truncated form of a predicted inner membrane permease subunit belonging to a resistance-nodulation-division (RND) family transporter system, while emrA encodes a component of a major facilitator superfamily (MFS) multidrug efflux pump. Mutations in emrA resulted in single amino acid substitutions (L271P or A102T), which likely conferred resistance by enhancing efflux activity.

Download:

• PNG
  larger image
• TIFF
  original image

Table 2. Mutations identified in six independent spontaneous mutants resistant to cilnidipine, isolated after growth on solid medium containing 25 µM cilnidipine. The locations of the bepE frameshift mutations and their effects on the open reading frame are illustrated in Fig 3.

https://doi.org/10.1371/journal.pgen.1011795.t002

bepE is one of many genes that have undergone pseudogenization in the B. ovis lineage through indel mutations that cause frameshifts [18,19,21]. Cilnidipine-resistant isolates carried single-base deletions near the original pseudogene frameshift site that extended the open reading frame (ORF) from 1,689–3,156 nucleotides (Fig 3A–B). The resulting “mutant” bepE alleles encode proteins aligning with full-length bepE homologs in B. melitensis, B. abortus, B. suis, B. canis, and other Brucella species (Fig 3C-D), effectively reconstituting a functional gene. To assess the functional impact of bepE ORF restoration, we compared growth of wild-type (WT) B. ovis and a bepE-reconstituted mutant (ΔG at position 1,500) in liquid culture with 25 μM nicardipine or cilnidipine. In untreated media, both strains grew similarly (S3 Fig). While a terminal density assay showed no effect of 25 μM nicardipine on B. ovis (Fig 1), a kinetic growth assay revealed a modest lag in the growth of WT cultures treated with nicardipine; the restored bepE mutant showed no lag in the presence of nicardipine (S3 Fig). WT cultures failed to grow over 72 h in the presence of 25 μM cilnidipine, whereas the bepE mutant resumed growth after a 24 h lag and reached densities comparable to untreated WT by 72 h (S3 Fig). These results demonstrate that nicardipine and cilnidipine can directly inhibit B. ovis growth, though to different degrees, and that restoration of the bepE pseudogene to encode an open reading frame that matches other Brucella species results in drug resistance.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 3. B. ovis cilnidipine-resistant mutants harbor single-nucleotide deletions that resurrect the open reading frame of the pseudogene bepE.

(A) Diagram of the bepDE locus (top) and alignment of bepE nucleotide sequences from wild-type B. ovis, four cilnidipine-resistant mutants, and B. abortus (bottom). The 2 bp deletion that led to the pseudogenization of B. ovis bepE, and the single base deletions in the cilnidipine-resistant mutants are highlighted. (B) Schematic of the bepE open reading frame B. abortus 2308, B. ovis ATCC 25840, and a B. ovis cilnidipine-resistant mutant. The site of the 2 bp frameshift in wild-type B. ovis compared to the other Brucella species, and site of the restoring frameshift mutations are indicated. (C) Alignment of the partial BepE amino acid from wild-type B. ovis ATCC 25840, four cilnidipine-resistant mutants and B. abortus. Differences from the consensus of the genus, exemplified by B. abortus, are highlighted in red. (D) Alignment of the entire BepE protein sequence from representative Brucella species lineages and one B. ovis cilnidipine-resistant mutant. Differences from the consensus are indicated in black.

https://doi.org/10.1371/journal.pgen.1011795.g003

Restoration of the B. ovis bepE open reading frame confers broad resistance to cell envelope stressors in vitro

B. ovis is more sensitive to compounds that disrupt the cell envelope than other Brucella species [39,40], but the genetic basis for this increased sensitivity has not, to our knowledge, been previously described. We reasoned that the bepE reversion mutations and the emrA point mutations that confer dihydropyridine resistance might also protect against additional envelope stressors. To test this, we evaluated the sensitivity of these mutants to the detergents sodium dodecyl sulfate (SDS) and CHAPS, as well as to the secondary bile acid deoxycholate, which has detergent-like properties. To further explore their stress resistance profiles, we also assessed the growth of wild-type B. ovis and the bepE and emrA mutant strains in the presence of carbenicillin, a β-lactam antibiotic that targets the cell wall. Wild-type (WT) B. ovis was highly sensitive to 0.005% (w/v) SDS, 0.009% (w/v) deoxycholate, and 0.005% (w/v) CHAPS, and moderately sensitive to 3 μg/ml carbenicillin. In contrast, bepE-restored mutants exhibited resistance to all four compounds at these concentrations (Fig 4A). Mutations in emrA had more limited effects: both emrA mutants (L127P and A102T) were as resistant to SDS as the bepE mutants and showed improved CHAPS tolerance compared to WT, though to a lesser degree than the bepE-restored strains. Only emrA(A102T), but not emrA(L127P), conferred partial resistance to deoxycholate (Fig 4A). These results indicate that pseudogenization of bepE contributes to the documented cell envelope stress sensitivity of B. ovis relative to other classical Brucella species.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 4. Restoration of B. ovis BepE function confers detergent resistance and enhanced tolerance of dihydropyridines during macrophage infection.

(A) Growth of the wild-type (WT) B. ovis and cilnidipine-resistant mutants on TSA blood plates without additives or with detergents or carbenicillin. Cultures of each genotype were normalized by optical density, 10-fold serially diluted, and spotted onto TSA blood plates that were untreated or contained 0.005% SDS, 0.009% deoxycholate, 0.005% CHAPS, or 3 µg/ml carbenicillin. (B) Bacteria recovered from THP-1 macrophage-like cells infected with WT B. ovis or a B. ovis bepE restored mutant. (C, D) Bacteria recovered from infected THP-1 cells treated with 25 µM nicardipine or 25 µM cilnidipine at (C) 24- or (D) 48- h post infection. Infections were performed three times; values are means ± SD from three independent trials. Statistical significance was calculated at 24 hpi and 48 hpi with an unpaired t-test (*, P < 0.05; **, P < 0.01).

https://doi.org/10.1371/journal.pgen.1011795.g004

Restoration of the B. ovis bepE reading frame confers resistance to dihydropyridines in the intracellular niche

The enhanced resistance of bepE-restored B. ovis mutants to drugs and envelope stress in vitro prompted us to examine their impact on bacterial replication and susceptibility to nicardipine and cilnidipine within the host environment. To assess whether bepE restoration affects intracellular proliferation, we infected THP-1 macrophage-like cells with either wild-type (WT) B. ovis (carrying the bepE pseudogene) or a strain with a reconstituted functional bepE allele. The bepE-restored strain showed a modest increase in CFUs at 24 hours relative to WT, but both strains exhibited comparable intracellular burdens at 48 hours post-infection (Fig 4B).

We next asked whether restoration of the bepE open reading frame confers resistance to dihydropyridines in the intracellular niche. Infected macrophages treated with nicardipine or cilnidipine showed a pronounced difference in bacterial burden between strains: the bepE-restored mutant exhibited approximately 10-fold higher CFUs at 24 hours and nearly 100-fold higher CFUs at 48 hours compared to WT B. ovis (Fig 4C–D). Despite this substantial increase in drug tolerance, bepE restoration did not fully rescue growth under drug treatment to levels observed in untreated cells at later time points, indicating that additional host- or drug-associated factors contribute to the intracellular activity of dihydropyridines. Importantly, reduced bacterial recovery was not due to host cytotoxicity (S4 Fig), consistent with our initial high-throughout screen results. Together, these results demonstrate that restoration of bepE enhances B. ovis resistance to dihydropyridine Ca²⁺ channel blockers in macrophages, although this protection is partial and depends on the time point post infection.

bepE contributes to bile acid and penicillin G resistance in B. abortus

Frameshift mutations in the B. ovis bepE pseudogene restored the open reading frame found in other Brucella species, conferred resistance to dihydropyridines, and enhanced survival under conditions of cell envelope stress. To determine whether bepE similarly influenced chemical resistance in other Brucella, we investigated its role in B. abortus, a closely related zoonotic pathogen that carries an intact bepE gene (locus tag BAB_RS17475; BAB1_0323). Attempts to delete bepE using standard allelic exchange with sacB/sucrose counterselection were unsuccessful. Instead, we inactivated the gene via single-crossover recombination, generating a mutant (∆bepE) that expresses only the first 103 amino acids of BepE. We assessed the sensitivity of WT and ∆bepE strains to nicardipine, cilnidipine, and the membrane-disrupting agents deoxycholate and CHAPS (1% w/v) on solid media (Fig 5A). While 1% deoxycholate modestly impaired WT growth, ∆bepE was unable to grow under this condition, consistent with prior findings in B. suis [41]. In contrast, bepE disruption did not affect sensitivity to CHAPS or the dihydropyridines under the tested conditions.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 5. Loss of BepE function renders B. abortus sensitive to deoxycholate.

(A) Colony-forming units of B. abortus WT and ∆bepE (BAB_RS17475) enumerated after growth on TSA blood medium alone (Untreated) or containing 25 µM cilnidipine, 25 µM nicardipine, 1% deoxycholate (w/v), or 1% CHAPS (w/v). Dotted line indicates limit of detection. (B) Deoxycholate dose-response curves with IC₅₀ values. Growth of B. abortus WT, ∆bepE, and the complementation strain ∆bepE glmS::bepDE in liquid medium with deoxycholate was assessed by optical density after 48 h. Values represent mean ± SD of three independent trials, each normalized to the untreated control. (C) Growth of serially-diluted B. abortus strains spotted onto TSA blood plates with or without 0.1% deoxycholate (w/v).

https://doi.org/10.1371/journal.pgen.1011795.g005

To confirm that the deoxycholate sensitivity of the ∆bepE strain was due to loss of BepE function, we integrated a single copy of the bepDE operon at the glmS locus for genetic complementation. In both liquid and solid media, the ∆bepE mutant showed increased sensitivity to deoxycholate (0.1–1% w/v) and genetic complementation restored resistance to WT levels (Fig 5B–C). Together, these results demonstrate that BepE specifically contributes to B. abortus resistance to deoxycholate, a secondary bile acid abundant in human and ruminant bile. BepE may thus play a pathogen-protective role when Brucella species enter through the oral route.

To identify additional chemical stress conditions in which bepE impacts B. abortus fitness, we screened WT and ΔbepE strains on Biolog phenotype microarrays. Penicillin G emerged as a strong differential hit, markedly inhibiting ΔbepE but not WT growth (S5A Fig). Spot-titration on penicillin-G agar verified this result: deletion of bepE increased β-lactam susceptibility by approximately five log₁₀ units, and ectopic expression of bepDE in this mutant background restored resistance (S5B Fig). Differential resistance to other common antimicrobials was less dramatic, though the ΔbepE strain was more sensitive to chemical treatment in the Biolog plate overall (S5A Fig). Collectively, these data show that BepE helps B. abortus withstand both bile acid and beta-lactam antibiotic stress, highlighting its important role in defending the cell envelope against chemically diverse stresses.

bepE is dispensable for intracellular growth but confers B. abortus resistance to cilnidipine treatment in host cells

We next tested whether loss of bepE affects B. abortus fitness in macrophage infection models. To do this, we used two distinct macrophage systems that differ in their physiological characteristics and potential relevance to Brucella infection. THP-1 cells are a well-established model of circulating monocyte-derived macrophages, while fetal liver alveolar-like macrophages (FLAMs) are a primary ex vivo model that more closely resembles tissue-resident macrophages [42]. Because B. abortus encounters both monocyte-derived and tissue-resident macrophages during the course of host colonization, we evaluated bacterial replication in both systems to capture this range of possible host cell environments. In THP-1 macrophage-like cells, ∆bepE colony-forming units (CFU) recovered at 2, 24, or 48 hours post-infection were not significantly different from those of wild-type or the genetically complemented strain (∆bepE glmS::bepDE) (Fig 6A). Similarly, in FLAMs, the growth and survival of the ∆bepE mutant matched that of the wild-type strain throughout the infection time course (S6 Fig).

Download:

• PNG
  larger image
• TIFF
  original image

Fig 6. B. abortus bepE does not contribute to survival in THP-1 macrophages but confers enhanced resistance to cilnidipine in the intracellular niche.

(A) B. abortus CFU recovered from THP-1 macrophages after infection with WT, ∆bepE, and the complementation strain ∆bepE glmS::bepDE. (B) WT B. abortus cells recovered from infected THP-1 macrophages untreated or treated with 25 µM nicardipine or 25 µM cilnidipine. (C, D) CFUs of WT, ∆bepE, and ∆bepE glmS::bepDE recovered from THP-1 macrophages at 24 (C) or 48 hpi (D). Infected macrophages were treated with 25 µM nicardipine or 25 µM cilnidipine. Infections were performed three times; values are means ± SD from three independent trials. Statistical significance was assessed at 48 hpi using two-way ANOVA on log₁₀ transformed CFU values, followed by Tukey’s multiple comparison’s test compared with the untreated WT group (*, P < 0.01; **, P < 0.001).

https://doi.org/10.1371/journal.pgen.1011795.g006

We further tested whether bepE influences B. abortus sensitivity to dihydropyridines in host cells. Using a luciferase-expressing B. abortus strain to monitor intracellular bacteria in THP-1 cells, we found that B. abortus was more sensitive to nicardipine (IC₅₀ = 4.4 µM) than to cilnidipine (IC₅₀ = 13.5 µM) (S7A Fig). Consistent with this, significantly fewer wild-type B. abortus CFUs were recovered from nicardipine-treated THP-1 cells compared to untreated controls (Fig 6B). Deletion of bepE from B. abortus did not alter sensitivity to nicardipine but did increase sensitivity to cilnidipine: the ∆bepE strain showed a significant reduction in CFU recovery at 48 hours post-infection relative to wild-type (Fig 6D), although no difference was observed at 24 hours (Fig 6C). The complemented strain restored resistance to WT levels (Fig 6D). We conclude that bepE contributes to B. abortus resistance to cilnidipine in the intracellular niche. Importantly, in axenic conditions, both WT and ∆bepE B. abortus were insensitive to nicardipine and cilnidipine (up to 25 µM) (S7C-D Fig). B. abortus is therefore more sensitive to these dihydropyridines in the intracellular environment, owing either to host-directed activity or to increased susceptibility to the drugs within its intracellular niche.

Dihydropyridines as intracellular pathogen inhibitors

Previous small-molecule screens identified dihydropyridine calcium channel blockers, including nicardipine, as inhibitors of B. abortus replication in host cells [15,16]. Through an unbiased small-molecule screen, we show here that treating B. ovis-infected macrophage-like cells with micromolar concentrations of nicardipine or cilnidipine significantly reduces bacterial fitness within the intracellular niche. These results extend the anti-infective activity of dihydropyridines to multiple Brucella species and macrophage models. Although nifedipine and the non-dihydropyridine calcium channel blocker verapamil have previously demonstrated intracellular anti-mycobacterial or anti-Brucella activity [36,43,44], neither compound emerged as a hit in our B. ovis screen at the concentrations tested.

Beyond Brucella, human case control studies have found that the use of dihydropyridine calcium channel blockers is associated with a reduced risk of active tuberculosis [24], suggesting broader clinical potential for this drug class against intracellular pathogens. Indeed, dihydropyridines have been reported to inhibit several intracellular bacteria, including Mycobacterium tuberculosis [36], Coxiella burnetii [16], and Legionella pneumophila [45] highlighting their promise as broad-spectrum intracellular pathogen inhibitors. However, despite these encouraging findings, the cellular and molecular mechanisms by which calcium channel blockers exert their antimicrobial effects (on both host cells and bacteria) remain poorly understood. Some data indicate these drugs may disrupt bacterial efflux systems and potentiate sensitivity to other chemical stressors. For example, the dihydropyridine amlodipine exhibits synergy with imipenem against multidrug-resistant Acinetobacter baumannii by inhibiting bacterial efflux [46]. Similarly, the non-dihydropyridine calcium channel blocker verapamil has shown synergistic effects with the anti-tubercular drugs clofazimine and bedaquiline [47]. We have not yet explored potential synergies between dihydropyridines and standard-of-care antibiotics for treatment of brucellosis, but our results raise the possibility that dihydropyridines could be integrated into combination therapies to improve treatment outcomes.

Nicardipine treatment alters metal homeostasis in macrophages

The effects of calcium channel blockers on THP-1 macrophage physiology and metal homeostasis are not well defined, though L-type Ca²⁺ channels are reported to be expressed in this human cell line [48,49]. To investigate how nicardipine affects intracellular metal levels in THP-1 macrophage-like cells, we used triple quadrupole inductively coupled plasma mass spectrometry (ICP-QQQ). While most metals remained unchanged, nicardipine treatment caused a significant increase in total intracellular calcium (Fig 2). Previous studies have shown that blocking voltage-gated Ca²⁺ channels can paradoxically enhance calcium influx in macrophages, leading to upregulation of pro-inflammatory pathways and improved clearance of M. tuberculosis [36]. Calcium signaling also influences macrophage polarization, with distinct calcium channel activities promoting either a classically activated (M1) or alternatively activated (M2) state [50]. As B. abortus has been reported to promote M2 polarization to support its intracellular survival at later infection stages [51], it is possible that increased calcium in nicardipine-treated macrophages may favor M1 polarization and activate antimicrobial pathways.

In addition to increasing intracellular calcium, nicardipine treatment also elevated manganese levels in THP-1 cells (Fig 2). Manganese (Mn²⁺) is reported to activate antiviral pathways such as STING [38], and plays an important role in antibacterial immunity, including host defense against M. tuberculosis [37]. Thus, nicardipine may enhance antimicrobial responses by modulating Mn²⁺ availability, potentially creating a less permissive intracellular environment for Brucella. Notably, Brucella tightly regulates manganese homeostasis, and perturbation of this balance reduces bacterial fitness both in vitro and in vivo [52].

The specific host pathways and ion channels responsible for the observed calcium and manganese shifts following drug treatment remain undefined. Identifying these systems could reveal new opportunities for host-directed therapeutic strategies. We note that while the ICP-QQQ method we used enables accurate quantification of total intracellular metal levels, it does not resolve how metals are spatially distributed among subcellular compartments or how this distribution changes upon treatment. It is possible that nicardipine alters metal compartmentalization, leading to localized effects on host or pathogen physiology. Despite this limitation, our data show that nicardipine significantly reshapes the macrophage metallome (Fig 2) without compromising cell viability (S4 Fig), supporting a potential role for host metal modulation in its anti-Brucella activity.

Pseudogenization of bepE underlies B. ovis sensitivity to dihydropyridines and envelope stressors

Although dihydropyridine-induced changes in host metal levels could influence Brucella infection dynamics by altering macrophage physiology, we found that these compounds also directly inhibit B. ovis growth in axenic culture (Figs 1B and S3). We selected for spontaneous B. ovis mutants capable of growing in the presence of cilnidipine, the more axenically potent of the two dihydropyridines tested, and discovered two classes of mutations: non-synonymous substitutions in a predicted emrA-family efflux gene, and single-nucleotide deletions in bepE, a pseudogene encoding a subunit of an RND-family transporter [41,53]. Deletions in bepE restored its reading frame, yielding a full-length gene homologous to functional bepE loci in other Brucella species.

B. ovis mutants with restored bepE alleles exhibited increased dihydropyridine resistance in the intracellular niche, but they remained partially sensitive to drug at 48 hours post-infection, indicating that BepE-mediated transport processes alone are insufficient to fully protect the bacterium within host cells. This residual sensitivity suggests that dihydropyridines may also impair Brucella fitness through host-directed mechanisms. Consistent with this, a B. abortus bepE deletion mutant was as resistant as wild-type in axenic culture across a range of nicardipine and cilnidipine concentrations (S7 Fig) but became significantly more susceptible to cilnidipine when inside THP-1 cells (Figs 6 and S7). Wild-type B. abortus was also more sensitive to nicardipine than cilnidipine in an intracellular setting (Fig 6); neither drug inhibited B. abortus growth in axenic culture up to 25 µM (S7 Fig). Together, these results support a model in which dihydropyridines act, at least in part, by modulating host pathways that restrict intracellular Brucella growth or survival.

The precise mechanism of direct nicardipine and cilnidipine toxicity to B. ovis remains undefined, and the lack of resistance mutations outside transport-related genes suggests that cilnidipine may act on essential pathways or exert its effects through multiple mechanisms. B. ovis is the only classical Brucella species to have lost a functional bepE gene (Fig 3), and our data demonstrate that bepE pseudogenization strongly contributes to the documented sensitivity of B. ovis to chemical stressors relative to other Brucella spp. [39,40]. B. ovis has undergone extensive genome reduction [19,54] and we speculate that bepE loss of function reflects restriction of this species to the niche of the ram reproductive tract, where exposure to cell envelope stressors may be reduced [55].

Function and evolutionary flexibility of bepE in the Brucella genus

Although bepE contributes to chemical stress resistance, it was dispensable for intracellular survival in two physiologically distinct macrophage models (Figs 6 and S6). These results are consistent with prior work in B. suis, where ∆bepE mutants replicated normally in cultured epithelial and phagocytic cells. In contrast, ∆bepC mutants (which lack a shared outer membrane component used by multiple RND-family efflux systems, including BepE) exhibit reduced spleen colonization in mice [41,56]. Thus, bepE may support fitness when Brucella faces specific immune pressures in vivo. Unlike in B. suis, we were unable to generate clean bepE deletions in B. abortus using sacB/sucrose counterselection and instead used a single-crossover strategy to disrupt the bepE locus. Supporting the idea that this efflux system may have broader functional importance, saturated transposon mutagenesis studies in B. abortus identified the periplasmic adaptor bepD (BAB_RS17470; BAB1_0322) and bepC (BAB_RS20535; BAB1_0963) as essential for growth in complex medium [57]. Together, these observations suggest that the BepDE RND-efflux system may have additional, yet uncharacterized, roles in Brucella physiology beyond protection against xenobiotic compounds.

Our results raise broader questions about the function and evolutionary flexibility of bepE across the Brucella genus. While bepE is pseudogenized in B. ovis, it clearly retains the potential for functional reactivation. This raises the possibility that bepE is a “contingency” locus [58,59] that is nonfunctional under typical conditions, but capable of being reactivated under certain selective conditions in a host. Such genomic flexibility may allow Brucella to balance the evolutionary pressure for genome streamlining with the need to preserve adaptability. Future studies are necessary to determine whether the presence of a frameshifted bepE allele in B. ovis reflects selective pressure to maintain a reactivatable transport function or simply represents an early stage in the irreversible degradation of this gene during host restriction.

Bacterial strains and growth conditions

Brucella ovis 25840 and Brucella abortus 2308 were grown on tryptic soy agar (TSA; Difco Laboratories) plates supplemented with 5% (v/v) sheep blood (Quad Five) or in brucella broth (BB; Difco Laboratories) dissolved in milliQ water for liquid cultures. B. ovis and B. abortus cells were incubated at 37°C with 5% (v/v) CO₂ supplementation.

All Escherichia coli strains were grown in liquid LB or on LB solidified with 1.5% (w/v) agar. Top10 and WM3064 strains were incubated at 37°C. WM3064 was grown with 30 µM diaminopimelic acid (DAP) supplementation. The growth medium contained 50 µg ml⁻¹ kanamycin or 20 µg ml⁻¹ chloramphenicol when necessary. Primer, plasmid, and strain information are available in S2 Table.

Plasmid and strain construction

1. (i). Deletion strain construction. To inactivate bepE in B. abortus (gene ID BAB_RS17475), a 528-bp internal fragment was cloned into pNPTS138-cat (a suicide vector that carries a chloramphenicol resistance gene). The resulting plasmid was used to disrupt the target gene by a single crossover insertion. Positive colonies were selected on sheep blood agar (SBA) plates containing 3 µg ml⁻¹ chloramphenicol. The obtained mutant strains are predicted to produce only the first 103 amino acids of BepE.
2. (ii). Complementation strain construction. To build plasmids for genetic complementation, the coding sequences of bepD (BAB_RS17470) and bepE were PCR amplified, starting ~300 bp upstream and ending ~50 bp downstream of the bepD/E operon. The PCR product was purified and inserted into plasmid pUC18-mTn7 by restriction enzyme digestion and ligation, followed by chemical transformation into E. coli TOP10 cells. After sequence confirmation by Sanger sequencing, the resulting pUC18-mTn7-bepD/E plasmid was transformed into chemically competent E. coli WM3064. The plasmid was co-conjugated into B. abortus strains with pTNS3, a suicide helper plasmid expressing the Tn7 integrase gene. B. abortus colonies carrying the integrated mTn7-bepD/E construct at the glmS locus were selected on TSA blood plates containing 50 µg ml⁻¹ kanamycin.
3. (iii). Bioluminescent strain construction. Luminescent B. ovis and B. abortus strains were generated from wild-type B. ovis ATCC 25840 and B. abortus 2308 parent strains by integration of a pUC18-mTn7-gentamicin plasmid harboring the luxCDABE operon at the glmS locus (the mini-Tn7 plasmid was a gift from H.P. Schweizer). The original gentamicin resistance marker was disrupted by Tth111I restriction enzyme digest (New England Biolabs) and replaced with a kanamycin cassette by Gibson assembly (S2 Table). The pUC18-mTn7-luxCDABE-kanamycin plasmid was co-conjugated into B. ovis and B. abortus strains with the suicide helper plasmid pTNS3 expressing the Tn7 integrase gene, as above for complementation strain construction. Positive B. ovis and B. abortus colonies carrying the integrated mTn7-luxCDABE luminescence construct were selected on TSA blood plates containing 50 µg ml⁻¹ kanamycin.

Small-molecule screening

1. (i). Macrophage infection. The drug screen was conducted at the Michigan State University Assay Development and Drug Repurposing Core (ADDRC). Briefly, THP-1 cells were seeded using a liquid handler (BioTek MultiFlo) dispensing 25µl of host cells at a titer of 2 x 10⁴ cells per well with 50 ng mL⁻¹ phorbol myristate acetate (PMA) to induce differentiation into macrophage-like cells in white 384-well plates (Corning). THP-1 cells were incubated at 37°C in 5% (v/v) CO₂ for 72 h. In parallel, luminescent B. ovis cells were streaked out on TSA blood plates and incubated at 37°C in 5% CO₂ for 72 hours. Immediately preceding the addition of luminescent B. ovis to differentiated THP-1 cells, screening compounds from the Prestwick Chemical Library dissolved in 2 mM DMSO stock solutions were added to the assay plates by a 150 nanoliter automatic robotic pin transfer (Beckman Coulter Biomek FX) for a final concentration of 5 μM. Then, luminescent B. ovis cells were resuspended in RPMI medium supplemented with 10% (w/v) heat-inactivated fetal bovine serum (HI FBS) (HyClone); host cells were infected by transferring 25 µl of luminescent B. ovis cell suspension using a liquid handler (BioTek MultiFlo). Plates were centrifuged for 5 min at 150 x g and incubated for 1 h at 37°C in 5% (v/v) CO_2. Following incubation, 10 µl of fresh medium containing 300 ug ml⁻¹ of gentamicin was added to each well and the plate was incubated for 24 hours at 37°C and 5% (v/v) CO₂ for 48 h before evaluating B. ovis luminescence using a plate reader (Biotek Synergy Neo Plate Reader). THP-1 cell viability was then determined using a fluorescent Gly-Phe-AFC peptide (MP Biochemicals). Positive hits from this intracellular B. ovis screen were defined as small molecules that diminished the B. ovis luminescence signal to at least 35% of that for DMSO (vehicle)-treated samples and did not affect THP-1 viability. A Z’ factor of 0.417 was achieved for intracellular inhibitory screening.
2. (ii). Axenic growth. Luminescent B. ovis cells were resuspended from a 72-h-old plate in BB (Difco Laboratories), and 60 µl of cells were dispensed into each well of 384-well plates by liquid handling (Biotek MultiFlo). Small molecules were then transferred from compound screening libraries using automated robotic pin transfer (Beckman Coulter Biomek FX). Assay plates were then incubated at 37°C and 5% (v/v) CO₂ for 48 h before evaluating B. ovis luminescence on a plate reader (Biotek Synergy Neo Plate Reader). A Z’ factor of 0.473 was achieved for axenic inhibitory screening.

Growth curves

BB was inoculated with Brucella cells from 48- to approximately 72-h-old TSA blood plates at cell densities ranging from OD₆₀₀ = 0.05 to OD₆₀₀ = 0.1. Growth was assessed by measuring OD₆₀₀ on a spectrophotometer (Genesys 30 Visible Spectrophotometer). Growth curves were conducted at least three independent times with two technical replicates in each experiment. Representative curves are shown for each set of strains. Where indicated, 25 µM nicardipine (Cayman Chemical) and 25 µM cilnidipine (Cayman Chemical), dissolved fresh in DMSO (Thermo Scientific), were added to BB at the start of the growth experiment.

Dose-response growth assay in axenic culture

Two-fold serial dilutions of compounds were prepared manually in 96-well plates, with final concentrations of each compound ranging from 0.2 µM to 50 µM suspended in 100 µl BB. From a 48-h-old TSA blood plate, B. ovis or B. abortus cells were resuspended in BB at densities ranging from OD₆₀₀ = 0.05 to OD₆₀₀ = 0.1. Brucella cell suspensions were transferred into 96-well plates (100 µl per well) containing compound dilutions for a final volume of 200 µl. Plates were incubated without shaking at 37°C and 5% (v/v) CO₂ for 48 h; absorbance was read spectrophotometrically at OD₆₀₀ with a plate reader (Tecan Spark 10M Multimodal Plate Reader or Tecan Infinite M200 Pro microplate reader).

Bacterial phenotype microarray

B. abortus strains (wild type 2308 and ∆bepE) were resuspended in BB from a 48-h-old TSA blood plate to an optical density of OD₆₀₀ = 0.05. Brucella cell suspensions were transferred into 96-well Biolog plates (100 µl per well) (Biolog) [60]. Plates were incubated without shaking at 37°C and 5% (v/v) CO₂ for 72 h; absorbance was read spectrophotometrically at OD₆₀₀ with Tecan Infinite M200 Pro microplate reader.

Forward genetic selection for cilnidipine-resistant B. ovis mutants

To identify spontaneous cilnidipine-resistant mutants, we grew three independent cultures of wild-type B. ovis ATCC 25840 overnight in BB, normalized the optical density to OD₆₀₀ ≈ 0.2, spread 100 µl from each culture onto TSA blood plates containing 25 µM cilnidipine, and incubated the plates for 72 h at 37°C and 5% (v/v) CO₂, at which point a small number of colonies (3–5) emerged on each plate. To confirm resistance to cilnidipine, these isolates were restreaked onto TSA blood cilnidipine-containing plates, and individual colonies were picked and grown in BB containing 25 µM cilnidipine at 37°C and 5% (v/v) CO₂.

DNA extraction, amplification, and quantification

Genomic DNA was extracted following a standard guanidium thiocyanate protocol. Cilnidipine-resistant mutants were streaked onto TSA blood plates; individual colonies were picked and cultivated in BB overnight. One milliliter of culture was centrifuged at 12,000 rpm for 20 s at room temperature; the resulting pellet was washed with 0.5 ml phosphate-buffered saline (PBS), pH 7.4. The pellet was resuspended in 0.1 ml TE buffer, pH 8.0 (10 mM Tris-HCl, pH 8.0 [Fisher Bioreagents], 1 mM EDTA pH 8.0 [Fisher Bioreagents]), to which 0.5 ml of 0.5% (v/v) sarkosyl was added. Following a 15-min incubation at 60°C, 0.25 ml of cold 7.5 M ammonium acetate (Fisher Bioreagents) was added. After incubation for 10 min on ice, 0.5 ml of chloroform (Fisher Bioreagents) was added, and samples were vortexed and centrifuged at 13,000 x g for 30 minutes at 4°C. The aqueous phase was mixed with 0.54 volume of cold isopropanol (Fisher bioreagents) and incubated at room temperature for 15 min before centrifugation at 13,000 x g for 10 minutes. The pellet was washed three times in 70% (v/v) ethanol before being resuspended in TE buffer plus RNase A, at which point the DNA concentration was determined spectrophotometrically (Thermo Scientific NanoDrop One/One Microvolume UV-Vis Spectrophotometer).

Whole-genome DNA sequencing

Genomic DNA (gDNA) from the parent B. ovis strain and gDNA from six independent spontaneous mutants that grew on TSA blood plates containing 25 µM cilnidipine were purified using the standard guanidium thiocyanate-based procedure described above. DNA library preparation and sequencing were performed at SeqCenter (https://www.seqcenter.com). Reads were mapped to the Brucella ovis ATCC 25840 genome (chromosome 1 and chromosome 2 RefSeq accession numbers NC_009505 and NC_009504, respectively), and polymorphisms were identified using breseq [61] (https://github.com/barricklab/breseq).

Cell culture

THP-1 macrophage-like cells were grown to a maximum titer of 1 x 10⁶ cells ml⁻¹ in complete RPMI 1640 medium supplemented with 2 mM glutamine (GIBCO), and 10% (v/v) HI FBS (HyClone). Fetal liver alveolar macrophages (FLAM) were prepared as reported by Thomas et al. [62]. Briefly, fetal liver cells were harvested as described [63] and grown in complete RPMI 1640 medium supplemented with 2 mM glutamine (GIBCO), 10% (v/v) HI FBS (HyClone), 30 ng ml⁻¹ recombinant mouse GM-CSF (PeproTech), and 20 ng ml⁻¹ recombinant human TGF-β1 (PeproTech). At 70–90% confluency, cells were lifted using PBS with 10 mM EDTA at 37°C for 10 minutes, followed by gentle scraping. After 1 week, adherent cells acquired a round, alveolar macrophage–like morphology, at which point stocks were frozen for infection studies.

Macrophage infections

1. i). THP-1 infection. THP-1 cells were seeded at a titer of 1 x 10⁵ cells per well in 96-well plates, and phorbol myristate acetate (PMA) was added at a final concentration of 50 ng µl⁻¹ to induce differentiation into macrophage-like cells for 72 h prior to infection. B. ovis and B. abortus cells were resuspended from a 48-h-old plate in RPMI supplemented with 10% (v/v) HI FBS and added to tissue culture plates on the day of infection at a multiplicity of infection (MOI) of 100. Plates were centrifuged for 5 min at 150 x g and incubated for 1 h at 37°C in 5% (v/v) CO₂. The medium was removed, and fresh medium containing 50 µg ml⁻¹ gentamicin was supplied, followed by a 1-h incubation. Following gentamicin incubation, the medium was removed and replaced with fresh medium containing gentamicin and/or 25 µM nicardipine or 25 µM cilnidipine. For the enumeration of colony-forming units (CFUs), THP-1 cells were washed once with PBS, pH 7.4 and then lysed in H₂O for 10 min at room temperature at 2 h, 24 h, and 48 h post infection. Lysates were serially diluted, spotted onto TSA blood plates, and incubated at 37°C in 5% (v/v) CO₂ for 48 h to enumerate CFUs.
2. ii). FLAM infection. FLAM cells were seeded at a titer of 5 x 10⁴ cells per well in 96-well plates 1 day prior to infection. B. abortus cells were resuspended from a 48-h-old plate in RPMI supplemented with 10% (v/v) HI FBS and added to tissue culture plates on the day of infection at a MOI of 100. Plates were centrifuged for 5 min at 150 x g and incubated for 1 h at 37°C in 5% (v/v) CO₂. The medium was removed, and fresh medium containing 50 µg ml⁻¹ gentamicin was supplied followed by a 1-h incubation. The medium was removed and replaced with fresh medium containing gentamicin and/or 25 µM nicardipine or 25 µM cilnidipine. For the enumeration of CFUs, FLAM cells were washed once with PBS, pH 7.4 and then lysed with H₂O for 10 min at room temperature at 2 h, 24 h, and 48 h post infection. Lysates were serially diluted, spotted onto TSA blood plates, and incubated at 37°C in 5% CO₂ for 48 h to enumerate CFUs.
3. iii). Intracellular dose-response. THP-1 cells were seeded at a titer of 1 x 10⁵ cells per well in 96-well plates, and PMA was added at a final concentration of 50 ng µl⁻¹ to induce differentiation into macrophage-like cells for 72 h prior to infection. B. ovis and B. abortus cells expressing the lux operon were resuspended from a 48-h-old plate in RPMI supplemented with 10% (v/v) HI FBS and added to tissue culture plates on the day of infection at a MOI of 100. The medium was removed, and fresh medium containing 50 µg ml⁻¹ gentamicin was supplied and incubated for 1 h. The medium was removed and replaced with fresh RPMI supplemented with 10% (v/v) FBS and containing 50 µg ml⁻¹ gentamicin and various concentrations of each calcium channel blocker (nicardipine or cilnidipine). Luminescence units were read at 48 h post infection using a Tecan Infinite M200 Pro microplate reader in the MSU University Research Containment Facility when working with B. abortus or on a Tecan Spark microplate reader when working with B. ovis. The IC₅₀ values were determined from dose-response plots of log(inhibitor) versus normalized responses using GraphPad Prism software.

Colorimetric THP1 XTT cell proliferation assay

THP-1 cells were seeded at a titer of 1 x 10⁵ cells per well in RPMI supplemented with 10% (v/v) FBS and containing 50 ng µl⁻¹ PMA to induce differentiation into macrophage-like cells in 96-well plates (Corning) for 48–72 h. Varying concentrations of compounds were prepared manually and transferred to plates containing THP1 cells with final concentrations of each compound (suspended in RPMI and 10% [v/v] FBS) ranging from 0.2 µM to 50 µM. After incubation, 50 µl XTT (sodium 3′-[1-(phenyl aminocarbonyl)-3,4-tetrazolium]-bis (4-methoxy6-nitro) benzene sulfonic acid hydrate) (Roche) was added to each well and incubated for 4 h at 37°C and 5% (v/v) CO₂. Following incubation, absorbance of the reaction product was measured at 570 nm and 650 nm.

Agar plate/stress assays

After 2 days of growth on TSA blood plates, B. ovis or B. abortus cells were collected and resuspended in sterile PBS to an OD₆₀₀ = 0.3. Each strain was serially diluted tenfold in PBS. Two to five microliters of each dilution was spotted onto either TSA blood, TSA blood containing either 0.005–1% (w/v) SDS, 0.009–1% (w/v) CHAPS, 0.009–1% (w/v) deoxycholate, or 1 µg ml^-1 penicillin G benzathine (Sigma). After 3 days of incubation on TSA blood, growth was documented photographically.

Inductively coupled plasma mass spectrometry

1. i). Macrophage treatment. THP-1 cells were seeded at a titer of 1 x 10⁷ cells in 100-mm dishes with RPMI supplemented with 10% (v/v) HI FBS and containing 50 ng µl⁻¹ PMA to induce differentiation into macrophage-like cells for 72 h. The medium was then removed and replaced with fresh medium alone or with 25 µM nicardipine and incubated for 48 h at 37°C in 5% (v/v) CO₂.
2. ii). Cell harvesting and ICP-MS sampling. The medium was removed and cells were washed with PBS, pH 7.4, twice. TrypLE Express Enzyme (1x) (Gibco) without phenol red was added to the plates and incubated for 5–10 min. Cells were gently detached from the plate surface and fresh RPMI supplemented with 10% (v/v) HI FBS was added. Cell solutions from the dishes were removed and transferred to new 15-ml or 50-ml tubes. Cells in suspension were gently mixed and transferred to 15-ml metal-free tubes. Gadolinium-DOTA (Gd-DOTA) MRI contrast agent (40 µM final concentration) was spiked into each 15-ml metal-free tube before centrifugation at 400 × g for 3 min at 4°C. After the cells were pelleted, 0.4 ml of each supernatant was transferred into a new 15-ml metal-free tube, then the remaining supernatants were removed. Gd-DOTA, a membrane-impermeable complex, remains confined to the extracellular space [64]. By spiking samples with Gd-DOTA, we can quantify the residual media volume in an isolated cell pellet, enabling accurate determination of intracellular element contents by subtracting extracellular elements in the retained media. Furthermore, this ICP-MS sample preparation method minimizes disruptions to metal homeostasis by eliminating washing steps and maintaining cells in their original culture media during sampling.

The resulting tubes were dried at 70°C overnight. Concentrated nitric acid (67–69%, 0.18 ml) was added to the tubes, which were then incubated at 70°C overnight until the pellet was completely acid-digested. Following the acid-digestion, Milli-Q water was added to the tubes to yield a 3% (v/v) nitric acid matrix. The elemental concentration of all digested samples and calibration standard were determined on a parts per billion (ppb, µg/L) scale using an Agilent 8900 Triple Quadrupole ICP-MS (Agilent) equipped with an Agilent SPS 4 Autosampler, integrated sample introduction system (ISIS), x-lens, and micromist nebulizer. Instrument tuning and ICP-MS calibration stand preparation were performed as previously described [65]. The isotopes selected for analysis of samples were ³¹P, ²³Na, ²⁴Mg, ³²S, ³⁹K, ⁴⁴Ca, ⁵¹V, ⁵²Cr, ⁵⁵Mn, ⁵⁷Fe, ⁵⁹Co, ⁶⁰Ni, ⁶³Cu, ⁶⁶Zn, ⁷⁵As, ⁸⁰Se, ^{111 Cd}, and ¹⁵⁷Gd.