Lactate dehydrogenase is the Achilles' heel of Lyme disease bacterium Borreliella burgdorferi

Abstract

As a zoonotic pathogen, the Lyme disease bacterium Borreliella burgdorferi has evolved unique metabolic pathways, some of which are specific and essential for its survival and thus present as ideal targets for developing new therapeutics. B. burgdorferi dispenses with the use of thiamin as a cofactor and relies on lactate dehydrogenase (BbLDH) to convert pyruvate to lactate for balancing NADH/NAD⁺ ratios. This report first demonstrates that BbLDH is a canonical LDH with some unique biochemical and structural features. A loss-of-function study then reveals that BbLDH is essential for B. burgdorferi survival and infectivity, highlighting its therapeutic potential. Drug screening identifies four previously unknown LDH inhibitors with minimal cytotoxicity, two of which inhibit B. burgdorferi growth. This study provides mechanistic insights into the function of BbLDH in the pathophysiology of B. burgdorferi and lays the groundwork for developing genus-specific metabolic inhibitors against B. burgdorferi and potentially other tick-borne pathogens as well.

Importance: Lyme disease (LD) is the most commonly reported tick-borne illness in the U.S. and Europe, and its geographic distribution is continuously expanding worldwide. Though early LD can be treated with antibiotics, chronic LD is recalcitrant to antibiotic treatments and thus requires multiple courses of antibiotic therapy. Currently, there are no human vaccines nor prophylactic antibiotics to prevent LD. As the causative agent of LD, Borreliella burgdorferi has evolved unique metabolic pathways, some of which are specific and essential for its survival and thus present as ideal targets for developing new therapeutics. By using an approach of genetics, biochemistry, structural biology, drug screening, and animal models, this report provides evidence that lactate dehydrogenase can be a potential target for developing genus-specific metabolic inhibitors against B. burgdorferi and potentially other tick-borne pathogens as well.

Keywords: Borrelia burgdorferi; Lyme disease; antibiotics; drug screening; lactate dehydrogenase.

Fig 1

BbLDH enzyme kinetics and inhibition constants of gossypol against BbLDH. (A) BB_0087 is allosterically regulated by FBP. The addition of 3 mM FBP enhanced the binding affinity of BbLDH to pyruvate. Mutation of H171A, a predicted key residue in the FBP binding site, abolished the allosteric activation of FBP on BbLDH. (B and C) The addition of FBP increased the V_max of BbLDH for NADH and NAD⁺ using a varying concentration of cofactors. (D) BbLDH has an optimal working pH range between 5.0 to 7.0. (E) BbLDH has an optimal working temperature of 37°C using a varying concentration of pyruvate substrate. (F) Enzyme kinetics of inhibition on BbLDH by gossypol. BB_0087 catalyzed the reaction of lactate to pyruvate, which could be inhibited by gossypol in a dose-dependent manner. Of note, NADH standards of 0 (blank), 2.5, 5, 7.5, 10, and 12.5 nmol per well were generated using the spectrophotometric assay at OD₃₄₀. The linear regression and enzyme kinetic analyses were performed in GraphPad Prism 10. S, substrate; V, velocity.

Fig 2

Structure of BbLDH. (A) 2.1 Å resolution crystal structure of BbLDH tetramer with NADH and oxamate (PDB 9DQ9). (B) Superimposition of BbLDH crystal structure without FBP (tan, PDB 9DQ9) and with FBP (green, PDB 9DQ7) (RMSD = 0.230 Å, 314 atom pairs). (C) Interacting residues in the NADH and oxamate binding site pocket. (D) Salt bridge and hydrogen bonding residues that interact with FBP in BbLDH (chains A and C, pink and yellow).

Fig 3

BB_0087 is essential for the in vitro growth of B. burgdorferi. (A) To create an inducible vector for conditional knockout of bb_0087, the full-length bb_0087 gene was cloned into an isopropyl β-D-1-thiogalactopyranoside (IPTG)-inducible vector (50) with a C-terminal FLAG tag, generating pJSB-BB_0087FLAG. The final construct was transformed into the wild-type B31 A3-68 parental strain. (B) To construct an in-frame deletion mutant in WT carrying pJSB-BB_0087FLAG, the primer pairs P₃/P₄ and P₅/P₆ were used to amplify the upstream and downstream flanking regions of bb_0087. The primer pair P₇/P₈ was used to amplify a promoterless kanamycin resistance cassette (kan). bb_0087 was in-frame replaced by kan via the PCR fusion technique with the primer pair P₃/P₆. The resulting PCR amplicon was cloned into a pGEM-T-easy vector forming BB_0087::kan. (C) Immunoblotting analysis of WT and 87^mut in the presence or absence of IPTG from the growth curve in panel D at day 10. The same amounts of WT and 87^mut whole-cell lysates were analyzed by SDS-PAGE and then probed with antibodies against BbLDH and DnaK. Of note, the band detected in the 87^mut strain was larger than its native form in the WT due to the addition of the FLAG tag. (D) Growth analysis of WT and 87^mut under a routine laboratory condition (34°C/pH 7.4) with and without 1 mM IPTG addition to induce the expression of BbLDH. Cell numbers were enumerated daily until cells entered the stationary growth phase (~10⁸ cells/mL). Cell counting was repeated in triplicate with three independent samples, and the results are expressed as means ± SEM.

Fig 4

BbLDH contributes to the infectivity of B. burgdorferi. To determine the role of BbLDH in the pathophysiology of B. burgdorferi, a needle infection study was conducted using BALB/c mice. WT or 87^mut strains (10³ spirochetes) were subcutaneously inoculated into BALB/c mice and sacrificed 3 weeks after injection. Mice infected with 87^mut were divided into two groups (four mice per group), one group received regular drinking water while the other received IPTG-supplemented water as described (51) for the duration of the infection study. (A) Ear tissues were harvested for quantitative reverse transcription polymerase chain reaction (qRT-PCR) as described (52). (B) The blood was collected for a seroconversion test. Each lane represents serum obtained from individual mice under each group as indicated on the top of the image. qRT-PCR data are presented as mean flaB transcripts over 10⁵ of mouse β-actin ± SEM, significant difference (*P < 0.05; **P < 0.01). ns, no significance.

Fig 5

BbLDH activity can be inhibited by gossypol and four novel LDH inhibitors. Four lead compounds with anti-LDH activities were identified through high-throughput screening from the Natural Product Set IV library. (A) Chemical structures of the four lead compounds with inhibitory effect on BbLDH activity along with their molecular weights. Gossypol was included for comparison. Enzyme kinetics of inhibition of BbLDH by 0–100 µM of (B) gossypol, (C) 14975, (D) 45923, (E) 114344, and (F) 350085 using varying concentrations of pyruvate as a substrate. Data are presented as mean velocity ± SEM from two independent experiments.

Fig 6

Impact of LDH inhibitors on the growth of B. burgdorferi. Growth analysis of wild-type B. burgdorferi in the presence of 0 (dimethyl sulfoxide [DMSO]) to 300 µM of (A) gossypol, (B) 14975, (C) 45923, (D) 114344, or (E) 350085. Cell numbers were enumerated every 2 days using a Petroff-Hausser counting chamber until cells entered the stationary growth phase (~10⁸ cells/mL). Cell counting was repeated in triplicate with two independent samples, and the results are expressed as means ± SEM.

Fig 7

Docking analysis of gossypol. (A) Top docking pose of gossypol to an NADH- and oxamate-bound BbLDH model. Left: overview of the BbLDH dimer model with the electrostatic potential surface; right: the gossypol binding site location relative to the NADH and oxamate cofactors. (B) Hydrogen bonding interactions between gossypol and BbLDH. A ligand interaction diagram (left) and stick model (right) for the top-ranked gossypol pose are shown.

Fig 8

Impact of LDH inhibitors on the growth of HeLa and telomerase immortalized gingival keratinocytes (TIGKs) cell lines. Growth analysis of an immortalized cervical cancer cell line, HeLa and telomerase-immortalized human gingival epithelial cell line (TIGKs) in the presence of (A) gossypol, (B) 45923, or (C) 350085. Cells were treated with the indicated concentration of inhibitors for 72 hours in a 96-well plate format, and cell viability was determined using the crystal violet assay as described (57). The growth inhibition study was performed in triplicate per concentration of inhibitors and in duplicate of technical and biological sets. Percentages of viable cells were normalized to mock-treated wells, and statistical analysis was performed using analysis of variance. The results are expressed as means ± SEM from two to three biological replicates, significant differences (*P < 0.05; **P < 0.01, ***P < 0.001, and ****P < 0.0001).