Leptospira glyceraldehyde-3-phosphate dehydrogenase (LiGAPDH): a cell-surface plasminogen binding protein

TL;DR

This study shows that a protein from Leptospira can bind to a human blood protein, which may help the bacteria invade and spread in the body.

Contribution

The paper reports a novel function of Leptospira GAPDH in binding human plasminogen.

Findings

LiGAPDH is surface-exposed and interacts with plasminogen.

The LiGAPDH-plasmin complex degrades fibrinogen and vitronectin.

Plasmin bound to LiGAPDH degrades the C5 α-chain but not C3b.

Abstract

Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) is classically recognized as a glycolytic enzyme that catalyzes the conversion of glyceraldehyde 3-phosphate into D-glyceryl 1,3-bisphosphate. However, it also exhibits “moonlighting” functions, serving roles unrelated to metabolism. Notably, this multifunctional protein, which lacks a conventional membrane anchor, is present on the surface of many prokaryotic and eukaryotic cells. In this study, we demonstrate that Leptospira interrogans GAPDH (LiGAPDH) is surface-exposed and interacts with plasminogen. In the presence of the exogenous activator uPA, plasminogen is converted into its active form, plasmin. The LiGAPDH-plasmin complex can degrade fibrinogen (α and β chains) and the 75-kDa form of vitronectin over time. Interestingly, plasmin, when bound to LiGAPDH, completely degrades the C5 α-chain but does not affect C3b. The functional characterization of moonlighting proteins and the identification of host molecules they interact with may offer insights for understanding the mechanisms of invasion, dissemination, and immune evasion employed by pathogenic leptospires.

In this study, we report a novel function of GAPDH from Leptospira: its ability to bind human plasminogen, a key component of the host coagulation system.

Figures

Cellular localization of GAPDH on Manilae L495 and Patoc 1 by transmission electron microscopy. Pathogenic (L. interrogans Manilae L495) and saprophytic (L. biflexa Patoc) leptospires were firstly incubated with anti-GAPDH or pre-immune sera obtained from rabbit, followed by incubation with anti-rabbit IgG conjugated to 10 nm colloidal gold. The analysis was performed using transmission electron microscopy. Note that the preparations were not contrasted and that gold particles were not observed when bacteria were incubated with pre-immune serum. Bars, 0.2 µm.

GAPDH and plasminogen interaction. (A) GAPDH interacts with plasminogen. SDS-PAGE: GAPDH (37 kDa), LigAC (63 kDa, positive control), and BSA (68 kDa, negative control) were subjected to SDS-PAGE and transferred to a nitrocellulose membrane. The membrane was incubated with purified plasminogen (50 µg), followed by incubation with antiplasminogen and secondary antibody conjugated to peroxidase and revealed by chemiluminescence with the ECL kit. (B) Plasminogen bound to GAPDH is converted to functionally active plasmin. GAPDH or BSA (10 µg/ml) immobilized on a microtiter plate were incubated with plasminogen (20 µg/ml). After washes, uPA (3 U) and the chromogenic substrate D-valyl-leucyl-lysine-ρ-nitroanilide dihydrochloride (0.25 μg/well) were added and incubated for 4 h. Data represent the mean absorbance value at 405 nm ± the standard deviation of three independent experiments, each performed in duplicate (P < 0.0001, Student’s t test). PLG (Plasminogen), uPA (urokinase-type plasminogen activator), PAI-1 (plasminogen activator inhibitor type 1), and Sub (substrate).

Degradation of human fibrinogen, vitronectin, C3b, and C5 by plasminogen bound to GAPDH. Plasminogen (20 µg/ml) was added to recombinant GAPDH (10 µg/ml) immobilized on microtiter plates. After washes, 500 ng of fibrinogen (A), vitronectin (B), C3b (C), or 1 µg of C5 (D), and 3 U of uPA were added and incubated for the indicated times points. Samples were separated by SDS-PAGE, transferred to nitrocellulose membranes, and probed with specific antibodies: antifibrinogen (A), antivitronectin (B), anti-C3 (C), or anti-C5 (D). Controls omitting plasminogen and uPA were included. Fragments subject to degration are indicated by arrows, and degradation products are indicated by asterisks.

Full text

Introduction

Moonlighting proteins are a group of multifunctional proteins that, besides carrying out their primary canonical functions, also play unrelated roles in diverse biological processes. This group of proteins includes enzymes, transcription factors, molecules with structural functions, and those involved in protein folding or synthesis (Jeffery 2017, Huberts and Van Der Klei 2010). Moonlighting proteins are present in various cellular compartments and are sometimes expressed across different cell types and species. This diversity extends to their functional versatility, with many acting as cytosolic enzymes or chaperones and performing additional roles in other cellular compartments. Some moonlighting proteins are also capable of carrying out multiple functions within the same compartment independently. These functional shifts can be triggered by several mechanisms, such as temperature changes, interactions with the cell membrane, secretion into the extracellular space, interactions with DNA or RNA, fluctuations in the cellular concentration of ligands, substrates, or cofactors, and other factors that enable the protein to perform the most crucial function for the cell or organism at that moment (Jeffery 2009, 2017).

In bacteria, moonlighting proteins may assume pathogenic functions contributing to host colonization (attachment, invasion, and biofilm formation), modulation and evasion of immune responses, and intracellular survival (revised in Liu and Bhunia 2024; Foulston et al. 2014). The mechanisms behind the secretion and surface anchoring of these proteins remain unclear, as they generally lack classical signal sequences or anchoring motifs. Proposed explanations for their atypical surface localization include increased membrane permeability caused by cellular damage and co-transport alongside conventionally secreted proteins. Additionally, electrostatic interactions are believed to play a role in anchoring these proteins to the bacterial surface (Liu and Bhunia 2024).

Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) is one of the most extensively studied moonlighting proteins in bacteria. It has a primary role in glycolysis, where it catalyzes the conversion of glyceraldehyde-3-phosphate to 1,3-bisphospho-D-glycerate. In a large range of microorganisms, including pathogenic Escherichia coli (EHEC and EPEC), Listeria monocytogenes, Mycobacterium tuberculosis, Mycoplasma genitalium, Mycoplasma pneumoniae, Staphylococcus aureus, Streptococcus agalactiae, Streptococcus pyogenes, and Streptococcus pneumoniae, GAPDH has been shown to moonlight on the surface, where it interacts with host proteins. Plasminogen, host ECM, complement molecules, epidermal growth factor, lactoferrin, transferrin, mucin, and erythrocytes are examples of GAPDH receptors in the host (Henderson 2017, Kopeckova et al. 2020). Besides displaying nonenzymatic functions on the bacterial surface, GAPDH has additional noncanonical functions intracellularly and in the extracellular space (Kopeckova et al. 2020). In the cytosol, it seems to participate in DNA repair processes (Povirk 1996, Ferreira et al. 2015) while extracellularly it exerts immunomodulatory effects on B and T cells, induces interleukin-10 (IL-10) production leading to suppression of neutrophil recruitment (Madureira et al. 2007, 2011), and interferes with Rab5-mediated endosome formation (Alvarez-Dominguez et al. 2008).

Recently, the crystal structure of GAPDH from Leptospira interrogans (LiGAPDH)—the etiological agent of leptospirosis—was solved. Furthermore, it has been shown that LiGAPDH can bind and sequester the anaphylatoxin C5a, potentially reducing the chemotatic response (Navas-Yuste et al. 2023). LiGAPDH is among the 20 most abundant extracellular proteins, according to an exoproteome study conducted by Eshghi et al. (2015). Given its abundance in the extracellular environment, it is possible that the sequestration of C5a by GAPDH is relevant and may interfere with the host’s innate immune response mediated by the complement system.

In the present study, we aimed to further explore the nontraditional functions of LiGAPDH, with a particular focus on its roles on the bacterial surface. Our results show that this moonlighting protein from Leptospira interacts with plasminogen and that the generated plasmin degrades fibrinogen, vitronectin, and the alpha chain of C5. It is therefore assumed that this interaction contributes to the processes of invasion and immune evasion of this spirochete, similar to what has been reported for other medically important bacteria (reviewed in Liu and Bhunia 2024; Henderson and Martin 2011).

Materials and methods

Bacterial strains and growth conditions

Leptospira interrogans serovar Manilae strain L495 and Leptospira biflexa serovar Patoc strain Patoc 1 were provided by Prof. Elsio Wunder (Department of Pathobiology and Veterinary Science, University of Connecticut, Storrs, CT, USA) and by Prof. Marcos Bryan Heinemann (Laboratory of Bacterial Zoonosis, Department of Preventive Veterinary Medicine and Animal Health, School of Veterinary Medicine, University of São Paulo, SP, Brazil). Both strains were cultivated at 29°C for 7 days, under aerobic conditions in liquid Ellinghausen-McCullough-Johnson-Harris (EMJH) medium (Difco) supplemented with Leptospira EMJH Enrichment (BD).

Purified proteins, sera, and antibodies

Plasminogen, fibrinogen, vitronectin, urokinase plasminogen activator (uPA), plasminogen activator inhibitor-1 (PAI-1), the chromogenic substrate D-valyl-leucyl-lysine-ρ-nitroanilide dihydrochloride, rabbit antihuman fibrinogen, and rabbit antihuman vitronectin were purchased from Sigma-Aldrich, Co., USA. Purified C3b and C5, and goat anti-human C3 or C5 polyclonal antibodies were purchased from Complement Technology, Inc., USA. All secondary peroxidase-conjugated antibodies were acquired from Sigma-Aldrich, Co., USA.

Cloning, expression, and purification of LiGAPDH

The gene sequence corresponding to LiGAPDH was amplified by PCR from the genomic DNA of Leptospira interrogans serovar Copenhageni strain L1–130 using the synthetic oligonucleotides F-5′-TTTTGGATCCACCAGAATAGCCATCAACGGATTTG-3′ and R-5′-TTTAAGCTTTTAACCTTTTTTCGCCATATAGCG-3′, which contain appropriate restriction sites. The fragment was purified, initially cloned into the pGEM-T Easy vector, and subsequently subcloned into pET-28a. Recombinant LiGAPDH expression was performed using E. coli BL21-DE3 transformed with LiGAPDH/pET-28a. Colonies grown on LB/kanamycin plates were used to inoculate 10 ml of LB/kanamycin and incubated at 37°C, 130 rpm, for 18 h. This preculture was transferred to 1 L of the same medium and grown until OD600 reached 0.6–0.8. Expression was induced with 1 mM IPTG, followed by 4 h incubation at 37°C. Cells were harvested by centrifugation (3000 × g, 15 min, 4°C), resuspended in 100-mM Tris, 500-mM NaCl, and lysed using a French Pressure Cell Press. The lysate was centrifuged (13 000 × g, 15 min, 4°C), and the soluble fraction was purified by metal affinity chromatography using Fast Protein Liquid Chromatography (ÄKTA system, GE Healthcare) with an imidazole gradient. Protein purity was assessed by 12% SDS-PAGE.

Cellular localization of GAPDH by immunogold labeling

Cultures of Leptospira strains grown in EMJH were washed (centrifugation at 2300 × g for 5 min) with PBS and fixed with 2% formaldehyde in PBS, for 30 min at room temperature. The preparations were then washed with PBS containing 0.2% BSA (PBS-BSA) and blocked with the same solution for 30 min. Following the blocking step, preparations were incubated with the primary antibody (anti-GAPDH produced in rabbit or preimmune serum) at a dilution of 1:10 in PBS, for 16–20 h at 4°C. Preparations were washed with PBS-BSA and incubated with the secondary antibody (goat antirabbit IgG conjugated to 10-nm colloidal gold particles—Sigma-Aldrich, Co., USA) diluted 1:10 in PBS for 3 h at room temperature. After final washes with PBS, pellets were resuspended in distilled water, and the suspensions were immediately applied onto 200 mesh square nickel grids previously coated with 0.5% Formvar (EMS, USA) in chloroform. After left at room temperature for 2 min, excess liquid on the grid was removed with filter paper, and the grid, air-dried. Preparations were then analyzed by transmission electron microscopy (TEM-LEO 906E, Zeiss, Germany), at the Structural and Functional Biology Laboratory of the Butantan Institute.

Interaction of LiGAPDH with plasminogen by far-Western blot

Purified recombinant LiGAPDH was subjected to SDS-PAGE (12%) under nonreducing conditions and transferred to a nitrocellulose membrane. Nonspecific binding sites were blocked with PBS-Tween 0.05% (PBS-T) containing 5% BSA for 18 h at 4°C, after which the membrane was washed with PBS-T. The membrane was then incubated with 50 µg of plasminogen diluted in PBS for 90 min at room temperature. Subsequently, the membrane was washed five times with PBS-T, and incubated with anti-plasminogen at a dilution of 1:5000 for 1 h at room temperature and under shaking. Following incubation, the membrane was washed three times with PBS-T and incubated with horseradish peroxidase-conjugated goat anti-rabbit IgG at a dilution of 1:5 000 for 1 h at room temperature. After washes, positive signals were detected by enhanced chemiluminescence (SuperSignal® West Pico Enhanced Chemiluminescent Substrate, Pierce). LigAC and BSA were respectively included as positive and negative controls.

Evaluation of the ability of LiGAPDH-bound plasminogen to be converted into plasmin

Recombinant LiGAPDH was used to coat 96-well plates (1 µg/well). Plates were incubated for 16–20 h at 4°C. After blocking nonspecific binding sites using 3% gelatin for 2 h at 37°C, purified plasminogen (2 µg/well) was added and plates were incubated for 1 h at 37°C. After removing unbound plasminogen by washing three times with PBS-T, 3 U of uPA and 0.25 µg/well of chromogenic substrate D-Val-Leu-Lys p-nitroanilide dihydrochloride were added in a final volume of 100 µL. After 4 h incubation at 37°C, absorbance was read at 405 nm. BSA was used as a negative control.

Assessment of fibrinogen, vitronectin, C3b and C5 degradation by LiGAPDH-bound plasmin

Recombinant LiGAPDH (10 µg/ml) was immobilized in 96-well plates, and nonspecific binding sites were blocked using 3% gelatin diluted in PBS. Plasminogen (20 µg/ml) was added to the wells and incubated for 1 h at 37°C. After three washes with PBS-T, uPA (3 U), along with fibrinogen (500 ng/well), vitronectin (500 ng/well), C3b (500 ng/well), or C5 (1 µg/well) were added to the wells. Reactions were incubated at 37°C, and samples were collected at 0, 2, and 4 h. Fibrinogen and vitronectin were subjected to 12% SDS-PAGE, while C3b and C5 were subjected to 10% SDS-PAGE, both under reducing conditions. Proteins were transferred to nitrocellulose membranes and the nonspecific binding sites were blocked with PBS-T containing 10% calcium-free lyophilized skimmed milk for 18 h at 4°C. Membranes were subsequently washed with PBS-T, and the cleavage fragments were detected using rabbit anti-fibrinogen, rabbit anti-vitronectin, goat anti-C3, and goat anti-C5 polyclonal antibodies diluted 1:5 000 in a 5% calcium-free lyophilized skimmed milk solution in PBS-T. After incubations with the primary antibodies, membranes were washed with PBS-T and incubated with the secondary antibodies (1:5000 diluted anti-rabbit IgG or 1:10 000 diluted anti-goat IgG, both conjugated with HRP), for 60 min at room temperature, under agitation. Positive signals were detected by chemiluminescence using SuperSignal® West Pico Enhanced Chemiluminescent Substrate, Pierce.

Results

LiGAPDH is localized on the cell-surface

In addition to its classical glycolytic role in the cytosol, LiGAPDH has been identified as one of the most abundant proteins in the extracellular environment (Eshgi et al. 2015). In this study, we investigated whether LiGAPDH is also present on the surface of Leptospira cells using transmission electron microscopy. Immunogold labeling confirmed the surface localization of LiGAPDH in both pathogenic (Manilae L495) and nonpathogenic (Patoc I) strains (Fig. 1).

Cellular localization of GAPDH on Manilae L495 and Patoc 1 by transmission electron microscopy. Pathogenic (L. interrogans Manilae L495) and saprophytic (L. biflexa Patoc) leptospires were firstly incubated with anti-GAPDH or pre-immune sera obtained from rabbit, followed by incubation with anti-rabbit IgG conjugated to 10 nm colloidal gold. The analysis was performed using transmission electron microscopy. Note that the preparations were not contrasted and that gold particles were not observed when bacteria were incubated with pre-immune serum. Bars, 0.2 µm.

LiGAPDH interacts with human plasminogen

As previously mentioned, both Gram-negative and Gram-positive bacterial species display GAPDH on their surface and exploit it to bind host plasminogen, a strategy that aids in tissue invasion, immune evasion, and pathogenesis. In this study, we demonstrate by ligand-affinity blotting that recombinant LiGAPDH also binds human plasminogen, similar to LigA, our positive control previously shown to interact with this key component of the coagulation cascade (Castiblanco-Valencia et al. 2016) (Fig. 2A). Plasminogen bound to LiGAPDH was subsequently converted into active plasmin, which was detected using the chromogenic substrate D-valyl-leucyl-lysine-p-nitroanilide. Negative controls included the addition of plasminogen activator inhibitor-1 (PAI-1) or the omission of urokinase-type plasminogen activator (uPA), plasminogen, or both (Fig. 2B).

GAPDH and plasminogen interaction. (A) GAPDH interacts with plasminogen. SDS-PAGE: GAPDH (37 kDa), LigAC (63 kDa, positive control), and BSA (68 kDa, negative control) were subjected to SDS-PAGE and transferred to a nitrocellulose membrane. The membrane was incubated with purified plasminogen (50 µg), followed by incubation with antiplasminogen and secondary antibody conjugated to peroxidase and revealed by chemiluminescence with the ECL kit. (B) Plasminogen bound to GAPDH is converted to functionally active plasmin. GAPDH or BSA (10 µg/ml) immobilized on a microtiter plate were incubated with plasminogen (20 µg/ml). After washes, uPA (3 U) and the chromogenic substrate D-valyl-leucyl-lysine-ρ-nitroanilide dihydrochloride (0.25 μg/well) were added and incubated for 4 h. Data represent the mean absorbance value at 405 nm ± the standard deviation of three independent experiments, each performed in duplicate (P < 0.0001, Student’s t test). PLG (Plasminogen), uPA (urokinase-type plasminogen activator), PAI-1 (plasminogen activator inhibitor type 1), and Sub (substrate).

Plasminogen bound to LiGAPDH degrades fibrinogen, vitronectin, and C5

Once recruited to the surface of bacteria and converted into its active form, plasmin(ogen) functions as a broad-spectrum protease that may act on diverse host molecules. Fibrinogen and fibrin clots are primary targets of plasmin. Extracellular matrix proteins, basement membrane, and complement components are also among host molecules degraded by plasmin, thereby promoting tissue invasion and barrier penetration. The proteolytic activity of plasmin(ogen) bound to LiGAPDH was evaluated by assessing its ability to degrade fibrinogen, vitronectin, and the complement components C3b and C5. After 2 h of incubation, both the α- (73 kDa) and β- (60 kDa) chains of fibrinogen were completely degraded (Fig. 3A). Similarly, the 78-kDa isoform of vitronectin was efficiently cleaved by LiGAPDH-bound plasmin (Fig. 3B). In contrast, the α- and β-chains of C3b remained intact under the same conditions (Fig. 3C), whereas the α-chain of C5 was fully degraded after 2 h of incubation (Fig. 3D).

Degradation of human fibrinogen, vitronectin, C3b, and C5 by plasminogen bound to GAPDH. Plasminogen (20 µg/ml) was added to recombinant GAPDH (10 µg/ml) immobilized on microtiter plates. After washes, 500 ng of fibrinogen (A), vitronectin (B), C3b (C), or 1 µg of C5 (D), and 3 U of uPA were added and incubated for the indicated times points. Samples were separated by SDS-PAGE, transferred to nitrocellulose membranes, and probed with specific antibodies: antifibrinogen (A), antivitronectin (B), anti-C3 (C), or anti-C5 (D). Controls omitting plasminogen and uPA were included. Fragments subject to degration are indicated by arrows, and degradation products are indicated by asterisks.

Discussion

To date, four moonlighting proteins have been identified and characterized in Leptospira. The first to be described was elongation factor Tu (EF-Tu), which functions not only in protein synthesis but also as a surface-exposed plasminogen receptor and complement factor H (FH)-binding protein, contributing to immune evasion (Wolff et al. 2013). The second, enolase, was shown to play a role in immune evasion by interacting with the complement regulators C4b-binding protein (C4BP) and FH, in addition to recruiting host plasminogen (Nogueira et al. 2013, Salazar et al. 2017). More recently, the chaperonin GroEL, classically involved in protein folding under stress conditions, was found to function as a surface adhesin, capable of binding plasma and extracellular matrix components, and of stimulating the production of proinflammatory cytokines by macrophages (Ho et al. 2021). Finally, LiGAPDH was shown to act extracellularly as an immune evasion factor by sequestering C5a and thereby preventing—or at least delaying—the anaphylatoxin’s ability to recruit nearby phagocytes (Navas-Yuste et al. 2023).

In this study, we demonstrate that LiGAPDH localizes not only to the cytosol and extracellular milieu but is also present on the surface of leptospires, where it is presumed to mediate interactions with host molecules. Given that plasminogen binding is a well-established characteristic of moonlighting proteins in a variety of bacterial species (Kopeckova et al. 2020), this study sought to investigate the interaction between LiGAPDH and this pivotal component of the coagulation cascade.

LiGAPDH interacts with human plasminogen, which is converted into its active form, plasmin. The ability of plasmin to degrade physiological substrates was evaluated. LiGAPDH-bound plasmin time-dependently degraded the α and β chains of fibrinogen, as well as the 78-kDa isoform of vitronectin. Regarding complement system molecules, which are already known to be targeted by plasmin (Barthel et al. 2012), it is interesting to note that plasmin bound to LiGAPDH completely cleaved the α chain of C5, but had no effect on C3b. In the context of Leptospira–host interactions, those findings may have key potential implications. Fibrinogen degradation leads to reduced fibrin clot formation, which normally acts as a physical barrier to infection, whereas vitronectin degradation may impair extracellular matrix integrity and disrupt host tissue architecture. As a consequence, bacterial dissemination through host tissues is facilitated. Thus, by binding plasminogen and facilitating its activation to plasmin, LiGAPDH could co-opt the host fibrinolytic system, providing the bacterium with proteolytic power without needing to produce its own proteases. In brief, plasmin, hijacked by the bacterium, can lead to vascular leakage and tissue damage.

Complete degradation of the C5 α-chain by LiGAPDH-bound plasmin may have significant implications for Leptospira’s ability to evade the complement system. Under normal circumstances, complement activation leads to the cleavage of C5 at a specific arginine-leucine bond, producing C5a, a small (∼11 kDa), potent anaphylatoxin released into the fluid phase (DiScipio et al. 1983), and C5b, a larger fragment that initiates the formation of the membrane attack complex (MAC). Degradation of the C5 α-chain disrupts the generation of functional C5b, thereby preventing MAC assembly. As a result, Leptospira may be protected from complement-mediated lysis, enhancing its survival within the host. Therefore, LiGAPDH targets complement C5 in two ways: by binding and sequestering C5a (Navas-Yuste et al. 2023) and by degrading C5 α-chain, thus behaving as an immune evasive protein.

In summary, plasminogen emerges as a common target of moonlighting GAPDH on bacterial surfaces, indicating that its recruitment may confer a considerable advantage for the dissemination and immune evasion of pathogens (Kopeckova et al. 2020). Bacteria often exploit a repertoire of moonlighting proteins—frequently exhibiting redundant or additive functions—to achieve their primary objective upon host entry: effective dissemination and colonization of target organs. Given its localization in multiple cellular compartments, LiGAPDH is presumed to contribute not only to its canonical role in metabolism but also to Leptospira invasion and modulation of host immune responses. Further studies are warranted to elucidate additional contributions of this moonlighting protein to Leptospira pathogenesis.