A stable systemic infection of methicillin-resistant Staphylococcus aureus (MRSA) in cynomolgus macaques produces extended window for therapeutic intervention

Abstract

Introduction: Staphylococcus aureus is a common gram-positive commensal that, upon entering the bloodstream, can cause devastating illness and death within hours or days. Methicillin-resistant S. aureus (MRSA) infections, now a leading cause of bloodstream infections worldwide, pose significant challenges due to their rapid progression, high mortality rates, and limited therapeutic options. While there are prevalent small animal models of experimental MRSA infection, there has been minimal development of larger mammalian models capable of recapitulating clinical aspects of human systemic MRSA infection.

Methods: Following a pilot study to determine the optimal dose and route to establish systemic MRSA infection, we challenged six cynomolgus macaques with 10⁹ colony-forming units MRSA (lineage USA300) via intravenous (IV) route. Animals were monitored closely up to 8 days for physiological, immunological, and cellular endpoints. Histopathology was performed on tissues collected 2 and 8 days after infection.

Results: An IV dose of 10⁹ CFU MRSA USA300 in cynomolgus macaques produced bacteremia resulting from multifocal invasive infections, elevated markers of systemic inflammation, as well as weight loss, fever, and hemodynamic changes consistent with bloodstream infection. Hematological analyses demonstrated neutrophilic leukocytosis, lymphocytopenia, monocytosis, and mild thrombocytopenia. We observed a robust cytokine response, including TNF-α, IL-6, G-CSF, and IL-1RA, peaking 6 h post-infection. Flow cytometry immunophenotyping revealed dynamic shifts in circulating monocyte subpopulations, and histopathological analysis demonstrated multi-organ damage with significant findings in the kidneys, heart, liver, and lungs. By 8 days post-infection, moderate to severe myocardial, renal, and hepatic dysfunction were evident, supported by changes in clinical chemistry biomarkers. None of the animals required euthanasia before the scheduled date of termination.

Discussion: In this study, we establish a non-human primate model of systemic MRSA infection that allows for the characterization of MRSA pathogenesis and evaluation of therapeutics over a period of days rather than hours. This model successfully recapitulates key aspects of human MRSA bloodstream infections, providing a valuable platform for evaluating therapeutic interventions and understanding disease mechanisms.

Keywords: MRSA; S. aureus bacteremia; Staphylococcus aureus; bacterial infection model; cynomolgus macaque; nonhuman primate; sepsis.

FIGURE 1

Pilot dose-finding study for systemic NHP MRSA infection model. Cynomolgus macaques were challenged with 10¹⁰–10⁷ CFU MRSA via IV or IP route and monitored for up to 7 days. Physical exams and blood collections were performed frequently between 0 and 48 hpi. (A) Percent change (%Δ) in body weight compared to baseline (0 hpi). (B) Change in rectal body temperature compared to baseline (0 hpi). (C) Mean arterial pressure (MAP) calculated from systolic and diastolic cuff readings. (A–C) The pilot animal that received a 10¹⁰ CFU dose via IV route (dark red) succumbed to infection between 5 and 18 hpi. (D) Absolute white blood cell counts from CBC hematology analysis of the pilot animal challenged with 10¹⁰ CFU MRSA via IP route.

FIGURE 2

IV challenge of NHPs with 10⁹ CFU MRSA produces bacteremia, elevated CRP, and clinical signs of systemic infection. Six cynomolgus macaques were infected with 10⁹ CFU MRSA via IV route. Three of the animals were monitored for 2 dpi and then euthanized; the remaining three animals were monitored up to 8 dpi and then euthanized. (A) Bacteremia was assessed via semi-quantitative bacteriology of whole blood. The upper limit of quantification (LOQ) for this bacteriological assay was 300 CFU/mL (black dotted line). Data are shown as the mean ± SD of two replicate TSA plates. Data are interleaved for clarity. (B) Concentrations of CRP in serum determined via CRP-ELISA. (C) Percent changes (%Δ) in body weight are shown calculated based on baseline body weight recorded at 0 hpi. (D) Changes (Δ) in body temperatures are shown calculated on baseline body temperatures recorded rectally at 0 hpi. (E) Mean arterial pressure (MAP) was estimated from systolic and diastolic blood pressure data recorded during physical exams. Gray areas indicate values outside the normal clinical ranges for cynomolgus macaques.

FIGURE 3

Changes in circulating white blood cell subtypes following IV MRSA challenge. Complete blood count (CBC) analyses were performed on an IDEXX analyzer for absolute (A) total white blood cells (leukocytes), (B) neutrophils, (C) lymphocytes, (D) monocytes, (E) eosinophils, and (F) platelets. (A–F) Statistical comparisons were performed using repeated-measures, mixed-effect analyses with Dunnett’s post hoc multiple comparisons test; *P < 0.05, **P < 0.01. Asterisks indicate p-values of multiple comparisons against the baseline “Pre” timepoint, and are shown below the x axis. Gray areas indicate values outside normal clinical ranges for cynomolgus macaques.

FIGURE 4

Host cytokine/chemokine immune response of cynomolgus macaques following IV infection with 10⁹ CFU MRSA. Concentrations of 37 cytokines, chemokines, and growth factors were characterized in cynomolgus macaque serum at baseline (Pre) through 8 dpi on the ProcartaPlex platform. (A) Heat map showing log-normalized change (Δ) in mean fluorescence intensity (MFI) of 21 analytes that were detected above the lower limit of detection in > 50% of animals. Serum concentrations of TNF-α (B), IL-6 (C), IFN-γ (D), G-CSF (E), IL-1RA (F), IP-10 (G), MCP-1 (H), MIG (I), and MIP-1α (J), are shown in pg/mL. (B–J) Statistical comparisons were performed using repeated-measures, mixed-effect analyses with Dunnett’s post hoc multiple comparisons test; *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001. Asterisks indicate p-values of multiple comparisons against the baseline “Pre” timepoint and are shown below the x axis for clarity.

FIGURE 5

Monocyte subpopulations in whole blood of cynomolgus macaques infected with 10⁹ CFU MRSA via IV route. Immunophenotyping performed by flow cytometry on whole blood collected throughout the study reveals unique behavior of circulating monocyte populations. (A) Representative gating strategy for monocyte subpopulations. (B) Total monocytes shown as frequency of CD45 + parent population. (C) Classical monocytes shown as frequency of monocyte population. (D) Intermediate monocytes shown as frequency of monocyte population. (E) Non-classical monocytes shown as frequency of monocyte population. (B–E) Statistical comparisons were performed using repeated-measures, mixed-effect analyses with Dunnett’s post hoc multiple comparisons test; *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

FIGURE 6

Histologic sections of MRSA-challenged cynomolgus macaque kidney harvested from DN508 and DN233 at 2 dpi. (A) H&E stain (200x) of multifocal tubular degeneration (arrows) with regeneration and inflammation. (B) H&E stain (400x) of tubular epithelial necrosis (asterisk) with eosinophilic granular or inflammatory cell casts (arrowhead). (C) H&E stain (400x) of peritubular or tubular neutrophilic inflammation (arrows).

FIGURE 7

Histologic sections of MRSA-challenged cynomolgus macaque heart tissue harvested at 2 dpi. (A,B) H& E stain of cardiac section from animal DN508 showing moderate neutrophilic inflammation (asterisk) with coalescing area of myocardial necrosis (arrows) at (A) 100x, and (B) 200x. (C,D) Successive cardiac sections (400x) harvested from animal DN508 showing (C) H&E strain showing foci of neutrophilic inflammation and Splendore-Hoeppli structures (arrows) and (D) hematoxylin/Gram stain with moderate numbers of gram-positive cocci (asterisk).

FIGURE 8

Histologic sections of MRSA-challenged cynomolgus macaque kidney, liver, and lung harvested at 8 dpi. (A) H&E stain (200 x) of kidney section from animal DL606 showing severe regional area of glomerular vascular thrombosis with renal parenchymal infarction/necrosis (asterisks), vascular wall necrosis, and fibrin deposition. (B) Hematoxylin/Gram stain (400x) of sequential kidney section from animal DL606 showing abundant neutrophilic inflammatory cells (arrowhead), and gram-positive cocci (arrows). (C) H&E stain (200x) of liver section from animal DN362 showing moderate neutrophilic or mixed cell inflammation with a primarily periportal distribution (asterisk), reactive hypertrophy and hyperplasia of sinusoidal Kupffer cells/macrophages, and expansion of the hepatic sinusoids with reactive appearing cells of histiocytic or endothelial origin (arrowheads). (D) H&E stain (400 x) of liver section from animal DN361 showing increased intravascular neutrophils and monocytic infiltrates with mild scattered single cell hepatocellular necrosis (arrow). (E–F) H& E stain of lung section from animal DN361 at 200x (E) and animal DL606 at 400x (F) showing minimal multifocal and regionally extensive areas of alveolar hemorrhage with alveolar septal necrosis (arrowheads) and numerous intravascular fibrin microthrombi (arrows). A mixed cell inflammation, consisting primarily of neutrophils and monocytes, expanded the alveolar septae; increased intravascular inflammatory cell infiltrate (asterisk) was observed.

FIGURE 9

Clinical chemistry markers of liver, kidney, and other organ function are disrupted throughout MRSA infection of cynomolgus macaques. Analyte concentrations and ratios were determined by serum biochemistry panel on IDEXX platform. (A) Concentration of aspartate transferase (AST) in units per liter. (B) Concentration of alanine transaminase (ALT) in units per liter. (C) Concentration of total bilirubin (TBIL) in mg/dL. (D) Ratio of albumin to globulin (A/G) concentrations. (E) Ratio of blood urea nitrogen (BUN) to creatinine (CRE) concentrations. (F) Concentration of creatine kinase (CK) in units per liter (A–F) Statistical comparisons were performed using repeated-measures, mixed-effect analyses with Dunnett’s post hoc multiple comparisons test; *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. Asterisks indicate p-values of multiple comparisons against the baseline “Pre” timepoint and are shown below the x axis for clarity. Gray areas indicate values outside the normal clinical ranges for cynomolgus macaques.