Bicarbonate correction of ketoacidosis alters host-pathogen interactions and alleviates mucormycosis
Teclegiorgis Gebremariam, … , Scott G. Filler, Ashraf S. Ibrahim
Teclegiorgis Gebremariam, … , Scott G. Filler, Ashraf S. Ibrahim
Published May 9, 2016
Citation Information: J Clin Invest. 2016;126(6):2280-2294. https://doi.org/10.1172/JCI82744.
View: Text | PDF
Research Article Infectious disease

Bicarbonate correction of ketoacidosis alters host-pathogen interactions and alleviates mucormycosis

Abstract

Patients with diabetic ketoacidosis (DKA) are uniquely predisposed to mucormycosis, an angioinvasive fungal infection with high mortality. Previously, we demonstrated that Rhizopus invades the endothelium via binding of fungal CotH proteins to the host receptor GRP78. Here, we report that surface expression of GRP78 is increased in endothelial cells exposed to physiological concentrations of β-hydroxy butyrate (BHB), glucose, and iron that are similar to those found in DKA patients. Additionally, expression of R. oryzae CotH was increased within hours of incubation with DKA-associated concentrations of BHB, glucose, and iron, augmenting the ability of R. oryzae to invade and subsequently damage endothelial cells in vitro. BHB exposure also increased fungal growth and attenuated R. oryzae neutrophil-mediated damage. Further, mice given BHB developed clinical acidosis and became extremely susceptible to mucormycosis, but not aspergillosis, while sodium bicarbonate reversed this susceptibility. BHB-related acidosis exerted a direct effect on both GRP78 and CotH expression, an effect not seen with lactic acidosis. However, BHB also indirectly compromised the ability of transferrin to chelate iron, as iron chelation combined with sodium bicarbonate completely protected endothelial cells from Rhizopus-mediated invasion and damage. Our results dissect the pathogenesis of mucormycosis during ketoacidosis and reinforce the importance of careful metabolic control of the acidosis to prevent and manage this infection.

Authors

Teclegiorgis Gebremariam, Lin Lin, Mingfu Liu, Dimitrios P. Kontoyiannis, Samuel French, John E. Edwards Jr., Scott G. Filler, Ashraf S. Ibrahim

×

Figure 7

BHB administration in mice lowers blood pH, increases GRP78 expression in target organs, and specifically enhances susceptibility to mucormycosis.

Options: View larger image (or click on image) Download as PowerPoint
BHB administration in mice lowers blood pH, increases GRP78 expression i...
(A–C) Normal mice were injected 5 times with 5 mg BHB (n = 4 mice per group) before determination of the blood pH (A), the concentration of BHB in the serum (B), and the expression of GRP78 in target organs (C). *P < 0.05 vs. normal mice. (D and E) BHB-treated or normal mice (n = 10 for BHB-treated and 8 for normal mice from 2 experiments) were infected intratracheally with R. oryzae spores, and survival (D) and tissue fungal burden (n = 5 for normal mice and 7 for BHB-treated mice) in lungs or brain after 48 hours of infection (E) were determined as primary and secondary end points, respectively. Delivered R. oryzae inocula were 8 × 103 and 1 × 103 spores for the survival and tissue fungal burden experiments, respectively. (F) Expression of R. oryzae CotH expression in mouse organs was also determined using the same organs processed for tissue fungal burden. ‡P < 0.05 vs. the corresponding organ from normal mice. BHB-treated or neutropenic mice (n = 10 per arm) were also infected with A. fumigatus via inhalation (delivered inoculum 1.1 × 103 conidia), and survival of mice was determined (D). Uninfected BHB-treated (n = 8) or uninfected neutropenic mice (n = 5) were included in the survival studies as control arms (D). **P < 0.001 for: normal R. oryzae–infected mice vs. R. oryzae–infected BHB-treated mice; uninfected BHB-treated mice vs. R. oryzae–infected BHB-treated mice; A. fumigatus–infected BHB-treated mice vs. R. oryzae–infected BHB-treated mice; or A. fumigatus–infected BHB-treated mice vs. A. fumigatus–infected neutropenic mice. Data in A–C, E, and F are expressed as median ± interquartile range. y axis value in E represents the lower limit of detection of the assay.