Letter to the Editor Lactate-Induced IL-8 Pathway in Endothelial Cells—Response Fr ed erique V egran, Emmanuel Seront, Pierre Sonveaux, and Olivier Feron In a recent publication (1), we have documented the capacity of lactate to enter endothelial cells via the monocarboxylate transporter MCT1 and to signal by stimulating an autocrine proangiogenic NF-kB/interleukin-8 (IL-8) pathway. In this mechanistic study, the reason why we used human umbilical vein endothelial cells (HUVEC) in our in vivo experiments with human tumor xenografts was that there is no clear-cut IL-8 homolog in mouse. We therefore preferred working with endothelial cells of human nature (i.e., HUVEC) that we injected together with human tumor cells in Matrigel plugs to study the lactate-driven interplay between these cells. We read with interest the letter of Pinheiro and colleagues about their failure to detect significant MCT1 expression in human tumor-associated endothelial cells. First, we would like to stress that in HUVECs and other tested endothelial cells, the metabolic use of lactate could only be marginally documented contrary to what we and others have reported in tumor cells (2, 3) and oxidative muscle fibres or brain cells (4). One central message of our article (1) was that lactate may however still represent a signaling molecule in these largely glycolytic cells and thus, this also infers that MCT1 does not need to be highly expressed in endothelial cells to support the observed proan- giogenic effects. A faint expression of MCT1 can notably be observed in some endothelial structures from the work of Pinheiro and colleagues. Second, a limited sensitivity of their antibody may also account for the lack of clear MCT1 expres- sion in endothelial cells of tumor blood vessels (vs. strong staining of clustered tumor cells). Furthermore, discrimination between the plasma membrane and cytosolic locations of MCT1 in sparse tumor-associated endothelial cells is compli- cated by the intracellular location of the epitope of the MCT1 antibody (Millipore) used by Pinheiro and colleagues. In our hands, this antibody works nicely for immunofluorescent studies on intact cells and xenografted human tumors (1, 3) and immunohistochemical staining of tumor cells expressing abundant amounts of the transporter. For paraffin-embedded tumor sections, we designed another antibody that we vali- dated against fibroblasts recombinantly expressing human MCT1, -2, -3, and -4. The MCT1 specificity was further Authors' Affiliation: Universit e catholique de Louvain, Pole of Pharma- cology & Therapeutics (UCL-FATH), Angiogenesis & Cancer Research Laboratory, Institut de Recherche Exp erimentale et Clinique (IREC), Brus- sels, Belgium Corresponding Author: Olivier Feron, UCL/SSS/IREC/FATH, 53 Avenue E. Mounier, B1.53.09, Brussels, B-1200, Belgium. Phone: þ32 2 764 5264; Fax: þ32 2 764 5269; E-mail: olivier.feron@uclouvain.be doi: 10.1158/0008-5472.CAN-12-0012 Ó2012 American Association for Cancer Research. Figure 1. Representative MCT1 immunostaining in human bladder cancer where significant vascular staining is observed using peroxidase (A and B) or DyLight 488– conjugated (green fluorescent; C, D) secondary antibodies. D, vascular costaining using CD31 antibodies and Alexa 568–conjugated (red fluorescent) secondary antibody. Arrowheads indicate MCT1 expression in red blood cells trapped inside blood vessels. A B C D Cancer Research www.aacrjournals.org 1903 Downloaded from http://aacrjournals.org/cancerres/article-pdf/72/7/1903/2677501/1903.pdf by guest on 29 October 2023