Spurious tumor necrosis factor-[alpha] and interleukin-6 production by human monocytes from blood collected in endotoxin-contaminated vacutainer blood collection tubes.
To the Editor:
Cell-mediated immunologic responses to various inflammatory and infectious disease processes have provided important insights into the biology of the immune system. As a measure of immune function, various cytokines are currently being measured in serum/plasma samples or in tissue culture supernatants after antigen/mitogen stimulation (1-3). When testing for proinflammatory cytokines such as tumor necrosis factor-[alpha] (TNF-[alpha]) and interleukin-6 (IL-6), it is essential that all reagents and supplies are free from contamination with endotoxin and other bacteria-derived components with cell-stimulatory activity. For example, certain commercial preparations of fetal bovine serum and heat shock protein have been shown to contain endotoxin at concentrations that induce TNF[alpha] production by monocytes/ macrophages (4, 5).
[FIGURE 1 OMITTED]
Vacutainer[R] blood collection tubes (Becton Dickinson) are routinely used by clinical laboratories to collect anticoagulated whole-blood samples for functional immunologic testing. During a study examining mitogen-induced proinflammatory cytokine production, we discovered that 5-mL glass sodium heparin Vacutainer Tubes (lot no. 3224585) were mitogenic and capable of inducing TNF[alpha] and IL-6 production by monocytes. Blood from heparin-containing Vacutainer Tubes was stimulated in culture with lipopolysaccharide (LPS), and the production of TNF[alpha] and IL-6 was determined by intracellular cytokine staining and flow cytometry. Because of the logistics of the study, blood was left in Vacutainer Tubes at room temperature for 6-12 h before being cultured. Much to our surprise, we found that 24% of monocytes (CD14+ cells) were positive for TNF[alpha] and/or IL-6 when whole blood was cultured without LPS (Fig. 1A, left panel). Cytokine expression by monocytes increased further, to 38%, when cultures were supplemented with LPS (Fig. 1A, right panel). When whole blood from the same donor was collected in another lot (lot no. 4029101) of 5-mL glass sodium heparin Vacutainer Tubes, very few monocytes were positive for TNF[alpha] and/or IL-6 when cultured without LPS (Fig. 1B, left panel). In contrast, almost one-half of the monocytes were positive when cultured with LPS (Fig. 1B, right panel). These findings indicated that the blood collection tubes, and not one of the culture reagents, was responsible for the proinflammatory cytokine expression by monocytes.
Whole blood collected in 5-mL Vacutainer Tubes (lot no. 3224585) was also cultured for 20 h with and without LPS, and cell-free tissue culture supernatants were tested for TNF[alpha] (R&D Systems) and IL-6 (BioSource International) by immunoassay. Mean TNF[alpha] and IL-6 values in supernatants from unstimulated cultures (three donors) were 1354 and 17 952 ng/L, respectively, which was similar to mean concentrations when stimulated with LPS (TNF[alpha], 1225 ng/L; IL-6, 16192 ng/L). In contrast, measured TNF[alpha] and IL-6 concentrations were <5 ng/L in supernatants from unstimulated cultures from blood collected in 10-mL heparin tubes. To determine whether the contaminant was endotoxin, we filled Vacutainer Tubes with saline and tested them for endotoxin by the Limulus Amebocyte Lysate test (Bio-Whittaker, Inc.). The 5-mL heparin Vacutainer Tubes (lot no. 3224585) were positive for endotoxin (10 ng/L), whereas another lot of 5-mL tubes (lot no. 4029101), 4-mL plastic heparin tubes, and 10-mL glass heparin tubes were negative for endotoxin (<5 ng/ L).
The manufacturer of the Vacutainer Tubes was notified regarding these findings and indicated that we were the only user experiencing problems with this particular lot of 5-mL heparin tubes. It is possible that the mitogenic effect of these contaminated blood collection tubes might not be obvious if only plasma cytokines are measured in samples that are rapidly processed after collection because endotoxin-induced production of TNF[alpha] and IL-6 by monocytes is time-dependent (6, 7). Nevertheless, we recommend that an adequate supply of blood collection tubes from a single lot should be obtained before starting studies that measure cytokine production. We further suggest that blood collection tubes should be screened to rule out potential contaminants that can adversely influence cytokine results.
Contaminated tubes can easily be identified by storing unprocessed blood samples overnight at room temperature before measuring TNF[alpha] or IL-6 by immunoassay. If the tubes contain significant amounts of endotoxin, then monocytes will be stimulated to produce proinflammatory cytokines, and the plasma or serum will contain high cytokine concentrations. As an example, we tested plasma from the contaminated 5-mL heparin tubes (lot no. 3224585) that were stored unprocessed overnight and found TNF[alpha] and IL-6 concentrations of 242 and 4398 ng/L, respectively. In contrast, another lot (lot no. 4029101) of 5-mL heparin tubes and several different-sized heparin and EDTA tubes had undetectable plasma concentrations of TNF[alpha] and IL-6 (<5 ng/L) when stored unprocessed overnight, indicating that endotoxin contamination was limited to one lot and size of Vacutainer heparin blood-collection tubes.
(1.) Wallis RS, Lederman HM, Spritzler J, Devers JL, Georges D, Weinberg A, et al. Measurement of induced cytokines in AIDS clinical trials using whole blood: A preliminary report. Clin Diagn Lab Immunol 1998;5:556-60.
(2.) Aziz N, Nishanian P, Taylor JMG, Mitsuyasu RT, Jacobson JM, Dezube BJ, et al. Stability of plasma levels of cytokines and soluble activation markers in patients with human immunodeficiency virus infection. J Infect Dis 1999;179: 843-8.
(3.) Redwine L, Dang J, Hall M, Irwin M. Disordered sleep, nocturnal cytokines, and immunity in alcoholics. Psychosom Med 2003;65:75-85.
(4.) Kirikae T, Tamura H, Hashizume M, Kirikae F, Uemura Y, Tanaka S, et al. Endotoxin contamination in fetal bovine serum and its influence on tumor necrosis factor production by macrophage-like cells J774.1 cultured in the presence of the serum. Int J Immunopharmacol 1997;19: 255-62.
(5.) Gao B, Tsan M-F. Endotoxin contamination in recombinant human heat shock protein 70 (Hsp70) preparation is responsible for the induction of tumor necrosis factor a release by murine macrophages. J Biol Chem 2003;278:174-9.
(6.) Mathiak G, Kabir K, Grass G, Keller H, Steinringer E, Minor T, et al. Lipopolysaccharides from different bacterial sources elicit disparate cytokine responses in whole blood. Int J Mol Med 2003;11:41-4.
(7.) Leroux-Roels G, Offner F, Philippe J, Vermeulen A. Influence of blood-collecting systems on concentrations of tumor necrosis factor in serum and plasma [Letter]. Clin Chem 1988;34: 2373-4.
Najib Aziz 
Michael R. Irwin 
Sally S. Dickerson 
Anthony W. Butch  *
 Department of Pathology and Laboratory Medicine David Geffen School of Medicine at UCLA  Cousins Center for Psychoneuroimmunology UCLA Neuropsychiatric Institute and  Department of Psychology UCLA Los Angeles, CA 90095-1732
* Author for correspondence. Fax 310-794-4864; e-mail firstname.lastname@example.org.