Should Soluble CD40 Ligand Be Measured From Serum or Plasma Samples?
To the Editor:
We were surprised to read the article by Blake et al1 on the relationship of soluble CD40 ligand (sCD40L) and lipid accumulation in carotid atheroma. We recently reported an association between inflammatory markers and early onset carotid atherosclerosis.2 Of the patients noted in that report, we had plasma samples from EDTA-anticoagulated blood, stored frozen for 1 to 3 years. Although the product information and manual of the Bender MedSystem’s ELISA kit clearly states that plasma preparations are not suitable for the essay,3 we decided to run a test on some of these plasma samples for sCD40L. We also used duplicate plasma samples, together with fresh serum and plasma of 5 healthy subjects obtained less than 4 hours before the ELISA test, in the same ELISA plate. We were not surprised but were disappointed, as no activity could be detected either in the frozen plasma samples of patients or in the fresh or recalcified plasma samples of the healthy subjects. The method and the plate worked properly, as sCD40L activity was measured with small variation between duplicates in all of the serum samples (7.0±1.5 ng/mL, mean±SD). We had to conclude and admit that, as stated in the product information, EDTA-anticoagulated plasma samples—either frozen or freshly collected—are not appropriate for the test.
After this failure, we came across the article of Blake et al,1 in which the authors referred to their previous report for the method of sCD40L determination.4 They used plasma samples from EDTA-anticoagulated blood, and the same Bender MedSystem’s ELISA kit for the measurements4 as we did. We checked the methods and found that they used 5-times diluted plasma samples, so we assumed that either the EDTA concentration or something else in our undiluted plasma samples might have caused our failure. Therefore, in a second attempt in 5 subjects, we measured sCD40L activities in parallel serum, plasma, and 5-times diluted plasma. In these 5 subjects, sCD40L activity could be measured in all samples from the serum (2.95±1.17 ng/mL), but no activity was found in the majority of plasma samples. Just to obtain values, we calculated mean and standard deviation, which were 0.36±0.51 ng/mL for undiluted plasma samples and 2.88±2.01 ng/mL for the 5-times diluted plasma samples.
Looking only at the group means, the results might seem as there is no difference between measurements from serum and diluted plasma. However, if we look at the individual data (Figure) it can be seen that although there is good agreement between the duplicates of serum samples (an average 5.9% difference from the mean), the variation is large between the diluted plasma samples of the same individuals, and in 3 of the 5 samples, zero activity was measured in one or both of the duplicates.
In agreement with recent reports5,6⇓ and the product information of the ELISA kit,3 our results also show that measuring sCD40L from EDTA-treated plasma samples is unreliable. Therefore, we also suggest measuring sCD40L activities from serum and not from plasma samples.
- ↵Blake GJ, Ostfeld RJ, Yucel EK, Varo N, Schonbeck U, Blake MA, Gerhard M, Ridker PM, Libby P, Lee RT. Soluble CD40 ligand levels indicate lipid accumulation in carotid atheroma: an in vivo study with high-resolution MRI. Arterioscler Thromb Vasc Biol. 2003; 23: e11–e14.
- ↵Magyar MT, Szikszai Z, Balla J, Valikovics A, Kappelmayer J, Imre S, Balla G, Jeney V, Csiba L, Bereczki D. Early-onset carotid atherosclerosis is associated with increased intima-media thickness and elevated serum levels of inflammatory markers. Stroke. 2003; 34: 58-63.
- ↵Bender MedSystems. Human sCD40L Instant ELISA. Product information and manual. Vienna, Austria: MedSystems Diagnostics GmbH; 2001: 23.
- ↵Schonbeck U, Varo N, Libby P, Buring J, Ridker PM. Soluble CD40L and cardiovascular risk in women. Circulation. 2001; 104: 2266–2268.
- ↵Semb AG, van Wissen S, Ueland T, Smilde T, Waehre T, Tripp MD, Froland SS, Kastelein JJ, Gullestad L, Pedersen TR, Aukrust P P, Stalenhoef AF. Raised serum levels of soluble CD40 ligand in patients with familial hypercholesterolemia: downregulatory effect of statin therapy. J Am Coll Cardiol. 2003; 41: 275–279.
- ↵Aukrust P, Muller F, Ueland T, Berget T, Aaser E, Brunsvig A, Solum NO, Forfang K, Froland SS, Gullestad L. Enhanced levels of soluble and membrane-bound CD40 ligand in patients with unstable angina: possible reflection of T lymphocyte and platelet involvement in the pathogenesis of acute coronary syndromes. Circulation. 1999; 100: 614–620.
We share the concern of Dr Bereczki and colleagues regarding the importance of using appropriate specimens in the currently available sCD40L ELISAs (Bender MedSystem and R&D) and have referred to this important issue in a recent publication (Schonbeck et al1).
Several issues are pertinent to the concerns of Dr Bereczki et al:
1. First and foremost, it is of utmost importance to avoid activation of platelets during the post-harvesting sample procedure, because sCD40L can be released within minutes in high concentrations from this cell type. Indeed, platelets are commonly considered the predominant source of sCD40L in the bloodstream. Thus, the sCD40L concentrations obtained in serum, as generated by clotting of whole blood, should be interpreted with care. Indeed, in a previous report, Garlichs et al2 used serum to measure sCD40L levels and reported an average of 25 ng/mL sCD40L, even in healthy individuals (typically thought to have sCD40 levels <2 ng/mL). Such concentrations might be considered the “absolute blood level of sCD40L,” combining freely circulating sCD40L and CD40L expressed in circulating cells, namely platelets, T cells, and mononuclear phagocytes. Therefore, the authors (and other groups with whom the authors have communicated in recent years) consider the use of serum specimens to determine circulating sCD40L levels to be questionable.
2. In light of the above, it is commonly considered crucial to remove platelets from blood preparations without activation if one desires to determine the levels of freely circulating sCD40L. Indeed, the authors have performed extensive studies comparing levels of sCD40L in serum, platelet-rich plasma, platelet-poor plasma and plasma centrifuged at 2000g (the technique the authors used). Platelet counts and sCD40L levels detected in these healthy subjects are shown in the Table, demonstrating that, in our hands, removal of platelets is crucial for appropriate analysis.
3. The authors have informed Bender MedSystems about these circumstances, suggesting respective modifications in their protocol.
4. Using plasma, the authors have furthermore compared Heparin, Citrate, and EDTA as anticoagulants. Although Citrate provided the highest reproducibility, EDTA also yielded reliable and reproducible read-outs (variation coefficient<10%). These data were verified in “spiking experiments,” in which recombinant sCD40L was added to plasma samples of healthy or atherosclerotic subjects and recovery rates determined.
5. Finally, the data provided by Bereczki et al appear somewhat unclear to the authors. Their Figure indicates that serum and diluted plasma samples (stated to have been prepared in a fashion similar to that used by the authors) indeed provide comparable data (with cases 1, 3, and 5, demonstrating rather marginal standard variations). Also, it remains to be determined why the high standard deviations were observed in only 2 of 5 samples; if EDTA comprises a more generic problem for the ELISA, values would be expected “all over the place.” Unfortunately however, Bereczki et al do not provide any information regarding the platelet contamination in their preparations, information that might be crucial to understand their findings better.
The authors hope that the above discussion reveals the rationale of avoiding the use serum in our studies on sCD40L and, furthermore, affirms the reliability of the data provided in our original manuscript.
- ↵Schönbeck U, Gerdes N, Varo N, Reynolds RS, Horton DB, Bavendiek U, Robbie L, Ganz P, Kinlay S, Libby P. Oxidized low-density lipoprotein augments and 3-hydroxy-3-methylglutaryl coenzyme A reductase inhibitors limit CD40 and CD40L expression in human vascular cells. Circulation. 2002; 106: 2888–2893.
- ↵Garlichs CD, John S, Schmeißer A, Eskafi S, Stumpf C, Karl M, Goppelt-Struebe M, Schmieder R, Daniel WG. Upregulation of CD40 and CD40 ligand (CD154) in patients with moderate hypercholesterolemia. Circulation. 2001; 104: 2395.