| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2004;78:46-52
© 2004 The Society of Thoracic Surgeons
a Cerebrovascular Research Center, Cleveland, Ohio, USA
b Neurological Surgery, Cleveland, Ohio, USA
c Thoracic and Cardiovascular Surgery, The Cleveland Clinic Foundation, Cleveland, Ohio, USA
Accepted for publication November 25, 2003.
* Address reprint requests to Dr Janigro, Cerebrovascular Research Center, Department of Neurological Surgery, NB20, The Cleveland Clinic Foundation, 9500 Euclid Ave, Cleveland, OH 44195, USA
e-mail: janigrd{at}ccf.org
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and methodsResultsCommentConclusionsAcknowledgmentsReferences |
|---|
METHODS: Systemic and pericardial cavity blood from 5 patients undergoing CABG were sampled before, during, and after cardiopulmonary bypass (CPB). A commercially available enzyme-linked immunosorbent assay (ELISA) kit was used to quantify S100ß. Two-dimensional gel electrophoresis, Western blot, and mass spectroscopy were also performed to identify S100â and other proteins.
RESULTS: Mean S100ß levels measured by ELISA, systemic and pericardial cavity blood were (in ng · mL–1) 1.0 ± 0.46 and 111 ± 71 before CPB, 0.6 ± 0.11 and 113 ± 54 during CPB, and 1.7 ± 0.64 and 101 ± 42 after CPB, respectively. However, gel electrophoresis and Western blot analysis revealed proteins other than S100ß to be present in the pericardial cavity giving a falsely elevated serum S100â levels measured by immunoassay. Mass spectroscopy of identified potential candidates revealed contaminants including haptoglobin I precursor, apolipoprotein A-1 precursor, complement factor B precursor, and complement C3 precursor.
CONCLUSIONS: S100ß immunoassays are not specific for S100â and give a falsely elevated reading due to contaminants from the surgical field that cross react with the assay's antibody. This does not appear to be an issue in nonsurgical patients. Caution must be exerted when evaluating immunodetection results for low-abundance proteins under conditions where contamination of the sample is likely.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentConclusionsAcknowledgmentsReferences |
|---|
Specificity of the cellular release site and timing of a marker's appearance are also important issues. To be clinically useful, serum levels of the putative marker must closely correlate with a well-defined pathologic condition and be released by cells that are associated with the pathologic sequelae. In the case of S100ß, conflicting results from cardiac surgery have suggested that either the site of release of these markers is not fully understood or that the tests used may have intrinsic specificity problems [5]. In fact, (1) levels of S100ß in serum are influenced by manipulations of peripheral hemodynamics by surgical procedures, such as in cardiac surgery, that do not necessarily affect the brain, and (2) the levels observed after cardiac surgery are comparable to those observed with serious brain damage [5, 10].
S100ß can also be used as a marker of BBB leakage [2, 6]. In the central nervous system (CNS), the function of S100ß and its cellular distribution are fairly well understood [11]. S100ß is located primarily in astrocytes and can be released to the perivascular space and extravasate immediately after BBB opening. Very low levels compared to the brain were found in non-CNS tissues [5, 11– 14]. Clinical studies have shown that steady state levels of S100ß in serum following maximal opening of the BBB reach the theoretical plateau of 0.34 ng · mL–1 [6]. Thus, serum values that exceed this threshold may imply an ongoing process of brain damage or accumulation of S100ß from nonbrain sources [6].
As mentioned above, the most exorbitant increases in S100ß blood levels have been observed during cardiopulmonary bypass (CPB) procedure [5, 10, 15], where S100ß has been used, although with very limited success, to monitor the possibility of ongoing brain injury. While the procedure itself does not commonly lead to permanent brain damage, S100ß values measured in the blood of CPB patients were comparable to the serum levels found after lethal head injury [16]. It was concluded that sources other than the CNS may contribute to these abnormally elevated values or that S100ß can be released from the brain at levels previously undocumented. Given that the S100ß serum levels obtained during CPB were significantly elevated compared to what was found experimentally and estimated theoretically, we tested the hypothesis that elevated S100ß levels during surgical procedures were due to inherent pitfalls in the testing strategy used. We investigated (1) the possibility that sources other than astrocytes can contribute to S100ß blood levels during CPB, and (2) the possibility that a contaminating agent(s) may give exaggerated S100ß serum values after CPB procedure.
|
Material and methods |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentConclusionsAcknowledgmentsReferences |
|---|
Anesthesia was induced with a combination of midazolam 3 to 5 mg (Abbott Laboratories, North Chicago, IL), fentanyl (Abbott Laboratories, North Chicago, IL) up to 0.25 mg, and thiopental 3 to 5 mg · kg–1 (Abbott Laboratories, North Chicago, IL), intravenously. It was subsequently maintained using inhaled isoflurane 1% to 2% (Abbott Laboratories, North Chicago, IL) supplemented with fentanyl up to a total of 1 mg. Muscle relaxation was achieved with pancuronium (GensiaSicor Pharmaceutical, Irvine, CA) up to a total of 10 mg. Cardiopulmonary bypass was instituted utilizing a conventional modular roller pump system (Sarns 8000 Terumo Cardiovascular, Ann Arbor, MI) and polyvinyl chloride tubing (Jostra, Anaheim, CA). The circuit also included a 25 µm arterial filter (Jostra, Anaheim, CA), adult membrane oxygenator (Jostra, Anaheim, CA), and an integrated, hardshell, venous-cardiotomy reservoir (Jostra, Anaheim, CA). The entire circuit was heparin coated. The circuit was primed with approximately 1,500 mL of Plasma-Lyte A (Baxter Healthcare Corp., Deerfield, IL), 250 mL 20% mannitol (Baxter Healthcare Corp., Deerfield, IL), 10,000 U porcine-intestinal heparin (Elkins-Sinn, Inc., Cherry Hill, NJ), and 15 meq sodium bicarbonate (Abbott Laboratories, North Chicago, IL). Operations were performed through a median sternotomy. Standard cannulation for CPB was performed with ascending aorta cannulation and dual-stage cannulation of the right atrium, after administration of 300 U · kg–1 of heparin. Cardiolegia cannulas were placed both antegrade and retrograde. Once on CPB, the arterial flow was adjusted to 2.0 to 2.4 L · min–1 · m–2, and blood pressure was maintained between 50 and 70 mm Hg, with administration of phenylephrine (Abbott Laboratories, North Chicago, IL) if required. The aorta was cross clamped and cold blood cardioplegia (Buckberg solution) was administered either antegrade or retrograde or in combination every fifteen minutes and was delivered at a ratio of four parts blood to one part crystalloid (4:1). Body temperature was either kept normothermic or allowed to drift to 34°C during CPB. After the bypass grafts were placed, patients were weaned from CPB and decannulated and heparin was reversed with protamine (American Pharmaceutical Partners, Schaumburg, IL). Blood samples, approximately 5 mL/draw, were collected before, during, and after CPB from the systemic circulation and pericardial cavity.
Immunoassay determination
The LIA-mat Sangtec 100 kit (Bromma, Sweden) was used to quantitate S100ß. The kit uses a monoclonal two-site immunoluminometric methodology (sandwich principle). The assay discriminates between the A1-subunit and B-subunit of the S100 protein through the use of three patented monoclonal antibodies (SMST 12, SMSK 25, and SMSK 28) and the sensitivity is 0.01 ng · mL–1.
Protein filtration
One milliliter of sample was pipetted into a centrifugal filter unit (50 kDa) and centrifuged at 7,500xg using a swinging bucket rotor (Beckman Instruments, Palo Alto, CA) for approximately 20 to 30 minutes. If necessary, centrifugation time was increased until equal volumes were achieved in both the upper and lower chambers of the centrifugal filter unit (0.5 mL each).
Gel electrophoresis
Sodium dodecylsulfate (SDS) and non-SDS polyacrylamide gel electrophoresis (PAGE) were performed by standard methods [17]. In non-SDS PAGE samples were analyzed in nondenaturing conditions. To address this, denaturating agents were excluded from the sample buffer and the running buffer and during the preparation of the gel. Fifty milliliters of sample buffer (5x) were obtained by mixing 15.5 mL of 1 mol/L tris-hydrochloric acid pH 6.8, 9.5 mL of water, 25 mL of glycerol, and a trace amount of bromophenol blue. An electrophoresis buffer (1x) was obtained by dissolving 3 g of tris-hydrochloric acid and 14.4 g of glycine in 1 L of deionized water and adjusting to a final pH of 8.3.
Western blot and two-dimensional gel electrophoresis
Identification of S100ß protein was performed by Western blotting techniques as described elsewhere [7]. Two-dimensional gel electrophoresis was performed according to the methods reported in our previous work [7]. Two different primary antibodies to S100ß were used: sheep antibovine (1:1,000, QED Bioscience Inc.) and mouse antibovine (1:1,000, Sangtec; Bromma, Sweden). After washes, the transfer membrane was incubated with secondary horseradish peroxidase: antisheep IgG (rabbit) (1:5,000, Calbiochem, LA Jolla, CA) or antimouse IgG (rabbit) (1:5,000, Dako Corporation, Carpinteria, CA) for 2 hrs. Specific blots were visualized by enhanced chemiluminescence reagent (ECL plus, Amersham Pharmacia Biotech, UK).
Mass-spectroscopy
We used a LC-MS system Finnigan LCQ-Deca ion trap mass spectrometer system with a Protana microelectrospray ion source interfaced to a self-packed 10 cm x 75 um id Phenomenex Jupiter C18 reversed-phase capillary chromatography column. Data were analyzed by using all CID spectra collected in the experiment to search the National Center of Biotechnology Information (NCBI) nonredundant database with the search program TurboSequest. All matching spectra were verified by manual interpretation. The interpretation process was also aided by using the programs Mascot and Fasta to perform additional searches, as needed.
Statistical methods
Data are presented as mean ± standard error of mean (SEM). Analysis of variance (ANOVA) was used to determine significance. Origin 7.0 (Microcal) was used for statistical analysis.
|
Results |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentConclusionsAcknowledgmentsReferences |
|---|
|
The low MW fraction displayed no significant changes in systemic blood, while the high MW fraction gave increasingly high values up to a final value of 1.7 ± 0.6 ng · mL–1 (Fig 1, panel C).Cavity samples produced a highly elevated signal, as measured by ELISA, during the time of the CPB in the more than 50 kDa fraction (113 ± 54 ng · mL–1) compared to the less than 50 kDa fraction (25 ± 6 ng · mL–1, Fig 1, panel D),suggesting that regardless the nature of the signal measured by ELISA, the cavity region produced the greatest quantity of immunosignal. Taken together, these results suggested that a high MW molecule, other than S100ß, may appear in the systemic circulation after recirculation of the cavity blood contents.
Probing further with the Western blot technique under nondenaturating conditions, a high MW band appeared in all the samples: the dashed arrow in Figure 2 refers to nonspecific binding found in all samples, while the continuous arrow shows an additional, well-defined band that was stronger in the cavity (indicated by C) samples throughout the procedure. To quantify the patterns of S100ß release under both conditions, we measured S100ß in the same samples by ELISA; the values obtained are shown at the bottom of Figure 2. Note that there was a consistent dissociation of cavity versus peripheral (indicated by P blood readings, the former giving much higher values. While elevated S100ß values measured by ELISA correlated with the presence of high MW bands, signals attributable to S100ß proper on the gel remained largely unchanged throughout the experiment (see low MW bands corresponding to the S100ß standard loaded for comparison). Significant increases in the amount of measurable S100ß proper were evident only in serum samples taken hours after the surgery (asterisks, Fig 2).
|
|
Mass spectroscopy identification of the high MW polypeptide fragments led to a short list of eight potential candidates but only four of these were repetitively detected in the cavity fluid (Table 1). These proteins represent the possible contaminants affecting the S100ß determination during CPB.
|
|
Comment |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentConclusionsAcknowledgmentsReferences |
|---|
First, we have shown that filtration of serum proteins from either peripheral blood or cavity fluids reveals a surprising amount of immunodetectable S100ß in the high MW fraction. This was unexpected because (1) S100ß is a low MW protein, and (2) the exceedingly high levels of S100ß measured during CPB procedures were previously attributed to release of S100ß by non-CNS sources (eg, bone marrow, fat tissue, etc) [5]. Our results do not completely rule out a contribution of extra-brain S100ß to serum levels, but suggest that this confounding factor plays, at best, a small role. Whatever the nature of this abnormal signal, our results clearly demonstrated that its origin is indeed extracranial, and that dilution in serum occurs during extracorporeal circulation. This has been directly demonstrated by others using the astute strategy of comparing serum S100ß levels under different hemodynamic and hemodilution conditions [5].
The second set of experiments demonstrated a further dichotomy between S100ß detection by MW-independent approaches and protein electrophoresis (Figs 2 and 3). Thus, we were able to demonstrate that potentially cross-reacting proteins were present in the cavity samples. These were clearly distinct from S100ß, as shown by direct comparisons with standards preloaded on the gel. In agreement with the experiments described above, these signals were clearly predominant in the high MW range. While the results discussed so far strongly argued in favor of a significant lack of specificity of the MW-independent approach commonly used for diagnostic purposes, it was still possible that linkage of S100ß to other proteins could have occurred.
The likelihood of protein-protein interaction is far from remote, in particular in body fluids. Perturbations of equilibria (pH, temperature, etc) have powerful disruptive effects on these delicate interactions. To rule out that this was not the cause underlying the observed differences between samples, we subjected bidimensional blots to proteomic analysis, an approach we previously used for similar purposes [7]. The data clearly demonstrated that the large immunosignals revealed by the anti-S100ß antibodies were not attributable to S100ß or other proteins of the S100 family. In fact, none of the sequences derived were matched to these peptides.
An important question is how our findings impact the interpretation of data obtained with immunodetection of S100ß. The following section tries to address this issue, keeping in mind that serum S100ß has been used to predict both blood-brain barrier leakage and(or) brain damage.
Significance of serum S100ß
A large corpus of clinical and experimental results have shown that serum S100ß can be a sensitive peripheral marker of brain damage; historically, most of the work was done in patients undergoing cardiothoracic surgeries [2, 18]. It was generally assumed that the specific brain protein S100ß is released systemically after cerebral lesion. More recently, this view was challenged and experiments have shown that cerebral lesions are not necessary to produce S100ß elevation in serum. Thus, Kapural and colleagues [2] have shown that S100ß levels can be indicators of BBB function even in the absence of neuroglial damage. It is thus not surprising that a quantitative relationship exists between serum S100ß levels and specific cerebrovascular or parenchymal pathologies [6, 19]. In severe head injury [16, 20, 21] and stroke patients [22], S100ß levels correlate with clinical outcome. For example, patients who died after head injury had significantly higher serum S100ß values compared with those who survived (2.7 µg · L–1 vs 0.5 µg · L–1) [16, 20, 21].
If the values observed in our study were the result of S100ß released in response to neuroglial dysfunction, then it would be difficult to accept serum values obtained after CPB procedures with nondetrimental neurologic outcome [5, 23]. Several explanations were offered to justify this discrepancy, including contamination by non-CNS sources during surgery [5]. Our results confirm the presence of contaminants, but these were non-S100ß proteins presumably released by extra-CNS tissue.
Despite these results, the notion that S100ß released by the brain into the peripheral circulation has diagnostic values remains valid. The presence of high MW cross reactants warrants the development of filtration steps before the immunodetection step. Alternatively, a different test (or antibodies) can be developed. The problems we unveiled, however, seem to be only present during surgical procedures. In fact, exorbitantly elevated S100ß values have not been described in "intact" patients, including those affected by either neurologic disorders or brain tumors [6, 7, 14, 19]. Our findings could also explain the discrepancy between neurologic patients affected by trauma or trauma plus various bodily injuries [24, 25]. It is in fact likely that, in the latter population, proteins other than S100ß were measured as a consequence of multiple traumatic injuries.
Significance and sources of contaminants
Our extensive search for non-S100ß proteins responsible for spurious results led to a short list of candidates that were repetitively detected in the cavity fluids. Haptoglobin combines with free plasma hemoglobin to prevent the loss of iron through the kidneys. Apolipoprotein A-1 (APOA-1) precursor is the source of APOA-1, major components of plasma high density lipoprotein (HDL), which participates in the reverse transport of cholesterol from tissue to the liver for excretion by promoting cholesterol efflux. Interestingly, one of the precursor proteins that our proteomic analysis unveiled (APOA-1) was shown to be expressed in human brain, where it may correlate with cognitive function by direct or indirect HDL mechanisms [26]. Complement factor B and complement factor C3 precursors are linked to proinflammatory changes and are thus expected to increase in the surgical field. Their predominantly peripheral origin rules out CNS sources.
While filtering cavity samples at 50 kDa promptly removed the offending protein from immunodetection, mass spectroscopy analysis revealed molecular weights close to S100ß. This apparent inconsistency can be easily explained by the fact that all these protein species readily assemble with abundant protein (eg, hemoglobin).
|
Conclusions |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentConclusionsAcknowledgmentsReferences |
|---|
|
Acknowledgments |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentConclusionsAcknowledgmentsReferences |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentConclusionsAcknowledgmentsReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  M. Carrier, A. Denault, J. Lavoie, and L. P. Perrault Randomized controlled trial of pericardial blood processing with a cell-saving device on neurologic markers in elderly patients undergoing coronary artery bypass graft surgery. Ann. Thorac. Surg., July 1, 2006; 82(1): 51 - 55. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Watson, S. M. Shirreffs, and R. J. Maughan Blood-brain barrier integrity may be threatened by exercise in a warm environment Am J Physiol Regulatory Integrative Comp Physiol, June 1, 2005; 288(6): R1689 - R1694. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. P. Grocott and G. B. Mackensen Apolipoprotein E Genotype and S100{beta} After Cardiac Surgery: Is Inflammation the Link? Anesth. Analg., June 1, 2005; 100(6): 1869 - 1870. [Full Text] [PDF]
|
|
|
|
 |
|
 H. P. Grocott S100{beta} and postcardiac surgery neurological dysfunction: reasons to disregard any link Can J Anesth, April 1, 2005; 52(4): 441 - 442. [Full Text] [PDF]
|
|
|
|
|
|
M. Kanbak, F. Saricaoglu, and T. Ocal REPLY Can J Anesth, April 1, 2005; 52(4): 442 - 443. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
