| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Abstracts of Oak Ridge Posters |
(1 James Graham Brown Cancer Center and 2 Department of Pathology and Laboratory Medicine, University of Louisville, Louisville, KY;
aaddress correspondence to this author at: Department of Pathology and Laboratory Medicine, University of Louisville, Louisville, KY 40292; fax 502-852-1771, e-mail jmiller{at}louisville.edu)
Human plasma is a complex fluid that contains more than 3000 individual proteins and peptides in quantities ranging from nanograms to tens of grams per liter (1)(2). The plasma proteome and peptidome(3) hold great promise for disease diagnosis and therapeutic monitoring(4)(5). In recent years proteomics has focused primarily on analysis of low-abundance proteins by use of high-resolution methods such as 2-dimensional electrophoresis(6) and mass spectrometry(7). These methods have identified changes in the composition of low-abundance proteins and peptides in plasma that correlate with particular diseases. Typically no single protein emerges from such analyses as a wholly reliable biomarker. Instead changes in the pattern of proteins and peptides often serve as the best diagnostic indicator for a particular disease. In addition, many components of the peptidome were found to be complexed with more abundant serum proteins, especially human serum albumin and immunoglobulins. These findings led to the concept of the "interactome"(8).
Currently, the routine clinical laboratory procedures that are used to assess major components of the proteome include protein electrophoresis (PE) and immunofixation electrophoresis. Modern differential scanning calorimetry (DSC) provides a direct means for detecting what is perhaps the most fundamental property of biochemical reactions, heat changes, and can reliably measure heat changes of 0.1 µcal. In a typical DSC experiment, an aqueous solution of protein is heated at a constant rate in the sample cell alongside an identical reference cell that contains only the solvent (buffer). Thermal balance between the sample and reference cells is maintained by electrically powered feedback heaters. Any chemical process in the sample cell that absorbs or releases heat results in a thermal imbalance. The power applied by the feedback heaters provides a direct measure of heat capacity changes accompanying thermally induced denaturation and dissociation reactions and is recorded as a thermogram in the form of excess specific heat capacity vs temperature. Here we present preliminary data demonstrating that DSC provides a new method for studying the plasma/serum proteome and generates unique signatures for samples from healthy and diseased individuals.
Plasma samples from patients with cervical cancer and moderate cervical dysplasia [cervical intraepithelial neoplasia (CIN) category II] were obtained from the Gynecological Cancer Repository of the James Graham Brown Cancer Center. Normal plasma samples were purchased from Innovative Research, and plasma samples from patients with other diseases were purchased from BBI Diagnostics. Samples were stored at –80 °C before analysis. Pure human plasma proteins were purchased from Sigma-Aldrich and Calbiochem. Samples (100 µL) were dialyzed for 24 h at 4 °C against 10 mmol/L potassium phosphate, 150 mmol/L NaCl, and 3.8 g/L sodium citrate, pH 7.5, to ensure complete solvent exchange. Samples were then filtered through 0.45-µm cellulose acetate filters, diluted 25-fold with the dialysis buffer, and run (1–2 h) in a MicroCal capillary DSC to obtain a thermogram. The total protein (TP) concentration of the processed samples was measured by the bicinchoninic acid method (Pierce). Thermograms are plotted as excess specific heat capacity (cal/°C · g) vs temperature. We also performed PE on the samples (Helena Laboratories). In this study we used only frozen and thawed plasma samples, but we found no significant differences between fresh and frozen plasma or between plasma and serum except for the lack of fibrinogen in the latter (see the Data Supplement that accompanies the online version of this article at http://www.clinchem.org/content/vol53/issue11). We used Origin 7.5 software (OriginLab) to calculate 2 thermogram variables, the area (A) and the 1st moment (M1). M1 is approximately the weighted center of the thermogram and is calculated as 
Tf(T)dT.
The thermogram of a sample from a patient with stage IVB cervical cancer is shown in Fig. 1
along with the mean of 15 normal samples. The shaded region around the normal thermogram indicates the SD at each temperature. The 15 normal samples had a mean A of 5.04 cal/g (SD 0.25, CV 5.0%) and a mean M1 of 67.8 °C (SD 0.59, CV 0.86%). In addition, 24 h after obtaining the thermograms shown, we reanalyzed 1 of the normal samples and the cervical cancer patient sample shown in Fig. 1
(samples were stored at 5 °C before the 2nd analysis). The A in the 2 samples increased 7.8% and 2.3%, respectively, which is comparable to the SD obtained for the mean of 15 normal samples, and the M1 increased slightly, by 0.36% and 0.33%, respectively. The patient sample thermogram was markedly different from that of the normal sample. The area of the patient sample (4.91 cal/g) was within the reference interval, but the M1 (69.7 °C) was abnormally high. From thermograms of pure proteins (see the online Data Supplement) we have begun to identify some peaks. The major peak in the normal thermogram at 63 °C is apparently due to albumin, and the small peak at 51 °C is due to fibrinogen. In the cervical cancer patient thermogram, the signal from albumin appears to be greatly decreased or shifted to higher temperatures. The albumin concentration of this sample was within the reference interval, and the PE pattern was remarkable only for anincreased gamma globulin fraction (see the online Data Supplement). Therefore, the albumin peak appears to have been shifted, perhaps because of stabilization by bound ligands. The peak at 61 °C revealed by the shifted albumin appears to be haptoglobin.
|
We ran thermograms (data not shown) on additional plasma samples from patients with CIN II and early stages of cervical cancer. Samples from patients with CIN II (n = 4) had slightly abnormal thermograms. Samples from patients with early stage cervical cancer (n = 3) had moderately abnormal thermograms with a slight degree of albumin shifting. Compared with thermograms of samples from healthy individuals, thermograms of samples from diseased patients showed distinctive shifts as the disease progressed from CIN II, through early stage cervical cancer to the critically ill stage IVB cervical cancer (Fig. 1
). Compared with the DSC thermograms, PE patterns showed only subtle changes throughout the progression of the cancer.
We also ran DSC on 1 plasma sample each from patients with Lyme disease (LD), rheumatoid arthritis (RA), and systemic lupus (SL). The thermogram of the LD sample was markedly abnormal, resembling the cervical cancer IVB pattern, but with a smaller haptoglobin peak. The thermogram of the RA sample was similar to the thermograms of the CIN II sample, but slightly more abnormal. The SL thermogram was also similar to the cervical cancer IVB thermogram, but with an increased haptoglobin peak. PE patterns for these samples were generally unremarkable. PE results showed that the LD sample had a low TP and slightly low gamma globulin fraction, the RA sample was completely normal, and the SL sample had a low TP and gamma globulin fractions. Also, the PE pattern of the SL sample had a prominent peak on the cathodic side of the 
2-globulin fraction where haptoglobin migrates. This result in the PE was consistent with the increased haptoglobin peak in the thermogram, but insufficient to cause the 2-globulin fraction to be above the reference range.
The thermogram provides a specific signature for each sample, reflecting the protein composition of the sample, and thus offers a unique window and physical basis with which to view the plasma proteome. Every protein has, under a given set of buffer conditions, a characteristic denaturation thermogram that is unique and provides a fundamental thermodynamic signature for that protein. A primary DSC thermogram is an extensive property of a protein solution and as such is directly proportional to the mass of the protein in solution. The thermogram of a protein mixture will be the sum of all of the individual protein thermograms, weighted according to the mass of each component. Moreover, DSC is exquisitely sensitive to binding interactions. When a ligand binds to a protein, it stabilizes that protein, raising its melting temperature. Proponents of the interactome concept argue that in disease states, low molecular weight proteins or peptides unique to that disease increase in concentration in plasma, forming complexes with the more abundant proteins, particularly albumin and the immunoglobulins (8). Such interactions would alter the denaturation thermograms of those proteins, producing characteristic changes in observed thermograms relative to the normal signature (Fig. 1
). Changes in DSC thermograms are likely to be far more dramatic than are changes in concentration that are observed by chemical and immunochemical methods and PE.
We conclude that DSC of plasma samples may become a useful diagnostic procedure. Ongoing studies will explore the range of thermogram patterns in a wide variety of diseases. In addition, a greater number of samples from disease-free individuals will be examined to more rigorously define a normal thermogram. The effects of specific interactions with small protein or peptide components and other ligands (e.g., bilirubin) that may be present in plasma/serum in the disease state, as well as protein modifications (e.g., glycation), will also be examined. We will correlate the results from our calorimetric assay with clinical diagnoses as well as the results from PE and immunochemical assays for specific plasma proteins. When discrepancies are found, diseased tissues will be examined histologically to confirm or refine the clinical diagnoses.
Acknowledgments
Grant/funding support: This work was supported by National Cancer Institute Grant R44 CA103437 (to J.B.C.).
Financial disclosures: None declared.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
