Patent Application
20130177927
1. Field of the Invention
The present invention relates to proteins which are markers for esophageal cancer.
Esophageal cancer is a deadly disease, and one of the ten most common cancers worldwide. For reasons that are not clear, there has been a rapid increase in the incidence of esophageal adenocarcinoma. The occurrence of a change in the esophagus, referred to as Barrett's mucosa, in which a premalignant metaplastic epithelium replaces the normal squamous epithelium following gastroesophageal reflux, is the major risk factor for this deadly cancer. Reducing the high mortality associated with this cancer will require simplified means of diagnosing Barrett's mucosa and identifying, at an early stage, changes in Barrett's mucosa that indicate progression to cancer.
2. Description of Related Art
At the present time, esophageal cancer is diagnosed primarily by biopsy. Unfortunately, by the time the cancer is diagnosed it is often far advanced. Survival after diagnosis is poor.
Thus, a need exists for the diagnosis of esophageal cancer at an early stage, or even a pre-malignant stage. Markers which correspond to the advance of the illness may be used to monitor therapeutic regimens.
By comparison of 2-D gels showing proteins from normal esophagus, normal gastric mucosa, Barrett's mucosa, and esophageal cancer, a set of proteins have been identified in the different source tissues. These proteins provide information on the pathogenesis of esophageal cancer, and may have utility as markers to monitor therapeutic regimens.
The proteins can also be purified and used as immunogens to generate antibodies which can be used as diagnostic reagents. Antibodies which specifically bind the proteins can be used in an immunoassay of body fluids to detect the proteins, thereby providing diagnostic information and/or information to monitor therapeutic regimens. The body fluids used in the immunoassay method may be samples of blood or serum, urine, saliva, tears, or other body fluids.
In addition, the proteins and products derived therefrom have therapeutic applications.
Tissue from 27 patients was obtained for 2-D analysis from tumor tissue and normal esophagus. For patients in whom the tumor occurred in the context of Barrett's mucosa, Barrett's tissue was also obtained. (An interesting pathological feature of Barrett's mucosa is that it resembles gastrointestinal type of tissue rather than esophageal tissue, where it is found. The lower end of the esophagus resembles gastric mucosa rather than true esophageal tissue so that tumors that arise in that location need to be compared to normal esophagus as well as gastric tissue as control.) For tumors that occurred at the lower end of the esophagus, normal gastric mucosa was also obtained. All specimens were obtained at the time of initial surgery. Therefore, for every patient two control tissues were obtained, in addition to the cancer: normal esophagus and Barrett's for half of the patients, and normal esophagus and stomach for the other half.
Carrier ampholyte-based 2-D gels that cover the pH range of approximately 3.5-10.0 were prepared for all specimens.
Tissue was solubilized by addition of lysis buffer consisting of (per liter) 8 M urea, 20 ml of Nonidet P-40 surfactant, 20 ml of ampholytes (pH 3.5-10), 20 ml of 2-mercaptoethanol, and 0.2 mM of phenylmethylsulfonyl fluoride in distilled deionized water. Approximately 30 μl aliquots containing 70 μg of protein were loaded on individual gels.
Because isoelectric focusing is sensitive to charge modification, it is important to minimize protein alternations (e.g., proteolysis, deamidation of glutamine and asparagine, oxidation of cystine to cystic acid, carbamylation) that can result from improper sample preparation. Once solubilized, samples may be stored frozen at −80° C. for short periods (<1 month) without significant protein modification).
2-D PAGE was done as previously described (Strahler et al, Journal of Clinical Investigation, 85:200-207, 1990). In most cases aliquots were immediately applied onto isofocusing gels. First-dimension gels contained 50 ml of ampholytes per liter (pH 3.5-10). Isofocusing was done at 1,200 V for 16 h and 1,500 V for the last 2 h. 20 gels were run simultaneously. For the second-dimension separation, an acrylamide gradient of 11.4-14.0 g/dl was used. Protein spots in gels were visualized by the silver-staining technique of Merril et al. (Merril et al, Science, 211:1437-1438, 1981).
In addition to generating 2-D patterns that were carrier ampholyte-based, a second set of patterns using immobilized pH gradients were generated as a complementary set.
Samples were prepared as for the CA gels discussed in Example 1. For first dimension separation an immobilized pH gradient covering the separation range of pH 4-10. The second dimension is the same as for the CA gels of Example 1.
IPG gels are prepared using derivatives of acrylamide having carboxyl or tertiary amino groups with specific pK values. A linear pH gradient is prepared from a dense, acidic solution and a light, basic solution using a two-chamber microgradient former. The pH gradient is stabilized during polymerization of the Immobiline-acryl-amide-bisacrylamide matrix by a co-linear gradient of glycerol. Formulations of buffering Immobiline mixtures with titrating Immobiline for the pH limit solutions for narrow pH gradients (1 pH unit) or for broad pH gradients (>1 pH unit, up to 6 pH units) (Gianazza et al, Electrophoresis 6:113 (1985) and LKB application Note 324 (1984)) have been published.
The second dimension separates proteins on the basis of molecular weight in an SDS gel. An 11.5 to 14% T (2.6% cross-linking) acrylamide gradient provides effective separation of proteins of mass from 15,000 to 100,000. Proteins outside this range are less well resolved. Proteins with molecular weight less than 10,000 Da electrophorese close to the dye front and are not resolved.
Each gel was scanned in a 1024×1024 pixel format, where each pixel can have one of 256 possible values representing different degrees of intensity. Spot lists for study images are matched to spot lists of master images so that the result is a hierarchy of matched protein spots. The purpose of the matching is to link the same polypeptide spot through the hierarchy to allow assessment of its present, quantitative variation and specificity, as described in Strahler et al, 1990 (J. Clin. Invest. Vol. 85, pp 200-207). For comparison, between gels, of the amount of individual proteins, an adjustment process is utilized. The integrated intensity of detected polypeptides, measured in units of optical density per square millimeter, is adjusted relative to the intensity of reference polypeptides that are ubiquitously expressed. The adjustment is made to compensate for any variation between gels due to protein loading or staining.
Proteins that are Overexpressed in Barrett's Dysplasia Compared to Normal Tissue.
The number of spot differences between Barrett's and normal esophagus was enormous. It was very clear that Barrett's samples were much more similar to gastric mucosa samples, so spots that were larger in Barrett's compared to Gastric mucosa were identified.
Some of the spots present in Barrett's and not in gastric tissue were present in normal esophagus, and these are probably less interesting. They appear at the bottom beginning with spot 6. It should be noted that the spots that are bigger in Barrett's are also big in esophageal adenocarcinoma, since the two sample types represent essentially the same tissue but differ with respect to malignancy status.
Spot 102 is bigger in Barrett's than in normal tissues analyzed.
The large spots 109, 110, 111 are nearly always present in larger amounts in Barrett's than in either gastric mucosa or normal esophagus. They are also large in esophageal adenocarcinoma.
Spot 33 is always a pretty big spot in Barrett's, and occasionally is about the same in gastric mucosa, but often it is smaller. It would not deserve mention, except that it appears to be larger in esophageal adenocarcinoma.
Spot 61 is nearly always bigger in Barrett's. A spot just to the right is usually larger in gastric mucosa.
Spot 112 is larger in most Barrett's
Spot 113 is larger in Barrett's, but also present as an obvious spot in gastric mucosa. (the spot just to the right of 113 is large in gastric mucosa).
Spot 117 is potentially interesting. It is difficult to find in many gastric mucosas because it is in an area of many small spots.
Spot 115 looks pretty good and is near 117.
Spot 118 is usually bigger in Barrett's. It is quite basic and therefore difficult to compare in all gels.
Spot 44 looks good, but has already been found as an interesting spot because it seems bigger in esophageal adenocarcinoma.
Spot 120 is particularly large in Barrett's and esophageal adenocarcinoma.
Spot 121 is an enormous, gray, very basic, 18 kD spot in Barrett's and esophageal adenocarcinoma. It is so far to the right on the gel that it is sometimes tough to figure out if it is there or off the gel.
Spot 122 is bigger in Barrett's.
Spot 123, very low and acidic, is part of a family of spots bigger in Barrett's.
Spot 124 looks good. It is possible that it is a phosphorylated P18 (“c”).
Spots 6, 7, and 8 are bigger in Barrett's, but these are HUGE spots in normal esophageal.
Spot 116 is usually bigger in Barrett's but appears large in the esophagus.
HSP27 spots are bigger in Barrett's usually, but they are even larger in normal esophagus.
Spot 119 looked exciting, but it is a huge spot from the esophagus.
Summary of spots across Gastric Mucosa (GM), Cancer of the Cardia (CC), Barrett's (B), and Esophageal Adenocarcinoma (EA).
The first 5 spots were first noticed comparing EA vs B, the remainder from comparing B vs GM. (Perhaps 115 can be noticed in EA vs B as well.)
Barrett's Spots that can be Confidently Identified in IPGs:
The comparison of Barrett's to gastric mucosa and normal esophagus for the same patients described above identified some 15 spots that were thought to be increased in Barretts, using gels prepared with isoelectric focusing (IEF) in the first dimension. All of these except for one were present in equal or increased amounts in esophageal adenocarcinoma samples. Efforts have been made to identify these same protein spots in Immobilized pH gradient (IPG) gels using as guides the spot location on IEFs, spot color using the same silver staining protocol, and whether a similar difference between Barrett's and control material (normal esophagus, gastric mucosa) was observed in IPGs.
Nine of the spots were confidently identified. They had spot numbers 113, 109-111 (a group of spots counted as a single protein), 117, 118, 120, 122, 123, 124, and 44 (which is among the spots reported as being larger in esophageal adenocarcinoma).
Spots 61 and 121 were also confidently identified. The remaining four protein spots (38, 102, 112, 115) were tentatively identified in IPGs.
This component of the study was initiated by analyzing the patterns of five patients that appeared to be the most outstanding. This is a comparison of Barrett's versus adenocarcinoma. The comparison led to the following findings:
Subsequently other gels were reviewed for the above spot differences. All the gels were analyzed but the data was tabulated for only a subset of the clearest gels.
The table below shows only a subset of the spots above, in that it includes only the best spot differences. Asterisks appear next to those that were thought to be the most interesting among these.
Initially acidic-loaded IPG's (immobilized pH gradient gels) were run for three patients (cOO71-80) and basic-loaded IPG's for three different patients (clO25-34) and analyzed separately. As in previous experiments, there are many unique differences for each pair of samples from an individual, some due to varying amounts of various sample impurities, others perhaps idiosyncratic alterations which are true differences but are not shared among most tumors. The difficulty is therefore in finding differences shared among most tumors.
Two new sets of basic-loaded IPG's were made from which the three best pairs (tumor, Barrett's) of patterns were selected in order to determine which candidate spots exhibited the most reproducible spot changes. From this review there appear to be six spots which are most worth pursuing for sequencing. That these spots are more worthy than some of the others is partially colored by their intensity in the tumors, as this relates to how easily they might be purified and sequenced. Also, in the case of spots 15 and 51 the fact that these spots were previously identified as interesting differences in IEF gels helped keep them on the list.
There are figures containing close-up sections of gels for three patients from one run, and two more from a second run (slightly different conditions), with Barrett's sample on the left and tumor sample on the right.
It follows that studies were performed using two separate methods of first dimension separation, namely isoelectric focusing gels (IEFs) and Immobilized pH gradient gels (IPGs). We had first singled out 5 spots (IEF5, IEF15, IEF29, IEF31, IEF51) as the most consistent differences in IEF samples, where we searched for spots that were absent or very nearly absent in tissues other than the esophageal adenocarcinoma (Barrett's, gastric mucosa, normal esophagus). Subsequently in studies of IPGs, 6 very interesting spots were identified which were larger in the esophageal adenocarcinoma samples (IPG29, IPG56, IPG57, IPG53, IPG51, IPG15, the last two being identical to spots IEF42 and IEF44 respectively). Spots IEF5, IEF29, and IEF31 were tentatively identified in the IPG gels.
The table below summarizes which spots were increased (in both types of gels) in the esophageal adenocarcinoma compared to Barrett's. “+” indicates clear increase, “−” indicates no difference, and “?” implies difficulty scoring the spot, usually due to being obscured by another large spot, or in the case of spot IPG57 because the spot was not well separated from the dyefront (bottom) of the second dimension gel.
Spot 56 actually appears as a pair of spots which do seem related, as their sizes appear to be correlated.
Gels of secreted proteins were prepared from fresh tissue pieces from two new patients for normal esophagus, gastric mucosa, and Barrett's, as well as control gels separating proteins in the media used to culture the cells. These patterns of “secreted proteins” mixed with media proteins (serum) are promising. However, the limitation here is that this study requires fresh samples and therefore the number of samples is bound to be small. The results to date are that a group of spots numbered 109-111, which were found to be increased in Barrett's relative to normal esophagus and gastric mucosa, are quite large in the Barrett's “secreted proteins” samples compared to control samples.
There are a total of at least some twenty spots that are of interest to sequence and identify. The spots are eluted from the gels and subjected to sequence analysis.
A number of potential markers described above on the basis of their estimated isoelectric point (pl) and molecular weight, have been further characterized with respect to their N-terminal sequence. This information has led to the identification of some of the proteins based on their matching with protein databases and to novel sequences for others. The data is shown below.
Spots which have been Sequenced
The proteins eluted from the gels as in Example 6, or peptide fragments thereof, may be used as immunogens for the production of antibodies. The antibodies may be polyclonal antibodies or may be monoclonal antibodies. The antibodies are made by methods known to those skilled in the art. Antigen-binding fragments of the antibodies are prepared by methods known to those skilled in the art. Antibodies or antibody fragments with very high affinity and specificity may be used for immunological tests for markers of esophageal cancer.
Antibodies or antibody fragments with very high affinity and specificity are used to test body fluids for the presence of the proteins identified above. The body fluids tested may be blood or serum, urine, saliva, tears or other body fluids. The immunoassay may be of any type known to those skilled in the art. The assay may be homogeneous or heterogeneous. The assay may be used with automated machines, or may be functional in lower technology environments.
Such a test provides a simple and rapid test to aid in diagnosis of these serious diseases.
The antibodies or antibody fragments produced in Example 7 may be conjugated with a radioactive tag and injected into a patient. With appropriate imaging techniques the tumor can be located using the radioactively conjugated antibody. If the amount of radioactivity attached to the antibody is increased considerably, or the antibody or antibody fragment is conjugated to a toxin, the conjugate can be used to kill tumor cells in vivo. The antibody or fragment provides the targeting function, and the toxin or radioactivity kills the cells which are targeted by the antibody.
The gene corresponding to tumor specific proteins identified by the method of the present invention may be isolated and identified by techniques known to those skilled in the art. The gene can then be inactivated by molecular biological techniques and replaced into the body by gene therapy. Alternatively, anti-sense molecules can be made to genes of the tumor specific markers, and the anti-sense molecules can be used as therapeutics. By either of the above methods known to those skilled in the art, the tumor specific gene expression is decreased.
| Number | Date | Country | |
|---|---|---|---|
| 20070148713 A1 | Jun 2007 | US |
| Number | Date | Country | |
|---|---|---|---|
| 60018659 | May 1996 | US |
| Number | Date | Country | |
|---|---|---|---|
| Parent | 10393992 | Mar 2003 | US |
| Child | 11549050 | US | |
| Parent | 09508097 | Dec 2000 | US |
| Child | 10393992 | US |
