| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Enzymes and Protein Markers |
Departments of
1
Clinical Biochemistry and
2
Urology, Skejby University Hospital, DK 8200 Aarhus N, Denmark.
a Author for correspondence. Fax 45 89 49 60 18; e-mail orntoft{at}kba.sks.aau.dk.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The definition of cutoff values for biological markers above which disease or relapse after treatment is to be suspected is difficult. This is partly because of an interindividual variation, related to obvious causes like sex, age, hormonal status, and others but also to individual differences in genetic makeup. Very few studies have investigated the possible effect of different genotypes on the concentration of biomarkers encoded for by genes that have a high frequency in the population for heterozygosity for nonfunctional alleles. Previously, the concentration of Ca 19-9 has been shown to be increased in bladder cancer patients (3)(4). The intention of this study has been to evaluate the influence of Lewis allelotype and secretor phenotype on the concentration of Ca 19-9 in urine from healthy individuals and bladder cancer patients.
The tumor marker Ca 19-9 recognizes the carbohydrate structure
sialyl-Lewis a (sialyl-Lea) (5), which is
synthesized by a sialyltransferase and the Lewis transferase (Fig. 1
). The latter is encoded by the Lewis gene (FUT3) and
is an 
(1,3/1,4)-L-fucosyltransferase responsible for the
synthesis of both Lea and Leb. Because of the
competition for enzyme acceptors, the concentrations of Ca 19-9 in
serum is low among secretors (who make Leb) and high among
nonsecretors (6) (Fig. 1
).
|
Several missense mutations have been found in the Lewis gene, and when these mutations present in the homozygous form, they can cause inactivation of the enzyme (7)(8)(9)(10)(11). Furthermore, a gene-dosage-like effect has been shown to exist for healthy individuals heterozygously mutated in the Lewis gene, leading to a lower concentration of the tumor marker Ca 19-9 in serum compared with homozygous wild-type individuals (12). Consequently, we have proposed that cutoff values for Ca 19-9 measurements be redefined on the basis of secretor and Lewis genotyping (12).
The aim of this study was to compare the urinary concentrations of genotype-interpreted carbohydrate tumor marker Ca 19-9 in bladder cancer patients and healthy individuals and, subsequently, to compare the Ca 19-9 concentrations in the cancer patients with T-category, histologic grade, and presence of concomitant urothelial atypia. To investigate the possibility of using Ca 19-9 measurements in urine as a biomarker for bladder cancer, we undertook this study.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
At the time of primary transurethral treatment and (or) after recurrences, biopsy specimens were taken from apparently normal urothelium at preselected sites as described previously (13). All urine samples were screened for leukocytes and nitrite, and positive samples were excluded from the investigation.
T-classification according to UICC (14) was based on biopsy specimens obtained at transurethral resection and findings at bimanual palpation under general anesthesia. The tumors were graded according to the cellular criteria of Bergquist et al. (15). Urothelial dysplasia of selected site biopsies were graded similarly.
Apparently healthy volunteers (8 men and 13 women) with no known bladder diseases were included as a reference group. All 21 were blood-typed as Lewis-positive secretors (Leb) by routine blood bank procedures. Nine (43%) were homozygous wild type, and 12 (57%) were heterozygous mutated in the Lewis gene.
All procedures were approved by the local scientific ethical committee.
lewis genotyping
DNA was extracted from leukocytes by using Puregene DNA Isolation
kit (Gentra Systems, Inc., Los Angeles, CA) according to the
manufacturer's protocol. Lewis genotyping for the mutations T59G and
C314T was done by using PCR-cleavage assays as described previously
(12). For the performance of PCR-cleavage reactions for
the mutations T59G and C314T, the sequences of the sense and antisense
oligonucleotide primers, the annealing temperature of the PCR
reactions, the sizes of the products obtained, and the restriction
enzymes used for the cleavage of the PCR products, respectively, were
as follows: T59G, VE1 mms/EL3as (7)
(5'-ccatggcgccgctgtctggccgccc-3'/5'-gggagtggtgtcctgtcgggaggacccact-3'),
65 °C, 140, MspI; and C314T, EL3 s (7)/VE4as
(5'-agtgggtcctccccgacaggacaccactcc-3'/5'-gttggacatgatatcccagtggtgcacgat-3'),
63 °C, 204, NlaIII. PCR reactions were performed with 25
pmol of each primer, 1.25 mmol/L deoxynucleotide triphosphates, and 0.5
unit of Taq polymerase in 25 µL of buffer (50 mmol/L KCl, 10 mmol/L
Tris base, pH 8.3, 2 mmol/L MgCl2). Negative PCR controls
(amplification without template DNA) were included in all experiments.
For T59G, the expected fragments sizes after cleavage with
MspI of the mutated allele were 24 and 116 bp. For C314T,
the expected fragment sizes for the mutated allele were 105, 34, and 65
bp. The digested products were separated by 4% and 3% agarose gel
electrophoresis, respectively (Metaphor® agarose, FMC; Fig. 2
).
|
secretor typing of urine
All incubations were at room temperature. Urine samples were
centrifuged, and the supernatants were used for the assay. MaxiSorp
plates (Nunc) were coated overnight with capture lectin wheat germ
agglutinin (Sigma Chemical Co.) 5 mg/L in coating buffer. The coated
wells were emptied, washed with washing buffer (phosphate-buffered
saline, pH 7.2, plus 1 mL/L Tween) three times, and then blocked for 30
min with 100 µL of coating buffer containing bovine serum albumin, 20
g/L. After blocking and a wash, urine samples were diluted in washing
buffer to a creatinine content of 1.5 mmol/L and added to the wells.
After incubation for 1 h, the wells were washed with washing
buffer and incubated for 1 h with the detecting antibodies
anti-Lea (BioClone; Ortho Diagnostic Systems) and
anti-Leb (Seraclone; Biotest AG). After washing,
peroxidase-conjugated rabbit anti-human IgM (DAKO), diluted 2000-fold
in washing buffer, was added, and the plates were incubated for 1
h. The plates were developed with o-phenylenediamine (DAKO),
7 g/L in 0.1 mol/L sodium citrate, pH 5.0, containing 0.3 mL/L
H2O2. The reaction was stopped after 2–3 min
by the addition of 1 mol/L sulfuric acid. Absorbance at each well was
read at 492 nm, and the results were calculated after subtraction of
the blank values obtained in the absence of the detecting antibodies
anti-Lea and anti-Leb. All assays were
performed in duplicate. Positive controls for the detecting antibodies
consisted of urine obtained from apparently healthy adults, typed as
secretors and nonsecretors, respectively, on fresh blood samples, by
standard agglutination methods performed in a specialized blood
grouping laboratory at Aarhus University Hospital (Skejby,
Denmark).
The performance of the assay was tested by typing urine from 31 healthy
subjects (24 secretors and 7 nonsecretors) and 33 bladder cancer
patients (25 secretors and 8 nonsecretors) and comparing the results
with convenient secretor typing on erythrocytes. No overlap between
secretors and nonsecretors was observed in healthy individuals and in
bladder cancer patients (Table 1
). Intraassay variation (CV) for Lea and
Leb ELISA readings was determined on duplicate analyses on
a series of clinical specimens as 12.9% (0.0–35.4%) and 2.2%
(0.1–4.2%) for secretor-positive healthy subjects (n = 10),
3.8% (0.4–6.4%) and 8.1% (0.0–20.2%) for nonsecretor healthy
subjects (n = 6), 12.5% (1.2–21.1%) and 1.3% (0.0–4.5%) for
secretor-positive bladder cancer patients (n = 10), and 2.1%
(0.6–8.0%) and 10.8% (1.8–35.5%) for nonsecretor bladder cancer
patients (n = 10). All subjects were Lewis positive.
|
determination of ca 19-9 and creatinine
Urine samples were centrifuged, and the supernatants were assayed
for concentrations of tumor-associated antigen Ca 19-9 by using a
solid-phase, two-site chemiluminescence enzyme immunometric assay for
use with the Immulite Automated Analyzer (Diagnostic Products Corp.)
according to the manufacturer's instructions. Intraassay variation
(CV) was determined on duplicate analyses on a series of clinical
specimens as 5.9% in the range of <20 kU/L (n = 16), 4.8% in
the range of 20–200 kU/L (n = 12), and 6.7% in the range of
>200 kU/L (n = 8). Interassay reproducibility for urine was
tested by two successive assays (double determinations) of a series of
concentrations (nine high, >200 kU/L, and nine low, <20 kU/L) during
1 week. CVs were 4.7% (range, 0.5–10.6%) and 5.0% (range,
0.0–10.1%) for high and low concentrations, respectively. Intraassay
and interassay CVs for urine samples were comparable with those
reported for serum pools. Urinary content of creatinine was determined
by using a routine alkaline picrate method. To allow for different
concentrations of the urinary samples, Ca 19-9 concentrations were
expressed as Ca 19-9/creatinine ratios. Ca 19-9 was expressed as kU/L
and creatinine as mmol/L.
statistical analysis
Distributions were compared by using the Mann–Whitney
U-test. Unless otherwise stated, P <0.05 was
considered statistically significant.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Secretor typing of the urine by a novel lectino-immuno assay showed
that 60 patients (79%) were secretors, whereas 16 (21%) were
nonsecretors (Fig. 3
). Secretor typing on urine could not be performed in the five
Lewis-negative patients.
|
As a result of the Lewis genotyping for the two mutations that are representative of the mutated Lewis alleles present in the Danish population, bladder cancer patients and reference individuals could be classified as either homozygous wild-type, heterozygous mutated, or homozygous mutated in the Lewis gene (12). Five of 81 patients (6%) were genotyped as homozygous mutated, i.e., Lewis negative. Fifty-one (63%) were homozygous wild-type, and 25 (31%) were heterozygous mutated in the Lewis gene. The distribution of genotypes agrees with the Hardy–Weinberg theorem on the assumption that the frequency of genuine Lewis negatives is ~7%. The five Lewis-negative individuals had no detectable Ca 19-9 in urine and were excluded from data processing.
As shown in Fig. 4
, Ca 19-9 concentrations were significantly higher (P
=0.004) in reference subjects who were homozygous wild-type in the
Lewis gene compared with those who were heterozygous mutated in the
Lewis gene (Mann–Whitney U-test). All reference subjects
were secretors.
|
Urine samples from secretor-positive bladder cancer patients showed a
significantly (P <0.01) higher concentration of Ca 19-9
than did the secretor-positive healthy individuals (Fig. 5
). On the basis of a comparison of Ca 19-9 in nonsecretor and
secretor patients, ignoring nonsignificant differences in tumor stage
and grade between the two groups, an examination of Ca 19-9 ratios
showed a considerably higher concentration (P <0.01) in
urine from nonsecretors (Fig. 5
).
|
Secretor-positive cancer patients were then grouped according to the
Lewis genotypes, heterozygous mutated and homozygous wild-type.
Statistical analyses indicated a higher Ca 19-9 ratio in homozygous
wild-type patients than in heterozygous patients (P =0.06;
Fig. 6
).
|
From all patients, selected site biopsies were taken from eight
predefined sites in the bladder when the bladder tumor was removed.
However, to compare Ca 19-9 concentrations in urine from patients with
and without concomitant dysplasia, we included only the 52 patients who
were both Lewis positive and secretors, and who had their urine samples
taken the same day they underwent surgery. From the pathology
examination of the eight bladder biopsies, we judged that 30 patients
(58%) had concomitant urothelial dysplasia, 7 of whom (23%) had
carcinoma in situ, and 22 patients (42%) had normal urothelium.
Secretor-positive bladder cancer patients with concomitant epithelial
dysplasia showed significantly (P =0.02) higher Ca 19-9 in
urine than did cancer patients with normal epithelium (Fig. 7
). When the patients were divided the second time into four
groups according to Lewis genotype, a pattern of increasing Ca 19-9
concentrations appeared (Fig. 8
).
|
|
Ca 19-9 ratios were not correlated to tumor grade. The patients with invasive tumors were too few in number for an evaluation of the correlation between Ca 19-9 concentrations and tumor stage.
We conclude that genotype evaluation of Ca 19-9 measurements in urine is a prerequisite for the interpretation of the concentration of this tumor marker. Furthermore, an increased concentration of Ca 19-9 ratio in urine seems to reflect both the presence of concomitant urothelial dysplasia and the presence of a tumor. On this basis, Ca 19-9 ratio in urine may serve as a biomarker and help identify the patients with field defect of the bladder, i.e., patients who need to be followed closely.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We have shown that Ca 19-9 ratio in urine is increased in patients with bladder cancer, and that the highest concentrations are found in urine from patients whose selected site biopsies from apparently normal epithelium showed concomitant urothelial dysplasia. The Ca 19-9 concentrations in urine from healthy individuals as well as from bladder cancer patients depend on the individual Lewis genotype and secretor type and can be interpreted correctly only when the types are known.
The amount of Ca 19-9 in urine seems to reflect not only the presence of tumor(s) but also the existence of urothelial dysplasia or carcinoma in situ in biopsies from the bladder mucosa. We do not know the origin of urinary Ca 19-9 at present. Although it is found in ducts of the prostate gland and in seminal fluid (16)(17), it more probably originates from the urothelium or the kidney.
In the human bladder, marked tumor-associated changes in blood group antigen expression have been demonstrated (18)(19)(20). ABH and Lea antigens are expressed in normal urothelium but are often replaced by Leb and Ley in high-grade invasive tumors (21)(22). Thus, one would not expect a high amount of sialyl-Lea from these tumors, which agrees with our present data. In normal urothelium, sialyl-Lea is present only in the luminal umbrella cell layer. However, in carcinoma in situ and in papillomas, the sialyl-Lea structure is expressed in all cell layers of the urothelium (21). This increased expression in carcinoma in situ alterations could explain the increased Ca 19-9 in urine from patients who, in addition to the tumor, have concomitant dysplasia or carcinoma in situ of a fraction of the urothelial mucosa. At present, these cases of field disease are diagnosed on biopsies taken at eight preselected sites in the bladder mucosa. Often, there is more than one dysplastic biopsy, indicating that a relatively large area of the mucosa is diseased. It is, however, difficult to get a quantitative impression of the extent of the dysplastic field on the basis of these eight biopsies. The genotype-interpreted Ca 19-9 measurement might be a marker that could supply such quantitative or semiquantitative information on the extent of the field change and its progression or regression during the urothelial disease. The presence of some overlap between the two groups, however, requires that additional studies are performed to determine the clinical applicability of the tumor marker. Other markers are presently approved by the Food and Drug Administration for bladder cancer testing. Future studies may show whether one marker is sufficient or whether combinations of markers will have better clinical performance.
Thus far, the carbohydrate tumor marker Ca 19-9 (or GI-MA) has been used mainly for serological diagnosis and follow-up of gastrointestinal and pancreatic malignancies (23). A few studies have also investigated the use of this tumor marker in the serum and urine of patients with bladder cancer (3)(4)(24). These data need to be reinterpreted in light of the secretor and Lewis genotypes of these patients. Very high serum or urine concentrations are, of course, abnormal, and in a follow-up situation the patient serves as his own reference. However, declining values of Ca 19-9 may present a problem when the concentration close to the normal reference interval is reached. The cutoff amount defining pathologically increased values is quite different among secretors and nonsecretors and for the groups internally among Lewis heterozygous and Lewis homozygous individuals. The recent cloning of the secretor gene (FUT2) and the identification of enzyme-inactivating mutations and gene fusion (25)(26)(27)(28)(29)(30) may even make it possible to classify the secretors as homozygous wild-type and heterozygous at the FUT2 locus, once the geographical distribution of inactivating mutations has been documented. Finally, the sialyltransferase adding the sialic acid to the carbohydrate chain could also be involved in a gene-dosage dependent effect. This is, however, not very likely, as several sialytransferases are present and may have overlapping functions.
The antigenic determinant for Ca 19-9, sialyl-Lea, serves as a ligand for the endothelial leukocyte homing receptor E-selectin (31)(32), and it is suggested that E-selectin-mediated binding of tumor cells to the endothelium is involved in the hematogenous metastasis of tumor cells (33). Whether Ca 19-9 in urine indicates tumor cells with a high density of sialyl-Lea on the cell membrane is unknown. However, if it is true, these tumor cells might have an ideal surface for binding to activated endothelial cells, and on the basis of this mechanism, might be prone to make hematogenous metastasis.
The synthesis of Ca 19-9 is complex because there are three genes
involved: the secretor gene, the gene encoding the sialyltransferase,
and the Lewis gene. The gene products are glycosyltransferases, which
in turn will add monosaccharides to the growing chain of carbohydrates
on glycoproteins or glycolipids. We have shown previously that the
Lewis glycosylation of circulating glycoproteins is gene-dosage
dependent because individuals with two intact alleles synthesize
markedly more than individuals with one allele. In this report, we
extend this observation to Lewis glycosylation of glycoproteins in
urine. We state the hypothesis that the same difference between
heterozygous and homozygous wild-type individuals would be expected,
when Ca 19-9 concentrations are increased because of cancer disease. In
favor of this hypothesis is the recent finding that the accumulation of
Ca 19-9 in colorectal tumors is mainly caused by the activation of
-2,3-sialyltransferase and not by the Lewis enzyme
(34). Furthermore, Lewis enzyme activity in malignant
bladder tissue does not differ from the enzyme activity in normal
bladder tissue (35). -2,3-Sialyltransferase activity
has not been investigated in bladder cancer. The increased
concentration of Ca 19-9 in urine might, on the basis of the data
presented above, reflect a shift in glycosylation pathway. This might
favor the synthesis of sialyl-Lea as terminal structure or
the increased synthesis of the glycoproteins that carry
sialyl-Lea as side chains, or both.
Further investigations are needed to evaluate whether high concentrations of creatinine-adjusted Ca 19-9 are specific for bladder cancer and concomitant urothelial dysplasia, or whether this is merely a diffuse marker for an abnormal urothelial mucosa.
Prospective comparative studies of the ability of Ca 19-9 and routine procedures to signal bladder cancer and (or) predict clinical course are also needed to define the exact role of Ca 19-9 in diagnosis and follow-up of bladder cancer. Such investigations may also elucidate whether biochemical alterations in the form of raised Ca 19-9 concentrations may be detectable before abnormal pathology.
In conclusion, the findings in this report draw attention to the possibility of a general genotype interpretation of results in clinical chemistry. The determination of biomarker concentrations in body fluids could be influenced by the individual genotype for the biomarker, because heterozygous individuals may produce less biomarker than homozygous individuals in normal as well as in diseased cells. Perhaps the future will show not only age- and sex- corrected reference intervals but also genotype-corrected reference intervals. In short, we believe we are in an era where individual biological variation is being explored by genetic tools, in hopes that it may lead to a more precise interpretation of clinical chemistry results in individual patients.
|
Acknowledgments |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
(1,3/1,4)-Fucosyltransferase (FucT-III) gene is inactivated by a single amino acid substitution in Lewis histo-blood type negative individuals. Biochem Biophys Res Commun 1993;196:624-631.
[Web of Science][Medline]
[Order article via Infotrieve]
(1/1,4)-fucosyltransferase gene deficiency (FUT3) found in Lewis-negative Indonesian pedigrees. J Biol Chem 1994;269:20987-20994.
(1,3/1,4)-L-fucosyltransferase (FUT3) gene mutations on enzyme activity, erythrocyte phenotyping and circulating tumor marker sialyl-Lewis a levels. J Biol Chem 1997;271:32260-32268.
(1,2)fucosyltransferase gene (FUT2). J Biol Chem 1995;270:4640-4649.
1,2-fucosyltransferase gene with the Lewis (a+b+) phenotype: a potential molecular basis for the weak secretor allele (Sew). Biochem J 1995;:329-332.
[Web of Science][Medline]
[Order article via Infotrieve]
(1,2)fucosyltransferase gene among Lewis(a+b-) non-secretors. Biochem Biophys Res Commun 1996;222:390-394.
(1,2)-fucosyltransferase gene deficiency in a Japanese population: a fusion gene generated by unequal crossover responsible for the enzyme deficiency. Am J Hum Genet 1996;:331-342.
[Web of Science][Medline]
[Order article via Infotrieve]
(1,2)fucosyltransferase gene (FUT2). Vox Sang 1996;70:21-25.
[Web of Science][Medline]
[Order article via Infotrieve]
2–3Sialyltransferase associated with the synthesis of Ca 19-9 in colorectal tumors. Cancer 1996;77(Suppl):1694-1700.
[Medline]
[Order article via Infotrieve]
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  E. M. Vestergaard, H. O. Hein, H. Meyer, N. Grunnet, J. Jorgensen, H. Wolf, and T. F. Orntoft Reference Values and Biological Variation for Tumor Marker CA 19-9 in Serum for Different Lewis and Secretor Genotypes and Evaluation of Secretor and Lewis Genotyping in a Caucasian Population Clin. Chem., January 1, 1999; 45(1): 54 - 61. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
