Immunoassays for the Quantification of uPAR

The cell surface plasminogen activation system has an active role in cancer invasion. Thus, both uPA and PAI-1 have been shown to be prognostic markers mainly in breast cancer [117-119]. Therefore, it was early proposed that measurement of uPAR might also be of value in identifying patients with aggressive tumors and hence poor prognosis. This provided impetus for the establishment of immunoassays to be used for sensitive and specific measurements of uPAR in tumor tissue lysates and blood from cancer patients once anti-uPAR antibodies were available. In the different enzyme-linked immu-nosorbent assays (ELISAs) described in this chapter, uPAR(I-III) is used as the standard but in many cases the biological samples contain a mixture of uPAR forms. Therefore, the concentrations measured should be expressed as molar concentrations, not as ng/ml or ng/mg. In order to compare the amounts of uPAR measured in donor plasma samples by the different immu-noassays, the amounts are expressed in moles. For simplicity, the ELISA formats that have been used in the prognostic studies described in Section 4 will be referred to as E1, E2, and so on. The immunoassays, the antibodies employed, and the uPAR forms they measure are summarized in Table 1.

The first ELISA designed for uPAR measurements with an analytical sensitivity of 19.5 pmol uPAR/liter was developed using two mAbs, R8 for catching and biotinylated R2 for detection [116]. These mAbs recognize nonoverlapping epitopes on domain III of uPAR and therefore, this ELISA measured all known forms of uPAR except uPAR(I). A 20% decrease in the recovery was reported when uPA/uPAR complexes were measured. This assay was never used on patient material because of the

TABLE 1

uPAR Forms Measured by Different Immunoassays

Immunoassay

uPAR variants measured

aC-Ab: anti-rsuPAR pAb bD-Ab: R3, R5; uPAR(I) R2; uPAR(III)

uPAR(I-III), uPAR(II-III), uPAR(I) uPAR(l-III)/uPA complex

C-Ab: anti-rsuPAR pAb D-Ab: R3, R5; uPAR(I) R2; uPAR(III)

uPAR(I-III), uPAR(II-III), uPAR(I) uPAR(I-III)/uPA complex

C-Ab: anti-suPAR pAb D-Ab: anti-suPAR pAb

uPAR(I-III), uPAR(II-III), uPAR(I) uPAR(l-III)/uPA complex

C-Ab: R2; uPAR(III) D-Ab: anti-rsuPAR pAb

uPAR(I-III), uPAR(II-III), uPAR(l-III)/uPA complex

C-Ab: anti-rsuPAR pAb D-Ab: R2; uPAR(III)

uPAR(I-III), uPAR(II-III), uPAR(l-III)/uPA complex

C-Ab: anti-rsuPAR pAb D-Ab:IIIF10; uPAR(I)

uPAR(I-III), uPAR(I)

C-Ab: anti-rsuPAR pAb D-Ab: HD13.1

Not known

C-Ab: R2; uPAR(III) D-Ab: R3; uPAR(I)

uPAR(I-III)

C-Ab: R2; uPAR(III) D-Ab: R23; uPAR(II-III)

uPAR(I-III), uPAR(II-III)

[108]

TR-FIA 3 C-Ab: R5; uPAR(I) D-Ab: R3; uPAR(I)

uPAR(I)

[108]

a C-Ab = coating antibody. b D-Ab = detecting antibody.

low sensitivity [116]. In order to increase the sensitivity and be able to measure all uPAR forms collectively, we designed an ELISA using a rabbit anti-suPAR pAb for catching and a combination of three biotinylated mAbs for detection. The antigen used for raising the pAb was recombinant suPAR1-277. The mAbs used for detection were R2 and the two domain I-specific antibodies, R3 and R5. This assay thus detects all forms of uPAR, including uPAR(I), and is referred to as E1 herein. The analytical sensitivity was increased more than 10 times to 1.6 pmol uPAR/liter [87]. E1 was first validated for measuring uPAR in breast cancer tissue extracts and lysates and later for use with blood plasma samples [98]. The median level of suPAR in a plasma pool from 39 healthy individuals was 28 pmol/liter as measured with E1.

The E1 ELISA was further optimized for use with plasma samples by replacing the biotinylated mAbs and end-point peroxidase measurements with nonbiotinylated mAbs and a rabbit anti-mouse pAb coupled to alkaline phosphatase. This kinetic ELISA enabled accurate calculation of the uPAR concentration from multiple time-point measurements of the linear color development obtained with a phosphatase substrate system [120]. The use of nonbiotinylated mAbs was prompted by the finding that biotinylation of mAbs could change not only their affinity but also their specificity. These problems were found to be most pronounced when the ratio of biotin to mAb was high and plasma samples were analyzed [121]. When this kinetic ELISA, called E2, was used for the measurement of uPAR in plasma samples from 93 healthy individuals, the median concentration was 39 pmol suPAR/liter [120]. Mizukami et al. developed another ELISA, E3, employing an anti-suPAR pAb raised against Pi-PLC-released uPAR from U937 cells. Using this pAb for catching and detection, the E3 ELISA determined a median concentration of suPAR of 65 pmol/liter in a plasma pool from 20 healthy volunteers [97]. The E3 ELISA should hypothetically have measured all forms of uPAR but it used biotinylated antibodies for detection without reporting the biotin/antibody ratio. It is therefore unknown how much influence the biotinylation had on antibody affinity and especially specificity.

The assay E2 was later inverted, using the high-affinity mAb R2 as catching antibody while the anti-suPAR pAb was employed for detection. This was also a kinetic ELISA. This ELISA, E4, quantifies uPAR(I-III), uPAR (II-III), and uPA/uPAR complexes and gives a lower background signal in EDTA plasma samples compared to the E2 assay [41]. However, serum collections are more commonly used in clinical studies than plasma, and when serum samples were analyzed with E4, rather high-non-specific signals were occasionally observed. This problem was solved by adding heparin to the assay buffer, which prevented nonspecific signals. The exact mechanism is unknown but the highly charged heparin polyanion probably binds to the solid phase not allowing the molecules responsible for the unspecific signal to bind [122]. This kinetic assay, E5, uses the anti-suPAR pAb for catching and the mAb R2 for detection [40, 122] and like the E4 assay, it detects all uPAR forms except uPAR(I). The analytical sensitivity of E5 is 0.13 pmol/liter and the median suPAR concentrations in citrate plasma and serum pools from healthy individuals were 75 and 85 pmol/liter, respectively [122].

Kotzsch et al. have developed two uPAR ELISAs measuring different forms of uPAR both using a pAb, raised in chicken against suPAR1-277

as the catching antibody. One of the ELISA formats, E6, uses the domain I-specific mAb, IIIF10, as the detecting mAb. E6 measures nonoccupied uPAR(I-III) and liberated domain I. The other ELISA, called E7, employs a mAb HD13.1 raised against suPAR1-277 for detection. Since the domain specificity of this mAb is not identified, it is unknown which uPAR variants the assay measures [38].

Two GPI-anchored and three soluble forms of uPAR have been identified in human tissue and body fluids (Fig. 4). The amounts of these in tissues and circulation will reflect not only the level of cellular expression but also the activity of the cleavage enzymes, including those of the plasminogen activation system. Thus, the levels of the cleaved uPAR forms may have a stronger prognostic significance than merely the total uPAR content. To enable studies of the prognostic potential of the individual uPAR forms, we designed specific time-resolved fluorescence immunoassays (TR-FIAs) using different combinations of mAbs. The detection system using time-resolved fluorescence was crucial, since biotinylation of detecting mAbs would cause problems with both sensitivity and specificity, when measuring plasma and possibly also serum samples [121]. TR-FIA 1 quantifies only nonoccupied uPAR(I-III), while TR-FIA 2 measures nonoccupied uPAR(I-III) and uPAR(II-III). The molar concentration of uPAR(II-III) can thus be calculated by subtracting the molar quantities measured by TR-FIA 1 from those measured by TR-FIA 2. TR-FIA 3 determines the amount of uPAR(I) in a sample. The specific detection of liberated uPAR(I) in the assay was obtained by adding an inhibitor, AE120, that prevents mAb R3 from binding to uPAR(I-III), but allows R3 binding to the liberated domain I [108]. The detection limits for both TR-FIA 1 and 2 were 0.3 pmol/liter and for TR-FIA 3, it was 1.9 pmol/liter. In a citrate plasma pool from healthy volunteers, 42 pmol/liter of uPAR(I-III) and 26 pmol/liter of uPAR(I) were measured and the calculated uPAR(II-III) content was 39 pmol/liter. This corresponds to the ELISA-measured amount using E5, which determined the total amount of suPAR in citrate plasma from healthy individuals to 75 pmol/liter compared to 81 pmol/liter as measured with TR-FIA 2, since the E5 ELISA does not measure uPAR(I).

  1. uPAR in Cancer
  2. 1. Breast Cancer

Thus far, no domain I-specific mAbs that detect uPAR in paraffin sections are available. The mAbs that have proven useful for immunohistochemistry have epitopes residing in domains II and III and none of these recognize uPAR(II-III) selectively [28, 33]. Therefore, the signals obtained after immunohistochemical staining of a tissue sample will originate from both uPAR(I-III) and uPAR(II-III) and will be referred to as uPAR in the following. In breast cancer tissue, the localization of uPAR was studied by three independent research groups using different antibodies, though obtaining consistent results [28, 33, 123]. In the first study, mAb R2, reacting with an epitope in the C-terminal part of domain III, stained uPAR in sections of formalin-fixed paraffin-embedded tissue from diagnosed invasive ductal breast carcinomas obtained from 40 women [33]. We found uPAR immuno-reactivity in 34 of the 40 cases. In 32 of the cases, uPAR was detected on macrophages immediately surrounding the malignant epithelium. In 5 of the 34 positive specimens, uPAR was found on cancer cells. There was a weak positive signal in most cases from tissue neutrophils, but only in 2 of the 34 positive cases a weak staining of endothelial cells was obtained. Normal appearing epithelium and normal female breast tissue were negative [33]. The localization of uPAR mainly to macrophages, some cancer cells, and a few endothelial cells was confirmed by the two other studies, in which paraffin-embedded tissue sections from invasive breast carcinomas (59 and 20 cases) were stained using an anti-uPAR pAb in one study and an mAb specific for domain II in the other [28, 123]. The immunohistochemical localization of uPAR in different cancer forms is summarized in Table 2.

In an attempt to optimize the conditions for measurements of uPAR in breast tumor tissue, the amounts extracted using three different buffers were compared [87]. A Triton X-114-containing buffer routinely used for extraction of GPI-linked uPAR from cells [53] solubilized the highest amounts of uPAR from the breast cancer tissue. The second most efficient was a Triton X-100-containing acetate buffer, pH 4.2, which was optimized for the extraction of uPA [124]. The least efficient buffer contained no detergent and is generally used for preparations of cytosols [125]. Whereas the first two buffers will extract both GPI-anchored and soluble forms of uPAR, the cytosol buffer is only capable of extracting soluble uPAR forms. Surprisingly, ELISA measurements, using E1, of cytosols and acidic Triton X-100 buffer extracts from 505 primary breast tumors revealed that uPAR levels above

TABLE 2

Immunohistochemical Localization of uPAR in Cancer

TABLE 2

Immunohistochemical Localization of uPAR in Cancer

Cancer

uPAR positive cell types

References

Breast

Macrophages, cancer cells, endothelial cells

[28, 33, 123]

Colorectal

Macrophages, neutrophils, cancer cells

[34]

Liver

Macrophages, myofibroblasts, cytokeratin 7 positive hepatocytes

[128]

Prostate

Macrophages, neutrophils

[140]

the median in cytosolic extracts were significantly associated with a shorter overall survival. The amounts of uPAR extracted with the acidic Triton X-100 buffer had less prognostic impact. In a group of node positive post-menopausal women, cytosolic uPAR was found to be a very strong predictor of overall as well as relapse-free survival [37]. In a different study measuring cytosolic extracts from 878 primary breast tumors, the prognostic significance of these forms of uPAR was confirmed. The E1 ELISA was slightly modified in this study using nonbiotinylated mAbs and an HRP-conjugated goat anti-mouse pAb for detection [126]. Higher levels of uPAR were measured in steroid hormone receptor negative tumors in this study, but the uPAR levels were found to be unrelated to menopausal status or grade of differentiation. The performances of E6, E7, and a commercial uPAR ELISA have been compared in another study. Only the combined levels of uPAR(I-III) and uPAR(I) measured with E6 in tumor tissue extracts correlated with disease-free survival in the 199 patients tested [38]. It would be interesting to compare the amounts measured by E6 with those measured by TR-FIA 1 and 3 described above and to investigate if the prognostic impact of either uPAR(I-III) or uPAR(I) alone would be more significant.

Measuring prognostic parameters such as uPAR in tumor tissue extracts or lysates requires that fresh tumor tissue is available from surgery or a biopsy. The tissue has to be frozen without delay and the proteins extracted prior to analysis. This requires considerable cooperation between surgeons, patholo-gists, and the immunoassay laboratory. Since the pathologist must have priority in selection of material for diagnosis and also because a tumor is rather heterogeneous, the piece that is available for extraction and immunoassay might not be completely representative of the whole tumor. Peripheral blood on the contrary is considerably more homogeneous and easier to collect. In the studies mentioned above, the level of soluble uPAR forms extracted from breast cancer tissue with the cytosol buffer was shown to have prognostic significance [37]. It was therefore reasonable to propose that the soluble forms found in blood might also be prognostic markers. A collection of sera made from preoperatively collected blood from 274 breast cancer patients without evident distant metastases was analyzed using E5, measuring all uPAR forms except domain I [40]. From 188 of these patients, cytosols were prepared from the primary tumor. Measurements of sera from 174 female blood donors revealed that there was an association between the age of the donor and the suPAR level with an increase of the suPAR level with age. The suPAR measurements of the patient sera were accordingly adjusted with respect to the age of the patient. Even though the mean suPAR level in sera from the breast cancer patients (124 pmol/liter) was not much higher than that in sera from the healthy volunteers (98 pmol/liter), the age-adjusted serum suPAR levels in the patients were significantly higher compared with those of the donors. A significant correlation was found between high serum suPAR levels in breast cancer patients and poor outcome and this was independent of lymph node status, tumor size, and estrogen receptor status. There was no correlation between the cytosolic and serum suPAR levels in samples from 188 of the patients [40].

5.2. Colorectal Cancer

In colon cancer, uPAR is localized mainly to tumor-infiltrating macrophages and neutrophils at the invasive front of the primary tumor [34]. Of the 30 cases of colonic adenocarcinoma investigated, a strong signal of immuno-reactivity was seen in all cases at the invasive foci. Some cancer cells were stained in 19 of the cases. uPAR was detected with mAb R2, employing mAb R4 as a control. The uPAR-expressing cell types were verified and the subcellular localization of uPAR was determined by an immunoelectron microscopical analysis of 12 cases of colon cancer [127]. In this study, mAbs R2 and R4 and an anti-uPAR pAb were used with identical results on frozen sections. The localization of uPAR to the plasma membranes of macrophages indicates that they were engaged in cell surface plasminogen activation. The uPAR positive cancer cells displayed uPAR on the plasma membrane and also in their rough endoplasmic reticulum, indicating intra-cellular synthesis. In contrast to the light microscopy study, some reactivity was detected on the plasma membrane of some endothelial cells. In several fibroblasts, immunoreactivity was found in the lumen of the endoplasmic reticulum with little expression along the plasma membrane, which is a typical staining pattern for cells that actively secrete protein [127].

In hepatocellular carcinoma, uPAR was also found on macrophages, on myofibroblasts, and in a few cases on cytokeratin 7 positive hepatocytes. Identical results were obtained whether the specimens were stained with mAb R2, R4, or the pAb against suPAR1-277 [128]. Interestingly, uPAR and uPA were often colocalized.

In a study including tumor tissue resected from 161 colorectal cancer patients, proteins were extracted from the tissue using a Tween 80-containing buffer. The uPAR content was measured using an ELISA, where the domain specificity of the catching mAb was not revealed and thus it is unknown which uPAR forms were measured. The study, however, identified the uPAR concentration as an independent and significant prognostic factor for 5-year overall survival [129]. In a more recent study where tumor tissue extracts from 71 colorectal cancer patients were analyzed using another ELISA, high levels of uPAR (>91 fmol uPAR/ng protein) were significantly associated with shorter overall survival (p = 0.0248). The levels of uPAR were also found to correlate with the presence of liver metastases as patients with tumors and liver metastases had higher levels of uPAR than those without [130].

suPAR was detected in plasma from Duke's stage D colorectal cancer patients and the suPAR levels were elevated as measured with the E2 ELISA [120]. This study was expanded and the prognostic significance explored by measuring preoperatively obtained EDTA plasma from 591 patients with colorectal cancer using the E4 ELISA. Overall survival was significantly related to plasma suPAR levels. An arbitrary cut point equal to the median of all patients (45 pmol uPAR/liter) divided the patients with Duke's stage B, C, and D tumors [131, 132] into statistically different survival groups. Since the survival of these patients was recorded as death of all causes, the survival of an age- and gender-matched population was used for comparison. Interestingly, for patients with Duke's stage B cancer and with suPAR levels below the cut point, the survival did not differ from the age- and gender-matched normal population [41]. Thus, in the most clinically difficult patient group for decision making, measurements of plasma suPAR were able to identify patients with the highest risk. In a multivariate analysis that included stage and other known prognostic indicators of colon cancer survival, an elevated preoperative suPAR level was an independent prognostic indicator for shorter overall survival [41].

High expression of uPAR has been reported in gastric cancer [133]. Heiss et al. demonstrated a prognostic value of uPAR in bone marrow aspiration biopsies from gastric cancer patients with cytokeratin positive tumor cells in the bone marrow at the time of primary surgery. Staining with a uPAR mAb showed that the presence of uPAR positive tumor cells in the bone marrow predicted early relapse and was a strong independent prognostic marker [134, 135].

5.3. Lung Cancer

The prognostic significance of uPAR has been studied in tumor tissue extracts from patients with non-small cell lung cancer (NSCLC). When the 228 patients were subgrouped into squamous cell carcinomas (n = 84), adenocarcinomas (n = 106), and large cell carcinomas (n = 38), the median level of uPAR was 83.3 fmol/mg protein in the adenocarcinomas and 45.3 and 42.0 fmol/mg protein in the squamous and large cell carcinomas, respectively [136]. In this study, tumor extracts were made using the acidic Triton X-100-containing buffer [124] and uPAR measured with the E1 ELISA. For uPAR measured in extracts from the patients with squamous cell carcinoma, there was a significant correlation between high uPAR level and short overall survival (p = 0.038) [39]. The highest uPAR levels were measured in extracts of the 106 adenocarcinomas. These levels did not correlate with survival

[136]. This was also the case for uPAR in tumor extracts from large cell carcinomas, possibly due to the low number of patients [39]. When 63 of 77 extracts from the squamous cell carcinoma patients were reanalyzed using TR-FIA 3, the measured amounts of domain I were found to be associated with survival. The prognostic impact of uPAR(I) was stronger compared to that of total uPAR measured previously [35]. Intact uPAR was not measured in this study since the recovery was very poor when the assay was validated in the tumor tissue extracts. The reason for this could be the presence of uPA in these extracts, which will form complexes with uPAR(I-III) and therefore cannot be measured by TR-FIA 1 [35,108]. The prognostic significance of the combined amounts of domain I and uPAR(I-III) in tumor extracts from NSCLC patients was demonstrated by Werle et al. using the E6 ELISA [137].

The median suPAR level, as measured with the E1 ELISA in citrate plasma from 17 patients with NSCLC, was significantly increased compared to that in citrate plasma from 30 healthy donors, while there was no increase in the uPAR level in citrate plasma from 14 patients with small cell lung cancer (SCLC) [138]. In another study using a different ELISA, elevated levels of circulating suPAR were found both in serum samples from patients with NSCLC and SCLC [139].

5.4. Prostate Cancer

In prostate cancer, uPAR is located on neutrophils and macrophages as demonstrated in tissue from 16 patients with histologically confirmed prostate adenocarcinoma [140]. uPAR protein was detected in specimens from all patients. The antibody used for immunohistochemical staining was an anti-suPAR1-277 pAb [87] and the staining pattern of the control mAb R2 was identical. In 8 of the 16 cases of prostate cancer, uPAR was identified in a subpopulation of macrophages mainly found in the interstitial tissue. In all 16 cases, the anti-uPAR pAb stained intravascular neutrophils, while none of the cases demonstrated staining in cancer cells. When benign prostate hyperplasia (BPH) from nine patients were analyzed, eight patient samples had uPAR positive macrophages, which were located in the lumen of glands, intraluminal, and in one case in the interstitial tissue. In seven cases, intravascular neutrophils stained positive for uPAR [140].

A possible correlation between serum suPAR levels and prostate cancer disease progression was investigated in serum from 72 treatment naive patients with confirmed prostate cancer [141]. suPAR was elevated in serum from prostate cancer patients compared to that from healthy individuals, and the mean concentrations were 127 ± 39 pmol/liter in healthy controls (n = 70) and 179 ± 91 pmol/liter in the patient sera (n = 72). No age dependence of the suPAR level in sera from male donors was found.

When patients were divided according to their serum suPAR content, the overall survival of patients with elevated suPAR levels was significantly shorter than that of patients with serum suPAR levels indistinguishable from those of healthy controls [141].

The presence of the different suPAR variants in EDTA plasma from a prostate cancer patient was analyzed by size exclusion chromatography and measurement of the fractions obtained with the three different TR-FIAs. This EDTA plasma contained detectable amounts of uPAR(I) in contrast to an EDTA plasma pool from healthy donors. The levels of suPAR measured with TR-FIA 2 were considerably higher and eluted at a later position in the patient sample than in the donor plasma, thus indicating the presence of suPAR(II-III) in the patient sample [108]. In a recent study, the concentrations of uPAR(I) as well as the calculated suPAR(II-III) were found to be significantly elevated in serum samples from patients with prostate cancer compared to the concentrations in serum from men with benign prostatic conditions. Specific measurements of uPAR(I) were found to improve specificity of prostate cancer detection [142].

5.5. Gynecologic Cancer

High uPAR levels in breast, colorectal, and lung cancers predict short overall survival. However, the opposite was found for uPAR in tumor tissue from ovarian cancer patients [143]. Proteins in this study were extracted from the tumor tissue using the Triton X-100-containing acidic buffer [124]. The levels of uPAR, measured with the E4 ELISA, were lower in benign as compared to invasive or borderline tumors. However, among the malignant tumors, the more advanced and poorly differentiated tumors contained lower levels of uPAR than the well-differentiated, less advanced tumors.

The uPAR forms present in blood, ascites, and cystic fluids from ovarian cancer patients have been extensively characterized and their levels measured [21, 82, 99, 144]. Using the E2 ELISA, preoperatively sera taken from 87 patients with ovarian cancer were measured and compared to the serum levels in 40 age-matched healthy women. The levels of suPAR were increased in sera from the patients [144]. However, when the patients were ranked according to their FIGO classification [131], the highest levels were found in stage 2 patients (111 ± 20 pmol/liter) and then the levels decreased with increasing FIGO stage (98 ± 10 pmol/liter in stage 3 and 72 ± 7 pmol/liter in stage 4). Nevertheless, high levels of suPAR in preoperatively collected sera were found to correlate with poor survival. This is in contrast to the results obtained using tumor tissue extracts from ovarian cancer patients discussed above [143]. The amount measured in sera from healthy controls (n = 40)

was 49 ± 3 pmol/liter and the suPAR levels were independent of the age of the donor [144]. Similar results were obtained in another study where citrate plasma samples from 53 ovarian cancer patients were analyzed [145]. However, no relation to prognosis was observed when EDTA plasma from 47 women with FIGO stage 3 ovarian cancer was analyzed using the E5 ELISA, possibly due to the smaller number of samples studied [146].

Ovarian cancer patients with progressed disease often present with ascites/ peritoneal fluid. In some women, ovarian cysts are detected containing cystic fluid. The concentrations of suPAR in these body fluids were compared with those in serum made from peripheral blood and blood aspirated from the surface veins on the tumor in 77 patients admitted for surgery of ovarian tumors [21]. In this study, elevated levels of suPAR were measured in serum from peripheral blood and tumor blood in the patients with more advanced disease. However, the concentrations of suPAR in the body fluids were quite different, in serum the measured concentrations were between 46 and 98 pmol/liter, in ascites/peritoneal fluid concentrations were between 293 and 586 pmol/liter, and in cystic fluids the concentrations were even higher, that is 651-8468 pmol/liter. The concentrations of suPAR in cystic fluids clearly separated benign and malignant cysts with predictive values above 90%. The levels of suPAR in cystic fluids could therefore be used in the early diagnosis of ovarian cancer patients. The suPAR in the cystic fluids was present both in intact and cleaved forms and at least some of the suPAR(I-III) was not occupied by uPA [21]. In another study, tumor tissue, serum, ascites, and urine from ovarian cancer patients were analyzed for their content of the different uPAR forms. Whereas all of tumor lysates, ascites, and urine contained uPAR(I-III) and uPAR(II-III), domain I was only present in urine samples. In serum, only intact suPAR was detected [82]. The antibodies used for identification were mAb R3 (domain I) and mAb R2 (domain III).

5.6. Leukemia

The uPAR expression on leukemic cells was compared to that on peripheral blood and bone marrow cells from healthy individuals in two studies using flow cytometric analysis. Both studies used the domain I-specific anti-Mo3f (also termed 3B10) and VIM-5 mAbs [147, 148]. The uPAR expression on neutrophils and monocytes in peripheral blood from healthy donors was verified in both studies and found uPAR expressed neither on resting B nor on resting T lymphocytes. uPAR expressed on monocytes and neutrophils could bind the added ATF of uPA [147], showing that at least some of the uPAR molecules present on these cells were unoccupied. When blasts from AML patients were analyzed, they were found to express uPAR and the amount of uPAR correlated with the French-American-British (FAB) myeloid leukemia classification [149]. Most uPAR was found on cells of the M5 subgroup. Interestingly, intracellular pools of uPAR were detected in many blast cells of lower FAB classification, even though no uPAR was detected on the surface of these cells. Only 12% of AML patients had uPAR positive blast cells, suggesting that uPAR could be used for leukemia typing. In bone marrow, CD34 negative myelomonocytic precursor cells expressed uPAR, whereas all CD34+ cells from bone marrow and mobilized peripheral blood CD34+ cells were uPAR negative [148].

In addition to the cellular expression on malignant blast cells in AML, elevated levels of suPAR were found in plasma from leukemia patients [18]. In a longitudinal study, in which patients receiving chemotherapy were monitored, it was demonstrated that the suPAR level in plasma from patients with AML correlated with the number of circulating tumor cells and that these were reduced after chemotherapy. In plasma from AML patients, suPAR(II-III) was detected in addition to intact suPAR. This is in contrast to findings in plasma from healthy individuals and from the ovarian cancer patients described above [144]. suPAR(II-III) was also present in plasma made from bone marrow aspirates. The other cleaved form, uPAR(I), was only identified in urine. Lysates of the leukemic cells contained both intact uPAR and uPAR(II-III). The amounts of suPAR(II-III) in plasma and uPAR(I) in urine were decreased following chemotherapy. In healthy controls, intact uPAR was detected in lysates from mononuclear cells in blood and suPAR(I-III) in plasma and bone marrow aspirates, while suPAR(II-III) was detected in urine [18].

The granulocyte colony-stimulating factor (G-CSF) induces hematopoie-tic stem cell (HSC) mobilization into peripheral blood and this stimulation also induces uPAR expression on some cell types but not on others [19]. It was confirmed, using mAbs R2 and R4, that CD34+ cells with or without G-CSF stimulation and B and T lymphocytes do not express uPAR [147,148]. However, uPAR expression was increased on myeloid precursors and CD14+ monocytes in donors after G-CSF administration. G-CSF induced an increase in the amount of uPAR(I-III) but not of uPAR(II-III) on the surface of peripheral blood mononuclear cells. The content of suPAR in serum collected from the donors was also increased after G-CSF stimulation and both intact and cleaved forms of suPAR were present. suPAR(II-III) containing amino acids 88-92 is able to induce monocyte chemotaxis by activating the low-affinity receptor for fMLP, as discussed in Section 2.3 [107]. This implies that the upregulation of uPAR by G-CSF results in production of the chemokine suPAR(II-III), resulting in increased mobilization of the CD34+ HSCs into the peripheral blood [19].

The Smoker's Sanctuary

The Smoker's Sanctuary

Save Your Lungs And Never Have To Spend A Single Cent Of Ciggies Ever Again. According to a recent report from the U.S. government. Centers for Disease Control and Prevention, more than twenty percent of male and female adults in the U.S. smoke cigarettes, while more than eighty percent of them light up a cigarette daily.

Get My Free Ebook


Post a comment