METCAM/MUC18 is a new early diagnostic biomarker for the malignant potential of prostate cancer: Validation with Western blot method,enzyme-linked immunosorbent assay and lateral flow immunoassay

Abstract

BACKGROUND:

METCAM/MUC18 expression was increased with the malignant progression of prostate cancer and also a bona fide metastatic gene, capable of initiating and driving the metastasis of a non-metastatic human prostate cancer cell line to multiple organs.

OBJECTIVE:

We explored if METCAM/MUC18 was detectable in human serum and a novel biomarker to predict malignant propensity of prostate cancer.

MATERIALS AND METHODS:

Two antibodies were identified by Western blot analysis having the highest sensitivity and specificity to establish calibration curves from the recombinant METCAM/MUC18 proteins. They were used in ELISA and LFIA to determine the METCAM/MUC18 concentrations in serum samples from 8 normal individuals, 4 BPH patients, 1 with PIN, 6 with high-grade prostate cancer, and 2 treated cancer patients.

RESULTS:

Serum METCAM/MUC18 concentrations were statistically significantly higher in the patients with PIN and prostate cancer than those with BPH, the treated patients and normal individuals. The LFIA results were statistically better than ELISA and Western blot methods. Serum METCAM/MUC18 concentrations were in direct proportional to most of serum PSA concentrations.

METCAM/MUC18 antigen was detectable in human serum samples. METCAM/MUC18 may serve as an early diagnostic biomarker for malignant potential of prostate cancer and at least complement the PSA test.

Keywords: METCAM/MUC18, Western blot, ELISA, gold nanoparticles, LFIA, malignant potential, prostate cancer

1. Introduction

Prostate cancer accounts for the 1 ${}^{\text{st}}$ leading cause of cancer in males and 4 ${}^{\text{th}}$ leading cause of cancer death in the world [1]. Prostate cancer in 10% of prostate cancer patients are aggressive and metastatic, killing the patients within one year of the initial diagnosis [2, 3]. The current dominant diagnosis for the cancer is to test an elevated serum PSA level. However, the diagnosis test has at least 20–25% false results [13]. Furthermore, the elevated level of PSA in the serum is not always predictive of the pathologic stage of the prostate cancer or the presence of a metastatic disease [4]. Many potential diagnostic markers for prostate cancer progression have been patented [5]; however, only two, such as Prostate Health Index and PCA3, are approved by Food and Drug Administration for the detection of prostate cancer and most of them cannot accurately distinguish between indolent and aggressive cancers [6]. As such, there is still an urgent need to search for a better marker for the early detection of the metastatic potential of prostate carcinomas. The new diagnostic marker may complement or replace the current serum PSA test. From our previous research we suggest that METCAM/MUC18 is a driver for the malignant progression of prostate cancer [7, 8, 9, 10, 11, 12] and may be used as a novel diagnostic biomarker for early detection of the metastatic potential of human prostate cancer and perhaps for distinguishing the aggressive cancers from indolent ones [13].

We have previously shown that the expression of METCAM/MUC18 antigen is positively correlated with the malignant progression of clinical prostate cancer, suggesting that it may be a novel diagnostic biomarker for predicting the metastatic potential of human prostate cancer [13]. Furthermore, we showed that it is also a bona fide metastatic gene, capable of initiating and driving the metastasis of a non-metastatic human prostate cancer cell line to multiple organs [9, 10, 11]. As such, it very likely plays a role in converting indolent prostate cancer to an aggressive one [13]. The purpose of this research is to use three immunological methods, such as the Western blot (WB) analysis [14], an enzyme-linked immunosorbent assay (ELISA) [14, 15], and a lateral flow immunoassay (LFIA) method [16, 17, 18], to investigate the above possibility that METCAM/MUC18 may be used as a novel diagnostic biomarker for predicting the metastatic potential of human prostate cancer and for differentiating the indolent prostate cancer from the aggressive one. We found that the serum huMETCAM/MUC18 concentrations were statistically significantly higher in the patients with PIN and with prostate cancer at different pathological stages than those with BPH, the treated and normal individuals [15, 18]. We also found that the serum huMETCAM/MUC18 concentrations were also directly proportional to serum PSA concentrations when PSA was $<$ 25 ng/ml. Furthermore, the results yielded from LFIA were statistically better than an ELISA and the Western blot method. We conclude that METCAM/MUC18 antigen is shown to be detectable in human serum samples by using three immunological methods and it may serve as an early diagnostic biomarker for clinical prostate cancer and at least complement the PSA test [18]. Preliminary results have been presented [19].

Table 1
Epitopes recognized by various antibodies

Antibodies	Size of METCAM/	Location of epitopes	Recombinant protein huMETCAM/	Sources
	MUC18 (WB)	(#amino acid)	MUC18 recognized by WB ${}^{*}$
Chicken Ab	113–150 kDa	212–374	M	Home-made
Goat Ab SC-18940	?	Internal region	Not detected (No Good)	Santa Cruz
Rabbit Ab SC-28667	110–130	586–646	C	Santa Cruz
Rabbit Ab MBS275688	$<$ 130 kDa	212–234	M	MyBiosource
Rabbit Ab MBS2518792	72 kDa	345–379	M (No good)	MyBiosource
Rabbit Ab MBS2529469 or	120 kDa	26–350	N	MyBiosource or
EPP11278				Elabsource

${}^{*}$ Recombinant proteins of human METCAM/MUC18: N terminus: aa# 1-212, M portion: aa# 212-374, NM portion: aa# 1-374, C-terminus: aa# 375-646.

2. Materials and methods

2.1 Materials

Chicken anti-human METCAM/MUC18 antibody was home-made [20]. Rabbit anti-METCAM/MUC18 antibody (SC28667) and goat anti-METCAM/MUC18 antibody (SC18940) were from Santa Cruz Biotech. Rabbit anti-METCAM/MUC18 antibodies (MBS2756 88 and MBS2518792) were from MyBiosource (Table 1). Instant non-fat dry milk powder (NFM) (8 g of protein per 23 g of dry powder, Sunny Select) was purchased at the Lucky Super Market, Novato, CA, USA. PCR adhesive PCR sealing film (cat # BL-6174) was from Basic Life. For LFIA tests, nitrocellulose membrane (Cat # M10-00101), sample pad (Cat # M10-01401), conjugated pad (Cat # M10-00701), absorbent pad (Cat # M10-01701), and plastic backing pad (Cat # M10-00400) were from Rega Biotech. Colloidal gold (40 nm) (Cat # RE-30013001) was from Rega biotech.

2.2 Preparation of recombinant METCAM/MUC18 proteins

The recombinant METCA/MUC18 proteins were obtained as described [20, 21]. In brief, the intact human or mouse METCAM/MUC18 cDNA was PCR-amplified and cloned to pGEM-T or pGEM-T Easy, then to pcDNA3.1 $+$ vector, and cleaved with different restriction enzymes, and cloned in-phase to a GST vector (pGEX-6P1, Pharmacia Biotech). The recombinant clones were over-expressed in E. coli BL21 cells, which were lysed with a French Press. The recombinant METCAM/MUC18-GST fusion proteins were purified by binding to an affinity column containing glutathione-S-Sepharose 4B and eluted from the column with reduced glutathione. If only the recombinant METCAM/MUC18 protein was to be obtained, the fusion protein bound on the affinity resin was treated with HRV-3C Prescission enzyme at 4 C overnight, eluted with PBS, and further purified with a Superdex-200 HR 10/30 column in an FPLC and the fractions were finally pooled and concentrated by Centricon or Centriprep-50. The protein concentration of each finally purified recombinant protein was determined by measuring the A ${}_{280}$ in a spectrophotometer and confirmed by a Biuret method (BioRad) and verified by staining after SDS-PAGE The recombinant proteins were kept frozen at $-$ 20 ${}^{\circ}$ C till use. Four purified recombinant proteins, NTD-GST (aa # 1-212), M (aa # 212-374) or M-GST, NM-GST (aa # 1-374), and C-terminus-GST (aa # 374-646), were obtained.

2.3 Preparation of human serum samples

Human whole bloods were obtained from voluntary donors with a consensus at CYCU with an approved IRB (20130510-20) or from patients at Department of Urology, NTU Hospital with an approved IRB (201406098RIND). 15 or 30 ml of whole blood per person was withdrawn into one or two 16 ml conical centrifuge tubes without heparin and allowed to coagulate at room temperature for 30–60 min. Then serum was obtained by centrifugation at 1,000 rpm (180 $\times$ g) in a S4180 swing out rotor in a table-top centrifuge for 20 min. The upper phase was further centrifuged for an additional 10 min. The clear yellowish supernatant was distributed in one-ml aliquots and centrifuged again at 14,500 rpm for one minute to remove residual RBC and WBC. The final clear yellowish serum was aliquoted in 0.5 or 1 ml portions in several 1.5 ml Eppendorf centrifuge tubes and stored in $-$ 20 C freezer until use for immunological assays.

Figure 1.

Determination of METCAM/MUC18 concentrations in human serum samples by the Western blot method. Each human serum sample was analyzed by the Western blot (WB) method as described in “Materials and Methods.” The intensity of huMETCAM/MUC18 band in each serum sample was determined by Image J and compared to that of the SK-Mel-28 cell lysate. The METCAM/MUC18 band is indicated with an arrow. All the numbers on top of each lane were the PSA concentration and all the numbers under each lane were the % intensity compared with that of the SK-Mel-28 cell lysate (100%) in lane 1.

2.4 Western blot analysis

Western blot was carried out [14] with slight modifications [7, 8, 9, 11, 12, 20, 21]. One volume of serum was mixed with 1/3 volume of 4X WB buffer and kept frozen till use. 10 $\mu$ l of each treated serum was mixed with an equal volume of 1X WB buffer, added 5 $\mu$ l of the 5X sample buffer, mixed, boiled for 5–10 min, loaded to the well of a slab gel, and subjected to electrophoresis. The gel was then electro-blotted to a nitrocellulose membrane. The METCAM/MUC18 band was identified by reacting with 50 $\mu$ g/ml (1/300 dilution) of our home-made chicken anti-METCAM/ MUC18 antibody [20, 21], 2 $\mu$ g/ml (1/500 dilution) of the rabbit anti-METCAM/MUC18 antibody (MBS 275688), 5 $\mu$ g/ml (1/200 dilution) of the rabbit anti-METCAM/MUC18 antibody (SC28667), or 6.7 $\mu$ g/ml (1/150 dilution) of the goat anti-METCAM/MUC18 antibody (SC18940), washed, subsequently reacting with 0.5 $\mu$ g/ml (1/2500 dilution) of an AP-conjugated rabbit anti-chicken antibody, goat anti-rabbit antibody, or rabbit anti-goat antibody (Chemicon), and finally with NBT/BCIP for a color reaction. The METCAM/MUC18 band on the membrane was imaged by an Epson Perfection V330 Photo scanner and quantified by the NIH Image J software. The relative intensity of METCAM/MUC18 in each serum sample on the nitrocellulose membrane after Western blot analysis (as shown in Fig. 1) was converted to concentration after comparing that to the known amount of METCAM/MUC18 in the SK-Mel-28 cell lysate (as a positive control). WB analysis of the serum METCAM/MUC18 was repeated 11 times each by two persons.

2.5 ELISA

Standard direct ELISA [14] was used with slight modifications [15]. One volume of serum was mixed with an equal volume of 2X CB (1/2X serum) or with an 1/3 volume of 4X CB (3/4X serum). 1X CB was composed of 0.1 M carbonate buffer, pH 9.6, 2X CB of 0.2 M carbonate buffer, pH 9.6, and 4X CB of 0.4 M carbonate buffer, pH 9.6. Most ELISA was carried out by the direct method and some by an indirect method [14, 15]. In brief, 0.4–6.3 ng of M protein, 0.42–6.77 ng of NM-GST fusion protein, or 0.23–3.6 ng of C-terminal-GST fusion protein in 0.05 ml of 1X CB buffer, or a triplicate of 0.05 ml of 3/4X human serum in 1X CB was pipetted to each of the 96-well, covered with an adhesive PCR seal- ing film, fixed for overnight at 4 ${}^{\circ}$ C, washed with DB-NFM (0.05% Tween 20, 0.25% BSA, 0.1% Na azide in 1X PBS containing 5% NFM) by using an ELISA washer, and incubated with 0.2 ml/well of DB-NFM for 1 hr at room temperature. After removal of the DB-NFM buffer, each well was added 0.05 ml of 25 $\mu$ g/ml (1/600 diluted with DB-NFM) chicken anti-METCAM/MUC18 antibody, or that of 2 $\mu$ g/ml (1/500 diluted with DB-NFM) rabbit anti-METCM/MUC18 antibody (MBS275688), and incubated at 37 ${}^{\circ}$ C for 30 min. Then, each well was washed twice with 0.2 ml of WB (0.05% Tween 20 in 1X PBS), added 0.05 ml of 0.4 $\mu$ g/ml (1/2500 diluted with DB-NFM) rabbit anti-chicken AP-conjugated Ab (AP162A), or goat anti-rabbit AP-conjugated Ab (AP132A), incubated at 37 ${}^{\circ}$ C for 1 hr, washed twice with 0.2 ml of WB, and rinsed twice with 0.05 ml of 10 mM Tris HCl, pH 9.5–0.5 mM MgCl ${}_{2}$ , and the plate was inverted on top of a paper towel. Color was developed by addition of 0.05 ml/well AP substrate solution (5 mg of p-nitrophenyl phosphate in 5 ml of 10 mM ethanolamine-0.5 mM MgCl ${}_{2}$ , pH 9.675). The absorbance of each well was read at 405 nm in an ELISA reader (BioTek) after 3 hrs and 16–20 hrs. The data were stored and analyzed. ELISA analysis of the serum METCAM/MUC18 was repeated 10 times each by three persons.

2.6 LFIA

LFIA is useful for immunoassay on a nitrocellulose strip/membrane [16, 17] and was carried out with slight modifications [18]. The test line contained about 26 ng (0.53 $\mu$ l) per 0.4 cm strip of the (50 $\mu$ g/ml) 1/300 diluted chicken anti-METCAM/MUC18 antibody with DB-NFM) and the control line contained about 5.3 ng (0.53 $\mu$ l) per strip of the 10 $\mu$ g/ml (1/100 dilution) secondary antibody of goat anti-rabbit antibody with DB-NFM were printed on a 30 $\times$ 2.5 cm membrane with a Agitest ${}^{\text{TM}}$ RP-100 printing machine (Rega Biotech), baked at 37 ${}^{\circ}$ C for 30 min, and kept at 30% humidity in a dehumidifier for one day before cut into 75 strips of 0.4 cm $\times$ 2.5 cm with a computer-operated automatic micro-cutter machine (model JS-101, HSHUENN, Taiwan). Each strip was used for the assembly of LFIA. Conjugation of gold nano-particles to a primary antibody (Cat #MBS275688) was carried out in ice bath (at 0 ${}^{\circ}$ C) as described next. 400 $\mu$ l (15.72 $\mu$ g) (A ${}_{525}$ $=$ 1.008, about 39.3 $\mu$ g/ml) of colloidal Gold (40 nm nanoparticles) was slowly dripped into a 390 $\mu$ l LF buffer, which contained 1X PBS, 10 mM Tris base, 0.15 M NaCl, 10 mM MgCl ${}_{2}$ , 1 mM ZnCl ${}_{2}$ , 10% BSA, 0.1% Triton X100, 0.02% Na Azide), added 10 $\mu$ l of 10 $\mu$ g/ml (1/100 dilution) the primary rabbit antibody (MBS275688), mixed, and shaken for 2–2.5 hours at 0 ${}^{\circ}$ C. The mixture was then centrifuged at 7,000 rpm for 20 min or 14,000 rpm for 5 min in a cooled high-speed micro-centrifuge (model Hettich Zentrifugen MIKRO 200R, Hettich Lab Technology, Germany). The pellet was washed with once with 400 $\mu$ l of cold 10 mM Tris.HCl buffer (pH 7.6), and finally suspended in 20 $\mu$ l of LF buffer. The absorption peak of the gold nanoparticles had a slight shift from 525 to 530 nm, indicating there was no aggregation after the antibody conjugation. 10 $\mu$ l of antibody (about 0.05 $\mu$ g)-gold nanoparticles (about 7.85 $\mu$ g) suspension was spotted on the conjugate pad and followed by 10 $\mu$ l of antigen or serum mixed with 50 $\mu$ l of LF buffer applied to the sample pad. The LFIA assembly was developed by 3–4 applications of 40–60 $\mu$ l of LF buffer, disassembled, and only the nitrocellulose membrane strip was allowed to develop color and dried at room temperature. The bands in the test line and the control line were scanned with an Epson Scanner and the intensities were determined by the Image J version 3.1. LFIA analysis of the control proteins was repeated 12 times and that of serum METCAM/MUC18 was repeated 7 times.

2.7 Statistical analysis

Standard deviation of each set of data and R2 of two different sets of data were analyzed by the build-in programs in the Excel. The one-way ANOVA method in the SSPS software was used for statistical analysis among several sets of data. The difference among different sets of data was considered statistically significant if $P$ value was $<$ 0.05.

Figure 2.

HuMETCAM/MUC18 concentrations in various human serum samples determined by the Western blot method and its relation with the status of patients. The average METCAM/MUC18 concentrations with standard deviations in the serum samples from the treated prostate cancer patients and patients with prostate premalignant PIN, or BPH, were compared to those from normal individuals. The number on top of each bar was the $p$ value, which was obtained by comparing the data to the normal individuals by statistical analysis as described in “Materials and Methods.”

Figure 3.

Relation of PSA to the huMETCAM/MUC18 concentrations determined by the WB method by using our home-made chicken antibody. The serum METCAM/MUC18 concentrations were compared to PSA. (A) shows 7 serum METCAM/MUC18 concentrations versus PSA (left panel) and PSA versus serum METCAM/MUC18 (right panel) when PSA was from 0.36 to less than 7 ng/ml, (B) shows 11 serum METCAM/MUC18 concentrations versus PSA (left panel) and PSA versus serum METCAM/MUC18 (right panel) when PSA was from 0.36 to less than 7 ng/ml, (C) shows 13 serum of that when PSA was from 0.36 to less than 25 ng/ml, and (D) shows 15 serum of that when PSA was from 25 to 1219 ng/ml.

Figure 4.

Established Standard curve of recombinant METCAM/MUC18 proteins determined by an ELISA by using our home-made chicken antibody (A) and the commercial rabbit antibody (MBS275688) (B). (A) shows the intensities of various recombinant METCAM/MUC18 proteins, versus their concentrations by using chicken antibody for ELISA. (B) shows those by using the rabbit antibody MBS275688 for ELISA.

3. Results

3.1 Determination of METCAM/MUC18 concentrations in human serum samples by using quantitative WB method

Figure 1 shows that human METCAM/MUC18 with a molecular weight of about 130 KDa was detectable in various human serum samples by WB analysis. Figure 2 shows the quantitative results that the METCAM/MUC18 protein concentrations in malignant prostate cancer patients were significantly higher than those in normal serums and BPH ( $p=$ 0.006). Figure 3 shows that there was a linear relationship between the serum METCAM/MUC18 concentrations and the serum PSA, when PSA was between 0.36 and less than 7 ng/ml (Fig. 3A and B) and an exponential relationship between them when PSA was between 0.36 and less than 25 ng/ml (Fig. 3C). However, there was no correlation between them when PSA was between 25–1219 ng/ml (Fig. 3D).

Table 2
Patient (male) characteristics ( $N=$ 19)

Characteristics	Age	No. of cases	%
Normal ( $N=$ 8)
Age (years) Range	28–75
Age (years) Median	54
BPH ( $N=$ 4)
Age (years) Range	68–72
Age (years) Median	70
PIN ( $N=$ 1)
Age (years) Range	63
Age (years) Median	63
Prostate carcinoma ( $N=$ 6)
Age (years) Range	50–78
Age (years) Median	69
Pathology (classification)
Normal		8	42
BPH		4	21.1
PIN		1	5.3
Prostate carcinoma		6	31.6
(Gleason scores of
3 $+$ 4, 4 $+$ 4, 4 $+$ 5)
Radiotherapy
Yes		2	30
No		4	70

Figure 5.

HuMETCAM/MUC18 concentrations in various human serum samples determined by using chicken antibody in the ELISA and its relation with the status of patients. The average METCAM/MUC18 concentrations with standard deviations in the serum samples from the treated prostate cancer patients and patients with prostate cancer, or BPH, were compared to those from normal individuals. The number on top of each bar was the $p$ value which was obtained by comparing the data from various patients to those from the normal individual by statistical analysis.

Figure 6.

Relation of PSA to the huMETCAM/MUC18 concentrations determined by ELISA by using our home-made chicken antibody (A) and the commercial rabbit antibody MBS275688 (B and C). The serum METCAM/MUC18 concentrations were compared to PSA. (A) shows 16 serum METCAM/MUC18 concentrations versus PSA by using our home-made chicken antibody, when PSA was from 0.27 to less than 25 ng/ml. (B) shows 15 serum METCAM/MUC18 concentrations by using the rabbit antibody for ELISA versus PSA, when PSA was from 0.27 to less than 25 ng/ml. (C) shows 17 serum METCAM/MUC18 concentrations by using the rabbit antibody for ELISA versus PSA, when PSA was from 0.27 to 1219 ng/ml.

Figure 7.

Established Standard curve of recombinant METCAM/MUC18 proteins determined by LFIA by using our chicken antibody. (A) shows the signals of NM-GST (the positive control protein) and C-terminus-GST (the negative control protein) on the test line of the membrane. The concentrations (ng/ml) of the recombinant proteins are indicated on the top of each lane. (B): The left panel shows the intensities of the two recombinant METCAM/MUC18 proteins (NM-GST and C-terminus GST) versus their concentrations by using chicken antibody for LFIA. The right panel shows the standard curve derived from the data in the left panel after subtraction of the intensity of the C-terminus GST from that of NM-GST.

Figure 8.

HuMETCAM/MUC18 concentrations determined by using chicken antibody for the LFIA in various human serum samples and its relation with the status of patients. (A) shows the signals of each serum sample on the test line of the membrane. The numbers on the top of each lane were the PSA concentrations (ng/ml). (B) shows the average METCAM/MUC18 concentrations with standard deviations in the serum samples from the treated prostate cancer patients, patients with prostate cancer, or BPH, were compared to those from normal individual. The number on top of each bar was the $p$ value from statistical analysis.

Figure 9.

Relation of PSA to the huMETCAM/MUC18 concentrations determined by LFIA by using our home-made chicken antibody. (A) shows 6 serum METCAM/MUC18 concentrations versus PSA, when PSA was from 0.02 to 0.85 ng/ml. (B) shows 10 serum METCAM/MUC18 concentrations versus PSA, when PSA was from 0.02 to 1219 ng/ml.

Figure 10.

PSA concentrations in various patients. The figure shows the average PSA concentrations with standard deviations in the serum samples from various persons. The number on top of each bar was the $p$ value from statistical analysis.

3.2 Identification of antibodies with the highest specificity and sensitivity to recognize the epitopes of human METCAM/MUC18 proteins [15]

A good antibody should recognize the full size huMETCAM/MUC18 in the lysates made from tumor cell lines of SK-Mel-28, DU145 and PC-3, and the proper recombinant proteins in the WB analysis. Table 1 shows the summary of the epitopes recognized by our home-made chicken antibody and four commercial antibodies used for the test. Our home-made chicken antibody and the rabbit antibody MBS275688 met our criteria and were further used for the following ELISA and LFIA. However, the other two commercial antibodies did not meet our criteria and were not used.

3.3 The human METCAM/MUC18 proteins in serum samples of various patients determined by an ELISA

We then determined the best concentrations of the two primary antibodies and secondary antibodies for obtaining optimal ELISA signals. We found that 25 $\mu$ g/ml of our home-made chicken antibody and 2 $\mu$ g/ml of the MBS275688 rabbit antibody were sufficiently sensitive as the primary antibodies and so as 0.4 $\mu$ g/ml of the secondary antibodies, such as the rabbit anti-chicken antibody and goat anti-rabbit antibody [15]. Third, we established a protein calibration curve for ELISA by using different concentrations of five recombinant proteins: the purified M (the middle region), and NM-GST (the N and M regions fused to GST) as the positive controls, and the purified NTD-GST (the N-terminus region fused to GST), C-terminus-GST (the C-terminus region fused to GST), and GST as the negative controls [15]. Figure 4 shows that our home-made chicken antibody could differentiate between the positive control and the negative control proteins with a 5-fold difference (left panel); in contrast, the commercial rabbit antibody MBS275688 could only differentiate the two control proteins at best with a 1.7-fold difference (right panel). Thus, the calibration curves of recombinant METCAM/MUC18 proteins were successfully established from our home-made chicken antibody and the commercial rabbit antibody, and furthermore, our home-made chicken antibody was better than the commercial rabbit antibody. By using these established calibration curves, it was possible to quantify the serum METCAMMCUC18 concentrations in 19 human serum samples obtained from normal individuals, BPH patients, and patients at different stages of prostate cancer (Table 2). Figure 5 shows that the METCAM/MUC18 protein concentrations in malignant prostate cancer patients were somewhat significantly higher than those in normal serums and BPH. Figure 6 shows that the serum METCAM/MUC18 concentrations were linearly proportional to most of the PSA concentrations when serum PSA concentrations were between 0.27 and $<$ 25 ng/ml when either our home-made chicken antibody (Fig. 6A) or the commercial rabbit antibody (MBS275688) (Fig. 6B) was used for ELISA. Furthermore, the serum METCAM/MUC18 concentrations were also linearly proportional to most of the PSA concentrations when serum PSA concentrations were between 25 and 1219 ng/ml (Fig. 6C).

3.4 The human METCAM/MUC18 proteins in serum samples of various patients determined by an LFIA [18]

Figure 7A shows that the intensity of the images in the test line of LFIA for the positive control recombinant protein was visually significantly stronger than that of the negative control protein. Figure 7B shows the quantitative results that our home-made chicken antibody could differentiate the positive control protein (NM-GST) from the negative control protein for a five-fold difference in intensity ( $P=$ 0.021), as shown in the left panel of Fig. 7B. After subtracting the background intensity of the negative control from the positive control, a calibration curve of the concentrations of the positive control recombinant METCAM/MUC18 protein versus the intensity was established (Fig. 7B, right panel). The calibration curve was then used for determining the METCAM/MUC18 concentrations in various human serum samples. Figure 8 shows that the average serum METCAM/MUC18 concentrations in prostate cancer patients were statistically significantly higher than those in normal individuals ( $p=$ 0.013), especially those with high-grade prostate cancer ( $p=$ 0.0004). Figure 9 shows the METCAM/MUC18 concentrations determined by LFIA were linearly proportional to that of PSA, when the PSA concentrations were between 0.02 and $<$ 0.85 ng/ml and also when the PSA concentrations were between 24 and 1219 ng/ml, but with a different slope.

4. Discussion

In this report, we presented evidence to show that it was possible to use three immunological methods to determine the serum METCAM/MUC18 concentrations and to differentiate those in the prostate cancer from normal individuals and patients with BPH. Furthermore, there was a positive correlation of METCAM/MUC18 with most of PSA in human serum. Our home-made chicken antibody was better than the commercial rabbit antibody MBS275688. The reason for our success was due to using chicken antibody for the tests instead of using antibodies made in mouse or rabbit, which had also been observed by one group [22]. In the near future we should expand our survey cohort in Taiwanese males and to increase sensitivity of our tests by using biotinylated antibodies [23, 24] to further improving the LFIA and by exploring the use of streptavidin-coated magnetic beads to enrich the antigen and removing the interference materials in the serum sample [24, 25]. Eventually, we may be possible to use the further improved LFIA to detect the presence of METCAM/MUC18 antigen in urine samples [26].

Each of the three methods has advantages and disadvantages. The major advantage of using the WB analysis was that the correct size of METCAM/MUC18 could be observed and quantified; however, the major disadvantages were that the quantitation was limited by the resolution of the Image J software, several costly electrophoretic equipment were required, and the procedures were tedious, time-consuming and not easily reproducible even with the trained personnel. The major advantages of using the ELISA method were that a large number of samples could be simultaneously processed on the 96-well plates and the quantitation sensitivity was excellent and highly reproducible, which was the gold standard for most diagnostic methods; however, the major disadvantages were that several costly equipment were also required and the procedures were time-consuming, tedious, and not easily reproducible with the train personnel. The major advantages of using LFIA were that the procedures were relatively simple and easy operation, user friendly and easily trained, rapid visual results with naked eyes, high reproducibility, and very cost-effective in comparison to the other two methods since no expensive equipment required; however, the major disadvantages were that the results were variable among materials from batch to batch, qualitative results were obtained rather than quantitative results (semi-quantitative), and resolution may be limited by different antibodies used.

In our hands, it was possible to use the LFIA for quantitative analysis, since among the three methods, the LFIA method was better than ELISA, which in turn was better than WB analysis, in that the standard deviations of the data in LFIA were much smaller and data were statistically better than the two other methods and the PSA test (as shown in Fig. 10). Furthermore, the linear proportionality between the METCAM/MUC18 and PSA concentrations in LFIA was better than the two other methods. Thus, LFIA has the high potential to be further improved to become a reproducible, simple, cost-effective diagnostic test for clinical purpose. At this time, we still could not differentiate malignant prostate cancers from indolent ones and also to monitor treatment outcome of the patients, which should await many years of follow-up checking our patients in the future.

Taken together, it is feasible and promising to use the METCAM/MUC18 as a diagnostic biomarker for developing a reliable, cost-effective, and rapid test for predicting the malignant progression of prostate cancer. The potential limitation for using the METCAM/ MUC18 as a diagnostic biomarker for predicting malignant potential of prostate cancer is that it is not prostate organ-specific, since it is also overly expressed in other cancers, such as angiosarcoma, breast cancer, gastric cancer, hepatocellular carcinoma, lung cancers, most melanoma, nasopharyngeal carcinoma type III, osteosarcomas, and pancreatic cancer; however, it is under expressed in some cancers, such as colorectal cancer, nasopharyngeal carcinoma type I, and ovarian cancer [27]. Nevertheless, it may be feasible to use the METCAM/MUC18 to at least complement the current biomarker PSA and to combine it with other prostate cancer biomarkers for predicting the malignant potential of prostate cancer [4].

Footnotes

Acknowledgments

We thank the supports from NSC (NSC-101-2320-B-033-003), Center for Biomedical Technology and Research Center for Circular Economy at CYCU, and the joined project of CYCU and Ten Chan General Hospital, Taiwan.

Conflict of interest

All the authors declare no conflict of interest.

Abbreviations

References

Siegel

R.L.

Miller

K.D.

and Jemal

, Cancer statistics 2019, CA Cancer J Clin 69 (2019), 7–34.

Wood

D.P.

Jr Banks

E.R.

Humphreys

McRoberts

J.W.

and Rangneker

V.M.

, Identification of bone marrow micro-metastases in patients with prostate cancer, Cancer 74(9) (1994), 2533–2540.

Catalona

W.J.

and Scott

W.W.

, Carcinoma of the prostate: a review, Journal of Urology 119(1) (1978), 1–8.

Liu

Yang

and Mao

, Current state of biomarkers for the diagnosis and assessment of treatment efficacy of prostate cancer, Discovery Medicine 27(150) (2019), 235–243.

Murphy

and Watson

R.W.

, Patented prostate cancer biomarkers, Nature Reviews Urology 9(8) (2012), 464–472.

Koharr

Petrovics

and Srivastava

, A rich array of prostate cancer molecular biomarkers: opportunities and challenges, Int J Mol Sci 20 (2019), 1813.

G.J.

Varma

V.A.

M.W.H.

Yang

Wang

S.W.C.

Liu

et al., Expression of a human cell adhesion molecule, MUC18, in prostate cancer cell lines and tissues, The Prostate 48 (2001), 305–315.

G.J.

, Chapter 7 The role of MUC18 in prostate carcinoma, in: Immunohistochemistry and in situ hybridization of human carcinomas Hayat

M.A.

ed., Vol. 2, Molecular pathology of lung carcinoma, breast carcinoma, and prostate carcinoma, Elsevier Sciences/Academic Press, New York, 2004, pp. 347–358.

G.J.

Peng

Wang

S.W.C.

Chiang

C.F.

Dillehay

D.L.

and Wu

M.W.H.

, Ectopical expression of human MUC18 increases metastasis of human prostate cancer cells, Gene 327(2) (2004), 201–213.

10.

G.J.

, METCAM/MUC18 and cancer metastasis, Current Genomics 6(5) (2005), 333–349.

11.

G.J.

Chiang

C.F.

Hess

W.J.

Greenberg

N.M.

and Wu

M.W.H.

, Increased expression of MUC18 correlates with the metastatic progression of mouse prostate adenocarcinoma in the TRAMP model, Journal of Urology 173(5) (2005), 1778–1783.

12.

G.J.

M.W.H.

and Liu

, Enforced expression of human METCAM/MUC18 increases the tumorigenesis of human prostate cancer cells in nude mice, Journal of Urology 185(4) (2011), 1504–1512.

13.

G.J.

, Human METCAM/MUC18 as a novel biomarker to drive and its specific SiRNAs to block the malignant progression of prostate cancer, J Cell Science and Therapy 6(5) (2015), 1000227. doi: 10.4172/2157-7013.1000227.

14.

Ausubel

F.M.

Brent

Kingston

R.E.

Moore

D.D.

Seidman

J.G.

Smith

J.A.

et al., In situ hybridization and immunohistochemistry, in: Current Protocols in Molecular Biology, New York: Green Publishing Associates and Wiley-Interscience Press, 1987; Section 14.

15.

H.W.

, Using immunological methods to determine human serum METCAM/MUC18 concentration for prediction of the pissible malignant potential of prostate cancer, M.S. thesis, Chung Yuan Christian University, Taoyuan, Taiwan, 2015. http://www.lib.cycu.edu.tw/thesis.

16.

Bahadır

E.B.

and Sezgintürk

M.K.

, Lateral flow assays: principles, designs and labels, Trends in Analytical Chemistry 82 (2016), 286–306.

17.

Parolo

de la Escosura-Muniz

and Merkoci

, Enhanced lateral flow immunoassay using gold nanoparticle loaded with enzymes, Biosensors and Bioelectronics 40 (2013), 412–416.

18.

C.K.

, Gold nanoparticles-based lateral flow immunoassay detection of METCAM/MUC18 concentration in human serum for its validation as a biomarker to predict the malignant progression of prostate cancer, M.S. thesis, Chung Yuan Christian University, Taoyuan, Taiwan, 2016. http://www.lib.cycu.edu.tw/thesis.

19.

R.J.C.

H.W.

C.K.

and Wu

G.J.

, A new early diagnostic biomarker for the malignant progression of prostate cancer, International Meeting on Chemical Sensors (IMCS2016), July 10–13, 2016, Jeju, Korea, 2016, Abstract # A-018.

20.

G.J.

M.W.H.

Wang

S.W.

Liu

Peng

et al., Isolation and characterization of the major form of human MUC18 cDNA gene and correlation of MUC18 over-expression in prostate cancer cells and tissues with malignant progression, Gene 279(1) (2001), 17–31.

21.

Yang

Wang

S.W.C.

Liu

M.W.H.

Armstrong

and Wu

G.J.

, Isolation and characterization of murine MUC18 cDNA gene, and correlation of MUC18 expression in murine melanoma cell lines with metastatic ability, Gene 265(1-2) (2001), 133–145.

22.

Zhong

Liu

Fong

and Zhong

, Development of a colloidal gold immunochromatographic strip for the rapid detection of soft-shelled turtle systemic septicemia spherical virus, J Virological Methods 221 (2015), 39–45.

23.

Xie

C.Y.

, Optimization of using gold nanoparticles-based lateral flow immunoassay for determination of the prostate cancer biomarker-METCAM/MUC18 concentration in human serum, M.S. thesis, Chung Yuan Christian University, Taoyuan, Taiwan, 2017. http://www.lib.cycu.edu.tw/thesis.

24.

Wei

Y.C.

, Comparing magnetic beads-enriched array display immunoassay with the two gold nanoparticles-based lateral flow immunoassay for determination of the prostate cancer biomarker-METCAM/MUC18 concentration in human serum, M.S. thesis, Chung Yuan Christian University, Taoyuan, Taiwan, 2018. http://www.lib.cycu.edu.tw/thesis.

25.

Chuang

Y.H.

, Using magnetic beads-enriched array display immunoassay and the gold nanoparticles-based lateral flow immunoassay for determination of the prostate cancer biomarker-METCAM/MUC18 concentration in human serum, M.S. thesis, Chung Yuan Christian University, Taoyuan, Taiwan, 2019. http://www.lib.cycu.edu.tw/thesis.

26.

Eskra

J.N.

Rabizadeh

Pavlovich

C.P.

Catalona

W.J.

and Luo

, Approaches to urinary detection of prostate cancer, Prostate Cancer and Prostatic Diseases 22(3) (2019), 362–381.

27.

G.J.

, METCAM/MUC18 promotes tumor progression and metastasis in most human cancers, in “Tumor Progression and Metastasis”, (A. Lasfar, editor) In Tech-Open Access Publisher, The Shard, 32 London Bridge Street, London SE1 9SG, United Kingdom ISBN: 978-1-78985-350-6. 2020, in press.

METCAM/MUC18 is a new early diagnostic biomarker for the malignant potential of prostate cancer: Validation with Western blot method,enzyme-linked immunosorbent assay and lateral flow immunoassay

Abstract

BACKGROUND:

OBJECTIVE:

MATERIALS AND METHODS:

RESULTS:

1. Introduction

Table 1 Epitopes recognized by various antibodies

2.1 Materials

2.2 Preparation of recombinant METCAM/MUC18 proteins

2.3 Preparation of human serum samples

2.5 ELISA

2.6 LFIA

2.7 Statistical analysis

3.1 Determination of METCAM/MUC18 concentrations in human serum samples by using quantitative WB method

Table 2 Patient (male) characteristics ( N = 19)

3.3 The human METCAM/MUC18 proteins in serum samples of various patients determined by an ELISA

3.4 The human METCAM/MUC18 proteins in serum samples of various patients determined by an LFIA [18]

4. Discussion

Footnotes

Acknowledgments

Conflict of interest

Abbreviations

References

Table 1
Epitopes recognized by various antibodies

Table 2
Patient (male) characteristics ( $N=$ 19)