JP2607492B2 - Human carcinoembryonic antigen - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION (Industrial Application Field) The present invention relates to human carcinoembryonic ant
igen; hereinafter abbreviated as human CEA) protein and a gene encoding the protein.

(Prior Art) Carcinoembryonic antigen (CEA) has been described by Gold and Freedman in 1965 as a human colon cancer.
It is found as an antigen commonly present in the digestive tract of 器 6 months of age (Gold, P. and Freedman, SO: J. Exp. Med., 121 , 43).
9,1965; Gold, P. and Freedman, SO: J. Exp.Med., 122 , 4
67, 1965), which initially attracted attention as an antigen specific to colorectal cancer. Subsequent studies have shown that CEA in the blood tends to occur more frequently in colorectal cancer patients, but it is also positive in other cancers,
It is also known that even a benign disease sometimes becomes positive, and even a normal person can detect a very small amount. At present, CEA is widely regarded as a substance that accompanies cancer, and is the most widely used tumor marker in clinical tests for early detection of tumors.

However, a significant problem in using CEA as a tumor marker is the presence of CEA-related antigens that are also present in normal tissues. CEA-related antigens are antigens that are very similar in protein chemistry to CEA and are considered to have a common antigenic determinant immunologically so that normal anti-CEA antibodies cannot be distinguished from CEA. Is a generic term for Representative CEA-related antigens include non-specific cross-antigens (nonss), which are glycoproteins with a molecular weight of about 90,000 found in the lungs and spleen of normal humans.
pecific cross-reacting antigen: hereinafter abbreviated as NCA)
(Von Kleist, S, et al., Proc. Natl. Acad. Sc.
i.USA, 69 , 2492,1972). Although the amino acid composition of NCA is similar to that of CEA, CEA contains relatively little methionine in NCA. According to the N-terminal amino acid sequence comparison, the amino acid sequence is identical to CEA up to the 24th position except that the valine at the 21st position is replaced with alanine (Engvall, E. et al., Pr.
oc.Natl.Acad.Sci, USA, 69, 2492,1972) . In addition,
NCA-2 found in fetal feces (Burtin, P. et al., J. Immun
ol., 111 , 1926,1973), NFA found in normal adult stool
(Normal fecal antigen) (Matsuoka, Y. et al., Gann, 64 , 20
3,1973). NFA consists of three molecular species, NFA-1, NFA-2 and NFCA (normal fecal cro
ss-reacting antigen), but none of its structure or physicochemical properties has been clarified (Kuroki, M. et al., Cancer Res., 41 , 713, 1981). Therefore, radioimmunoassay (RI
Various measurement kits such as A) or enzyme immunoassay (EIA) have been reported and are commercially available, but these kits have a problem that they react not only with CEA but also with CEA-related antigens. In other words, the antibody currently used for CEA measurement recognizes a common antigen portion of CEA and a CEA-related antigen, and thus has a drawback that the two cannot be measured separately. Furthermore, when the same CEA sample is measured with a conventional kit, there is also a problem that the measured value differs between the kits (Kuroki, M. et al., J. Immunol. Methods, 60 , 2).
21,1983).

In order to solve the above problems and to develop a method for specifically measuring only CEA, it is required to elucidate the structure of CEA and to produce a highly specific antibody. However,
CEA is a glycoprotein with a molecular weight of 180,000 to 200,000, 40 to 6 per weight
It is said to contain 0% sugar. Therefore, its high sugar content makes it difficult to analyze the structure of CEA, and only the amino acid sequence up to the 24th from the N-terminus has been disclosed (Glasman, JNS et al., BBRC, 85 , 209, 1978). And Ter
ry, WD et al., J. Exp. Med., 136 , 200, 1972). In addition, the sugar chain structure is hardly elucidated.

(Problems to be Solved by the Invention) To obtain an antibody that specifically reacts only with CEA, CEA
In addition, it is necessary to clarify the structure (amino acid sequence) of CEA-related antigens and find differences in their primary structures. Thus, the present inventors have conducted intensive studies to clone the human CEA gene and determine its DNA base sequence to clarify the entire amino acid sequence of human CEA.

(Specific Means for Solving the Problems) In order to clone the human CEA gene, the present inventors used cDNA extracted from the colorectal tissue of a colorectal cancer patient using cD
An NA library was prepared, the CEA gene was isolated and identified from the library, and its nucleotide sequence was analyzed and determined. Based on this elucidated nucleotide sequence, the entire structure of human CEA was elucidated, and the present invention was completed.

The gene encoding the human CEA protein of the present invention can be obtained as follows. First, tissue cells of a cancer patient,
Preferably, a cDNA library is prepared using mRNA produced by the colon tissue of a human colon cancer patient. The cDNA library may be prepared by a known appropriate method. For example, the method of Young et al. using a λgt11 phage vector (Youn et al.
g, RA et al., Proc. Natl. Acad. Sci. USA, 80 , 1194, 1983). The outline is shown in FIG. The desired clone (clone expressing human CEA protein) from the cDNA library (Escherichia coli) thus prepared
Can be performed using an existing (eg, commercially available) anti-CEA antibody. That is, the presence or absence of a reaction between the protein expressed by the prepared cDNA library and the anti-CEA antibody may be examined. Expression of the human CEA protein in Escherichia coli may be such that the CEA protein itself can be expressed.
It may be expressed as a fusion protein of a protein and another protein, preferably Escherichia coli β-galactosidase. In this case, the activity of β-galactosidase can be used as one indicator.

Next, phage DNA was isolated from the positive clone (Escherichia coli) screened as described above by a conventional method, and the target cDNA (human CEA cDNA) was analyzed by restriction enzyme analysis.
DNA) is inserted into the phage DNA, and the nucleotide sequence of the inserted cDNA is determined by a conventional method.
If a positive clone into which a cDNA of sufficient size has not been obtained is obtained, the cDNA obtained from a separate positive clone is used.
By binding NA to produce cDNA of desired size,
Alternatively, a new cDNA library is prepared, and a target clone is screened by the above-mentioned method using the insufficiently obtained cDNA obtained above as a probe. In the latter case, the cDNA library may be prepared by a known suitable method. For example, mRAN extracted from human cancer patient tissue cells (preferably E. coli patient tissue cells) is prepared by sucrose density gradient centrifugation. It can be prepared from the mRNA corresponding to about 2 kb to 5 kb by the Okayama-Burk method.

Next, the nucleotide sequence of the cDNA obtained as described above,
The human CEA gene and the translatable region (open reading region) corresponding to the amino acid sequence at the 24th position from the N-terminus of CEA, which were determined by the method of Sanger et al. The structure (amino acid sequence) of human CEA protein can be known. The human CEA gene whose nucleotide sequence has been determined in this way can also be obtained by chemical synthesis.

Furthermore, using the human CEA gene obtained as described above, a human CEA protein can be produced in a microorganism such as Escherichia coli or yeast or an animal cell. That is, an appropriate promoter region is added to the 5 'side of the human CEA structural gene obtained above, inserted into an appropriate plasmid, and then introduced into a microorganism such as Escherichia coli or yeast or an animal cell. And then culture. Such an operation can be performed using a known technique.

Further, by using the above-described fragment of the CEA gene (which may be chemically synthesized), specific antigen peptides excluding unnecessary peptide regions, for example, peptides common to CEA-related antigens, can be used for microorganisms and animal cells. It is also possible to manufacture with. And, using these peptides as antigens, human
A polyclonal or monoclonal antibody that specifically reacts only with CEA can be produced by a conventional method.

Hereinafter, the present invention will be described in more detail with reference to Examples.

Example (1) Preparation of cDNA library using λgt11 vector From 1.3 g of large intestine tissue of a colon cancer patient, Shagwin (Ch
irgwin, JM et al., Biochemistry, 18 , 5294, 1979).
1.37 mg of RNA was extracted using M guanidium thiocyanate. Next, 0.5M LiCl, 10mM E from 860μg of total RNA
Using 10 mM Tris-HCl (pH 7.2) containing DTA and 0.5% SDS as a binding buffer, 72 μg of poly (A) + RNA (mRNA) bound to oligo dT cellulose was separated (Nakazato, H .;
Et al., Meth. Enzym., 29 , 431, 1974). Then, poly (A) +
CDNA prepared using 10 μg of RNA and 0.25 μg of oligo (dT) _12-18 as a primer was incorporated into a λgt11 vector to prepare a cDNA library (Young, RA et al., Proc.
Natl. Acad. Sci. USA, 80 , 1194, 1983). Poly (A) + RNA1
About 48,500 recombinant phage plaques (cDNA library) were obtained per μg. FIG. 1 shows an outline of a method for preparing a cDNA library using the λgt11 vector.

(2) Separation and Analysis of Positive Clones by Antibody Screening Antibody screening was performed by the following method using a kit (Express-Blot ^™ Assay Kit) commercially available from Bio-Rad.

On the plate of about 300,000 phage plaques obtained in the above (1), impregnated with 10 mM IPTG, placed on an air-dried nitrocellulose filter, and inserted the gene product from the inserted cDNA (i.e. Protein). After about 2 hours, the filter was removed from the plate, and a TBS solution (20 mM Tris-HCl, pH 7.5, 500 mM NaC
l) Shake gently in for 5 minutes. Next, the filter is placed in a blocking solution (TBS solution containing 3% gelatin) for 2 hours.
Shake gently for 0-30 minutes. Next, the filter was washed by gently shaking twice in a T-TBS solution (TBS solution containing 0.05% Tween-20) for 5 minutes.

This filter was subsequently dipped into the primary antibody solution,
Shake gently all night. The primary antibody solution is
1/500 volume of anti-CEA antibody (primary antibody), 1/200 volume of 1% E. coli extract (Bio-Rad, catalog No. 170-3205) and 1% gelatin (Bio-Rad, catalog No. 170)
-6537). Usako anti-CEA antibody from Dako (DAKOPATTS, Denmark, code A115) was used as the primary antibody.

Next, the filter was gently shaken twice in the T-TBS solution for 5 minutes, then dipped in the secondary antibody solution, and gently shaken for 2 hours. The secondary antibody solution is a peroxidase-labeled goat anti-rabbit Ig antibody (Bio-Rad, Catalog N
o. 170-6513) with 1/3000 volume and 1% gelatin
-A TBS solution was used.

Subsequently, the filter was gently shaken twice in the T-TBS solution for 5 minutes, then dipped in the coloring solution, and gently shaken for 10 minutes. The coloring solution is a solution in which 50 mg of an HRP coloring reagent (Bio-Rad, Kataroku No. 170-3201) is dissolved in 16.5 ml of cold methanol, and 30% H ₂ O ₂ (hydrogen peroxide)
This is a solution in which 83.5 ml of a TBS solution containing 50 μm is mixed immediately before use.

As a result of the above operation, one purple-colored positive clone was obtained. This phage DNA named λKr40 was separated by a conventional method, and a cDNA fragment (about 2.4 kb) obtained by cutting at the restriction enzyme EcoR1 sites attached to both ends of the inserted DNA was subcloned into pBR322. With respect to the obtained plasmid pBRCEA5, the entire base sequence of the inserted cDNA was determined.

FIG. 2 shows the restriction enzyme cleavage site used for the nucleotide sequence determination of the inserted DNA region and the strategy for nucleotide sequence determination. In the figure, Pv is Pvu II, Ps is Pst I, S is Sac I, B is BamH I,
Hc indicates each restriction enzyme of Hinc II, and the upper number indicates the number of bases from the 5 'end. The blank rectangle indicates the region encoding mature human CEA in the translatable region, and the rectangle with a horizontal line to the left indicates the region encoding the leading peptide (signal peptide). In the figure, the direction of determination of the base sequence is indicated by an arrow. The strategy for sequencing was determined by the method of Dale et al. (RMKDale et al., Plasmids, 13 , 31, 19).
85) and the method of Sanger et al. (Proc. Natl. Acad. Sci. US
A, 74 , 5463, 1977). For λ Kr40, the solid line indicates that the method of Dale et al., the dotted line indicates that the nucleotide sequence was determined by the Sanger method after inserting the restriction enzyme fragment into M13, and in pCEA55-2 and pCEA80-11 described later, the arrow line from the vertical line (|
→) indicates the method of Sanger after the restriction enzyme fragment was inserted into pUC9, and the arrow (□ →) from the square indicates the base sequence by the method of Sanger using the synthetic oligonucleotide corresponding to each position indicated by the square as a primer. Indicates that the decision has been made.

The inserted DNA of λKr40 was composed of 2356 bases, and its DNA sequence contained a long translatable sequence (open reading frame) capable of encoding 531 amino acids (base sequence Nos. 501 to 2856 in FIG. 3). Number). However, ATG which is a codon of methionine which is an N-terminal amino acid was not found, and it was found that some N-terminal amino acids were deleted.

(3) Analysis Using Various Anti-CEA Antibodies To confirm that the cDNA obtained above contains the main part of the CEA gene, the same as (2) above using phage λKr40 containing the cDNA Methods, various commercial anti-CEA
The reactivity with the antibody (primary antibody) was examined. Anti-CE used
Antibody A was DAKOPATTS rabbit anti-CEA antibody (code A115) used for screening, Miles sheep anti-CEA antibody (NCA II absorption, code No. 64-733-1, sherotech sheep anti-CEA antibody). (Code No.AHP105A) and Selotech sheep anti-CEA antibody (NCA absorption, code No.A
HP105B). The secondary antibody was KPL (Kirke
gaard & Perry Laboratories Inc.) Peroxidase-labeled rabbit anti-sheep antibody (Catalog No. 04-23-06)
Was used. These primary and secondary antibodies were both used at a dilution of 1/500. A non-immune sheep antibody (Non Immune Serum) was used as a control.

FIG. 4 shows the results of analysis using this anti-CEA antibody. In the figure, 1 shows the results of the reaction with each antibody of the λKr40 clone, 2 and the clone (control) containing λgt11 into which cDNA was not inserted, and A shows the rabbit anti-CEA antibody of DAKOPATTS, and B shows the results.
Shows the reactivity with the sheep anti-CEA antibody from Miles, C and D show the reactivity with the sheep anti-CEA antibody from Serotech, and E shows the reactivity with the non-immune sheep antibody. In the figure, those with large black spots indicate that the reaction was positive. As is clear from the figure, λKr40 reacted with any of the anti-CEA antibodies and did not react with the non-immune sheep antibody. The phage clone λgt11 into which no cDNA was inserted did not react at all with the anti-CEA antibody. Thus, λKr40 reacts with both the Miles anti-CEA antibody absorbed by NCA II and the Serotech anti-CEA antibody absorbed by NCA, indicating that λKr40 has C
This shows that EA protein is expressed. However, as described above, λKr40 contains full-sized human CEAcDN
Since it was considered that A was not inserted, as shown in the following (4) and (5), examination was performed to obtain a complete cDNA.

(4) Preparation of cDNA Library by Okayama-Burke Method To 20 μg of poly (A) + RNA obtained by extracting from the colon tissue of the same E. coli patient as in (1) above, 50 μg of mouse ribosomal RNA as a carrier was added, and 15%- The mixture was subjected to a 30% sucrose density gradient centrifugation to fractionate the size. Subsequently, according to the Okayama-Burke method (Mol. Cell. Biol., 2 , 161, 1982), fractionated 2 to 5 kb of poly (A) + RNA (mRNA) 8.25 μm and 0.1.
By using 51 μg of vector primer DNA, c
A DNA library (plasmid) was prepared, and Escherichia coli DH1 strain was transformed. The transformant was plated on an LB-agar medium containing 50 γ / ml of ampicillin to obtain about 7,000 ampicillin-resistant transformants.

(5) Identification of CEAcDNA clone Approximately 7,000 colonies obtained in the above (4) were replicated on a nitrocellulose filter, and the colonies transferred on the filter were lysed with 0.5N NaOH and neutralized to pH 7.0. Then, the filter was immersed in 0.5 M Tris-HCl buffer (pH 7.0) containing 1.5 M NaCl and 3 × SSC (0.15 M NaCl, 0.015 M sodium citrate) for 5 minutes each. Finally, bacterial debris was removed with a paper towel, air-dried, and baked at 80 ° C for 2 hours.

Subsequently, the previously obtained Pvu II 534 bp fragment of the λKr40 clone (a mixture of fragments 566 to 1099 and 1100 to 1633 in FIG. 3) was labeled with α- ³² P-dCTP, Using this as a probe, hybridization was performed with the DNA molecule immobilized on the filter (William, I. et al., Proc.
tl. Acad. Sci. USA, 82 , 1585, 1985). The specific activity of the probe used was 2.3 × 10 ⁸ cpm / μg DNA, which was used for hybridization at a concentration of 5 × 10 ⁵ cpm / ml. Hybridization was performed using a 1 × Denhardt's solution (0.2%).
3 × SSC containing% BSA (Armor), 0.2% Ficoll (Sigma) and 0.2% polyvinylpyrrolidone (Wako Pure Chemical), 0.1% SDS and 50γ / ml salmon testis DNA (Pharmacia PL) For 16 hours at 65 ° C.
Next, the filter was subjected to 65 ° C. by 3 × SSC containing 0.1% SDS.
After washing and air drying, the film was exposed to an X-ray film. By this operation, eight positive clones were obtained. Eight clones of plasmid DNA were separated according to a conventional method, and DNA fragments were analyzed by restriction enzyme digestion. As a result, four clones (pCEA52, pCEA
55-11, pCEA55-2, pCEA80-11) are the previously acquired λKr
The clone was found to be extended to the 5 'side from 40. In addition, the first three clones (pCEA52, pCEA55-11,
Since pCEA55-2) was considered to be the same clone as a result of restriction enzyme analysis, the nucleotide sequence of the 5'-side extended region of the inserted cDNA was determined for the two clones pCEA55-2 and pCEA80-11.

FIG. 2 shows the restriction enzyme cleavage site used for nucleotide sequencing and the strategy for nucleotide sequencing, and FIG.
2 shows the nucleotide sequence of the DNA inserted into pCEA80-11 containing the entire sequence of the gene of Example 1 and the amino acid sequence corresponding to the nucleotide sequence. In FIG. 2, the direction of base sequence determination is indicated by an arrow. The nucleotide sequence was determined by the method of Sanger et al. (Proc. Natl. Acad. Sc
i.USA, 74 , 5463, 1977). As a result of the above base sequence determination, the base sequences of the 5′-side extension regions of both the two regions were completely identical except that the lengths of the 5′-ends were different (pCEA55-2).
Is 8 bases shorter at the 5 'end than pCEA80-11). Further, the nucleotide sequence of the region (120 bases) for which the nucleotide sequence was determined was completely identical to that of the previously obtained clone λKr40 (from the 500th to 620th positions in FIG. 3). Note that pCEA80−
Eleven clones (DH1 / pCEA80-11) have been named SBM289 and have been deposited with the National Institute of Advanced Industrial Science and Technology under the accession number FERM P-9119.

No start codon ATG is found in the determined DNA sequence, but TAG (2096-2098 in FIG. 3) starting from the 5 'end
There is a long translatable sequence (open reading frame) ending with ".", Which revealed that the sequence can encode 698 amino acids.

In FIG. 3, starting from Ser (−30), Ala (−)
The amino acid sequence of -30 → -1 ending in 1) is considered to correspond to a part of the signal peptide from its structure. Therefore, the human CEA structural gene starts with Lys (1). In fact, the amino acid sequence of 1-24 from Lys to the 24th Leu is the amino acid sequence of the N-terminal of CEA (JNSG
lassman et al., BBRC, 85 , 209, 1978).
However, the 25th to 30th amino acid sequences reported by the authors do not completely match the amino acid sequences predicted from cDNA. References showing the amino acid sequence of the tryptic peptide of CEA (Cancer Research, 38 , 219).
9, 1978), no sequence corresponding to the amino acid sequence predicted from the cDNA is found.

As a result of the above analysis, in particular, the λKr40 clone
In consideration of reacting with antibody A, the resulting cDNA is human C
It was determined to be a cDNA encoding EA.

These results indicate that human CEA is biosynthesized as a precursor of at least 698 amino acids, and that it passes through the endoplasmic reticulum membrane. It is thought that CEA is produced as a protein. The C-terminal 26 amino acids are rich in hydrophobicity, and this region is buried in the cell membrane, and the N-terminal 642 amino acids are considered to be extracellular. Therefore, using human CEA from which the amino acid sequence at the C-terminal part has been removed, it would be possible to produce human CEA-specific antibodies useful for clinical trials.

(Effects of the Invention) (1) Human CEA or various CEA fragments can be expressed in cells (Escherichia coli, yeast, animal cells, etc.) by using the gene of the present invention, and antibodies against them can be produced. Thereby, it is possible to find an antibody that reacts only with human CEA and does not react with CEA-related antigen. Alternatively, various CEA fragments can be produced by a synthetic method from the human CEA peptide sequence elucidated in the present invention, and a human CEA-specific antibody can be obtained in the same manner. By using the antibody thus obtained, a human CEA measurement kit with higher reliability than before can be produced.

(2) Using the human CEA gene obtained in the present invention,
Instead of conventional immunohistochemical methods using antibodies, human
Tissues such as tumors that produce CEA were
Itu) It can be detected by hybridization.

(3) CEA using the human CEA gene of the present invention as a probe
It is also possible to clone the gene for the relevant antigen. If those genes are cloned and the amino acid sequence is revealed, the differences between CEA and CEA-related antigens will be clearer.

(4) As described above, since human CEA-specific antibodies can be produced according to the present invention, cancer diagnosis, for example, cancer screening, definitive diagnosis, cancer progression determination, monitoring of treatment, and localization of cancer by labeled anti-CEA antibody Diagnosis is possible,
In addition, it is expected to be used for treatment such as missile therapy.

As described above, the field of use of the present invention is diversified, and its use value is great.

[Brief description of the drawings]

FIG. 1 is a diagram showing an outline of a method for preparing a cDNA library using a λgt11 phage vector. FIG. 2 is a diagram showing a strategy for determining a cDNA base sequence. 3) is a series of sequence diagrams showing the nucleotide sequence of the cDNA of human CEA and the amino acid sequence of the translatable region. FIG. 4 shows the results of analysis of the ability of the λKr40 clone and the control to produce CEA protein using an anti-CEA antibody. It is a photograph showing.

Continuation of the front page (56) References JP-A-63-119681 (JP, A) Proc. Natl. Acad. Sc i. USA, Vol. 73, No. 6, p. 2123-2127 (1976) Proc. Natl. Acad. Sc i. USA, Vol. 75, No. 4, p. 1670-1674 (1978) Cancer Res. 38, no. 3, p. 503-505 (1978) Mol. Immunol. , Vol. 1, P. 67-73 (1985)

Claims (3)

(57) [Claims] 1. The following formula I: The human carcinoembryonic antigen (c) encoded by the DAN sequence
arcinoembryonic antigen) A gene encoding an amino acid sequence containing at least a specific antigenic portion of a protein. 2. The following formula I: A plasmid comprising a gene encoding an amino acid sequence containing at least a specific antigenic portion of a human carcinoembryonic antigen protein encoded by the DAN sequence represented by: 3. The plasmid according to claim 2, which is pCEA55-2 or pCEA80-11.