NO850005L - NUCLEAR PREPARATION, PROCEDURE FOR CLEANING NUCLEAR ANTIGENS AND PROCEDURE FOR CLEANING ANTIBODIES - Google Patents

Description

The present invention relates to nucleoares

antigens found in a variety of human cancers

occurs, but which are not found in the corresponding healthy tissues,

as well as antibodies and antisera specific to these nucleolar antigens for diagnostic purposes. -

Previous investigations in experimental animals have

showed a presence of nuclear and nucleolar antigens in tumors

• which are not found in healthy tissue (R.K. Busch et al, Cancer Res.

34, 2362, 1974; Yeoman et al, Proe. Nati. Acad. Pollock. USA

73, 3258, 1976; Busch and Busch, Tumori 63. 347, 1977;

Davis et al., Cancer Res. _3_8, 1906, 1978; Marashi et al.,

Cancer Res. 3_9_, 59, 1979). In previous investigations by the present applicants, antibodies were produced for nucleoli

in normal and neoplastic cells in rats by immuno-

sering of rabbits (R.K. Busch et al, supra; Busch and Busch, supra; Davis et al, supra). Clear nucleolar fluorescence

was detected in the acetone-fixing cells by means of an indirect immunofluorescence. It was found that the immunoprecipitin bands in Ouchterlony gel formed with antisera to a Novikoff hepatoma nucleolar antigens that were extra-

heart from rat Novikoff hepatoma nucleoli differed from the corresponding immunoprecipitin bands obtained with liver nucleolar antigens and antiliver nucleolar antisera

(Busch and Busch, supra).

Additional specificity was demonstrated when anti-

tumor nucleolar antisera absorbed with liver nuclear extracts gave positive nucleolar fluorescence in Novikoff hepatoma ascites cells but not in liver cells. Equivalent-

finally, liver nucleolar antisera absorbed with tumor nucleolar extracts gave no detectable tumor nucleolar fluorescence, but gave positive fluorescence in living nucleoli (dvis et al, supra).

As the immunofluorescence assay indicated

that the differences that could be observed in the acetone-fixed tumor smears and in normal rat cell smears (especially after absorption of antisera with normal liver

nuclei and nucleoli), then attempts were made to use

these antisera for rat tumor nucleolar antigens at

testing of corresponding tissue samples that were derived from

tumors in humans. Investigations with antibodies

to small rodent tumor nucleoli showed that the positive immunofluorescence found there could not be found in tumor nucleoli from humans. Because of this, they started so

with a new series of experiments to find human nucleolar antigens. Positive immunofluorescence was then found in human tumor tissues with antisera and antibodies to these new human tumor nucleolar preparations. In these investigations, the antibodies were absorbed with placental nuclear soni-

cats as well as with fetal calf serum (Busch et al, 39,

3024, 1979; Davis et al, Proe. Nati. Acad. Pollock. US 76,

892, 1979; Smetana et al., Life Sci. 25<_>, 227, 1979).

The present invention is the result of investigations which were designed so that one could use these new human nucleolar antigens for the detection of a number of human neoplasms.

The following Table I gives an overview of the tumors in humans in which a clear nucleolar immunofluorescence was found with the antibodies to human tumours.

nucleoli. These investigations surprisingly and unexpectedly showed that many human tumors contain common nucleolar anti-

genes that show a positive immunofluorescence with antisera or immunoglobulin fractions of such antisera (Busch et al, supra).

In healthy tissue and in benign tumors and in inflammatory conditions negative results were generally obtained as indicated in the following Table II (Busch et al, supra).

These results that were originally obtained

by immunofluorescence, has since been confirmed and expanded using immunoperoxidase methods.

The present invention is based on the surprising and unexpected discovery that common nucleolar antigens are found in a number of types of human cancer cells,

but is not found in normal cells. These antigens are pro-

teins which may have gene control or other functions and which are always persistent during mitosis in a perichromoso-

template position. Important aspects of the invention are the discovery of common nucleolar antigens found in human cancer cells, the isolation and purification of these antigens, as well as the production of antisera and antibodies that are specific for these antigens as well as diagnostic methods using antisera and antibodies that are specific to these anti-

genes to detect human cancer cells, as well as a diagnostic device that either contains antibodies or antisera or both and where these are specific for these nucleolar antigens.

These antigens have been detected in a number of different types of cancer in humans, such as cancer of the central nervous system, of the intestinal tract, of the urinary tract, of the lungs,

in the skin, in blood-forming tissues and in endocrine and exocrine glands. Thus, malignant human cells include HeLa cells, prostate carcinoma, other types of carcinoma,

sarcoma and hematological neoplasm. These antigens can be extracted from the nuclei or nucleoli of human malignant cells. Said antigens have not been detected in corresponding healthy tissue. In diagnostic methods according to the present invention for detecting malignant cells, approx. 1% false negative and 3% false positive results.

The false negative results represent necrotic tumor-

tissue or non-reactive tumors of unknown origin.

The false positives represent two cases of "preneoplastic tissue" and weak positives occurred

occasionally in the focal areas of so-called "hyperplastic tissue".

Two focal positive areas were identified as "preneoplastic areas" or focal neoplastic transformation in inflammatory tissue of the intestinal tract.

The antigens consist of a main antigen and at least one

and possibly several minor antigens. The main antigen from human cancer cells itself has (a) a discrete isoelectric point from 6.0 to 6.7 and approx. 6.3 as determined by isoelectric focusing, pH 3-10, polyacrylamide gel;

(b) has an approximate molecular weight of from 50,000-60,000 daltons as determined by two-dimensional gel electrophoresis with an SDS (sodium dodecyl sulfate) second dimension; (c) is partially strongly bound to nuclear and nucleolar RNP

and is partially soluble in 0.01 molar tris-HCl, pH 8; (d) is both nucleolar and extranucleolar, but remains "intranuclear" or chromosomally associated during cell division.

The other type of antigen that has been detected has

a pl of approx. 6.0 (detected by the same procedure as for

clearly above) and its molecular weight is also from 50,000-60,000 daltons. It is possible that this second type represents a modified product of the first type, but it has not been determined whether they are structurally related. The last-

said type of antigen occurs in relatively smaller concentrations than the former.

The antigens are also present in nucleolar ribonucleoprotein (RNP) particles prepared by ultra-centrifugation of the Tris extracts, as described below. The antigen present in these particles is more strongly bound to proteins and ribonuclein

acid (RNA) than the antigen in the Tris-soluble fraction.

It is not entirely clear whether the antigens in the RNP particles

are identical to those found in the liquid, but their iso-electric point is the same and they absorb antibodies to the antigens. Using immuno-light microscopy could

also in cancer cells, nucleolar antigens were detected which were present in fibrils, presumably from the nucleolar ribonucleo-protein network. This network is extranucleolar structures that may represent elements from which said RNP particles arise.

It remains to be determined whether antigens represent a substance present in higher concentrations in cancer cells and in lower concentrations in healthy tissue or are fetal antigens as previously found in comparative studies of nuclear antigens in rat Novikoff hepatoma and in normal rat liver cells (Yeoman

et al, 1976).

All the steps performed to obtain and analyze samples of human tissue, blood and serum from tumors and other tissues of patients suspected of having cancer were recommended by the Human Research Committee at Baylor College of Medicine, Houston, Texas and affiliated hospitals. Sections of human tumors were obtained from frozen sections taken by surgery, biopsy or preserved in extreme cold, and they were essentially obtained from the Department of Pathology of the Houston Veterans Administration Medical Center, but also from The Michigan Cancer Foundation

in Detroit, Michigan as well as the Department of Internal Medicine, Charles University in Prague, Czechoslovakia. These sections were analyzed for the presence of nucleolar antigens by indirect immunofluorescence and immunoperoxidase techniques.

Purification of antigens was carried out by an extraction of nuclei or nucleoli with 10 mmol Tris HCl/0.1 mmol PMSF/pH 8 six times in a ratio of 20 volumes to 1 volume of nuclei or nucleoli. The extracts were first centrifuged at 27,000 x g for 10 minutes and then at 100,000 x g for 16 hours. Ammonium sulfate at 40% saturation was used to remove the pre-contaminant. The 40-100% ammonium sulfate fraction was collected by centrifugation and dialyzed against 20 mmol Tris HCl/pH 7.6. The antigen was chromatographed on DE-52 cellulose columns

(1 x 10 cm). The antigens were eluted with the 0.15 molar NaCl/0.1 mmol PMSF/pH 7.6 fraction. Isoelectric focusing gels were used to identify and purify the antigens. These contained 4% acrylamide/8M urea/2% ampholines (pH 3.5-10). The antigens with pI 6.3 and 6.0 respectively were then excised

of the gels. On the SDS (sodium dodecyl sulfate) gels, a major spot was found for each of these antigens.

HeLa cell nuclei or nucleoli were prepared

by collecting HeLa cells from Spinner culture flasks (7-8

liters). The cells should be in a log phase 7-8 x 10<5> cells/ml. The cells were centrifuged at 800 x g for 8 min to pellet cells. These were then suspended in (PBS) phosphate-buffered saline (0.15 M NaCl, 0.01 M phosphate, pH 7.2) by gentle homogenization with a soft Teflon rod and centrifuged at 800 x g for 8 min. The cells became so

washed once more with PBS and cell pellets were weighed. They were then suspended by gentle homogenization in 20 volumes of a reticulocyte standard buffer (RSB) at pH 7.4 and then set aside for swelling for 30 minutes on ice. The cells were then centrifuged at 1000 x g for 8 minutes and resuspended by careful homogenization in said RSB buffer plus 1/20 volume of the detergent Nonidet P40 (10% in RSB). The final volume of Nonidet was 0.5%. The cells were homogenized with a Dounce homogenizer with from 20-60 strokes until all the cells were broken up and the nuclei were freed and free of cytoplasm. The cells were then centrifuged at 1000 x g for 8 min, resuspended with gentle homogenization in 0.88 molar sucrose, 0.5 mmol Mg-acetate (20 x weight-volume) and centrifuged

centrifuged at 1500 x g for 20 minutes. The resulting pellets

which contained said HeLa nuclei were then used to produce

quiet the antigen extracts as described below.

Nuclei from other human malignant cells were prepared on

same way.

For isolation of nucleoli, said nuclear pellets were prepared as described above, suspended by careful homogenization in 0.34 molar sucrose, 0.5 mmol Mg-acetate, using 2 ml of sucrose for each gram of original cells. Said nuclei were sonicated (with a Branson ultrasound machine) at 10 second periods (and 10 second rest). Total time was between 60 and 110 seconds

there. The liberated nucleoli were followed by microscopic examination. To make said nucleoli visible, they were stained with Azur C (this solution contains 1%

Azure C in 0.25 molar sucrose). After the sound treatment period, the preparation should be free of nuclei. The sonicated fraction was lysed three times with a volume of 0.88 molar sucrose (without magnesium acetate) and centrifuged at 1500 x g for 20 minutes. The resulting pellet contained HeLa nucleoli that can be used as the immunogen.

Satisfactory purification was achieved by

using the above-mentioned method (Busch and Smetana, 1970), and light microscopy showed that the quality of these preparations was essentially satisfactory. However, an electron microscopic analysis indicated that there was a presence of chroma-

tin and nuclear contaminants. The main problem in a clean-

ning of these preparations is the limited amount of original HeLa cells in the culture, and this limits the number of purification steps that can be used. Nucleoli pro-

prepared from HeLa cell preparations whose weight varied between 5 and 10 g more than the h to 1 g amounts used in previous experiments, provided sufficient material for a satisfactory purification. The culture conditions for the HeLa cells and isolation of the placental nuclei were essentially the same as previously stated (Davis et al, 1979).

HeLa Tris extract was prepared by suspending

you mentioned HeLa nuclei in NaCl-EDTA buffer 10 x weight/volume,

1 g nuclei/10 ml buffer. (Buffer: 0.075 molar NaCl, 0.025

molar Na EDTA/pH 8, 1 mmol PMSF). Said phenylmethylsulfonyl fluoride (PMSF) was prepared in a 100 mmol concentration in isopropyl alcohol. This was then added to each solution before the extraction. The suspension was homogenized with a Dounce homogenizer with 20 strokes and then centrifuged

3000 x g for 5 minutes. The liquid above the precipitate was collected.

The above extractions were repeated on the nuclear

pellet twice more. The NaCl-EDTA extract was not used in the present antigen work, and was consequently discarded.

The nuclear pellet was suspended in 10 x w/v at

0.01 M Tris-HCl, pH 8, 1 mmol PMSF and homogenized again in said Dounce homogenizer with 20 strokes, although 0.01 M Tris-HCl, pH 7-9 is satisfactory. The liquid

was collected and placed on ice. During the Tris extraction, the nuclei were broken down and chromatin was released. The nuclear degradation was followed by means of microscopic sub-

search. Said pellet was resuspended in the Tris buffer and nuclei were set aside for swelling for 15 minutes on ice.

They were then again homogenized with 20 shocks and centrifuged

at 12,000 x g for 10 minutes. The liquid above the precipitate was collected. Said pellet was resuspended and had a white bulky appearance. It was again homogenized with 20 strokes and centrifuged at 27,000 x g for 30 minutes.

The liquid above the cells was collected and pooled with time-

ligers from the Tris extracts.

The Tris extracts were then concentrated with an Amicon UM-10 or PM-10 Diaflo membrane. Usually was

the volume at the beginning approx. 50 ml and this was concentrated to 4-5 ml. The final concentration of protein was around 4-5 mg/ml. The rabbit can be immunized with this Tris extract.

By using the above method prepared

one also extracts from HeLa nucleoli or from nuclei or nucleoli from other malignant human cells.

For the Tris immunogen, 250 µl of Tris extract

(4-5 mg/ml) prepared as above diluted with 250 µl PBS.

This was then mixed with Freund's diluent like this

this is described below for the nucleolar immunogen.

Antibodies were prepared by immunizing rabbits with HeLa cell nucleolar preparations on the following

method: said HeLa nucleoli were weighed (20-30 mg, wet weight) and suspended uniformly in 0.5 ml of 0.01 M phosphate buffered saline, pH 7.2. They were then mixed with 0.6 ml of

Freund's diluent (GIBCO) as follows: The suspended nucleoli were placed in a 5 ml syringe and

Freund's diluent in another. For each syringe

an 18-gauge needle whose point had been removed was attached.

The needles were then connected by means of a piece of polyethylene tubing, I.D. 0.04 7 (Clay-Adams). The contents of the syringes were then mixed until the preparation thickened and was

should be pushed through the tube.

The rabbits were shaved on their backs and injected intradermally at 6 sites at 0.1 ml/site. The remaining 0.4-0.5 ml was injected, half subcutaneously (under the loose skin of the upper back) and half intramuscularly (into the thigh muscle). The injections were given once a week for three weeks with similar amounts of nucleoli each time. The first blood draw was performed 7-10. day after the third week of immunization. A rabbit ear cup (Bellco) and a vacuum pump were used to collect the blood. The blood (approx. 45-50 ml) was

set for coagulation for 3-4 hours at room temperature. Serum-

the part was then taken out and centrifuged at 1000 g for 30 min.

ter (these sediments are free of red blood cells). The clear serum was collected and then absorbed (or kept frozen until absorption). The coagulated blood itself can be cooled overnight, and a couple of additional ml of serum will then be released. The serum from each blood draw was examined for the presence of nucleolar antibodies by an indirect immunofluorescence method.

Other non-human hosts (e.g. goat, sheep,

horse, chicken etc.) can be immunized with malignant human cell nucleoli preparations to elicit antisera or antibodies to the nucleolar antigens according to the present

invention. Antisera can also be prepared by immunizing non-human hosts with extracts (e.g. tris extract)

of malignant human cell nuclei or nucleoli.

Absorption of antinucleolar antiserum was

conducted by first absorbing the rabbit antiserum with 20% normal human serum and 20% fetal calf serum (GIBCO). 20%

normal human serum was added to the rabbit antiserum (4 ml/

20 ml) and incubated for one hour in a shaking water bath at 37°C.

The flask was removed and 20% fetal calf serum (4.8 ml/24 ml)

was added, and incubation was again carried out for one hour in a shaking bath at 37°C. The flask was taken and incubated for one hour

at room temperature with gentle stirring every 15 minutes.

It was then centrifuged at 15,000 x g for 30 minutes, and

the overlying fluid (absorbed serum) was removed and

saved. The absorbed serum was converted into immunoglo-

the buline (lg) using the procedure given for the (NH 2 ) 2 SO 4 ~precipitation of serum described in the following.

The Ig preparation from the nucleolar antiserum and

which was absorbed with normal human serum and fetal calf

serum was now absorbed with a normal human tissue (placenta or liver). With a corresponding volume of placental nuclear sonicate in PBS 7.2 (10-15 mg protein/ml) the absorbed nucleolar immunoglobulin (10 ml lg plus 10 ml nuclear sonicate) was added and incubated for one hour at 37°C in a shaking water bath. It was then incubated for a further hour at room temperature by gently mixing the contents of the bottle every quarter of an hour and then centrifuged at 15,000 x g for 30 minutes.

The supernatant was collected and absorbed by lg

was reprecipitated with (NH 4 ) 2 SO 4 as described. This Ig can be used as the final antibody product or can

ligers are purified by diethylaminoethyl (DEAE) cellulose chromato-

graphics in the following way:

Said Ig, which was in 0.01 molar phosphate-buffered saline pH 7.2, was dialyzed against 0.0175

molar phosphate buffer pH 6.3 (without salt solution). After dialysis, it was centrifuged at 2500 x g for 20 minutes.

The supernatant was added to the DEAE column (20 mg protein per gram cellulose, Whatman DE52). IgG was eluted from the column with 0.0175 molar phosphate buffer. After elution, the IgG fraction was dialyzed against 0.01 molar phosphate-buffered saline with pH 7.2.

The same procedure was used for the control serum which consisted of pre-immune serum obtained by drawing blood from the rabbit (or other non-human host animal) before the immunization started.

Rabbit immunoglobulin lg was prepared on

following way: a saturated (NH^)2SO4 solution (760 g/liter) was prepared and an equal volume of a cold (NH^^SO^) was added drop by drop to the antiserum with stirring.

A white precipitate formed and this was allowed to coalesce for 1\ to 2 hours on a magnetic stirrer in the ball. The precipitated antiserum was then centrifuged at 3000 x g for 20 minutes. The supernatant was removed and said pellet was resuspended in PBS, pH 7.2 (about half the volume of the original serum). The dissolved pellet was placed in a dialysis bag and dialyzed against 100 volumes of PBS overnight in the cold with magnetic stirring. The dialysis bag was then placed in fresh PBS (100 x the volume) the following morning and dialysis was continued for 6 hours. The immunoglobulin was carefully removed from the dialysis bag and centrifuged at 2500 x g for 20 minutes. The supernatant was collected.

The procedure described earlier

(RK Busch et al, 1974; Hilgers et al, 1972) for immunofluorescence was used in this study as follows: 150 µl of antinucleolar antiserum diluted 1:50 was placed on acetone-fixed HeLa cells or on fixed tissue samples (from Hilgers et al, 1972, RK Busch et al, 1974). It may be necessary to use more than 150 1, if the tissue sample covers a larger part of the slide. The dilution of the antiserum (As) depends on the strength of the antibody (Ab). Other dilutions can be used up to the point where the As and Ab dilution becomes too thin to get a positive reaction

against known positive cells (e.g. HeLa). The slides were incubated in a moist chamber for 45-50 minutes at 37°C (The moist chamber can consist of a larger petri dish to which moistened paper has been added.) After the incubation, the antiserum was washed off by careful addition of PBS, and the slides were placed in a holder and washed in PBS for one hour. The PBS was changed three times, ie after 15 minutes, 30 minutes and 45 minutes. The slides were taken out of PBS and dipped in distilled or deionized water ten times by rapid upward and downward movements.

They were then dried with cold air from a hairdryer for 2-3 minutes, being careful not to dry too strongly. 150 µl of fluorescein-labeled goat antirabbit antiserum (Hyland or Cappel) diluted 1:10 was then placed on the slides and incubated in a moist chamber for 30-35 minutes at room temperature. The second antibody was removed from the slides by gentle washing with PBS. The slides were then washed in PBS for one hour, changing the liquid three times, i.e. at 15 minutes, 30 minutes and 45 minutes after the experiment began, or after the first washing after 15 minutes, they can be placed in fresh PBS and stored in refrigerator overnight. After the last wash with PBS, the slides were dipped in deionized or distilled water ten times with rapid movements and dried in cold air using a hair dryer for 2-3 minutes. A solution of glycerol and PBS in a ratio of

1:1 was added to the cells or tissue sample and this was then covered with a glass coverslip. The sample can be kept for several months if the coverslip is closed with a sealant, such as clear nail polish, and then kept in a refrigerator or cooler. The slide was examined using a fluorescence microscope. Nucleolar fluorescence could not be observed with preimmune immunoglobulin or preimmune IgG fractions. The other types of immunological technique used are the same as those used in previous investigations (Kendall, 1938; Lowry et al, 1951; Dale and Latner, 1969; Laurell, 1972; Wallace et al, 1974; and Marashi et al eel,

1979). For analysis of the nucleolar location of the immunofluorescence that could be observed, the samples were taken out and in by phase contrast illumination during the fluorescence observation.

Instead of using fluorescein-labeled goat anti-rabbit serum, an immunoperoxidase method can be used. E.g. can 150 µl of peroxidase-labeled goat anti-

rabbit 1:10 or 1:20 dilution is added. Localized peroxidase activity can be detected by a variety of redox dye systems for light or electron microscopic examination.

Other enzymes can also be used as markers for this indirect method, and peroxidase and other enzymes can be used directly by labeling with primary antibody.

Karnofsky's incubation medium was then prepared as follows: sufficient diaminobenzidine (Sigma) was weighed out and suspended in 0.05 molar tris-HCl,

pH 7.6, so that the concentration was 0.5 mg/ml. One forward-

then prepared a hydrogen peroxide solution of 0.02% (in said 0.05 molar Tris-HCl buffer). 0.5 mg/ml DAB and

said 0.02% ^ 2°2 ^ equal parts'ie. in a ratio of 1:1

(this solution was freshly prepared each time

used and stored in the cold during use). From 200-300 µl of DAB and H202 ~ mixture was then added to the slide and this was incubated for 30 minutes in a moist chamber at room temperature.

After the incubation, the DAB and H 2 O 2 mixt

washed off the slide with 0.05 molar Tris-HCl, pH 7.6-

buffer to which 0.1 molar NaCl had been added. The slides were then given two 10 minute washes in 0.05 molar Tris-HCl pH 7.6

0.1 molar NaCl. Final processing of the slides is indicated in steps 11-14 (except PBS has been replaced with Tris.HCl). The final slide was examined using light microscopy.

HeLa cell preparations for immunofluorescence were prepared as follows: a stock of fixed HeLa cells was prepared by removing and washing with PBS, pH 7.2, actively growing cells from said HeLa culture flask. The cells were suspended so that there was at least 1.5 x 10 g cells/ml. A drop of this HeLa cell suspension was placed on each washed slide (washed with detergent, cleaned with distilled or deionized water and then cleaned with alcohol,

and then dried using heated air from a hair dryer) and lightly spread outwards and set aside to dry at room temperature (or in the cold overnight). The dry cells were fixed by placing the slides at 4°C

in acetone for 12 minutes. The preparation tubes were numbered

using a diamond-tipped pencil. The preparations were used

as positive controls on immunofluorescence.

The present investigation confirms that nucleo-

learn antigens are present in tumor cells, but are not

present in healthy tissue. The first investigations showed that both

in cell cultures in human tumors and in samples taken either at autopsy or at biopsy, clear nucleolar fluorescence was obtained with the double antibody technique

(indirect immunofluorescence), and a corresponding result

could not be obtained in a series taken from healthy tissue (Davis et al, 1979). In later investigations, more than 60 ond-

types of tumors examined, and a variety of tissue types were also assessed. It is of interest that this large variety of malignant tumors of ectodermal, endodermal and mesodermal origin all have a presence of one or more common nucleolar antigens (Table I).

Example 1

Normal tissue - 17 healthy tissue samples were found

no nucleolar fluorescence after incubation of said anti-

sera and antibodies with different fixed cell preparations

ter. It was of particular interest that neither the Malpighian layer in the skin nor the cells in the bone marrow nor Lieberkuhn's caves gave positive immunofluorescence with this method. Furthermore, a number of different healthy tissue samples taken near the neoplasms were also negative. Furthermore, a group of benign tumors was examined

include several types of thyroid adenoma,

and these were also negative (Table II).

Example 2

Inflammatory lesions To investigate

whether an inflammatory reaction was associated with these antigens, investigations were carried out on eight types of inflammatory tissue. In most of these, there was no detectable fluorescence in the cell nucleoli that were sub-

searched. However, it turned out that sections found in ulcerative colitis and samples taken from gastric ulcers had a posi-

tive nucleolar fluorescence. Remarkably, 2 out of 3 sections of the ulcerative colitis were negative, while one showed a definitively positive nucleolar fluorescence. From the ulcers that were examined, it turned out that one out of two sections had positive nucleolar fluorescence. These results are particularly interesting on the basis that these types of lesions often develop in a malignant direction. It was of particular interest to examine both the focal positive and negative areas of these sections when these were stained with hematoxylin and eosin. It then appeared that certain areas of these wounds not only had mitotic formation, but also

an accumulation of the epithelial tissue. These results suggest that these cells could possibly constitute preneoplastic lesions or carcinoma in situ. It is possible that the discovery of these fluorescent areas can facilitate decisions about whether to perform a surgical intervention, whether this is local or applies to larger areas of the wound surface.

Example 3

Artifact - In the stomach epithelium there was a

area of fluorescence in each cell that was non-nucleolar

•of origin, and this seems to represent a non-speci-

obtained localization of fluorescent antibody. In a cavity in the small intestine, a non-specific um-

have antibody in the form of aggregates, and in most cases an elimination of these aggregates was achieved

when the antibodies were filtered through a 0.45 um millipore filter. In a sample of breast tissue that was negative for nucleolar fluorescence, small non-specific immunofluorescent spots were obtained that were distributed without particular

localization with respect to cell morphology. The diameters of

these very small non-specific precipitates were from 0.5

to 0.1 ym compared to the nucleolar diameters in nuclei and nucleoli which vary from 4 to 6 ym.

Example 4

Fluorescence during phases of the cell cycle - The nucleolar fluorescence was very easily visible in the interphase nucleoli. In metaphase, no nucleolar fluorescence could be seen in the form of distinct areas, but this was syn-

between the chromosomes and in connection areas between the "nucleus" and the cytoplasm. As the nucleolus essentially

lig disappears during metaphase and rRNA synthesis stops in late prophase, so it is not surprising that the nucleolar fluorescence was not visible as a distinct om-

prevail in such cells (Tan and Lerner, 1972). However, the fact that traces of the immunofluorescent products were found throughout mitosis seems to suggest that the nucleolar sub-structures (more than the nucleolar products themselves) contain

keeps the antigens persistently epigenetic.

Example 5

Negative malignant tumors - In a series of malignant

in various types of tumors, negative areas were found to varying degrees throughout the sections. Usually these were correlated with either necrotic or abscess-like parts of neoplasms.

In a sample of a tumor from the brain, it was mass like

did not show any positive fluorescence, purely necrotic, and many leukocytes were present, but there was no defined struc-

trip. In an adenocarcinoma that had developed in the form of a metastasis in the brain, no positive fluorescence was obtained, and the reason is not clear. As 61 of 63 examined tumors had a positive nucleolar fluorescence, 97% of the examined series were positive. These investigations have now been extended to over 300 cancer samples from humans such as breast, prostate, lung and haematological tumors with similar results.

Example 6

Labeling - Direct immunochemical methods for detecting the antibodies include labeling of the primary antibody with one or more of the following elements: 125 131 14

a radioisotope for autoradiography such as I, I, C

or<3>H, a fluorescent dye such as fluorescein or tetramethylrhodamine for fluorescence microscopy, an enzyme that gives a fluorescent or colored product for detection

by fluorescence or light microscopy, or which provides an electron-dense product for detection by electron microscopy,

or an electron-dense molecule such as ferritin for direct visualization in the electron microscope.

Indirect immunochemical methods include

labeling the second antibody or another binding protein specific for the first antibody with a fluorescent dye, an electron-dense compound,

an enzyme that yields a product that can be detected by light microscopy, or fluorescence or electron microscopy, or a radioisotope that can be detected by autoradiography.

The indirect immunochemical methods for visualizing the antibodies include the application of hybrid primary or secondary antibodies or antibody fragments

ments (F(ab')2) where part of the hybrid antibody preparation is specific for the nucleolar antigens (hybrid primary antibody) or for the primary antibody (hybrid other antibody), and is partly specific for such labeling, e.g.

of the type mentioned in the above paragraph.

Labeled conjugated and non-conjugated antibodies

fer can be packaged separately in phosphate-buffered saline (PBS) or other buffered suspending agents for distribution as diagnostic specimens. Suitable suspending agents include glycerin, heparin or sucrose. Suitable buffers include barbital buffers, morpholine buffers, M0PS-3-(N-morpholino)propanesulfonic acid, hepes-N-2-hydroxyethylpiperazine-N-2-ethanesulfonic acid, Tris carbonate and the like.

Claims (3)

1. Nuclear preparation, characterized in that it includes Tris-HCl extracts of nuclei or nucleoli from human cancer cells. 2. Process for the purification of nucleolar antigens, characterized by performing a sequential purification using ammonium sulphate precipitation, DEAE column chromatography, Sephadex gel exclusion, cation exchange and calcium phosphate gel chromatography and preparative isoelectric focusing. 3. Method for purification of antibodies harvested from a non-human animal host, characterized in that the purification is induced by immunization with a nucleolar antigen.