Identification of differentiation markers in normal and virally transformed avian hematopoietic cells


Material Information

Identification of differentiation markers in normal and virally transformed avian hematopoietic cells
Physical Description:
xiv, 119 leaves : ill. ; 29 cm.
Liu, Juinn-Lin G., 1960-
Publication Date:


Subjects / Keywords:
Research   ( mesh )
Hematopoiesis   ( mesh )
Cell Transformation, Viral   ( mesh )
Disease Models, Animal   ( mesh )
Antibodies, Monoclonal   ( mesh )
Antigens, Differentiation   ( mesh )
Retroviridae   ( mesh )
Birds   ( mesh )
Department of Pathology and Laboratory Medicine thesis Ph.D   ( mesh )
Dissertations, Academic -- College of Medicine -- Department of Pathology and Laboratory Medicine -- UF   ( mesh )
bibliography   ( marcgt )
non-fiction   ( marcgt )


Thesis (Ph.D.)--University of Florida, 1990.
Bibliography: leaves 111-117.
Statement of Responsibility:
by Juinn-Lin G. Liu.
General Note:
General Note:

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 001476235
oclc - 22519072
notis - AGY8080
System ID:

This item is only available as the following downloads:

Full Text








This dissertation is dedicated to my parents. Without

their understanding, support and encouragement, it would

have never been possible for me to accomplish any goals in

my life.


Words cannot express my appreciation to my mentor, Dr.

Carlo Moscovici. Four years ago, I went to his office to

discuss my rotation project. Suddenly I was touched by a

poster on the wall, which depicted a newly hatched baby

chick, excited and curious about its new life, asking What

do I do ? It was nearly the mirror image of myself. The

very same question had struck me from time to time since I

arrived in the United States. Fortunately, he has guided

me, spiritually and intellectually, during the past four

years not only to become an independent researcher but a

more mature person as well. Nevertheless, we still have

some disagreement about my own differentiation pathway,

e.g., he never gives up the idea that jazz should be a

"growth factor" for me.

My work would not be complete without the great impact

of Dr. Giovannella Moscovici. She has enthusiastically

shared her knowledge and experience with me. I have learned

from her many techniques regarding cell biology and tumor

virology and have a better vision about life.

I also sincerely appreciate Dr. Paul Klein's help. His

generosity allowed me to use the space and materials in his

laboratory to complete the work on RIA and SDS-PAGE. He


also helped me interpret the results and select the MAbs.

In addition, I am indebted to Dr. Paul Linser for providing

me the opportunity to establish the SDS-PAGE and

immunostaining techniques in his laboratory at the Whitney

Lab. The atmosphere there was so wonderful that I used to

walk along the beach and listen to the songs of the ocean at

night. It really eased a lot of pressure built up during my

project. By the way, the enjoyment of reading a science

fiction novel "The Rapture Effect," a gift from Dr. Linser,

may also enhance my "healing" process.

I would like to extend my appreciation to the other

members of my committee, Dr. Edward Wakeland and Dr. Ammon

Peck, for their helpful discussions and constructive

comments on my work.

Finally, I am also grateful to our technician, Mr.

Gordon Thompson, for his technical assistance and

preparation of the materials; to Mrs. Linda Green, Sue

Hammack, Janice Odebralski and Sandra Wotoweic in the

Hybridoma Laboratory for helping me develop the hybridomas;

and to Mrs. Melissa Chen for her assistance with FACS

sorting and analysis.


ACKNOWLEDGEMENTS....................................... iii

LISTS OF TABLES ........................................ vii

LISTS OF FIGURES.......................... ............ viii

ABBREVIATIONS.......................................... X

ABSTRACT..................... ... ....... ...... .......... xii


1 INTRODUCTION AND BACKGROUND ....................... 1

Introduction ........................ ........ ...... 1
Normal Avian Hematopoiesis ...................... 3
Interaction of the Avian Leukemia Viruses with
Hematopoietic Cells............................. 19
Identification of Cell Surface Markers in Normal
Hematopoietic Cells and Tumor Cells by MAbs.... 30

CELL-TYPE SPECIFICITIES ......................... 38

Introduction.................................. 38
Materials and Methods................... ......... 39
Results ........... .. ..... ... ..... .. 49
Discussion........ ...... ........................ 61


Introduction............................... ... 72
Materials and Methods.......................... 73
Results. .................... .......... ......... 75
Discussion............. ......................... 81


Introduction.................................... 85
Materials and Methods............................ 86

Results.......................................... 89
Discussion...................... .......... ....... 98

5 CONCLUDING REMARKS ............................... 105

REFERENCES........................... ................ 111

BIOGRAPHICAL SKETCH................................... 118













Table 2-4.

Table 2-5.

Table 2-6.

Table 2-7.

Table 3-1.

Table 3-2.

Table 3-3.

Table 4-1.


Chicken hematopoietic precursor cells...... 12

The three different lineages of the avian
erythroid compartment ...................... 15

Avian defective leukemia viruses........... 21

The cell-type specificities of MAbs from
BM2 and BM2/L fusions....................... 51

The isotypes of selected MAbs.............. 51

RIA binding indexes of MAbs to different
cell types................................. 52

Summary of the cell-type specificities
of MAbs.................................... 60

Comparison of the proliferating potential
of BM2/C3A cells and 2.5 pg/ml
PMA-differentiated BM2/C3A cells ........... 62

Comparison of the proliferating potential
of 6C2 cells and 1mM butyric acid-treated
6C2 cells.................................. 65

RIA binding indexes of MAbs to normal CEF
and RAV-infected CEF...................... 68

Characterization of the target cells
sorted with 3D7-1C9 from the bone marrow... 78

Characterization of the target cells
sorted with 2E10-1E10 from the bone marrow. 80

Characterization of the negative
populations isolated with MAbs from the
yolk sac.................................. 80

The biochemical nature and antigenic
determinants of the differentiation
markers recognized by MAbs................. 97




Figure 1-1.

Figure 1-2.

Figure 1-3.

Figure 1-4.

Figure 2-1.

Figure 2-2.

Figure 2-3.

Figure 2-4.

Figure 2-5.

Figure 2-6.

Figure 2-7.

Schematic illustration of the chick
embryo at the 3rd day of embryogenesis....

Diagram of hematopoiesis...................

Inerference of the AEV with cells of
the erythroid lineage.....................

Interference of the AMV with cells of
the monocytic lineage .....................

Flow cytometric analysis of the 20%/50%
interface of a percoll gradient of 2-week-
old bone marrow cells .....................

APAAP immunoenzymatic staining of 20%/50%
interface of a percoll gradient of 2-week-
old bone marrow cells .....................

Flow cytometric analysis of the 50%/70%
interface of a percoll gradient of 2-week-
old bone marrow cells .....................

Flow cytometric analysis of the 30%/50%
interface of a percoll gradient of 4-day-
embryo yolk sac cells ....................

Flow cytometric analysis of the 20%/50%
interface of a percoll gradient of 2-week-
old bone marrow cells tagged with
2E10-1E10 .................................

Flow cytometric analysis of the 50%/70%
interface of a percoll gradient of 2-week-
old bone marrow cells tagged with
2E10-1E10 .................................

Reactivitities of MAbs to BM2 cells and
PMA-differentiated BM2 cells..............


Figure 2-8.

Figure 3-1.

Figure 3-2.

Figure 4-1.

Figure 4-2.

Figure 4-3.

Figure 4-4.

Figure 4-5.

Figure 4-6.

Figure 4-7.

Figure 5-1.

Reactivities of MAbs to 6C2 cells and
butyric acid-treated 6C2 cells............

Flow cytometric analysis of 3D7-1C9
tagged bone marrow cells before sorting...

Flow cytometric analysis of 3D7-1C9
tagged bone marrow cells after sorting....

Reactivities of MAbs to BM2/C3A cells with
enzymatic digestion and chemical

Reactivities of MAbs to 6C2 cells with
enzymatic digestion and chemical

Reactivities of MAbs to MSB1 cells with
enzymatic digestion and chemical

Silver stain of 10% SDS-PAGE analysis
of immunoprecipitates obtained from
unlabeled BM2/C3A cells ...................

Fluorography of 10% SDS-PAGE analysis
of immunoprecipitates obtained from
35 S-Methionine-labeled BM2/C3A cells......

Fluorography of 10% SDS-PAGE analysis
of immunoprecipitates obtained from
35 S-Methionine-labeled 6C2 cells..........

Fluorography of 10% SDS-PAGE analysis
of immunoprecipitates obtained from
35 S-Methionine-labeled MSB1 cells.........

Diagram of the specificities of MAbs
for the hematopoietic cells...............













ACS Anemic chicken serum

AEV Avian erythroleukemia virus

AHS gamma globulin-free horse serum

ALV Avian leukemia virus

AMV Avian myeloblastosis virus

APAAP Alkaline phosphatase and monoclonal anti-alkaline

BFU-E Burst-forming unit-erythroid

BPA Burst-promoting activity

CEF Chicken embryo fibroblast

CFU-E Colony-forming unit-erythroid

CFU-M Colony-forming unit-marrow

cpm Counts per minute

CSF Colony-stimulating factor

DLV Defective leukemia virus

DMSO Dimethyl sulfoxide

DMEM Dulbecco's modified Eagle's medium

EGF Epidermal growth factor

ELISA Enzyme-linked immunosorbent assay

FACS Fluorescence-activated cell sorter

FBS Fetal bovine serum

FITC Fluorescein-isothiocyanate




















Goat anti-mouse polyvalent IgA, G & M

Hank's balanced salt solution

Lymphoid leukosis virus


Monoclonal antibody

Myeloblastosis-associated virus

Macrophage-granulocyte colony-forming cell

Multiplicity of infection

Phosphate-buffered solution

Phorbol 12-myristate-13-acetate


P-nitrophenyl phosphate

P-phenylene diamine

Rabbit anti-mouse IgG

Rous-associated virus


Sodium dodecyl sulfate-polyacrylamide gel

Tris-buffered saline


Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy




May 1990

Chairman: Dr. Carlo Moscovici
Major Department: Pathology and Laborotary Medicine

Avian hematopoiesis has been an excellent model

for resolving numerous enigmas about growth and

development. The interaction of avian retroviruses with the

avian system has created a relatively new discipline of

onco-development which allows us to analyze abnormal tissue

growth and hematological disorders in a more sophisticated


The specific aim of this project is to identify

lineage-specific differentiation markers in normal avian

hematopoietic cells and transformation-associated

antigens in retrovirus-transformed cells by utilizing

monoclonal antibody techniques. Four groups of MAbs were

selected among nearly 5,000 supernatants from 10 fusions.

Characterization of the cell-type specificities was achieved


by radioimmunoassay, immunofluorescence staining, flow

cytometry and immunoenzymatic staining as well as FACStar

sorting or immunomagnetic bead separation followed by

colony-forming assays and transforming assays. Analysis of

MAbs revealed that 1) MAbs 1HlO-1F9, 2H1-2A10 and 3D7-1C9

are specific for transformation-associated antigens present

preferentially on BM2 cell lines rather than on normal

monocytic cells. The expression of these antigens was

diminished after BM2 cells were induced to differentiate.

2) MAb 1F7-1A3 recognizes BFU-E and CFU-E, AEV-transformed

yolk sac cells and MSB1 cells. 3) MAb 3F6-1E7 reacts with

the embryonic stem cell and precursor cell populations. The

expression of the marker recognized by MAb 3F6-1E7 was also

observed on some tumor cells, e.g., AEV-transformed yolk sac

cells, BM2 cells and MSB1 cells. 4) MAb 2E10-1E10 defines a

marker present on proliferating hematopoietic cells instead

of terminally differentiated cells, however, it starts

appearing only after the 4th day of embryogenesis.

Trypsinization, neuraminidase digestion and

deglycosylation treatment reduced the binding

specificities of MAbs 1H10-1F9, 2H1-2A10, 3D7-1C9 and

2E10-1E10. This suggests that the markers recognized by

these MAbs are glycoproteins and that sialic acids with or

without carbohydrates are contributing to the conformation

of the antigenic determinants. Conversely the antigenic


determinants for MAbs 1F7-1A3 and 3F6-1E7 must be strictly

proteins, since only trypsinization was able to inhibit

their binding specificities.

This study will permit investigations focusing on the

expression of these markers to bring us a step closer

toward the understanding of the mechanisms involved in

regulating proliferation and differentiation of

normal cells versus tumor cells.





Proliferation and differentiation of normal cells is

controlled by the interactions with other cell types, with

extracellular matrix and with regulatory molecules such as

growth factors and differentiation factors. Although the

regulatory mechanisms are extremely complicated, they have

been programmed in such a way as to maintain proliferation

and differentiation of the cells in a harmonic state. In

other words, the loss of cells from the stem cell

compartment by differentiation into committed progenitor

cells must be balanced by replenishment via self-renewal of

the stem cells. If too many stem cells undergo

differentiation, the stem cell reserve will rapidly become

exhausted; if too many stem cells undergo self-renewal

rather than differentiation, the production of mature cells

will drastically fall.

Cancer is believed to be a molecular disease resulting

from the deregulation of proliferation and differentiation,

i.e., the harmony has been short-circuited by the

constitutively triggered self-renewal machinery with or

without the blockage of the differentiation pathway.

It has been noticed that AEV-transformed avian

embryonic yolk sac cells can eventually undergo spontaneous

differentiation into mature erythrocytes (Jurdic et al.,

1985) and that spontaneous regression and differentiation of

human neuroblastomas are observed occasionally (Evans et

al., 1976). In addition, a variety of tumor cells have also

been shown to revert to a normal state under different

conditions despite the continued expression of activated

oncogenes. For example, the tumorigenicity of hybrids

formed between normal and tumor cells has been completely

suppressed (Stanbridge et al., 1982); embryonal carcinoma

(Pierce et al., 1979), neuroblastoma (Podesta et al., 1984),

B16 melanoma (Pierce et al., 1984) and murine leukemia

(Gootwine et al., 1982) have been converted into benign cell

lineages by their appropriate embryonic environments;

naturally occurring substances such as colony-forming

factors (Sachs, 1986), glia maturation factor (Lim et al.,

1986) and transforming growth factors (Sporn et al., 1986)

have been shown to be able to induce differentiation of

tumor cells; while a number of chemical agents such as

hexamethylene bisacetamide, retinoid acid, 5-azacytidine and

DMSO etc. can also induce terminal differentiation and/or

reverse the neoplastic phenotype of malignant cells (Bloch,

1984; Fresney, 1985). Moreover, in some cases, terminally

differentiated cells such as macrophages can still serve as

the target cells for transformation by a group of

retroviruses, namely AMV, MC29 and MH2 (Pessano et al.,

1979). All the information mentioned above suggests that

self-renewal and differentiation of cells are regulated by

separate mechanisms. The roles of various protooncogenes

and antioncogenes in these processes are yet to be


Normal Avian Hematopoiesis


The avian hematopoietic system has provided a unique

and interesting model to study mechanisms of the regulation

of cell proliferation and differentiation in normal versus

tumor cells. It has several distinct features compared to

that of mammals, including the presence of the bursa of

Fabricius which is involved in the differentiation of B

lymphocytes, the expression of class IV MHC antigens, coded

by the B-G region, on the surface of the mature erythrocytes

(Miller et al., 1982). In addition, the erythrocytes are

nucleated and oval-shaped, and the nucleated thrombocytes,

instead of the platelets, are responsible for the hemostasis

in the avian system. There are several advantages in using

the avian models. For instance, tolerance to foreign

antigens can be developed during early ontogeny (Hasek and

Hraba, 1955), and hematopoiesis can be studied both in vivo

and in vitro by using retroviruses as indicators for

specific precursors present within each lineage.

The most studied avian hematopoietic system is in the

chicken. The outbred SPAFAS line has been used in our

laboratory to carry out all the experiments for my

dissertation project. My work has focused mainly on the

erythroid and monocytic lineage and their interaction with

the avian retroviruses.

The Blood-Forming Organs

The first blood cells appear after 18 hours of

incubation in the blood islands disseminated in the

blastoderm. During embryogenesis, the yolk sac represents

the major hematopoietic organ until day 15. Erythropoiesis

in the spleen starts around day 9 and continues through day

16 to 18 with a peak at day 15. The bone marrow begins its

function on day 12 and becomes the main site of

hematopoiesis throughout adult life (Dieterlen-Lievre,


The yolk sac

The yolk sac becomes established during the third day

of embryogenesis and it originates in the extraembryonic

region consisting of a peripheral "area vitellina" which is

made up of ectoderm and endoderm only, and a central "area

vasculosa" consisting of all three germ layers (Figure 1-

1). The area vasculosa contains the blood islands and its

boundary is delineated by a circular blood vessel, the sinus

marginalis. As the area vitellina grows over the yolk, the

area vasculosa also increases in size and invades the area

vitellina. Eventually the latter disappears completely, and

the entire yolk sac is vascularized.

The bone marrow

In the adult chicken, the bone marrow remains the major

source for erythropoiesis, granulopoiesis and lymphopoiesis.

The spleen does not appear to play a role in hematopoiesis

in the adult. In the avian bone marrow, erythropoiesis

occurs in the lumen of the medullary sinuses, while

granulopoiesis and lymphopoiesis are compartmentalized

within the extravascular spaces (Campbell, 1967). In

addition, masses of lymphatic tissue with germinal centers

are also present (Campbell, 1967; Payne and Powell, 1984).

However, in the mammalian bone marrow, erythropoiesis is

confined to the extravascular spaces and there is no

lymphatic tissue at all.

The development of the bone marrow during ontogeny has

been studied in the chick embryo by Sorrell and Weiss (1980)

using light, scanning and transmission electron microscopy.

The bone marrow cells can be obtained from the chick embryo

as early as 12 days of incubation. Marrow at this stage is

Figure 1-1. Schematic illustration of the chick
embryo at the 3rd day of incubation. The yolk sac
is composed of the area vitellina and the area
vasculosa. As the area vitellina grows over the
yolk, the area vasculosa also increases in size
and invades the area vitellina. Eventually the
latter disppears completely and the entire yolk
sac is vascularized. 1, embryo; 2, area
pellucida; 3, area vasculosa; 4, sinus terminus;
5, area vitellina internal; 6, area vitellina





o !- \,

\, y '




richer in stem cells and non-committed progenitor cells than

the older bone marrow, but it already harbors cells which

are committed to specific hematopoietic lineages. The

latter cells are less numerous than in the adult bone marrow

and consist essentially of hemocytoblasts.

Other blood-forming organs

The spleen, the bursa, the liver and the thymus

function as additional hematopoietic organs. In the adult,

lymphopoiesis occurs mainly in the thymus and in the spleen,

whereas the bursa, from which the precursors of B

lymphocytes originate, is a transient granulopoietic organ.

Recent investigations (Cormier and Dieterlen-Lidvre,

1988) report that some intraembryonic sites may be another

source of hematopoietic stem cells in the developing embryo.

At 3-4 days of incubation, the wall of the dorsal aorta

surrounding the intraembryonic mesenchyme is found to be the

site from which hematopoietic progenitor cells emerge, i.e.,


The Differentiation Pathway of Hematopoietic Cells

The cells of the hematopoietic system arise by

proliferation and differentiation of the progenitor cells.

This process begins with multipotential stem cells which can

self-renew as well as undergo progressive differentiation to

progenitor cells committed to the particular lineages,

ultimately yielding mature blood cells (Metcalf and Moore,


Analysis of the stem cells and progenitor cells in the

different hematopoietic tissues has been useful in

clarifying the differentiation pathway as well as in

exploring the regulatory mechanisms of hematopoiesis.

Hematopoietic stem cells

The stem cell population is the fundamental base from

which all the major hematopoietic cell lines are derived.

This population is thus considered to be pluripotent in its

differentiation potential, giving rise to erythroid,

granulocytic, monocytic, megakaryocytic and lymphoid

lineages (Figure 1-2). However, the stem cells account only

for 0.01% of the total bone marrow cells in a normal mouse,

and for 0.003-0.004% in a normal chicken (Table 1-1). In

addition, since they are morphologically indistinguishable,

their existence can only be inferred by the progeny they


Efforts to identify chicken hematopoietic stem cells

have followed the protocol of the transplantation

experiments by Till and McCulloch (1961) whereby they have

identified the mouse hematopoietic stem cells. Samarut, et

al. (1976) transplanted normal chicken bone marrow into

irradiated chickens. Six days later, erythrocytic colonies

were observed only on the surface of the tibial marrow.


k 0H --0

*4 40 r4I U

0 4) 4 )i

4 $4 J 0 -- 0
E 40) O0
4 4 ** 4iH

,, o m o 4)
0 to 40 04 -4U
.4 -4 (0 I 0 I

U-4 0 > -E ,
)0 0 0 1 40

0 ...i-W 0I-
E4 UC > CP

S4)0 CO
o 4M u4 0H

E- ) 0 r0 I
S0 0 U

*0 0 4I W0 I
*r4 M 0 4 0 0
o p 4 oro 0
M Vl 0

0 *2V OW
0 4) .Q *0 -

0 4) 0 >4
-0 M 4 U --
0) in HC *. 0

00 0*)

*40 4Or-
0) 0 V 4-)

04 m 4) 1C 0
IO I H C00

H E0 0 Q .- 44 U
-r 0 (C4 0

W 0) VV 0 I 044

&1- 0 >1 0 rl(

7Q .


of C





ct 0
i C

? C

cc m


C2 ~










O ^
\ ll






















C. 0





- N


Table 1-1. Chicken hematopoietic precursor cells

Hematopoietic Frequency*
precursor cells Bone marrow Yolk sac Progeny

BFU-E 110-160 300-600 1,000-2,000

CFU-E 500-2,000 1,500-1,800 8-150

(early progenitor) 250-400 100-200 50-1,000

(late progenitor) 1,000-1,300 100-200 3-50

Source: Modified from Moscovici and Gazzolo (1982).
*Expressed as per 105 cells.

Each colony was originated from one single cell, i.e.,

colony-forming unit in the marrow (CFU-M). However, neither

macrophage-granulocytic colonies nor mixed types of colonies

were observed in the marrow of the irradiated chickens. It

is still unclear whether the medullar environment does not

favor the development of colonies other than the ones of the

erythroid lineage or if the CFU-M represents only the stem

cells at the earliest step of commitment in the erythroid


Committed progenitor cells

Committed progenitor cells are directly derived from

the stem cells and are each committed to a specific

differentiation pathway or lineage. Commitment is an

irreversible step whereby these cells have lost the

potential to generate hematopoietic cells of other lineages.

Most of the proliferative activity in the bone marrow seems

to occur in the progenitor cells committed to the production

of single or restricted ranges of hematopoietic cell types.

Only a small proportion of the pluripotential stem cells is

cycling at a given time. The progenitor cells of both

erythroid and myeloid lineages will be discussed in detail.

The erythroid lineage

The avian erythroid compartment consists of three

distinct lineages, namely the primitive, the intermediate

and the definitive lineage, respectively (Table 1-2). The

cells from the primitive lineage are produced by the early

blood islands and mature "in cohort" (Ingram, 1972). These

cells are released at a very immature state, but they

continue to divide and mature synchronously within the blood

vessels between 2 and 5 days of incubation. These cells are

called megalocytes because of their large size. They are

spherical with round nuclei and synthesize hemoglobins E

(embryonic) and P (primitive) which are specific for the

primitive lineage. At 5 days, cells of the intermediate

erythroid lineage begin entering the blood and eventually

supersede the primitive cells. At 7 days, the primitive

cells account for only 5% of the red blood cells, and after

12 days of incubation they are rarely encountered. The

cells of the intermediate lineage are observed from 5 to 6

days until 18 to 20 days. They synthesize specific

hemoglobin H (hatching). It is not until 18 to 21 days of

incubation that mature erythrocytes of the definitive

lineage start to appear in the blood circulation. They are

oval-shaped with oval nuclei and have hemoglobins A (adult)

and D (definitive) (Bruns and Ingram, 1973).

The progenitor cells of the definitive lineage are

morphologically unrecognizable. However, by the use of in

vitro colony forming assays in methylcellulose, two classes

of erythroid progenitor cells have been identified,

the colony-forming unit-erythroid (CFU-E) and the burst-

Table 1-2. The three different lineages of the avian
erythroid compartment

Primitive Intermediate Definitive

Appearance Day 2-7 Day 5-20 Day 18-Hatched

Progenitor Megaloblast? BFU-E, CFU-E BFU-E, CFU-E

Mature Cell Megalocyte Erythrocyte Erythrocyte

Hemoglobin* E: a* + e H: a* + pH A: aA + PA
P: i + p D: aD + pA

*Source: Modified from Bruns and Ingram (1973).
a-like globin: a*, ir, aD; P-like globin: e, p, pH, A^.
Abbreviation: E, embryonic; P, primitive; H, hatching;
A, adult; D, definitive.


forming uint-erythroid (BFU-E). The BFU-E give rise, after

six days in culture, to large aggregates made of several

benzidine-positive clusters containing about 1,000

erythrocytes (Samarut and Bouabdelli, 1980). These BFU-E

are highly sensitive to burst-promoting activity (BPA), and

are also dependent on high concentrations of erythropoietin.

The CFU-E proliferate to form one compact colony of 8

to 150 benzidine-positive erythrocytes after three days of

incubation (Samarut et al., 1979). The requirement for

erythropoietin in the development of CFU-E is lower than

that for BFU-E. The BFU-E and CFU-E can be detected in the

yolk sac as well as in the embryonic and adult bone marrow.

The BFU-E are also found in the blastoderm at the primitive

streak stage (18 hours of incubation), whereas the CFU-E are

not yet detectable (Samarut and Bouabdelli, 1980). The

BFU-E and CFU-E can be also distinguished by the expression

of two different cell surface antigens which are recognized

by polyclonal antisera (Gazzolo et al., 1980; Samarut et

al., 1979). An antigen specific to immature red blood cells

is present on the CFU-E but not detectable on the BFU-E.

Conversely, a chicken brain-related antigen is expressed on

the BFU-E and less expressed on the CFU-E.

In the murine system, subpopulations of the BFU-E at

different degrees of maturation have been observed (Gregory

and Eaves, 1978). This is not the case, however, for the

BFU-E in the chickens.

The myeloid lineage

Hematopoietic cells of granulocytic and monocytic

lineages are referred to as myeloid cells. In response to

infection, the progenitor cells in the bone marrow, i.e.,

granulocytic-macrophage colony-forming cells (GM-CFC), would

rapidly produce a large amount of mature granulocytes and

monocytes under the control of a variety of colony-

stimulating factors (CSFs). In the murine and human system,

four kinds of CSFs involved in myelopoiesis have been

characterized as IL3, GM-CSF, G-CSF and M-CSF. Conversely,

the specific CSFs in the chicken have yet to be identified.

Nevertheless, sources of CSFs can be furnished by using an

underlayer of macrophages (Graf et al., 1981), or by adding

to the semi-solid medium either egg albumin (Szenberg,

1977), or serum from endotoxin-injected chickens (Dodge and

Hansell, 1978) or a conditioned medium from chicken

fibroblast cultures (Dodge and Moscovici, 1973; Dodge et

al., 1975; Gazzolo et al., 1979). AMV-transformed cells

were shown also to be capable of producing CSFs (Silva et

al., 1974). Recently a myelomonocytic growth factor was

isolated from medium conditioned by a transformed macrophage

cell line (Leutz et al., 1984). This factor (MGF) promotes

the growth of macrophage colonies together with a minor

proportion of granulocytes.

The existence of chicken GM-CFC can be detected when

bone marrow cells are cultured in soft agar or

methylcellulose media. The wide range of the colony size

obtained suggests that these progenitor cells display

different degrees of maturity. The less mature CFC give

rise to clusters containing from 50 to more than 2,000 cells

(Dodge and Moscovici, 1973; Dodge et al., 1975), while the

more mature CFC produce colonies from 3 to 50 cells (Gazzolo

et al., 1980). Colonies which are composed mostly of

macrophages readily develop in a semi-solid medium

containing chicken serum and fibroblast-conditioned-medium

(Dodge and Hansell, 1978). Granulocytic colonies will

develop if the chicken serum is depleted from the medium and

the fibroblast-conditioned-medium is replaced by spleen-

conditioned-medium (Dodge and Sharma, 1985). This finding

suggests the existence of a factor similar to M-CSF in the

mouse and human. The monocytic colonies are composed of

scattered cells, whereas the granulocytic colonies are

dense. The cells from both colonies contain granules.

The GM-CFC have been observed in various chicken

hematopoietic tissues of embryonic and adult stages (Dodge

and Moscovici, 1973; Dodge et al., 1975; Szenberg, 1977).

Moreover, the CFC can be enumerated in the blastoderms

incubated for 24 hours (Moscovici et al., unpublished

results). A higher percentage of CFC was also found in the

embryonic spleen and bone marrow. The frequency dropped

rapidly once the chickens hatched. Interestingly, a peak of

CFC occurred in the bursa at 14 and 15 days of incubation,

indicating that the stem cells colonizing the bursa

differentiate first into myeloid elements and subsequently

into lymphoid ones (Szenberg, 1977).

Interaction of the Avian Leukemia Viruses with
Hematopoietic Cells

There have been many comprehensive reviews to date on

the molecular biology and the pathogenesis of the avian

leukemia viruses (ALVs) (Moscovici and Gazzolo, 1982; Graf

and Stdhelin, 1982; Bishop, 1983; Enrietto and Wyke, 1983;

Bister and Jansen, 1986). The ALVs belong to a taxonomic

subfamily termed oncovirinae (specifically, avian leukosis-

sarcoma group of type C RNA tumor viruses) within the family

of retroviridae (retroviruses) (Fenner, 1976). They

contribute to a variety of avian hematopoietic as well as

non-hematopoietic disorders. The ALVs can be divided into

two groups according to the pathological features: the

defective leukemia viruses (also known as acute leukemia

viruses) and the avian leukosis viruses (Hanafusa, 1977).

The defective leukemia viruses (DLVs)

These viruses induce various types of acute leukemia

within a few weeks after inoculation. They also cause

sarcomas and carcinomas in some cases. All the strains

known can transform the cells of specific hematopoietic

lineages in vitro. In addition, most of them transform

fibroblasts in culture as well, with the exception of AMV

and E26. Another distinct aspect of the DLVs is that they

are all replication-defective due to total or partial

deletions of the essential virion genes: gag, pol and env.

Consequently, they can produce infectious progeny only in

the presence of the avian leukosis helper viruses. The

deleted sequences are replaced by the viral oncogenes

(v-onc), i.e., v-myc, v-mil, v-erbA, v-erbB, v-myb, and

v-ets, which originated from transduction of mutated or

truncated forms of protooncogenes. Their gene products are

responsible for the transformation of the hematopoietic

cells. Based on the predominant response of the

hematopoietic system of the infected host and the major

types of oncogenes which they carry, three subgroups of

DLVs and their respective representatives can be

distinguished: i) the MC29 subgroup: myelocytomatosis

(v-myc), ii) the AEV subgroup: erythroblastosis (v-erb),

iii) and the AMV subgroup: myeloblastosis (v-myb) (Table

1-3). However, a more detailed description of the

interaction of the AEV and the AMV with the hematopoietic

system will be given since more data have been obtained in

the last decade.

The Lymphoid leukosis viruses (LLVs)

In contrast to the DLVs, the LLVs do not contain any

v-onc and they are fully competent for replication. Because

Table 1-3. Avian defective leukemia viruses

Subgroups Neoplasms Cell types
and Viral induced transformed
virus strains oncogenes in vivo in vitro

MC29 subgroup
Strain MC29 v-myc Myelocytomatosis, Myeloid,
Strain CMII v-myc endothelioma, macrophage,
Strain OK10 v-myc carcinoma epithelioid,
Strain MH2 v-myc, v-mil fibroblastic

AEV subgroup
Strain AEV-R v-erbA, v-erbB Erythroblastosis, Erythroid,
Strain AEV-H v-erbB fibrosarcoma fibroblastic

AMV subgroup
Strain AMV v-myb Monoblastosis Monocytic

Strain E26 v-myb, v-ets Erythroblastosis, Erythroid,
monoblastosis monocytic

Source: Modified from Bister and Jansen (1985).

of the absence of v-onc, the LLVs don't transform cells in

vitro and most strains induce predominantly lymphoid

leukosis in vivo only after a long latent period of several

months or longer by the activation of protooncogenes. In

the cases of chicken B-cell lymphomas induced by the LLVs,

the viral LTR regions containing an enhancer and a promoter

were found to be integrated very close to the c-myc gene

(Hayward et al., 1981). Most integration of LLVs result in

the separation of exons II and III of the c-myc gene from

the normal promoter and exon I (Payne et al., 1982; Shih et

al., 1984), causing a fifty-fold higher transcription of c-

myc RNA under the control of the LTR promoter than the 5

copies found in normal cells. In a minority of tumors, the

LTR integrated in the opposite orientation to that of the c-

myc and in one case the provirus was actually located at the

3' end of the c-myc (Payne et al., 1982). It is thought in

these rare events that the enhancer element in the viral LTR

probably increases the transcription from the normal c-myc

promoter. More recently, the LLVs have also been found

integrated at the c-erbB locus in chicken erythroblastosis

(Fung et al., 1983). An elevated level of c-erbB

transcripts was observed.

LLVs under different conditions, for example, chicken

genotype, site of injection, age of host, etc., may induce a

larger spectrum of diseases including osteopetrosis, anemia,

nephroblastoma and occasional fibrosarcomas and


The Avian Erythroleukemia Virus (AEV)

The oncogenes of the AEV

The AEV can induce erythroblastic leukemia and sarcomas

in infected birds within a short period of time (Graf and

Beug, 1978; Moscovici and Gazzolo, 1982). The virus also

transforms chicken fibroblasts and hematopoietic precursor

cells of the erythroid lineage in vitro (Graf et al., 1981).

It carries two oncogenes, namely v-erbA and v-erbB. The

v-erbB encodes a protein of 61 Kd which is glycosylated in

infected cells to higher molecular weight forms (Privalsky

et al., 1983). The gp65"er and gp68erb are localized in the

intracellular membrane, while the mature gp74rbB is found in

the plasma membrane (Privalsky and Bishop, 1984). The

latter protein represents a truncated form of the receptor

for epidermal growth factor (EGF) (Downward et al., 1984),

where the extracellular EGF-binding domain has been deleted,

but the region for tyrosine-specific protein kinase is

preserved (Gilmore et al., 1985). It has been postulated

that the transforming potential of v-erbB is due to the lack

of regulation by EGF resulting in the constitutive

activation of the tyrosine kinase (Hayman and Beug, 1984;

Privalsky and Bishop, 1984). However, the exact mechanism

of transformation by the tyrosine kinase activity is still


The v-erbA is a mutated truncation transduced from a

cellular multigene family encoding the thyroid hormone

receptor (Sap et al., 1986). The v-erbA protein,

p75gag-ert, is linked with the gag product, is unglycosylated

and has no kinase activity. It appears to be a DNA-binding

protein exhibiting distinct nuclear and cytoplasmic

subcellular locations (Boucher et al., 1988). Nevertheless,

it doesn't seem to bind thyroid hormone (Sap et al., 1986).

The transforming potential of v-erbA and v-erbB

It has been shown that v-erbB alone is necessary and

sufficient to induce cell transformation (Frykberg et al.,

1983; Sealy et al., 1983) by constructing a series of

deletion mutant viruses. The data corroborate the ability

of the product of v-erbB gp65-68erb to independently induce

erythroleukemia in chickens and transform fibroblasts in

vitro. In contrast, constructs which produce the v-erbA
p75ag-erbA only were incompetent for transforming activity in

vitro. However, other observations from studies of

temperature sensitive mutants (Beug and Hayman, 1984),

transductants of the c-erbB gene alone (Fung et al., 1983;

Yamamoto et al., 1983), suggest that v-erbA may play a

distinct role in maintaining the proliferation and

transformed phenotype of AEV-infected cells. It has been


shown that v-erbA possesses the capability to potentiate the

erythroid transformation not only by v-erbB but also by

other oncogenes (Frykberg et al., 1983; Kahn et al., 1986).

These phenomena could be partially due to the suppression of

the transcription of the anion transporter (band 3) gene by

the v-erbA proteins (Zenke et al., 1988). It is not until

recently, however, that the transforming potential of

v-erbA has been reevaluated. The XJ12 vector which carries

v-erbA oncogene in association with Neo R (neomycin

resistance) gene is shown to be able to transform bone

marrow cells in vitro (Gandrillon et al., 1989; Moscovici

and Moscovici, unpublished results).

The target cells for AEV

The target cells for AEV infection in the bone marrow

of the hatched chick are recruited within the BFU-E (Burst

forming unit-erythroid) compartment (Gazzolo et al., 1980).

After AEV infection, the transformed cells continue to

proceed within the differentiation pathway until they are

blocked at the stage of CFU-E (colony forming unit-

erythroid) (Samarut and Gazzolo, 1982). These cells express

the erythroid markers of the CFU-E, but have gained the

self-renewal potential to forego the fate of terminal

differentiation. Conversely, the target cells for AEV in

the embryo are within either the CFU-M or the pre-BFU-E or

in both compartments (Jurdic et al., 1985). The embryonic


transformed colonies are partially hemoglobinized even after

subcloning, which suggest that the AEV-transformed embryonic

cells can escape the blockage and undergo spontaneous

differentiation in contrast to the AEV-transformed adult

cells (Figure 1-3).

The Avian Myeloblastosis Virus (AMV)

The oncogene of the AMV

The AMV induces rapid myeloblastic leukemia in the

chickens and transforms hematopoietic cells of monocytic

lineages in vitro. It can be divided into subgroups A and

B, depending on the expression of envelope glycoproteins of

the helper viruses (Moscovici et al., 1975). Besides the

AMV transforming agent, this strain contains two

nondefective helper leukosis viruses, the myeloblastosis

associated virus 1 and 2 (MAV-1 & MAV-2) of subgroups A and

B, respectively (Moscovici and Vogt, 1968).

The oncogene of AMV, v-myb, encodes a transforming

protein p45m b which is located in the nucleus (Klempnauer et

al., 1984). It has been shown that p45*b has DNA- binding

activity (Boyle et al., 1985). The v-myb proteins are

associated with their nuclear sublocation. The exact

function of v-myb proteins remains to be clearly

established. However, recent results indicate that v-myb

along with c-myb may act as transcriptional activators

(Weston and Bishop, 1989).

Figure 1-3. Interference of the AEV with cells of the
erythroid lineage. The AEV only transforms the cells
at the BFU-E stage from the adult bone marrow.
Transformed cells are frozen at the CFU-E stage and
capable of self-renewal. Only cells at the pre-BFU-E
stage are the target cells for AEV transformation in
the embryonic yolk sac. Although the transformed cells
display the phenotypes of CFU-E, they will eventually
escape the blockage and spontaneously differentiate
into erythrocytes. The brain antigens (Br) are
expressed on the BFU-E more than on the CFU-E, the
immature antigens (Im) are found on the CFU-E and
erythroblasts, meanwhile only erythroblasts and
erythrocytes are capable of synthesizing the
hemoglobins (Hb).

AEV Infection
Erythropoiesis Adult Embryonic
Bone Marrow Yolk Sac









Br +

Im +
Br -

Br -

Im -
Br -




The target cells for the AMV

The characterization of AMV target cells by density,

velocity sedimentation, adherence and phagocytic activity

indicate that they are recruited among a wide range of cells

within the monocytic lineage from the stage of the

myelomonocytic progenitors, i.e., CFC (colony-forming cells)

(Gazzolo et al., 1979) committed toward the macrophage

lineage (Boettiger and Durban, 1984) to the terminally

differentiated macrophages (Moscovici and Gazzolo, 1982).

Regardless of the origin of the target cells, the AMV-

transformed cells are morphologically identical, and possess

the same functional and surface properties (Gazzolo et al.,

1979; Beug et al., 1979; Durban and Boettiger, 1981). They

are mostly nonadherent and round which diametered about 10

micrometers. Their large and eccentric nucleus is

surrounded by a rim of cytoplasm containing small granules.

The receptors for the Fc region of immunoglobulins are

expressed on the cell surface, whereas C3 receptors are not

present. Normal avian macrophages express both receptors.

Although the AMV-transformed cells can engulf latex

particles mediated by nonspecific receptors, phagocytosis

mediated by Fc receptors does not occur, i.e., these Fc

receptors are not functional. Acid phosphatase and

adenosine triphosphatase are also found in the cytoplasm and

on the membrane of the transformed cells. Moreover, after

treatment with phorbol 12-myristate-13-acetate (PMA), a


tumor promoter, the AMV-transformed cells became adherent to

the surface of the culture flask and eventually

differentiated into macrophages (Pessano et al., 1979).

There were no obvious alterations in terms of the expression

of the v-myb proteins in these PMA-differentiated cells.

However, the v-myb proteins were found to be located in the

cytoplasm instead of in the nucleus (Symonds et al., 1984).

The differentiation into macrophages was also obtained with

a temperature-sensitive mutant of AMV when the transformed

cells were shifted to a non-permissive temperature

(Moscovici and Moscovici, 1983).

In conclusion, cells from all stages of monocytic

lineages, from the committed progenitors to the mature

macrophages, may serve as the target cells for AMV. Once

the cells are transformed, they become frozen at a stage

between monoblasts and monocytes (Figure 1-4).

Identification of Cell Surface Markers in Normal
Hematopoietic Cells and Tumor Cells by MAbs


The cellular microenvironment plays a crucial role in

regulating the proliferation and differentiation of normal

cells. A cell will interact with adjacent cells and with

structural components of the extracellular matrix via the

external surface of its plasma membrane. Although little is

known about the mechanisms of cellular interactions, it is

well established that cell development requires that the

HO )0
r-I 44 w
a) H r

# O & I0 ,C
Sr-4 0 4- 0
0 02 0 r.
04) MA 0
OA.Q P r- W
0)0 ) 4)

-4 4 4 0 43
4.4- V 4+

0 t MO

01 O) M Cr

O i ~M t
UI+V 4.1 k-)W0
0 to 010 to M.

(D 00k0C

,4 4) V 0 U M
4-)C tC -4 r

0 to k 0) 4V k t
0 10 4 C 0
oU) C) (n
r-H )) 4)4 4 IU W5

0 H0) 9 I
SA it 4)V
,0 *H r- r-1I a >4
*4 r-H 10 O ) 0
> 0 0P vr > on
> U 04 0) M

) 0 4-) E
O44 S 4 k0
V. 0 4,C)H 3
9> 6 f-i g)
044 0 H 0
J + (D 0
a) *- ) > a) 0
r. 01 0)0
a) E-4 0 E-4 $4
a)a) >1 4) 015t

4 0 r. 0 04) 00
H4J *i Ni S0

S* 0 .C u 04

a) 44 a) to H-4
I 4 .1 W V k i ) )(
H -1 C Ut ) $4 > r.

$4i 4() Ao <-) Vo
Sr 4 0I4) r

rT4 0 V 00 NO4-1








0 0)


0 0

0 s0
S 8



CL cc aC
0 ) )
Su- o()

-1& I

C C)




8 0

E o c8
E ct 13L
Q.. <

surface membrane should be capable of receiving and

transmitting regulatory signals, i.e., growth factors and

differentiation factors from the microenvironment. Cancer

is a disease resulting from abnormalities in both cell

proliferation and differentiation characterized by increased

growth rate, prolonged survival, decreased adhesion, loss of

contact inhibition, increased invasiveness and motility,

expression of repressed antigens and escape from immune

surveillance (Wallach, 1968). All these phenomena have been

shown to be associated with alterations in structure and

function, and in particular with an aberrant glycosylation

of the cell surface membrane. As a result, cancer can be

regarded as a molecular disease of cell surface

glycoconjugates (Abe et al., 1983).

Cell surface glycoconjugates comprise a heterogeneous

group of compounds, all of which contain carbohydrate

N-glycosidically or O-glycosidically linked to protein

glycoproteinss) or O-glycosidically to lipid (glycolipids).

The predominant glycoconjugates are glycoproteins containing

at least 80% of all cell surface-located carbohydrate. It

is assumed that all membrane-bound proteins but only 10% of

membrane lipids are glycosylated (Shinitzky, 1984).

However, it is extremely difficult to identify specific

glycoconjugates expressed exclusively on tumor cell surface.

All identified tumor-associated markers to date are found to

be more or less expressed on normal cells at particular

stages of the differentiation pathway (Old, 1981).

The identification and characterization of cell surface

markers can provide us with invaluable information on

various aspects of differentiation and oncogenesis. First,

the identification of cell surface markers that are specific

for particular stages of differentiation and maturation will

enable us to follow the differentiation pathway in molecular

terms. Mechanisms regulating the expression of cell surface

markers will allow us to understand how functionally mature

cells are formed and what kind of structural elements are

necessary for functional differentiation. Secondly, they

could be used to study whether tumor cells are arrested at a

certain stage of differentiation. In addition, the studies

could also enable us to determine how the control of

proliferation and differentiation in tumor cells differs

from that of normal cells. Thirdly, differentiation markers

can be used for diagnostic and therapeutic purposes (Fukuda,


The introduction of hybridoma technology by Kohler and

Milstein (1975) has made a significant contribution in the

study of both normal and malignant cell surface markers.

MAbs have been used as powerful tools for the detection,

isolation and characterization of cell surface markers.

Compared to polyclonal antibodies, MAbs exhibit three major

advantages. First, they can be produced against relatively

impure antigens. Secondly, MAbs can be produced in much

larger (theoretically unlimited) quantities. Thirdly, they

are monospecific (i.e., they bind to a single epitope) and

thus MAbs recognized markers can be identified and

characterized individually.

Brief synopsis on the MAbs recognizing chicken cells of
erythroid and monocytic lineages

Cell surface markers of normal or transformed

hematopoietic cells in the human and murine systems have

been extensively studied by using monoclonal antibody

techniques. There are only a handful of MAbs which have

been developed against chicken hematopoietic cells, and none

of them are specific for embryonic precursor cells.

Hayman et al. (1982) developed a panel of MAbs against

the temperature-sensitive mutant of ts34 AEV-transformed

erythroblasts which had been grown at 41.5C for five days.

Three MAbs were chosen for further characterization. MAb

4.2A5 recognized erythrocytes and AEV-transformed

erythroblasts as well as granulocytes; MAb 4.5A5 reacted

with erythroblasts and retrovirus-transformed producer

cells, however, the authors did not address the specificity

tests on other types of normal cells; MAb 4.6C1 was specific

for erythroid cells at all stages.

Jurdic et al. (1982) produced a MAb, Sl-37, from the

fusion of spleen cells of a mouse immunized with AMV-

transformed cells. Sl-37 was shown to be specific for the

cells of the monocytic lineage. Its binding specificities

revealed by radioimmunoassay and flow cytometric analysis

were too low for further identification and


Miller et al. (1982) immunized mice with 1-day-old

erythrocytes from inbred line 003 and hybrid strain

Shaver-Starcross 288 and raised a MAb, MaEE1, against a 48Kd

antigen which was expressed on the erythrocytes of 1-day-

old peripheral blood and adult bone marrow. It was also

present in the retina, muscle tissues, liver and on

epithelia and lymphoid cells of young and adult chickens.

Sanders et al. (1982) were able to generate a MAb

(190-4) against 1-day-old erythrocytes from the SC strain

which recognized a 50Kd molecule expressed on the cell

surface of erythrocytes, reticulocytes, chicken embryo cells

and reticuloendotheliosis virus (REV)-transformed lymphoid

cells, but not on the AEV-transformed erythroleukemia cells.

Kornfeld et al. (1983) obtained five different groups

of MAbs by immunizing mice with normal macrophages and

myeloid cells transformed by MC29, AMV and E26. Only one

group was specific for myeloid lineage, predominantly

reacting with immature myeloid cells. However, the authers

failed to show the reactivities to normal cells except the


Trembicki and Dietert (1985) produced 4 MAbs against 1-

day cornell K-strain white leghorn chickens. MAb 10C6

detected a chicken fetal antigen (CFA) on 1-day-old chick

erythrocytes. MAbs 3F12 and 4C2 recognized chicken adult

antigen (CAA) on adult erythrocytes, whereas 9F9 reacted

with all peripheral erythrocytes from both Japanese quail

and chicken regardless of age.

Schmidt et al. (1986) immunized mice with plasma

membranes from the ts34 AEV cell line HD3 induced to

differentiate at 420C for five days. Only one out of eight

groups of MAbs was specific for erythroid lineage, reacting

only with reticulocytes.

The following chapters in this dissertation describe

the first systematic attempt using MAbs in combination with

other techniques to identify embryonic differentiation

markers on normal avian hematopoietic cells and

transformation-associated antigens on retrovirus-transformed





The lack of MAbs recognizing embryonic differentiation

markers on avian hematopoietic precursor cells may be due to

the use of inappropriate antigens for immunization. Because

in most cases 1-day-old erythrocytes were chosen as antigens

instead of embryonic cells, the MAbs developed were not

specific for the embryonic precursor cells. Therefore it

was decided to use different strategies for the immunizing

protocols in order to produce MAbs which identify embryonic

differentiation markers present on normal avian

hematopoietic cells of the erythroid and monocytic lineages.

Moreover the approaches used were designed to identify

transformation-associated antigens in retrovirus transformed

avian hematopoietic cells and to characterize the

biochemical properties as well as to study the biological

functions of these markers.

Several types of cells were used to immunize 6-week-

old BALB/c BYJ female mice: 1) normal 3-day-embryo yolk sac

cells, 2) normal 3-day-embryo megalocytestogether with AEV-


transformed nonproducer cells from 3-day-embryo yolk sac, 3)

BM2 cells (AMV-transformed nonproducer cells from embryonic

bone marrow), and 4) BM2/L cells (leukemogenic variant of

BM2 cell line). Theoretically, 3-day-embryo yolk sac cells

are rich in embryonic hematopoietic precursor

cells, and the AMV/AEV nonproducer cells possess not only

the transformation-associated antigens but also

oncodevelopmental markers which are present normally only on

the precursor cells. 10-14 days after each cell fusion,

supernatants of hybridomas were screened against a panel of

different types of cells as well as against yolk by indirect

radioimmunoassay (RIA) and enzyme-linked immunosorbent assay

(ELISA). Only hybridomas showing continuous production of

MAbs of potential interest were subcloned by the limiting

dilution method. The isotypes of the MAbs were determined

by the Ouchterlony (double diffusion) test using isotype

specific antisera. MAbs were purified from culture media of

the cloned hybridomas or from mouse ascites by using high-

salt protein A-sepharose chromatography.

Material and Methods

Normal Cells

17-somite blastoderm. The blastoderm cells were

obtained from the 17-somite stage at 2 days of incubation by

mechanical dissociation with gentle pipetting and dispersed

in a-MEM (GIBCO) containing 10% fetal bovine serum (FBS).

Cells were then filtered through a 1.5i nylon mesh (Tetko

Inc., New York) to remove any clumps (Moscovici et al.,


Yolk sac cells. The yolk sacs from 3rd or 4th day of

embryogenesis were dissected free of other embryonic

membranes, pooled and rinsed extensively with Tyrode's

solution (8.0 g/L NaCl, 0.2 g/L KC1, 0.05 g/L NaH2PO4*H2O,

1.0 g/L Glucose and 1.0 g/L NaHC03) to remove as much yolk

as possible. The tissue were minced with scalpels and

dispensed in a-MEM/10% FBS by gentle pipetting. The

resulting cell suspension was then washed by centrifugation

to remove residual yolk, after which the cells were

resuspended in media and passed through nylon mesh to obtain

a single cell suspension. The yolk sacs from 6th or 12th

day of embryogenesis were first minced thoroughly with

scalpels and then digested with 0.125% trypsin for 10-15

minutes at 370C (Moscovici et al., 1975). The cell

suspension was then washed by centrifugation and passed

through nylon mesh as above.

Bone marrow cells. Bone marrow cells were flushed out

of the tibias with BT-88 medium (GIBCO) containing 10%

tryptose phosphate broth, 5% calf serum and 5% chicken serum

by passing the cells through a syringe with a 22-gauge

needle 3 times (Jurdic et al., 1982). The cells were washed

once, resuspended in medium and then filtered through nylon



Buffy coat. The buffy coat (WBC) from peripheral blood

was harvested by Ficoll-Hypaque (Lymphocyte Separation

Medium; Litton Bionetics) gradient centrifugation at 2,000

rpm for 20 minutes.

Macrophages. The buffy coat obtained from heparinized

blood was seeded in BT-88 complete medium and 48 hours later

the attached cells differentiated into macrophages. These

cells were then detached from the petri dishes by adding the

C-PEG solution (8.0 g/L NaCI, 0.29 g/L KC1, 0.2 g/L KH2PO4,

0.763 g/L Na2HPO4, 0.2 g/L EDTA, 3.7 g/L NaHCO3 and 1.0 g/L

Glucose, pH 8.0) for 5 minutes.

Chicken embryo fibroblasts (CEF). The fibroblasts from

10-day embryos were prepared according to the procedure

described by Vogt (1969).

Transformed Cells

BM2/C3A cells. An AMV-transformed monoblastic

nonproducer cell line, GM727, was generated by in vitro

infection of 17-day-embryo bone marrow cells with AMV-B at a

low multiplicity of infection (m.o.i. of 10-2 to 103)

(Moscovici & Moscovici, 1980). GM727 cells were then

injected into 13-day embryos via the chorioallantoic vein.

Four weeks after the injection, no overt case of leukemia

was observed unless the chickens were challenged with helper

viruses such as MAV-2 or RAV-7 (Moscovici et al., 1982).

However, the injected transformed cells could be retrived

from the bone marrow, and cloned by an in vitro colony

assay. A cell line namely BM2/C3A was established. The

cells in this line all express the v-myb proteins but are

nonproducers and are nonleukemogenic.

BM2/L cells. The BM2/L cell line is a variant of

BM2/C3A cell line. It was obtained from a BM2/C3A-injected

bird which came down with leukemia involving liver, spleen

and heart (Moscovici and Moscovici, unpublished results).

The leukemic cells were reisolated and a new line was

established, namely BM2/L, which when reinjected into

chicken embryos induced a 90-100% incidence of leukemia.

6C2 cells. 6C2 is an AEV(RAV-2)-transformed

erythroleukemia producer cell line obtained from infection

of adult bone marrow cells in vitro (Beug et al., 1982).

MSB 1 cells. MSB 1, obtained from U.S.D.A. (East

Lansing, MI), is a lymphoblastoid producer cell line derived

from a splenic lymphoma of chicken with Marek's Disease

(Akiyama and Kato, 1974).


AMV-B. The AMV subgroup B (AMV-B) was derived from

standard laboratory stocks as described (Moscovici et al.,


AEV-A. The AEV subgroup A (AEV-A) is the ES-4 strain

of AEV (RAV-1) originally obtained from Dr. J.M. Bishop (San

Francisco, CA).

RAV-1 and RAV-2. The RAVs were prepared from our

laboratory stocks.

Discontinuous Percoll Gradient

2 ml of cell suspension was layered on top of a

discontinuous percoll gradient (20%/50%/70% for bone marrow

cells and 30%/50%/70% for yolk sac cells) and centrifuged 15

minutes at 2,500 rpm. It has been demonstrated under these

conditions that the 20%/50% interface from the bone marrow

cells and the 30%/50% interface from the yolk sac cells are

rich in the precursor cells and mononucleated cells, whereas

the 50%/70% interface consists mostly of erythroblasts and

other types of differentiated cells, and the cells from the

pellet are terminally differentiated erythrocytes.

Hybridoma Production

Hybridomas were produced according to the protocol

established in the Hybridoma Laboratory, Interdisplinary

Center for Biotechnology Research, University of Florida.

Spleen cells harvested from immunized mice were fused with

SP2/O myeloma cells at a ratio of 7.5:1 (spleen cells :

myeloma cells) using polyethylene glycol 1540. The

supernatants of growth-positive hybridomas from 96-well flat

bottom tissue culture plates were screened 12-14 days later

by the indirect RIA and ELISA methods. Hybridomas which


produced monoclonal antibodies with potential interest were

subcloned by the limiting dilution method.

Radioimmunoassay (RIA)

The cells used as targets in immunoassays were washed

with PBS buffer containing 1% AHS (gamma globulin-free horse

serum) and 0.02% sodium azide and resuspended to a final

concentration of 20 x 106 cells/ml in PBS/azide/5% AHS. 50

Al of cell suspension and 100 gl of hybridoma supernatant

was added to each well of 96-well flexible polyvinyl round-

bottom microtiter plates (Dynatech) for 45 minutes at 40C.

At the end of this incubation, plates were washed with

PBS/azide/l% AHS and centrifuged at 1,100 rpm 3 times.

50 Al of 1251-rabbit anti-mouse IgG (RaMIG), containing 1 x

105 cpm was then added and incubated with the cells for 45

minutes at 4C. Plates were washed with PBS/azide/AHS and

centrifuged 3 times again. Individual wells were cut free

from each plate with a hot-wire cutter, transferred into

plastic tubes and counted in a gamma counter (LKB-Wallace

Ria Gamma 1274, Pharmacia).

Binding index = mean cpm bound with specific MAbs
mean cpm bound with negative control MAb

Enzyme-Linked Immunosorbent Assay (ELISA)

96-well flat bottom immunoplates (Nunc, Denmark) were

coated with 50 p1 of yolk at a 1:40 dilution overnight and

then blocked with 1X PBS containing 0.02% sodium azide and


1% BSA for 1 hour at room temperature. 100 Ai of hybridoma

supernatants were then incubated in wells for 45 minutes at

room temperature followed by 100 4l alkaline phosphatase-

conjugated rabbit anti-mouse IgG (Sigma) at a 1:1000

dilution in PBS/azide/BSA for 45 minutes at room

temperature. The plates were finally incubated with 200 Al

p-nitrophenyl phosphate (P-NPP) (Img/ml) (Sigma) in pH 9.0

bicarbonate substrate buffer for 30 minutes to 2 hours in

the dark and read on an ELISA reader (Molecular Devices; V

Max). The plates were washed three times with PBS/azide/l%

Tween-20 in between the steps.

Immunofluorescence staining

Live cells. The cells were incubated with 1 ml MAb

supernatant for 30 minutes at 4C and washed before fixation

with 30 Al 37% formaldehyde in 1 ml PBS for 20 minutes at

40C. The cells were washed again after fixation. 200 pl

fluorescein-isothiocyanate (FITC) conjugated goat anti-

mouse polyvalent IgG, A and M (FITC-GAM) at 1:50 dilution in

PBS containing 1% normal goat serum was then added for 30

minutes at room temperature. After washing in PBS twice,

the cell pellet was resuspended with 200 Al PPD-Glycerol/

PBS (1 ml 10X PBS, 3 ml deionized water, 6 ml glycerol and 1

to 2 flakes of p-phenylene diamine) and dropped onto slides

to observe under the fluorescence microscope coverslipped.


Frozen sections. Chicken embryos from 3 and 6 days of

incubation were dissected. Tissues of no more than a few mm

thick were fixed in 4% paraformaldehyde fix (in 0.1 M sodium

cacodylate buffer, pH 7.2-7.4) for 5 hours and shifted into

30% sucrose in IX PBS overnight. Afterwards, the tissues

were embedded in OCT compound (Lab-Tek; Miles) and frozen

with liquid nitrogen. 10 Am Cryostat sections were then

mounted onto slides which were precoated with Histostick

(Accuraqe Biochemicals) and stored at -200C overnight prior

to use. The slides were incubated with primary antibodies

containing 0.3% triton-X in PBS for 30 minutes at room

temperature. They were then washed in a PBS bath for 5

minutes and incubated with FITC-GAM/0.3% triton-X/l% normal

goat serum for another 30 minutes at room temperature. The

slides were again washed in PBS for 5 minutes followed by

addition of a few drops of PPD-Glycerol/PBS and then

examined under fluorescence microscope coverslipped.

Flow Cytometry

Cells to be examined were washed with IX PBS/azide/AHS

and incubated with 1 ml MAb supernatant for 1 hour on a

rocker at 40C followed by incubation with 1:10 dilution of

FITC-conjugated sheep anti-mouse IgG [F(ab')2 fragment)

(Sigma) for 30 minutes on a rocker at 40C. Cells were

washed twice and resuspended in Hank's balanced salt

solution (HBSS) containing 1% FBS at a concentration of 2 to

3 x 106 cells/ml. 1 x 104 cells were then analyzed on a

FACStar-plus fluorescence-activated cell sorter (Becton-

Dickinson, Mountain View, CA) by the parameters of forward

light scatter and fluorescein fluorescence. Cellular

excitation was obtained with an emission wavelength of 488

nm at an output power of 0.25W for fluorescein fluorescence.

The FITC fluorescence emitted was filtered with a 530 nm

long pass interface filter and a 530 band pass filter. The

data were collected and analyzed by a Becton/Dickinson

Consort 30 Computer program (Braylan et al., 1982).

Immunoenzymatic Staining by APAAP (Alkaline Phosphatase and
Monoclonal Anti-Alkaline Phosphatase) Complex

2 x 105 cells in 0.2 ml were cytofuged onto slides by

Cytospin (Shandon Southern) at 300 rpm for 7 minutes. The

slides were air-dried at room temperature for 2 hours

followed by fixation with equal parts of acetone and

methanol for 5 minutes at 4C. 1 ml MAb supernatants were

added to the slides and incubated in a moist chamber for 30

minutes at room temperature. Anti-mouse immunoglobulins

(DAKOPATTS; at 1:25 dilution) were then incubated for 30

minutes followed by APAAP complex (DAKOPATTS; at 1:50

dilution) for another 30 minutes. The slides were washed in

a tris-buffered saline (TBS) bath for 1 minute in between

the steps. Finally, alkaline phosphatase substrate was

added onto the slides for 15-20 minutes and then washed off

first with TBS and then with tap water (Cordell et al.,

1984). Slides were counter-stained with Giemsa stain at

1:20 dilution for 5 minutes.

Benzidine Staining

10 pA H202 (30%) was added to one ml of the benzidine

solution (0.5 g benzidine in 100 ml 70% ethanol) immediately

prior to use. Just a few drops of the staining mixture were

deposited on the cell sample and left at room temperature in

the dark for 5-10 minutes.

Induction of Cell Differentiation

BM2 cells. 1 x 107 BM2 cells were treated with 10

gg/ml lipopolysaccharide (LPS) and 0.25 Ag/ml phorbol

12-myristate-13-acetate (PMA) or 2.5 Ag/ml PMA alone in a

100-mm petri dish. 3 days later, most of BM2 cells had

attached to the petri dish and had differentiated into


6C2 cells. 6C2 cells were treated with 1.0 mM butyric

acid for 3 days. Although 6C2 cells did not

differentiate into mature erythrocytes with butyric acid,

their proliferating potential had been arrested.

High-Salt Protein A-Sepharose Chromatography

Ascites (1:10 dilution) or hybridoma supernatants were

adjusted to contain 1.5M glycine, 3M NaCI (pH 8.9), filtered

through a 0.22 p millipore filter and run through a protein


A-sepharose column (Sigma) twice to allow binding of IgG to

the column. 5-10 column volumes of binding buffer (1.5M

glycine, 3M NaCI, pH 8.9) was then percolated through the

column to get rid of unbound proteins. Elution buffer (100

mM citric acid, pH 6.0) was then added to the column to

elute the IgG, followed by regeneration buffer (100 mM

citric acid, pH 3.0) to wash the column. Samples were

collected by a fraction collector, neutralized to pH 7.0-

7.2 with 1M tris buffer (pH 9.0) and read with a U.V.

spectrophotometer at wavelength 280 nm. The concentration

of IgG was calculated as: mg/ml = O.D. 280 nm / 1.4.

Collected fractions were then dialyzed against ix PBS/azide

buffer overnight.


Production, Subcloninq and Isotyping of the MAbs

The rationale for MAb production is simple and

straightforward, however, the goals turned out to be much

tougher to achieve than we originally expected, especially

from the fusions with spleens from mice immunized against 3-

day-embryo yolk sac cells and megalocytes. Of nearly 2,500

hybridomas derived from 5 fusions, the predominant antibody

specificities detected were for yolk components due to the

fact that the unavoidably large amount of yolk was

associated with the yolk sac cells. Only three MAbs

exhibited potential interest, namely, 2E10, 3F6 and 1F7.

Fusions of spleens from mice immunized against BM2

cells and BM2/L cells yielded a panel of MAbs with various

specificities against different types of hematopoietic cells

rather than just MAbs specific for BM2 or BM2/L cells (Table

2-1). Three MAbs, 1H10, 2H1 and 3D7, from group VI, which

displayed specificity for normal monocytic cells and AMV-

transformed cells, were chosen for further studies.

These selected hybridomas were then subcloned by

limiting dilution methods and isotyped by Ouchterlony double

diffusion tests (Table 2-2). MAbs 1H10-1F9, 2H1-2A10, 3D7-

1C9 and 2E10-1E10 as well as 3F6-1E7 are all IgG1 (n); only

1F7-1A3 is an IgM (n).

Cell-Type Specificities of MAbs

As demonstrated by RIA, flow cytometry and

immunofluorescence staining, MAbs 1H10-1F9, 2H1-2A10 and

3D7-1C9 exhibited specificity for monocytic cells with a

preferential reaction against BM2 lines. The RIA results

also revealed a slight reaction of these MAbs with cells

from the 20%/50% interface of a discontinuous percoll

gradient from 2-week-old bone marrow cells (Table 2-3).

Further analysis by flow cytometry (Figure 2-1) and APAAP

immunoenzymatic staining (Figure 2-2) confirm that about

10%-20% of the cells are expressing different degrees of the

differentiation markers recognized by this group of MAbs.

Conversely, no expression of these markers was observed in

Table 2-1. The cell-type specificities of MAbs from BM2 and
BM2/L fusions

Monoclonal Antibody Group
Cell Type II II III IV V VIa

Normal Cells

Granulocytic ++b + + -

Monocytic ++ ++ + + ++ +

Erythroid ++ + + -

Lymphoid ++ + -

Transformed Cells

AMV ++ ++ ++ ++ ++ ++

AEV ++ ++ + + -

aOnly three MAbs from group VI were selected for further
characterization because of their specificities for
normal monocytic cells and AMV-transformed cells.
++: RIA binding indexes >10.0; +: RIA binding indexes >5.0;
-: RIA binding indexes <2.0.

Table 2-2. The








isotypes of selected MAbs


IgG1 (K)

IgG1 (K)

IgGI (K)

IgG1 (K)

IgGI (1)

IgM (K)

Table 2-3. RIA binding indexes of MAbs to different cell types
Monoclonal Antibody
Cellsa 1H10-1F9 2H1-2A10 3D7-1C9 2E10-1E10 3F6-1E7 1F7-1A3

6C2 1.19 b 1.21 1.66 26.35 3.03 3.96

3DYS-AEV N.T. N.T. 1.00 6.48 5.38 3.04

6DYS-AEV N.T. N.T. N.T. 25.94 12.52 5.27

MSB1 1.26 0.59 0.34 14.06 7.78 5.17

BM2/C3A 43.55 38.80 38.01 9.91 9.61 1.25

BM2/REC1 34.09 50.83 57.77 8.03 N.T. N.T.

BM2L/A1 29.98 35.05 41.72 8.26 9.72 1.29

M0 4.89 6.85 5.87 N.T. N.T. N.T.

RBC(PB) 0.77 1.52 1.14 2.58 1.09 1.31

WBC(PB) 1.15 2.25 1.63 2.24 1.25 1.11

CEF 0.83 0.97 0.88 0.87 0.75 N.T.

2DYS 0.52 0.94 1.13 2.34 2.01 N.D.

3DYS 1.52 1.40 1.16 2.18 2.11 5.72

6DYS 0.68 1.60 0.84 3.27 1.12 N.T.

12DYS 0.69 1.28 0.62 7.59 0.52 N.T.

BM 20/50 2.88 3.48 2.89 4.76 1.98 N.T.

BM 50/70 1.62 2.39 1.91 7.49 1.63 N.T.

a6C2, AEV-transformed producer cell line from adult bone marrow;
3DYS-AEV, AEV-transformed 3-day-embryo yolk sac cells; MSB1,
Marek's Disease Virus-transformed lymphoblastoid producer cell
line; BM2, AMV-transformed non-producer cell line from embryonic
bone marrow; BM2/C3A, a subclone of BM2; BM2/REC1, a subclone
recovered from the bone marrow of BM2/C3A injected chick;
BM2L/A1, a leukemogenic variant of BM2/C3A; MO, macrophage; PB,
peripheral blood; CEF, chicken embryo fibroblast; BM 20/50,
20%/50% interface of a percoll gradient of 2-week-old bone
marrow cells; N.T., not tested.
bRIA binding indexes: see Materials and Methods for details.


0- I *11 ....I' 1 n '""1 O '"" '
Z 0 I T 1 1 1 I 1 0- I1 4 i0 o I I I I Fi l l"i 1 11 1; 1 11 1 "t 'l l I #i 4
10 10 1 10 10 10 12 10 1

300 300
1H10-1F9 2H1-2A10

0 -W kA.. r i... I r 1.. 1.11 1 1 11111
10 101 I02 103 14 10 11 10 102 10Q

Log Fluorescence Intensity

Figure 2-1. Flow cytometric analysis of the 20%/50%
interface of a percoll gradient of 2-week-old bone marrow
cells. The cells were labeled with negative control MAb,
1H10-1F9, 2H1-2A10 and 3D7-1C9 respectively. The background
fluorescence resulting either from nonspecific binding or
autofluorescence was present in 10% of the cells. 10-20% of
the cells were positive for MAbs 1H10-1F9, 2H1-2A10 and 3D7-


Figure 2-2. APAAP immunostaining of 20%/50% interface of a
percoll gradient of 2-week-old bone marrow cells (600x).
The cytofuges were treated with (A) negative control MAb,
(B) IH10-1F9, (C) 2H1-2A10 and (D) 3D7-1C9 respectively.
Positive cells were labeled with purple-reddish products
around the cell surface and/or inside the cytoplasm.

the cells from the 50%/70% interface (Table 2-3 and Figure


1F7-1A3 recognizes the AEV-transformed yolk sac cells,

MSB1 cells and 20% of the cells in the 30%/50% interface of

a discontinuous percoll gradient of yolk sac cells (Table 2-

3 and Figure 2-4). Interestingly enough, its reaction to 6C2

cells is amplified after the treatment with neuraminidase

(discussed in Chapter 4 in more detail).

3F6-1E7 detects a differentiation marker present mainly

on AEV-transformed yolk sac cells, BM2 cell lines and MSB1

cells as well as 10% of cells from the 30%/50% interface of

a percoll gradient of normal yolk sac cells (Table 2-3 and

Figure 2-4).

2E10-1E10 possesses a variety of cell-type

specificities such as 6C2 cells, AEV-transformed yolk sac

cells, MSB1 cells, BM2 cells, 6 & 12-day-embryo yolk sac

cells and 2-week-old bone marrow cells (Table 2-3, Figure

2-5 & Figure 2-6). However, there is no reaction to the

terminally differentiated hematopoietic cells.

The cell-type specificities of MAbs are summarized in

Table 2-4.

Induction of Cell Differentiation

Treatment of BM2/C3A cells with 0.25 Ag/ml PMA and 10

Ag/ml LPS or 2.5 pg/ml PMA alone for 3 days caused the

BM2/C3A cells to attach to the petri dishes and become

Negative control 3D7-1C9


1lt0 1Q 1 2 13 Ia4 10 1C 10 02a3 0104
300 _30_0__
S1H10-1F9 2H1-2A10


10 10 1 2 11 3 1 1 1 10 12 I0 1 4
Log Fluorescence Intensity

Figure 2-3. Flow cytometric analysis of the 50%/70%
interface of a percoll gradient of 2-week-old bone marrow
cells. The cells were labeled with negative control MAb,
1H10-1F9, 2H1-2A10 and 3D7-1C9 respectively. The results
showed no reaction of MAbs with these cells at all.

Negative control

t V3F6-1E7

E 1F7-1A3

10 1 1 102 103
Log Fluorescence Intensity

Figure 2-4. Flow cytometric analysis of the 30%/50%
interface of a percoll gradient of 4-day-embryo yolk sac
cells. Cells were tagged with negative control MAb, 1F7-1A3
and 3F6-1E7 respectively. The analysis showed that 3F6-1E7
label 10% of the cells, whereas 20% of the cells are
positive for 1F7-1A3.

Negative control



10 101 102 103
Log Fluorescence Intensity

Figure 2-5. Flow cytometric analysis of the 20%/50%
interface of a percoll gradient of 2-week-old bone marrow
cells tagged with MAb 2E10-1E10. Almost 80% of the cells
were positive for MAb 2E10-1E10.

Negative control



I ~ I lll iii II -I1 i I l in
100 101 102 103

Log Fluorescence Intensity

Figure 2-6. Flow cytometric analysis of the 50%/70%
interface of a percoll gradient of 2-week-old bone marrow
cells tagged with MAb 2E10-1E10. 2E10-1E10 recognized
nearly 70% of the cells.

Summary of the cell-type specificities of MAbs

Groups MAbs Specificities






BM2 cell lines
Normal monocytic cells

AEV-transformed yolk sac cells
BM2 cell lines
MSB1 (Marek's Disease)
10% of cells from the 30%/50%
interface of a percoll fraction
of 3-day-embryo yolk sac cells

AEV-transformed yolk sac cells
MSB1 (Marek's Disease)
20% of cells from the 30%/50%
interface of a percoll fraction
of 3-day-embryo yolk sac cells

6C2 cells
AEV-transformed yolk sac cells
MSB1 (Marek's Disease)
BM2 cell lines
6-day-embryo yolk sac cells
12-day-embryo yolk sac cells
2-week-old bone marrow cells

Table 2-4.

differentiated into macrophage-like cells. In addition,

their proliferating potential (Table 2-5) and ability to

bind MAbs 1H10-1F9, 2H1-2A10 and 3D7-1C9 were dramatically

decreased (Figure 2-7). Therefore, the differentiation

markers recognized by these MAbs may be candidates for

transformation-associated antigens in this system.

Although 6C2 cells were not induced to terminally

differentiate into erythrocytes with the treatment of ImM

butyric acid for 3 days, their proliferating potential had

been impaired (Table 2-6) and the expression of the 2E10-

1ElO-recognized marker was also reduced (Figure 2-8).

Reactivity to the Envelope Proteins of Retroviruses

The RAV-1 or RAV-2 infected CEF cells were used as

control to exclude the possibility that MAbs might react

with the envelope proteins of retroviruses which were

expressed on the cell surface. 3F6-1E7 and 2E10-1E10 were

shown to have no reaction to either CEF cells or RAV-

infected CEF cells (Table 2-7).


Four groups of MAbs were selected among nearly 5,000

hybridoma supernatants from ten fusions. Characterization

of their cell-type specificities was achieved by RIA,

immunofluorescence staining, flow cytometry and

immunoenzymatic staining. 1) MAbs 1H10-1F9, 2H10-2A10 and


Table 2-5. Comparison of the proliferating potential of
BM2/C3A cells and 2.5Ag/ml PMA-diffentiated BM2/C3A cells

BM2/C3Ab BM2/C3A + PMAb

Number of cells
after 3 days of 58 x 106 6 x 106
al0 x 106 cells were seeded per 100-mm petri dish.
bThe number of cells was expressed as the average from 6


- tO
-4- 0 )

*0 a

4) $ 4
m) 4

H *>

0 4)
O ) ,-4
r- I 1tH
a) a) >
u bo t

a 0)

.0 04. 4-1 +

c0 0 00

M-H a)

0H 0 4

rq C
10 41 "

14 0




o "

,it CO CM T-
sexepul 6ujpuig VIEI


Table 2-6. Comparison of the proliferating potential of 6C2
cells and 1mM butyric acid-treated 6C2 cells

6C2b 6C2 + Butyric acidb

Number of cells
after 3 days of 40 x 106 11.5 x 106
al0 x 106 cells were seeded in 60-mm petri dish.
bThe number of cells was expressed as the average from 6

r-i mn
S(0 (0
NM tP4-P10
U: 3...
S0 4) X
.4.4-) 0)
O E-4 0 0

W O1
a-) ric!t
9 >i*4 C
0 MC
4) .-

4 >0 0 *'
go 1 -H

$4 9
4 ) 4) -ri
0 0) 1-4 A
ov O'-4.
.0 cr-4
(8 O1c-
0 4J 0.4
*.-I p-4.)

$14 04

*E M W
4-i 0
> o I
v.ri m

-40 4W

H NiO 0
04) -I I'.
go u-

tv 4) Xa
c o in
0 4- 4)
H k 0t

rvi V 4)
r-4 >0
0) 4J 4 $4
A 4J X

".4) 4)
u> # *u (

o 14
0 00
4) 4 E-

*-I C
4J -4-

'0 4J.
4) -4 4-)

*N- S
c o U *MC
0 to rq go

40) 4) 0)

OH 0 4.

>00 -H
V 4J~
la 0
to (D -I 9 4)

(' $4AS>
H 04-

0 MH0 '04)
UM r-j N1 C 4-
)U IV 4O

H 0%0 E 0a







0 LO 0 C) 0 L) 0
CO C C v-- -

sexepul 6u!pul vIUI


Table 2-7. RIA binding indexes of MAbs to normal CEF and
RAV-infected CEF

Monoclonal Antibody
Cells 2E10-1E10 3F6-1E7

CEF 0.87 0.75

CEF (RAV-1) 1.08 0.75

CEF (RAV-2) 1.10 0.80


3D7-1C9 are specific for transformation-associated antigens

present preferentially on BM2 cell lines rather than on

normal monocytic cells. 2) MAb 1F7-1A3 recognizes AEV-

transformed yolk sac cells, MSB1 cells and 20% of cells from

the 30%/50% interface of a discontinous percoll gradient of

4-day-embryo yolk sac cells. 3) MAb 3F6-1E7 reacts with

AEV-transformed yolk sac cells, BM2 cells and MSB1 cells as

well as 10% of cells from the interface of a discontinous

percoll gradient of 4-day-embryo yolk sac cells. 4) MAb

2E10-1E10 defines a marker present on different tumor cells,

yolk sac cells and bone marrow cells instead of terminally

differentiated cells.

The expression of transformation-associated antigens

recognized by MAbs 1H10-1F9, 2H1-2A10 and 3D7-1C9 is

possibly up-regulated by v-myb proteins, but is diminished

after the BM2 cells are differentiated by PMA. Future

investigators may want to determine 1) whether the

expression of these antigens is regulated at the

transcriptional and/or the translational level or is due to

posttranslational modification, such as aberrant

glycosylation and/or sialylation, and 2) whether the

expression of these markers is essential for the

transforming process or if they are merely the by-products

of transformation.

Although 2E10-1E10 displays a wide range of cell-type

specificities, it labels primarily 6C2 cells and


AEV-transformed 6-day-embryo yolk sac cells and it does not

recognize terminally differentiated cells such as

erythrocytes, macrophages, lymphocytes and PMA-

differentiated BM2/C3A cells. These results lead us to

speculate that 2E10-1E10 may react only with proliferating

hematopoietic cells. This possibility was supported by the

observation that the binding specificities of 2E10-1E10 to

6C2 cells was diminished after treatment with butyric acid

which inhibited the proliferating potential of 6C2 cells

without the induction of differentiation. Moreover, this

marker starts appearing at the 4th day of embryogenesis

(discussed in Chapter 3 in more detail) and its expression

is enhanced after the cells are transformed by AEV.

LPS only induces the BM2 cells to attach to the petri

dishes but does not induce differentiation. As a matter of

fact, PMA alone can induce both the attachment and

differentiation of the BM2 cells even at a low concentration

of 0.25 Ag/ml. The reason why we added 10 Ag/ml LPS to the

PMA treatment was simply because we didn't want to see "any

fish sneak out of the net", as the Chinese proverb says.

Butyric acid has been shown to induce the

differentiation of the AEV strain R (RAV-2)-transformed

erythroleukemia cells from SC strain chickens (Nelson et

al., 1982), but was incapable of differentiating 6C2 cells

even though their proliferating potential had been

apparently impaired. This is possibly due to the

difference in the susceptibilities to butyric acid of

various avian erythroleukemia cell lines and/or sublines.

Immunoperoxidase staining procedures were not used

because significant amounts of endogenous peroxidase present

in granulocytes, erythroid cells and macrophages would give

rise to unwanted background staining, and it is rather

difficult to abolish this activity by exposing samples to

peroxidase inhibitors (such as H202 and methanol) without

causing antigenic denaturation (Cordell et al., 1984). In

contrast endogenous alkaline phosphatase activity survives

poorly in cytofuge slide preparation and any residual

activity may be selectively inhibited by including

levamisole in the enzyme substrate solution (Ponder and

Wilkinson, 1981). Therefore the APAAP immunostaining

technique was chosen for our studies. Wet slides without

any mounting fluid were then photographed because it was

observed that the nonaqueous mounting solution dissolved the

reaction product, whereas an aqueous mounting medium will

disrupt the Giemsa counter-stain.

The antigens recognized by MAbs 1F7-1A3, 2E10-1E10 and

3F6-1E7 could only be detected by RIA and immunofluorescence

staining and not by APAAP immunostaining technique. It is

very likely that these markers may be so fragile that they

were destroyed during the fixation (acetone and ethanol)

and/or any subsequent steps.




The avian hematopoietic precursor cells represent only

a small population in the blood-forming organs, however, it

is this small population that builds up the entire

hematopoietic repertoire of various types of specialized

blood cells with different functions. The nature of self-

renewal and commitment remains an enigma, but if the

precursor cells can be isolated from the heterogenous

population of hematopoietic cells for direct studies, it

would provide us with invaluable pieces of information to

solve the enigma. However, these cells are morphologically

unrecognizable, and until now no specific markers had been

identified to facilitate the purification of these cells.

One of the specific aims of this project was to develop MAbs

against differentiation markers specifically present on the

surface membranes of the avian hematopoietic precursor

cells. These cells have been purified and enriched by using

FACS or immunomagnetic beads as follows. Bone marrow cells

or yolk sac cells are treated with specific MAbs followed by

FITC-conjugated or magnetic bead coated with secondary



antibodies. Fluorescence-positive and -negative populations

are then separated by the FACS, whereas the magnetic bead-

bound cells are separated from the negative population by a

magnetic field (negative selection). The separated

fractions are then identified by indirect methods, such as

AMV/AEV transforming assays and BFU-E/CFU-E colony forming

assays. For example, if MAb-positive fractions produce

colonies derived from the CFU-E instead of the BFU-E, the

antigen must be expressed later than the BFU-E stage; if it

produces AMV-transformed colonies, but neither AEV-

transformed colonies nor BFU-E/CFU-E colonies are developed,

the MAb ought to be specific for the target cells for AMV,

i.e., cells of the monocytic lineage.

These approaches helped us to confirm the cell-type

specificities of the MAbs that we developed.

Materials And Methods

FACS Sorting

After being analyzed on a FACStar-plus sorter,

according to the procedure described in the previous

chapter, the cells were sorted into positive and negative

fractions at a rate of 1,000 cells/sec.

Immumomagnetic Beads Separation

10 x 106 Cells were tagged with 1 ml of MAb

supernatants for 1 hour at 40C on a rocker and then washed

with PBS/azide/1% AHS followed by incubation with magnetic

beads coated with sheep anti-mouse IgG, (Fc) (Dynabeads M-

450; Dynal Inc.) or magnetic particles coated with goat

anti-mouse IgM (Advanced Magnetics) at a bead to cell ratio

of 4 to 1 for 30 minutes at 40C on a rocker. Cells bound to

the beads were separated from the negative population that

remained in suspension by applying a cobalt steel magnetic

force (Dynal MPC-1; Dynal Inc.) for 1 minute (Cruikshank et

al., 1987). This process was repeated twice. It was not

possible to separate viable cells (positive selection) from

the magnetic beads because of technical limitations.

Retrovirus Transforming Assays

The transforming assays were performed according to the

protocol described by Moscovici et al. (1983). Briefly,

cells were infected with retroviruses at high m.o.i. The

virus adsorption was carried out 30 minutes at 4C and then

30 minutes at room temperature. 1 x105 AEV-infected cells

was seeded in 2 ml semi-solid a-medium containing 20% FBS,

10% ACS, 1% BPA, 0.1% 10'-M 9-mercaptoethanol, 0.1%

gentamycin and 25% methocellulose per 35-mm dish. Whereas,

1 x105 AMV-infected cells was mixed with 1 ml F12 overlay

medium containing 20% 2X F12, 6% calf serum, 2% chicken

serum, 10% tryptose phosphate, 1% 100X vitamins, 1% 100X

folic acid and 40% fibroblast-conditioned medium as well as

20% of 1.8% Bacto agar, and then overlayed on top of 2 ml

3.6% hard agar base in 35-mm petri dishes. The transformed

colonies were scored after 6-12 days of incubation.

BFU-E/CFU-E Colony Assays

0.5 ml of cell suspension were mixed with a-medium

containing 20% FBS, 10% ACS, 1% BPA, 0.1% 10'-M P-

mercaptoethanol, 0.1% gentamycin and 25% methocellulose and

seeded at 1x105 cells per 35-mm dish. The CFU-E were scored

after 3-4 days of incubation, whereas BFU-E were scored

after 6-7 days of incubation (Samarut and Bouabdelli, 1980).


FACS Sorting

In order to identify the target cells for the MAbs in

the bone marrow, FACS was utilized to separate the MAb-

positive population and MAb-negative population. The bone

marrow cells were analyzed before and after the FACS sorting

(Figure 3-1 and 3-2). Transforming assays and colony-

forming assays were performed on both negative and positive


The results obtained in the transforming assays and

colony-forming assays (Table 3-1) confirm that the 3D7-1C9

positive population (10%-20% of the cells from the 20%/50%

interface of a discontinuous percoll gradient of 2-week-old

bone marrow cells) represents the target cells for AMV,

i.e., cells of the monocytic lineage.

l10 101 102 10
Log Fluorescence Intensity

Figure 3-1. Flow cytometric analysis of 3D7-1C9-tagged bone
marrow cells before sorting. 20%/50% interface of a percoll
gradient of 2-week-old bone marrow cells were tagged with
3D7-1C9 MAb. The analysis showed that 10-20% of the cells
are positive for MAb 3D7-1C9.

0 1a o f bi 11 111 & I1
S 00 2 10

U 300

3D7-1C9 Positive Fraction

10 101 12 103 104

Log Fluorescence Intensity

Figure 3-2. Flow cytometric analysis of 3D7-1C9-tagged bone
marrow cells after sorting. 20%/50% interface of a percoll
gradient of 2-week-old bone marrow cells were sorted into
3D7-1C9 positive fraction and 3D7-1C9 negative fraction
followed by the FACS analysis. The 3D7-1C9 positive
fraction still contained about 20% negative cells due to
contamination during the sorting.

Table 3-1. Characterization of the target cells isolated
with MAb 3D7-1C9 from the bone marrow

Colonies 3D7-1C9 3D7-1C9
(1x105cells) positive negative
population population

CFU-E 34.0 5.0 1620.5 52.5

BFU-E 12.0 3.0 512.0 132.0

AEV-A 26.0 2.0 144.0 7.0

AMV-B 592.5 20.5 86.5 13.5

20%/50% interface of a percoll gradient of 2-week-old bone
marrow cells were tagged with 3D7-1C9 and then sorted by
FACStar followed by transforming assays and colony-
forming assays. The results show that 3D7-1C9 positive
population contain the target cells for AMV, i.e.,
cells of monocytic lineage, rather than BFU-E/CFU-E or the
target cells for AEV.


On the other hand, the results of colony-forming assays

and retrovirus transforming assays suggest that MAb 2E10-

1E10 not only recognize the target cells for AEV and AMV

but also the BFU-E and CFU-E (Table 3-2). This finding

prompted us to suggest in Chapter 2 that MAb 2E10-1E10 react

with proliferating hematopoietic cells.

Immunomagnetic Bead Separation

Cells from 30%/50% interface of a discontinuous percoll

gradient of 4-day-embryo yolk sacs were incubated with MAbs

3F6-1E7 or 1F7-1A3 followed by magnetic beads conjugated

with secondary antibodies. Only the negatively selected

population of MAbs were used for the assays. There was

about 60% reduction in the BFU-E and the CFU-E colonies from

1F7-1A3 negative population. However, no significant

difference in transformed colonies was found (Table 3-3).

In other words, MAb 1F7-1A3 recognizes erythroid cells at

the BFU-E and the CFU-E stages but not the pre-BFU-E stage,

since it does not recognize cells at the pre-BFU-E stage

which are the target cell for AEV in the yolk sac.

Conversely MAb 3F6-1E7 seems not to react with either the

BFU-E/CFU-E or the target cells for the retroviruses. It

probably recognizes a differentiation marker expressed on

the stem cell and precursor cell populations at an earlier

stage than pre-BFU-E one.

Table 3-2. Characterization of the target cells isolated
with MAb 2E10-1E10 from the bone marrow

Colonies 2E10-1E10 2E10-1E10
(1x105 cells) positive negative
population population

CFU-E 474.0 27.0 0.0 0.0

BFU-E 282.0 2.0 0.0 0.0

AEV-A 59.0 1.0 0.5 0.5

AMV-B 403.0 3.0 5.0 2.0

20%/50% interface of a percoll gradient of 2-week-old bone
marrow cells were tagged with 2E10-1E10 and then sorted by
FACStar followed by transforming assays and colony-
forming assays. The results reveal that MAb 2E10-1E10
recognize not only BFU-E/CFU-E, but also the target cells
for AEV and AMV.

Table 3-3. Characterization of the target cells isolated
with MAbs from the yolk sac

(1 x 105 Monoclonal Antibody
Cells) Control 3F6-1E7 2E10-1E10 1F7-1A3

CFU-E 3313.1142.5 2576.2232.4 1597.078.4 1487.5103.4

BFU-E 1211.9 11.9

AEV-A 135.6 6.4


50.0 5.0

899.2 67.3

115.0 7.0

46.0 2.0


108.1 5.2

36.8 2.4

501.1 23.8

102.0 11.0

44.0 3.0

30%/50% interface of a percoll gradient of 4-day-embryo yolk
sac cells were treated respectively with MAbs 1F7-1A3, 2E10-
1E10 or 3F6-1E7 followed by immunomagnetic bead separation.
The MAb-negative populations were collected to perform
colony-forming assays and transforming assays. See text for
more details.

The MAb 2E10-1E10 negative population in the 4-day-

embryo yolk sac cells exhibited nearly 50% reduction in the

BFU-E/CFU-E colonies and 30% in the AMV-transformed colonies

respectively (Table 3-3). Moreover, I have shown in Chapter

2 that MAb 2E10-1E10 has no binding specificities for 2- or

3-day-embryo yolk sac cell. These results indicate that the

marker identified by MAb 2E10-1E10 starts appearing after

the 4th day of embryogenesis.


The FACS sorting and immunomagnetic bead techniques

allow us to perform direct studies on MAb-positive and/or

MAb-negative populations. The results from colony-forming

assays and transforming assays showed that MAb 3D7-1C9 can

purify the target cells for AMV, i.e., cells of the

monocytic lineage, MAb 2E10-1E10 define a marker present on

proliferating hematopoietic cells and it starts appearing

only after the 4th day of embryogenesis and that MAb 1F7-1A3

recognize erythroid cells at BFU-E and CFU-E stages.

MAb 3F6-1E7 recognizes only some tumor cell lines and

10% of cells from the 30%/50% interface of a discontinuous

percoll gradient of 3- and 4-day-embryo yolk sac cells, and

it does not possess the specificities for the terminally

differentiated cells and the lineage-committed progenitor

cells. However there is no direct evidence yet to support

the idea that the 10% of the cells recognized by MAb 3F6-1E7

represent the stem cell and precursor cell populations in

the yolk sac. In order to prove this, there are two

obstacles that need to be overcome. One is to improve the

FACS sorting condition to prevent yolk sac cells from

lysing. The other is to establish a long-term culture of

normal avian yolk sac cells. Once these problems are

solved, it will become possible to obtain the long-term

culture of the FACS-sorted 3F6-1E7 positive cells and then

we can compare the results from the transforming assays and

colony-forming assays between the 3F6-1E7 positive

population with and without the long-term culture.

Theoretically, in the long-term culture, the stem cell and

precursor cell populations will proliferate and undergo the

normal differentiation program to become the committed

progenitor cells and mature cells. If the 3F6-1E7 positive

cells indeed represent these populations, the increased

number of retrovirus-transformed colonies and BFU-E/CFU-E

colonies will then be observed in the 3F6-1E7 positive cells

with long-term culture. Ultimately, 3F6-1E7 positive cells

should be capable of repopulating the bone marrow of

irradiated chicks.

It is still a puzzle why MAb 1F7-1A3 reacts with

lymphoblastoid cell line-MSBl. Maybe the marker recognized

by 1F7-1A3, which is normally present on the embryonic

BFU-E/CFU-E only, can be regarded as an onco-fetal antigen,

i.e., its expression being turned on in the AEV-transformed

yolk sac cells as well as in the MSB1 cells instead of in

other types of transformed cells. This interesting finding

also implies that the relationship between the erythroid and

lymphoid lineage may be closer than we originally thought.

The bone marrow cells were sorted into fluorescence-

positive and -negative populations using the FACStar-plus at

a rate of 1,000 cells/sec. At this rate it took

approximately 2-3 hours to collect the minimum workable

number of cells, i.e., 5x105 MAb-positive cells which

account for 10-20% of cells from the 20%/50% interface of a

discontinuous percoll gradient of 2-week-old bone marrow

cells. The results of flow cytometry, retrovirus

transforming assays and colony-forming assays showed that

there is a slight contamination of MAb-negative cells in the

positive population. To increase the purity of the

collection by reducing the analysis rate or by processing a

"two-run" procedure would be too time-consuming and the

viability or the behavior of the sorted cells might be


The yolk sac cells are extremely delicate and fragile

compared to other types of cells. Suffering from the

"abuse" of scalpel mincing, extensive washing, percoll

fractionation, as well as FACS sorting, the cell membranes

could have been damaged in such a way that only less than

10% of the sorted cells stayed intact, while the rest were

all lysed. We have tried different approaches such as

changing the sheath fluid, minimizing the laser power and

electric charge, reducing the centrifugation speed and

changing the collecting tubes etc. None so far gave us

satisfactory results.

Nevertheless, this is the pioneer study of the avian

hematopoietic system by FACS analysis. Once all the

conditions are standardized, it will become a tremendously

powerful tool to identify the stem cell and precursor cell

populations of the avian hematopoietic system, especially in

the yolk sac.

There are two disadvantages in using the immunomagnetic

beads for cell separation. First, its sensitivity is much

lower than that of the FACS sorting, i.e., it yields less

pure separation than FACS does. Second, because of the

strength of positive cells binding to magnetic beads,

trypsinization is needed to free the magnetic beads from the

cell surface. Cell surface receptors for retroviruses and

growth factors will be destroyed by the trypsinization, not

to mention the possibility that the yolk sac cells may

become more susceptible to lysis. As a result, few viable

cells remain for further study. Nevertheless, compared to

the FACS sorting, immunomagnetic bead separation is not as

time-consuming and has the advantage of not damaging the

yolk sac cells. Therefore, immunomagnetic beads separation

represents the most useful technique so far to study the

influence of MAbs in the yolk sac system.




There are two kinds of changes that contribute to the

expression of differentiation markers specific for a

particular cell type or lineage on normal hematopoietic

cells or to the expression of transformation-associated

antigens on retrovirus-transformed hematopoietic cells. One

is the appearance of new surface markers due to enhanced

synthesis at the transcriptional and/or translational level.

The other is the alteration in the structure of carbohydrate

groups attached to proteins or lipids at the

posttranslational level. These changes probably are

involved in the alteration of the interaction of particular

hematopoietic cells with other types of hematopoietic cells,

with the extracellular matrix or with stromal cells and in

the different response to growth and differentiation


Our initial attempt was to determine the nature and

molecular weights of these differentiation markers in order

to elucidate what kind of changes have occurred to these

markers recognized by MAbs. The experimental approach



toward the characterization of these differentiation markers

was via enzymatic digestion and chemical deglycosylation of

the cell surface. Trypsin was used to test if proteins were

portions of the antigens and endoglycosidase F and sodium-M-

periodate were used to determine whether they were

glycosylated. In addition, neuraminidase digestion was

performed to test for the presence of sialic acid. Western

blotting and biosynthetic labeling/immunoprecipitation

techniques were utilized to determine the molecular weights

of these markers.

Materials and Methods

Enzymatic Digestion

Trypsin. Cells were treated with 0.125% trypsin

(Sigma) in TBS buffer, pH 5.0, for 1 hour at 370C. Cells

were then washed, incubated with MAb supernatants followed

by RIA.

Neuraminidase. Cells were treated with neuraminidase

(from Clostridium perfringens; Sigma) at a concentration of

20 x 106 cells/U in TBS buffer, pH 5.0, for 1 hour at 370C.

Endoglycosidase F. Cells were treated with Endo F at a

concentration of 20 x 106 cells/U in TBS buffer, pH 5.0, for

1 hour at 370C.

Chemical Deglycosylation

2.5 mM sodium-M-periodate in PBS was added to BM2

Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EW9QFR38W_W070YL INGEST_TIME 2012-02-20T22:59:40Z PACKAGE AA00009073_00001