Evaluation of potential mechanisms for xenobiotic-mediated disruption of lymphocyte function and responsiveness

MISSING IMAGE

Material Information

Title:
Evaluation of potential mechanisms for xenobiotic-mediated disruption of lymphocyte function and responsiveness
Physical Description:
x, 180 leaves : ill. ; 29 cm.
Language:
English
Creator:
Masten, Scott Alexander, 1968-
Publication Date:

Subjects

Subjects / Keywords:
Prolactin -- physiology   ( mesh )
Tetrachlorodibenzodioxin -- pharmacology   ( mesh )
Tetrachlorodibenzodioxin -- toxicity   ( mesh )
Cocaine -- pharmacology   ( mesh )
Cocaine -- toxicity   ( mesh )
B-Lymphocytes -- drug effects   ( mesh )
Receptors, Aryl Hydrocarbon -- drug effects   ( mesh )
Antigens, CD19 -- drug effects   ( mesh )
Genre:
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1995.
Bibliography:
Includes bibliographical references (leaves 161-179).
Statement of Responsibility:
by Scott Alexander Masten.
General Note:
Typescript.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
oclc - 49846108
ocm49846108
System ID:
AA00011203:00001

Full Text










EVALUATION OF POTENTIAL MECHANISMS FOR XENOBIOTIC-MEDIATED
DISRUPTION OF LYMPHOCYTE FUNCTION AND RESPONSIVENESS













BY


SCOTT ALEXANDER MASTEN














A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY


UNIVERSITY OF FLORIDA

1995




























This dissertation is dedicated in loving memory to my grandmother and grandfather, and
also to Susie, my parents and sister















ACKNOWLEDGEMENTS


I would like to express my sincere thanks to my mentor, Dr. Kathleen Shiverick,

for her guidance and patience throughout the course of the completion of this work. What I

have gained in knowledge and capability under her direction, I will carry with me into a

hopefully long-lasting career in science.

I also wish to respectfully thank the members of my dissertation supervisory

committee, Dr. Stephen Baker, Dr. Stephen Roberts, Dr. Thomas Rowe, and Dr. Paul

Klein, for their advice and support. I also thank the past and present members of Dr.

Shiverick's laboratory, including Terry Medrano, Paul Saunders, Dr. Mary Vaccarello, Dr.

Phyllis Conliffe, Dr. Yara Smit, Grantley Charles, and Liyan Zhang for their help and for

providing a pleasant working atmosphere. I would also like to express my appreciation to

my fellow graduate students, including Bismarck Amoah-Apraku and Monica Sanghani,

and to all the faculty members in Department of Pharmacology for their help and support.

My sincere thanks also go to the administrative staff of the Department of Pharmacology,

especially Judy Adams, Barbara Reichert, and Cookie Mundorff.

In addition, I would like to acknowledge the following people for providing

materials or services essential to the completion of this research: Dr. William Millard, Dr.

Janet Karlix, Dr. Christopher Bradfield, Dr. Nancy Cooke, Dr. Thomas Tedder, Dr. Oliver

Hankinson, Dr. Paul Kelly, Dr. William Greenlee, and Dr. Mary Vaccarello.

Finally, I thank my fiance, Susie, for standing by me through long hours and

numerous occasions of unpleasantness as I pursued my Ph. D., and my parents, Kenneth

and Marsha Masten, for their continued love and support. Without them, this dissertation

would not have been possible and part of it belongs to each of them.















TABLE OF CONTENTS


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

KEY TO ABBREVIATIONS................................................vii

ABSTRACT ...................... ..........................................ix

CHAPTER 1: INTRODUCTION .......................................................

Objectives of this Research...........................................................
PRL and the Immune System........................................................2
Immune Effects of PRL In Vivo........................................... 3
Immune Effects of PRL In Vitro ............................................4
PRL and PRL Receptor Expression in Lymphocytes......................5
Immunomodulation by Cocaine .......................................................7
Biological Effects of TCDD.......................................................9
General Effects of TCDD................................. .............. 9
Effects of TCDD on Humoral Immunity and B Lymphocytes........... 12
Thesis Rationale....................................................................... 16
PRL as a Target for Cocaine and TCDD.................................. 16
Responsiveness of Human B lymphocytes to TCDD................... 17

CHAPTER 2: MATERIALS AND METHODS ...........................................20

M aterials............................................................................... 20
Chemicals and Biochemicals.............................. ......... ... 20
Radiochem icals ............................................................. 20
Antibodies and Hormones...................................................21
Oligonucleotides.............................................................. 21
Cell Lines and Culture Medium.......................................22
Recombinant cDNA Clones................................................ 23
General Methods .....................................................................23
Cocaine Treatment of Animals...........................................23
Isolation of Rodent Lymphocytes......................................... 24
Cell Proliferation Assay ................................................24
Measurement of Secreted IL-2............................................ 25
Measurement of Plasma GH, PRL, and IGF-1...........................25
ELISA for Secreted Human IgG and Rat Serum IgG................... 26
Sodium Dodecyl Sulfate Polyacrylamide Gel Electrophoresis
(SDS-PAGE) and Western Immunoblotting..................... 27
N-Terminal Sequence Analysis...........................................28
Metabolic Labeling and Immunoprecipitation........................... 28
Nb2 Lymphoma Cell Bioassay.............................................29
Isolation of RNA ............................................................ 29
Northern Blot Analysis ...................................................... 30
Preparation of Cytosol and Nuclear Extracts..............................31









Gel M obility Shift Assay .................................................. 31
UV Cross-Linking .............................................. ......... 32
Immunofluorescence Staining and Flow Cytometry Analysis........... 33
Quantitation of Autoradiographs........................ ........... .. 34
Statistical Analysis ............................. .................... ...34

CHAPTER 3: EFFECT OF TCDD ON LYMPHOCYTE PROLACTIN
PRODUCTION AND PROLACTIN STIMULATED MITOGENESIS......... 35

Introduction ................................................... ..................... 35
Results....... ..... ............................................................................... 37
Identification of Proteins Cross-Reacting with PRL Antiserum
in Human B Lymphocytes ......................................... 37
Effect of TCDD on Lymphocyte PRL-Immunoreactive Protein
Production......................................... ................. 38
Effect of TCDD on PRL-Stimulated Proliferation in Lymphocytes..... 39
D iscussion........................................ ....... ......................41

CHAPTER 4: EVALUATION OF IMMUNE PARAMETERS AND LYMPHOCYTE
PROLACTIN-IMMUNOREACTIVE PROTEIN PRODUCTION AFTER
CHRONIC ADMINISTRATION OF COCAINE TO PREGNANT RATS .... 57

Introduction .................................................. ..................... 57
Results ................................................................................... 59
Effect of Cocaine Treatment on Body and Lymphoid Organ Weights.. 59
Plasma Levels of IgG, PRL, GH, and IGF- ..........................60
Mitogen-Stimulated Proliferative Responses of Spleen Lymphocytes
and Thymocytes .......................................... ....... 60
IL-2 Production by Spleen Lymphocytes ................................. 61
Lymphocyte Production of PRL-Immunoreactive Proteins.............61
Northern Blot Analysis of PRL mRNA Expression .................... 62
D iscussion.................................................. ...........................63

CHAPTER 5: EVIDENCE THAT A MITOGEN-INDUCIBLE PROLACTIN-
IMMUNOREACTIVE PROTEIN IN RAT SPLEEN LYMPHOCYTES IS
ALDOLASE A................................................................ ...... 78

Introduction ................................................... .....................78
R esults.................................................................................. 79
PRL-Immunoreactive Proteins in Lymphocytes..........................79
N-Terminal Amino Acid Sequence of rPIP-43 .......................... 80
Specificity of the rPRL Antiserum for rPRL and Rabbit Aldolase A.... 81
Effect of Aldolase A in the Nb2 Lymphoma Assay ...................... 82
D iscussion................................................... .......................... 82

CHAPTER 6: CHARACTERIZATION OF THE AH RECEPTOR AND TCDD
RESPONSIVENESS IN HUMAN B LYMPHOCYTE CELL LINES..........95

Introduction .......................... ..................................... ......95
Results........................... ..................................................... 96
Human B Lymphocyte Cell Lines Express AhR and ARNT.............96
TCDD Induces CYP1Al Expression in PJS-91, But not IM-9 Cells... 97
Gel Mobility Shift Analysis of the Cytosolic AhR Complex........... 98
Gel Mobility Shift Analysis of the Nuclear AhR Complex............100









TCDD and Salt Concentration Dependency of AhR:DRE Binding
in Nuclear Extracts...................................... ......102
UV Cross-Linking Analysis of the Nuclear AhR Complex ...........102
D iscussion....................................... ...... .........................105

CHAPTER 7: TCDD RESPONSIVENESS IN HUMAN B LYMPHOCYTE CELL
LINES: EFFECT OF TCDD ON PROLIFERATION, IMMUNOGLOBULIN
PRODUCTION, AND CD19 EXPRESSION....................................122

Introduction ........................................................................ 122
Results...................... ......................................................... 123
Effect of TCDD on Spontaneous Cell Proliferation and
Immunoglobulin Production in IM-9 and PJS-91 Cells........123
Regulation of Gene Expression in B Lymphocytes by
TCDD: Approach................................................125
Effect of TCDD on Steady-State Levels of CD19 mRNA in IM-9
and PJS-91 Cells......................................... ......127
Flow Cytometric Analysis of CD19 Expression in IM-9 and
PJS-91 Cells ...................................................... 128
Oligonucleotides Corresponding to BSAP Binding Sites Compete
with the CYP1A1DRE for Binding to the Nuclear AhR
Complex ..............................................................131
TCDD Does not Alter Nuclear Protein Binding to the CD19-2
and 5'Sy2a Probes ........................................ .....132
Discussion.......................... ........ .................................133

CHAPTER 8: CONCLUSIONS AND FUTURE DIRECTIONS......................153

LIST OF REFERENCES ......................... ...... .........................161

BIOGRAPHICAL SKETCH ..............................................................180














KEY TO ABBREVIATIONS


AhR aryl hydrocarbon receptor
ARNT aryl hydrocarbon receptor nuclear translocator
BCGF B cell growth factor
BrdUTP 5-bromo-deoxyuridine triphosphate
BSA bovine serum albumin
BSAP B cell lineage-specific specific activator protein
C degrees celsius
cDNA complementary deoxyribonucleic acid
ConA Concanavalin A
CYP1AI cytochrome P-450 1A1
DMSO dimethyl sulfoxide
DNA deoxyribonucleic acid
DRE dioxin response element
DTT dithiothreitol
EBV Epstein-Barr virus
ELISA enzyme-linked immunosorbent assay
FBS fetal bovine serum
GH growth hormone
HIV human immunodeficeincy virus
HTLs human tonsillar lymphocytes
hr hours
IC50 50% inhibitory concentration
Ig(G, M) immunoglobulin (G, M)
IGF-1 insulin-like growth factor-1
IL-(2, 4, 5) interleukin-(2, 4, 5)
IFN-y interferon-y
ip intraperitoneal
iv intravenous
Kd dissociation constant
kDa kilodaltons








LPS lipopolysaccharide
MFI mean fluorescence intensity
Mr relative molecular mass
mRNA messenger ribonucleic acid
PBLs peripheral blood lymphocytes
PBS phosphate-buffered saline
PHA phytohemagglutinin
PMSF phenyl-methyl sulfonyl fluoride
(r, h)PRL (rat, human) prolactin
PRL-R prolactin receptor
PWM pokeweed mitogen
RIA radioimmunoassay
RNA ribonucleic acid
RT-PCR reverse transcriptase-polymerase chain reaction
SDS sodium dodecyl sulfate
SDS-PAGE sodium dodecyl sulfate polyacrylamide gel electrophoresis
SE standard error
SRBC sheep red blood cells
TCDD 2,3,7,8-tetrachlorodibenzo-p-dioxin
TGF-P transforming growth factor-P
TNF-a tumor necrosis factor-a
TRF T cell-replacing factor
TSST-1 toxic shock syndrome toxin-1














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


EVALUATION OF POTENTIAL MECHANISMS FOR XENOBIOTIC-MEDIATED
DISRUPTION OF LYMPHOCYTE FUNCTION AND RESPONSIVENESS

By

Scott Alexander Masten

December, 1995



Chairman: Kathleen T. Shiverick
Major Department: Pharmacology and Therapeutics


The first objective of this work was to investigate the potential role of prolactin

(PRL) in the immunotoxic action of the xenobiotics 2,3,7,8-tetrachlorodibenzo-p-dioxin

(TCDD) and cocaine. TCDD did not alter PRL-immunoreactive protein production by

human B lymphocyte cell lines nor the mitogenic response to PRL in the PRL-dependent

Nb2 T lymphoma cell line. Lymphocyte production of PRL-immunoreactive proteins was

significantly decreased in pregnant rats after in vivo cocaine treatment; however, decreased

PRL-immunoreactive protein production did not coincide with a change in other immune

parameters measured. A major PRL-immunoreactive protein induced after mitogenic

stimulation of rat spleen lymphocytes was found not to be a structural variant of PRL but

rather was identical to the glycolytic enzyme aldolase A. In summary, although prolactin

has been reported to have an important function within the immune system, lymphocyte-

derived prolactin does not appear to be a target for the immunotoxic actions of TCDD or

cocaine.








The second objective was to explore the use of human B lymphocyte cell lines to

characterize the responsiveness of human B lymphocytes to TCDD and to determine

potential mechanisms of TCDD-mediated immunotoxicity. TCDD did not significantly alter

spontaneous cell proliferation or immunoglobulin G production, and only induced

cytochrome P-450 1Al expression in one of two human B cell lines. In contrast, TCDD

did alter mRNA levels of CD 19, a B lymphocyte-specific signal transducing molecule. The

aryl hydrocarbon receptor (AhR) recognized a B lymphocyte transcription factor DNA

binding site in the promoter region of the human CD19 gene, data suggesting a possible

mechanism by which TCDD could modulate CD19 gene expression. Interestingly, there

were quantitative and qualitative differences in the DNA-binding form of the AhR between

human B lymphocytes and the HepG2 human hepatoma cell line. In summary, alteration

of CD19 expression may play a role in the humoral immunotoxicity of TCDD, and

furthermore, the B lymphocyte AhR may function in a manner different from what has

been demonstrated in other cell types. These findings have implications for evaluating

human health risks associated with TCDD exposure.














CHAPTER 1
INTRODUCTION


Objectives of this Research

A functional immune system is necessary to maintain adequate host resistance to

infectious diseases and malignancy. Over the past decade, there has been a growing

interest in the importance of examining the immunotoxic potential of xenobiotics (Luster

and Rosenthal, 1993). Most experimental techniques used in immunotoxicological studies

are based on the ability of the test animal, or cell cultures derived from the animal, to

respond to antigenic stimuli. The response usually measured is some effector function.

These approaches to screening xenobiotics for immunotoxicity, though efficient, do not

offer much insight into putative mechanisms of action, and in some instances do not

correlate with, or predict, changes in host resistance (Luster et al., 1993). Since a

functional immune response is the result of a highly complex interplay between a variety of

different cell types and the many different effector molecules they produce, this represents a

wide array of potential targets for immunotoxic action. In addition, this complicates the

search for a specific mechanism of action. This dissertation investigates potential

mechanisms by which known immunotoxic xenobiotics interfere with specific signal

transduction pathways and responsiveness in lymphocytes.

There is presently substantial evidence that prolactin (PRL) has an effector role in

the immune system. The first major objective of this research was to examine the possible

role of PRL as a target for xenobiotic-induced immunotoxicity. These studies examined the

effect of immunotoxic compounds on the production of PRL and PRL-related proteins by

normal rat lymphocytes and human B lymphocyte cell lines. PRL-specific responses in

lymphocytes were also examined. The compounds used in these studies were the








environmental contaminant 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD), and the drug of

abuse cocaine. The immunotoxic effects of TCDD and cocaine have been well

characterized, but a precise mechanism of action for each is still unknown.

Exposure to TCDD has been described in numerous reports to produce changes in

proliferation and/or differentiation in normal lymphocytes, and it is possible these changes

are the result of altered responses to extracellular signals. Furthermore, much less

information is available regarding the responsiveness of human lymphocytes to TCDD.

Therefore, the second major objective was to investigate whether the IM-9 and PJS-91

human B lymphocyte cell lines express the AhR and if so, whether these cell lines respond

to TCDD similar to normal lymphocytes and other cell types. Such studies provide an

opportunity to elucidate possible cellular mechanisms of action for the TCDD-induced

alterations in human B lymphocytefunction.

The next three sections of this Chapter present a brief review of the current literature

regarding the role of PRL in the immune system and the immunotoxic effects of cocaine

and TCDD.


PRL and the Immune System

PRL is a neuroendocrine hormone secreted by the anterior pituitary gland. There is

now evidence demonstrating that in addition to the pituitary, PRL is produced by the

placenta, uterus, brain and spinal cord, mammary gland, and lymphoid tissue (Sinha,

1995). The best described physiological effects of PRL are on the mammary gland and

reproductive tissues; however, PRL also has growth promoting effects on many tissues

(DeVlaming, 1979). Thus far, only one gene encoding PRL has been identified. In

mammals, the PRL gene encodes a 229 amino acid precursor polypeptide which is cleaved

to generate a 197-199 amino acid mature PRL molecule with a molecular weight of

approximately 23,000 (23K). In addition to the 23K PRL form, several other forms of

PRL can be detected in biological samples from various species (Sinha, 1995). PRL








variants of 16K, 21K, 25K, 45K, and >60K have been identified. PRL does undergo

post-translational modification, and it is likely that at least some of these variants result

from these modifications. Post-translational modifications which have been documented

include deamidation, cleavage, phosphorylation, sulfation, glycosylation, oligomerization,

and binding to plasma proteins (Sinha, 1995). Interestingly, some of these variants have

been shown to have biological activity and immunoreactivity different from 23K PRL, and

levels are altered during different pathological and physiological states (e.g. sleep,

pregnancy, breast feeding) (Sinha, 1995). It is currently not known exactly what role these

PRL variants have in vivo, and future research will determine if PRL variants are uniquely

responsible for some of the diverse physiological effects of PRL.


Immune Effects of PRL In Vivo

A possible role for PRL in the immune system was suggested by early experiments

where hypophysectomized (hypox) animals had impaired immune responses (Berczi et al.,

1981; Nagy and Berczi, 1981; Berczi and Nagy, 1982; Nagy et al., 1983; Berczi et al.,

1984). Similar changes were observed in animals treated with bromocriptine (Nagy and

Berczi, 1981; Berczi et al., 1984), a dopamine receptor agonist which potently inhibits

pituitary PRL secretion. These alterations included decreased production of IgG and IgM

against SRBCs (Berczi et al., 1981; Nagy et al., 1983), inhibition of an adjuvant arthritis

response (Berczi and Nagy, 1982; Berczi et al., 1984), and a decrease in the

dinitrochlorobenzene contact hypersensitivity response (Nagy and Berczi, 1981; Nagy et

al., 1983). These responses could be restored by syngeneic pituitary grafts or

administration of PRL, and in some cases, growth hormone (GH) was nearly as effective

(Nagy et al., 1983; Nagy et al., 1984). Additionally, hypox animals suffered from anemia,
leukopenia, thrombocytopenia, and decreased bone marrow RNA and DNA synthesis,

which could be returned to levels seen in control animals by PRL or GH treatment (Nagy

and Berzci, 1989). This study suggested a role for PRL in hematopoiesis; and subsequent








work further demonstrated that in hypox animals the decreased thymus and spleen weights,

as well as DNA synthesis in these organs, could be corrected by PRL administration

(Berczi et al., 1991). Other investigators have used the DW/J dwarf mouse, which lack

PRL and GH producing pituitary cells, and has numerous immunological defects (Dumont

et al., 1979). PRL treatment of these dwarf animals immunized with keyhole limpet

hemocyanin resulted in an increase in the number and proliferative capacity of antigen-

specific lymph node T cells (Murphy et al., 1993). In an earlier report, however, these

authors were unable to correct the deficiency of mature B cells in the bone marrow in these

animals with either PRL or GH administration (Murphy et al., 1992). There are also

reports describing decreased splenocyte mitogenic responses after treatment with

bromocriptine (Bernton et al., 1988), and cysteamine (Bryant et al., 1989), which depletes

pituitary PRL independent of dopamine receptors (Millard et al., 1982). In both cases,

PRL administration partially or completely restored these responses.


Immune Effects of PRL In Vitro

It has been more difficult to demonstrate direct actions of PRL on normal

lymphocytes. In mouse splenocytes, PRL augmented proliferation induced by

concanavalin A (ConA), phytohemagglutinin (PHA), and alloantigen (Montgomery et al.,

1988). The enhancement of mitogen-stimulated proliferation of murine splenocytes was

confirmed by one group (Spangelo et al., 1987), but not by others (Mukherjee et al., 1990;

Bernton et al., 1991). In ovariectomized rats, however; PRL increased splenocyte

proliferation when added alone and also enhanced the mitogenic response to interleukin-2

(IL-2) (Mukherjee et al., 1990). PRL has been shown to stimulate the cytotoxic activity of

natural killer cells purified from human blood (Matera et al., 1990), as well as to potentiate

the anti-IgM + IL-2-mediated stimulation of IL-2 receptor expression and immunoglobulin

production in normal human B cells (Lahat et al., 1993).








The best characterized actions on lymphoid cells have been described using rodent

T cell lines. These are the rat T lymphoma cell line, Nb2, which is dependent on PRL for

growth (Gout et al., 1980), and the mouse helper T cell clone, L2. In Nb2 cells, PRL

induces the transcription of genes for c-myc, P-actin, an HSP-70 homologue, ornithine

decarboxylase, and interferon regulatory factor-I (Yu-Lee, 1990; Yu-Lee et al., 1990). In

the L2 cell line, PRL is required in the nucleus for IL-2 driven cell proliferation (Clevenger

et al., 1991). PRL also induces interferon regulatory factor-1 in this cell line, as well as

enhancing IL-2 stimulated transcription of other growth related genes (Clevenger et al.,

1992). Unfortunately, no studies to date have demonstrated that these responses occur in

response to PRL in normal lymphocytes.


PRL and PRL Receptor Expression in Lymphocytes


Although PRL is generally thought to be strictly a pituitary gland-derived hormone,

PRL is also produced by several other tissues, including cells of lymphoid origin. There

are now many reports in the literature describing production of PRL and PRL-related

proteins by both human and rodent lymphocytes. The first demonstration was that ConA-

stimulated murine splenocytes secreted a factor which cross reacted to PRL antiserum and

was mitogenic for Nb2 cells (Montgomery et al., 1987). Subsequently, a PRL cross

reactive protein was identified in resting mouse splenocytes by immunocytochemistry, and

the intensity of staining was increased in ConA-stimulated cells (Kenner et al., 1990). In

this study, western blotting of lysates from ConA treated cells revealed a Mr 48,000 PRL-

immunoreactive protein. A protein of this size was also noted in ConA-stimulated murine
splenocytes and after proteolytic cleavage and peptide mapping of the gel-purified Mr

46,000 band, there were two fragments with similarity to pituitary PRL (Shah et al.,

1991). This is the strongest evidence thus far that a higher molecular weight lymphocyte

PRL-immunoreactive protein is structurally related to PRL. Also, murine thymocytes

constitutively expressed Mr 33,000 and 35,000 PRL-immunoreactive proteins, and a Mr








22,000 mitogen-inducible PRL-immunoreactive protein. The proteins are synthesized in

vitro and the Mr 22,000 protein is secreted into the media (Montgomery et al., 1990).

PRL-immunoreactive proteins have also been identified in human lymphocytes.

Human thymocytes stimulated with ConA synthesized Mr 11,000, 21,000, and 24,000

PRL-immunoreactive proteins (Montgomery et al., 1992), while peripheral blood

lymphocytes (PBLs) produced a Mr 27,000 protein (Montgomery et al., 1992), a Mr

60,000 (Sabharwal et al., 1992), and Mr 23,000 and 25,000 PRL-immunoreactive proteins

(Pellegrini et al., 1992). Among human cell lines, the IM-9 B lymphoblast line synthesizes

and secretes PRL (DiMattia et al., 1988), and AG-876 and Ramos, two Burkitt lymphoma

cell lines, also produce PRL-immunoreactive proteins (Baglia et al., 1991; Sabharwal et

al., 1992). It is important to note that none of the identified PRL-immunoreactive proteins

in either murine or human cells have been purified or sequenced, so the relation to pituitary

PRL is known. However, in three reports RT-PCR was used to amplify PRL mRNA

from ConA-stimulated human thymocytes (O'Neal et al., 1992), human PBLs (Sabharwal

et al., 1992), and purified human T and B cells from peripheral blood (Pellegrini et al.,

1992). These results confirmed that the PRL gene is expressed in lymphocytes.

Clearly, mouse and human lymphocytes are capable of producing PRL, which may

act in an autocrine/paracrine fashion to maintain lymphocyte growth and/or function. A

necessary part of this model is the presence of specific PRL receptors (PRL-R) on immune

cells. Indeed, PRL-R gene expression has been demonstrated by RT-PCR in human T and

B lymphocytes (Pellegrini et al., 1992), and mouse thymocytes and splenocytes (O'Neal et

al., 1991). Recently, PRL-R have also been identified by flow cytometry on rat

splenocytes (Viselli and Mastro, 1993), and mouse thymus, spleen, bone marrow, lymph

node, and blood lymphocytes (Gagnerault et al., 1993). The recent findings that the Nb2

PRL-R is associated with and can activate the serine-threonine kinase Raf-1 (Clevenger et

al., 1994) and the tyrosine kinase Jak-2 (Rui et al., 1994), which have been shown to be

involved in the signal transduction pathway for several hematopoetic growth factors








(Horseman and Yu-Lee, 1994), is a further suggestion that PRL is likely to play a role in

lymphocyte function.


Immunomodulation by Cocaine

The tropane alkaloid cocaine, extracted from the leaves of the coca plant,

Erythroxylon coca, is a powerful local anesthetic by virtue of the ability to block sodium

channels. Another well documented effect of cocaine is competitive inhibition of the

reuptake of catecholamine (norepinephrine and dopamine) and serotonin neurotransmitters

at nerve endings (Ritchie and Greene, 1990). Cocaine also alters the plasma levels of many

neuroendocrine hormones including ACTH, P-endorphin, corticosterone, and PRL (Van de

Kar, 1992). The mechanism of immunomodulation of cocaine is unknown at present, but

it is interesting to note that all of the above hormones have been ascribed some sort of

immunomodulatory role, and alteration in the level of these hormones may account for at

least some of the effects of cocaine on the immune system. The emerging awareness of the

reciprocal interplay between the neuroendocrine and immune systems has led to the

proposal that many compounds may lead to immunotoxicity secondary to a perturbation of

neuroendocrine function (Fuchs and Sanders, 1994). With this in mind, it is reasonable to

hypothesize that disruption of neuroendocrine hormone production by cocaine could be

responsible for some of the effects of cocaine on the immune system.

Cocaine use has been associated with increased occurrence of several types of

infections (Richter, 1993), some of which are likely the result of tissue damage at the site

of exposure to the drug or lifestyle modifications (poor nutrition and sexual promiscuity).

However, there have been epidemiological studies revealing an association between cocaine

use and seropositivity for human immunodeficiency virus (HIV) (Chaisson et al., 1989;

Anthony et al., 1991) that upon multiple regression analysis, could not be accounted for

by such confounding factors as needle sharing or sexual habits (Anthony et al., 1991).

Interestingly, cocaine was recently shown to increase HIV replication in








phytohemagglutinin (PHA) and cytomegalovirus activated human PBLs and this effect

could be prevented by including antibodies to transforming growth factor-3 (TGF-P) or

tumor necrosis factor-at (TNF-a) (Petersen et al., 1991; Peterson et al., 1992).

There is further evidence that cocaine alters immune function directly with in vitro

exposure. Cocaine is capable of affecting mitogen-stimulated proliferation of human PBLs

and mouse splenocytes. Very high concentrations (>25 gg/ml) were required to decrease

optimal ConA (Klein et al., 1988; Delafuente and Devane, 1991), or PHA (Klein et al.,

1988) stimulation of proliferation in human and mouse lymphocytes. Another study

reported no effect of cocaine on PHA stimulation of human or mouse cells at concentrations
up to 32 pg/ml (Luo et al., 1992). When suboptimal concentrations of PHA were used,

there was a slight decrease in proliferation and calcium mobilization in T cell enriched
human PBLs at lower (<10 gM) concentrations of cocaine (Klein et al., 1993). These

effects could be overcome by increasing the concentration of PHA or using unseparated

PBLs, suggesting that T lymphocytes are a sensitive target of cocaine. In another report,

cocaine potentiated proliferation, IL-2 secretion, and calcium mobilization in purified

human peripheral blood T cells stimulated with an antibody to the CD3 component of the T-

cell receptor (Matsui et al., 1992).

In vivo studies with cocaine have been carried out almost exclusively with mice and

the results of these studies are often contradictory. Some of the reported changes include

decreased thymocyte numbers (Ou et al., 1989; Di Francesco et al., 1992; Lopez et al.,

1992), decreased natural killer cell activity and in itro generated cytotoxic T cell activity

(Di Francesco et al., 1992), and decreased splenocyte cytokine production (Di Francesco et

al., 1992; Lopez et al., 1992). Antibody responses are also affected by cocaine treatment.

There are several reports of decreased antibody forming cells specific for sheep red blood

cells (SRBCs), a T-dependent antigen (Faith and Valentine, 1983; Holsapple and Munson,

1985; Ou et al., 1989). In another study, high doses of cocaine (40-50 mg/kg) given

several times a day increased antibody responses to both T-independent and T-dependent








antigens, however; a lower dose was ineffective in altering this response (Havas et al.,

1987). In rats, a T-independent antibody response was increased, while a T-dependent

response was increased only at low doses, and decreased at the highest dose (Bagasra and

Forman, 1989). Also in this study, there was a decrease in spleen T cells and an increase

in spleen activated B cells which may partially explain the observed alterations in antibody

responses. Host resistance to a tumor cell challenge was reported decreased (Ou et al.,

1989) or unchanged (Havas et al., 1987), while resistance to a bacterial challenge (S.

pneumoniae) was reported unaffected (Havas et al., 1987). Since the experimental design

in the studies cited above are in all but a few cases very different, it is difficult to draw

conclusions on possible target cells or mechanisms. However, it is clear that cocaine can

alter many aspects of immune function and warrants further investigation.


Biological Effects of TCDD

General Effects of TCDD

TCDD is the prototype compound for a large and diverse group of chemicals

referred to as halogenated aromatic hydrocarbons (HAHs), which include polychlorinated

dibenzodioxins (PCDDs), polychlorinated dibenzofurans (PCDFs), and polychlorinated

biphenyls (PCBs). TCDD is a ubiquitous environmental contaminant and is produced in

the manufacture and combustion of various chlorinated chemicals. The sources of release

of TCDD into the environment are primarily anthropogenic, although some PCDDs and

PCDF are produced during forest fires (Vanden Heuvel and Lucier, 1993). The major

source of these compounds is from municipal and medical waste incinerators, with some

HAHs produced during pulp bleaching, as intermediates in the manufacture of chlorine-

containing chemicals, and in the combustion of other organic materials (Vanden Heuvel and

Lucier, 1993). The estimated human exposure to PCDDs arid PCDFs is 1-3 pg/kg/day and

the half-life of TCDD is 5-10 years in humans and 10 years in soil (Vanden Heuvel and

Lucier, 1993). TCDD exerts a wide spectrum of toxic effects in laboratory animals.








Among these are immunotoxicity, tumor promotion in liver and skin, developmental

toxicity and teratogenicity, male and female reproductive toxicity, epithelial hyperplasia,

and carcinogenicity (primarily in the liver and thyroid gland) (Vanden Heuvel and Lucier,

1993). The only clearly documented toxic response to TCDD in humans is chloracne, a

condition characterized by hyperkeratization of the skin, although there is some evidence

for immunotoxicity, an increased risk of cancer, and developmental toxicity (Silbergeld and

Gasiewicz, 1989)

A schematic illustrating the current model of TCDD action is shown in Fig. 1-1.

TCDD acts by binding with high affinity to an intracellular protein, termed the aryl

hydrocarbon receptor (AhR). Although TCDD is the highest affinity ligand known,

originally thought to have a binding affinity in the subnanomolar range but now believed to

be in the picomolar range (Bradfield et al., 1988; Landers and Bunce, 1991), other HAHs

also bind to the AhR (Hankinson, 1995). Some PCDDs, PCDFs, PCBs, aromatic amines

and indolecarbazole derivatives all have appreciable affinity for the AhR. The unoccupied

and inactive AhR is located in the cytosol completed with the heat shock protein, hsp90,

and possibly other proteins (Okey et al., 1994). Ligand binding activates the AhR,

promoting the release of hsp90 and allowing the AhR to associate with another protein, the

aryl hydrocarbon nuclear translocator (ARNT). Both the AhR and ARNT are members of

the basic helix-loop-helix family of transcription factors (Hoffman et al., 1991; Burbach et

al., 1992; Ema et al., 1992). The AhR-ARNT heterodimer (referred to hereafter as the

AhR complex) is the transcriptionally active form of the AhR and binds to specific DNA

sequences termed dioxin response elements (DREs) containing the consensus sequence

TNGCGTG (Whitlock, 1993). The hallmark response to TCDD is transcriptional

induction of the cytochrome P-450 1A gene (CYPIA1). TCDD-dependent transcription

of CYPIA1 and genes for other drug metabolizing enzymes is mediated by binding of the

AhR complex to DREs present in the 5' flanking region (Okey et al., 1994). There are

reports of several other genes being regulated by TCDD, including the growth regulatory









genes transforming growth factor-a, TGF-l1 and 2, IL-lp, plasminogen activator

inhibitor-2, the epidermal growth factor receptor, and c-fos and c-jun; however, there is as

yet no conclusive evidence that these genes are under direct transcriptional control by the

AhR (Okey et al., 1994; Hankinson, 1995).

It is generally accepted that the toxicity of TCDD is mediated by the AhR. The two

approaches classically used to determine if a specific response is mediated by the AhR are

genetic and pharmacological. The genetic approach involves the use of strains of mice, or

the different complementation groups of the Hepa- murine hepatoma cell line, that vary in

their susceptibility to TCDD (as measured by the induction of CYP1AI). The classic high

responder mouse strain is the C57BL/J6 mouse which has an AhR with a 10-fold higher

affinity for TCDD than the classic low responder strain DBA/2 (Hankinson, 1995).

Demonstration that a toxic response segregates with the high responder mouse strain (or the

proper complementation group of the Hepa Ic cell line) is accepted as evidence that the

response in question is mediated by the AhR. The pharmacological approach uses the

structure-activity relationship for AhR ligands with affinities lower than TCDD. In this

case, lower affinity ligands should require higher concentrations to elicit a particular toxic

response (if at all). Another pharmacological approach is to use AhR antagonists to block a

toxic response; however, recent evidence has shown this method to be less conclusive

since the AhR "antagonists" identified thus far also have partial agonist activity. There is

some evidence suggesting non-Ah receptor dependent events in TCDD-induced toxic

responses (Landers and Bunce, 1991; Okey et al., 1994), however; these are not well

characterized and not will be addressed further.

Over 20 years ago, it was demonstrated that TCDD could cause atrophy of the

thymus in mice, rats, and guinea pigs (Vos et al., 1973). Since the thymus is where T cell

maturation occurs, these initial observations lead to many studies on the effects of TCDD

on cellular immunity (T cell-mediated), as well as other aspects of the immune system.

These studies have generated a very large database from which to design future








investigations aimed at characterizing a specific mechanism of action for TCDD. From the

data compiled thus far, it is clear that TCDD is capable of altering cell mediated, humoral,

as well as innate immunity with varying sensitivity (Holsapple, 1991). The degree of

immunotoxicity and the components of the immune system affected have been recognized

to differ with the species being tested (the mouse is by far the most commonly used),

although there are clearly some effects in all species tested. The in vivo humoral

immunotoxicity of TCDD also varies with the time of exposure. Effects on cell-mediated

immunity are especially pronounced after prenatal or perinatal exposure, while changes in

humoral immunity are most apparent after adult exposure (Holsapple, 1991). The role of

the AhR in the immunotoxicity of TCDD is well established, however; as with other

responses to TCDD, there is some evidence that TCDD may elicit immunotoxic responses

through non-AhR dependent mechanisms (Holsapple et al., 1986; Kerkvliet et al., 1990;

Davis and Safe, 1991; Kerkvliet and Burleson, 1994). It remains controversial as to what

role these may play in the immunotoxicity seen with TCDD exposure, nor is it clear

whether both Ah receptor dependent and independent events are important within the same

cell.


Effects of TCDD on Humoral Immunity and B Lymphocytes


TCDD exposure results in a decrease in serum antibody levels and stimulated

antibody production (Thomas and Hindsill, 1979; Hindsill et al., 1980; Vecchi et al., 1980;

Clark et al., 1981; Kerkvliet et al., 1985). These effects occur in the absence of

appreciable effects on lymphoid organ size, cell number, or cell viability. One of the most

sensitive endpoints of TCDD toxicity described to date is inhibition of the antibody

response to SRBCs in responsive mouse strains (Davis and Safe, 1988; Holsapple et al.,

1991; Narasimhan et al., 1994). Inhibition of the SRBC antibody response results after as

little as a single dose of 0.1 Rg/kg in the C57/BL6 mouse (Davis and Safe, 1988), a dose

several fold lower than that required to produce thymic atrophy. SRBC is considered to be








a T-dependent antigen because a functional antibody response requires the presence of

competent B cells, macrophages, and T cells (Mosier and Coppelson, 1968). Because of

the requirement for T cells in the SRBC antibody response and earlier work showing

significant effects of TCDD on T cell mediated responses, some authors have suggested

that inhibition of the antibody response is the result of a primary effect of TCDD on T cells.

Further work, however, demonstrated that antibody responses to the T-independent

antigens TNP-LPS and DNP-Ficoll were suppressed in a similar fashion as the SRBC

response over the same dose ranges of TCDD (Dooley and Holsapple, 1988). The

suppression of T-independent antibody responses has been reported by others (Vecchi et

al., 1980; Luster et al., 1988; House et al., 1990). In a series of separation/reconstitution

experiments using B or T cells isolated from control and treated mice and reconstituted in

various combinations, it was shown that the B lymphocyte is the primary target of TCDD

(Dooley and Holsapple, 1988; Dooley et al., 1990). This evidence supports the hypothesis

that TCDD is capable of directly affecting B cells, and strongly suggests that effects on the

B cell may be responsible for the suppression of antibody response after TCDD exposure.

With evidence that the B cell is a primary target for TCDD in vivo, several groups

have characterized the direct effects of TCDD on B cells stimulated in itr with various

mitogens and growth factors. Interestingly, the ability of TCDD to alter B cell responses

was found to differ with the method used to stimulate the cells. Pokeweed mitogen (PWM)

stimulates proliferation and immunoglobulin production in human and mouse spleen,

tonsil, and peripheral blood B cells as long as macrophages and T cells are present.

TCDD, at concentrations of 0.3-30 nM, had no effect on PWM-stimulated proliferation or

immunoglobulin production by mouse splenocytes or human tonsillar lymphocytes (Wood

et al., 1992). In this same study, however, background proliferation was significantly

suppressed at all concentrations of TCDD in both mouse and human lymphocytes. These

same authors also investigated the effect of TCDD on toxic shock syndrome toxin (TSST-

1) stimulated proliferation and immunoglobulin production in human tonsillar B








lymphocytes. TSST-1 is an exotoxin secreted by St .aph .uus which, in the presence of

irradiated T cells, forms a "bridge" between the T cell receptor and MHC Class II

molecules on B cells. This cross-linking stimulates proliferation and differentiation of B

cells (Mourad et al., 1989). TCDD had no effect on TSST-1 stimulated proliferation, but

inhibited IgM production by approximately 75% at a concentration of 0.3 nM (Wood and

Holsapple, 1993). In another study, purified murine splenic B cells were exposed to 1-50

nM TCDD and stimulated with anti-IgM alone or together with either B cell growth factor

(BCGF) or T cell replacing factor (TRF) (Luster et al., 1988). Antibodies to surface IgM

crosslink the B cell antigen receptor and mimic the signal delivered when the B cell

encounters antigen. Anti-IgM alone only stimulates proliferation but, when combined with

the helper T cell derived products BCGF or TRF, leads to immunoglobulin production

(Jelinek and Lipsky, 1987). The cytokines IL-4 and IL-5 are now known to be the active

proteins in these earlier BCGF and TRF preparations, respectively, although other proteins

are also present. TCDD had no effect on anti-IgM stimulated inositol phosphate formation,

or anti-IgM + BCGF-stimulated RNA and DNA synthesis (Luster et al., 1988).

Additionally, when TCDD-treated cells were stimulated with anti-IgM + BCGF + TRF,

there was a dose related suppression of IgM production. TCDD also decreased expression

of the plasma cell surface marker PC.2 without affecting expression of MHC Class II

antigen (Ia). These results implied that proliferation and activation of B cells (as measured

by increased expression of la) were minimally affected by TCDD, but differentiation into

plasma cells (measured by PC.2 expression) was impaired by TCDD exposure. Similar

results were also noted in human B cells (Wood et al., 1993). In these experiments, the

proliferative response of purified low density (preactivated) human tonsillar B lymphocytes

(HTLs) to LPS + TRF was suppressed 25% by 30 nM TCDD. Lipopolysaccharide (LPS)

+ TRF-stimulated IgG secretion in these cells was also inhibited to a similar extent by

TCDD. When high density (resting) HTLs were stimulated with anti-IgM and BCGF, with

or without TRF, TCDD had no effect on proliferation. These experiments suggest that the








effect of TCDD on B cell responsiveness to stimulatory factors varies with the activational

state of the cell. B cells also respond to certain growth factors only during specific stages

of activation and development (Bancherau and Rousset, 1993). A possible explanation of

these findings is that the effects of TCDD on the B cell are due to an inappropriate alteration

in the cellular programming which renders the cell unable to respond to the proper growth

factors. Alternatively, TCDD may interfere with the necessary response to growth factors

during specific states of activation. In support of the first possibility, the proliferative

response to anti-IgM, or phorbol ester + Ionomycin, both of which stimulate calcium

influx, was inhibited by 35-50% by TCDD in a dose related manner (Karras and

Holsapple, 1994). In another report, however, there was no effect of TCDD on anti-IgM

stimulated calcium mobilization (Karras and Holsapple, 1993). The suppression of the

phorbol ester + ionomycin response could be overcome by increasing the concentration of

ionomycin (Karras and Holsapple, 1994). Clearly, there is a change in calcium

homeostasis in TCDD-treated B cells that could be partially responsible for the observed

effects. Concerning the possibility that TCDD disrupts growth factor signalling pathways,

there are as yet no reports in the literature on the effects of TCDD on the B cell response to

a single specific cytokine or growth factor.

TCDD has also been shown to alter the pattern of protein phosphorylation in B

lymphocytes. Phosphorylation of cellular proteins is a key event in normal cellular

processes as well as in growth factor signalling. In murine splenic B cells, I nM TCDD

increased the phosphorylation of Mr 15,000, 29,000, 45,000, 52,000, and 63,000

proteins by 40-70% (Snyder et al., 1993). It was further shown that interferon-y (IFN-y)

also increased the phosphorylation of the same set of proteins, except for the Mr 15,000

protein, but the IFN-y-stimulated phosphorylation was of lower magnitude. When TCDD

and IFN-y were added together, phosphorylation of these proteins was decreased by 25-

40% relative to control levels. Interestingly, the same concentration of IFN-y was also able

to reverse the TCDD suppression of both the LPS-stimulated and SRBC antibody response








if added within 18 hr of the beginning of the culture period. In another study examining

protein phosphorylation in purified mouse spleen B cells, TCDD increased phosphorylation

of Mr 17,000, 38,000, 42,000, 50,000, 54,000, 61,000, 64,000, and 105,000 proteins

(Clark et al., 1991). The Mr 50,000, 61,000, and 64,000 proteins were also

phosphorylated after anti-IgM + BCGF treatment. This effect on protein phosphorylation

was characterized as being independent of protein kinase C. For the Mr 61,000 and

64,000 proteins, increased phosphorylation was found to occur within 5 min of TCDD

exposure and to be specific for tyrosine residues. When TCDD was added to cells pre-

stimulated for 36 hr with anti-IgM + recombinant IL-4, tyrosine specific phosphorylation

of the Mr 61,000 and 64,000 proteins was decreased by 45%. This is surprisingly similar

to the results in the previous study, where TCDD + IFN-y treatment of activated B cells

showed a lower level of phosphorylation than with either treatment alone. Additionally, the

molecular weights of two of these proteins are almost identical, and could represent the

same proteins. It is intriguing that the molecular weight of these proteins corresponds

roughly to that of the B lymphocyte src family tyrosine kinases lyn (53 kDa) and fyn (59

kDa), both of which are tyrosine phosphorylated after ligation of the B-cell antigen receptor

(Cambier and Campbell, 1992). Clearly, TCDD has the ability to alter protein

phosphorylation either alone or in response to growth and differentiation stimuli, and is

another possible mechanism by which TCDD could suppress humoral immunity.


Thesis Rationale

PRL as a Target for Cocaine and TCDD


From the work cited above it can be concluded that PRL is an effector molecule in

the immune system. It is at present unclear, however, what role lymphocyte-derived PRL

might have and, furthermore, there are no reports on the regulation of PRL production by

normal lymphocytes. Dexamethasone and the phorbol ester PMA have been shown to alter

PRL secretion in the human IM-9 cell line (Gellersen et al., 1988; Gellersen et al., 1989).








Interestingly, both of these compounds have significant effects on normal lymphocytes.

Therefore, other immunotoxicants may modulate lymphocyte PRL production or disrupt

PRL signalling and PRL-specific responses in lymphocytes. Investigation of this

possibility will increase our understanding of the importance of PRL in lymphocyte

function, as well as provide insight into novel mechanisms of action for potential

immunotoxic compounds.

A more thorough understanding of the effects of cocaine on the immune system is

important for two reasons. First, it is a widely abused drug and a significant percentage of

the population is being exposed; and secondly, there have been a number of

epidemiological studies revealing an association between cocaine use and the risk of

infection, most notably HIV.

The specific aims of this portion of the research are as follows:

Specific aim 1. Characterize PRL-immunoreactive proteins in lymphocytes.

Specific aim 2. Investigate the general immunotoxic effects of in vivo cocaine

administration to pregnant rats as measured by lymphoid organ size, lymphocyte mitogenic

responses, IL-2 production, and lymphocyte production of PRL.

Specific aim 3. Evaluate whether TCDD alters lymphocyte production of PRL-

immunoreactive proteins or PRL-stimulated mitogenic responses in lymphocytes.


Responsiveness of Human B lymphocytes to TCDD


The potency of TCDD in producing adverse effects in laboratory animals and our

lack of knowledge about how TCDD really affects cellular processes has been, and still is,

the impetus for continued investigation into the effects and mechanism of action of TCDD.

Additionally, due to the large amount of research already devoted to TCDD, this compound

can now serve as a model for future toxicology studies. This is especially important to the

discipline of immunotoxicology since TCDD can affect so many different components of

the immune system and may serve to define how this discipline can better be studied. Most








of the research devoted to the effects of TCDD on the immune system have been conducted

using rodents, or cells and cell lines derived from rodents. Therefore, studies examining

the actions of TCDD on human lymphocytes are meaningful, especially if such studies lead

to an increased understanding of how TCDD and other potential immunotoxic compounds

may alter the functional capacity of these cells.

The specific aims of this portion of the research are as follows:

Specific aim 4. Characterize the AhR and TCDD responsiveness in human B

lymphocyte cell lines.

Specific aim 5. Evaluate the effect of TCDD on proliferation, immunoglobulin

production, and gene expression in the IM-9 and PJS-91 human B lymphocyte cell lines.

Successful completion of specific aims 1-5 will increase our knowledge and

understanding of the role of PRL in immune function as well as the relationship between

cocaine, TCDD, and lymphocyte-derived PRL. This will be useful in designing new

methods for assessing immunotoxicity as well as possible strategies for enhancing immune

responsiveness. An increased understanding of the effects of TCDD on human B

lymphocyte function will provide a basis for determining the susceptibility of humans to

TCDD-induced humoral immunotoxicity, as well as formulating approaches to characterize

and quantitate the health risks associated with TCDD exposure in humans. These studies

will also provide information regarding the mechanism of action of TCDD in the

suppression of humoral immune responses.












gs-


m 0



Ira

il


m .0














-H
0i
wg -a






~i|g

*g'sgs














CHAPTER 2
MATERIALS AND METHODS

Materials

Chemicals and Biochemicals

Cocaine-HCl, ConA, PHA, PWM, S. typhimurium LPS, rat IgG, human IgG, p-

nitrophenyl phosphate, Histopaque 1077, 5-bromodUTP, and normal rabbit serum were

obtained from Sigma Chemical Co. (St. Louis, MO). Recombinant human IL-2 was

purchased from Genzyme (Cambridge, MA), and recombinant human insulin-like growth

factor-I (IGF-1) from Bachem California (Torrance, CA). TCDD was obtained from

Midwest Research Institute (Kansas City, MO) through the National Cancer Institute

Chemical Carcinogen Reference Repository. Oligo(dT)-cellulose and medium and

antibiotics for cell culture were from Gibco/BRL (Grand Island, NY), except horse serum

and fetal bovine serum (FBS), which were from Hyclone Laboratories (Logan, UT). Poly

(dI-dC):(dI-dC), and nucleotides were obtained from Pharmacia Biotech (Piscataway, NJ),

and rat IgG was from Fisher Scientific (Pittsburgh, PA). All other chemicals were of the

highest quality available from standard suppliers unless stated otherwise.

Radiochemicals

[3H-methyl] thymidine (5 Ci/mmole) and [y32P]dATP (3000 Ci/mmole) were

obtained from Amersham Life Sciences (Arlington Heights, IL), while [1251] Protein A (9

mCi/mg), Tran SLabelM (1000 Ci/mmole), [a32P]UTP (3000 Ci/mmole), and
[a32P]dCTP (3000 Ci/mmole) were from ICN Biomedical (Costa Mesa, CA).








Antibodies and Hormones


Anti-rat PRL-IC-5, anti-human PRL-IC-5, rat PRL-RP-3, rat PRL-B-7, rat PRL-

B-8-SIAFP, human PRL-B-1, human PRL-B-2-SIAFP, rat GH-RP-2, anti-rat IGF-1

(UB3-189), and antibodies for GH and PRL radioimmunoassay (RIA) were kindly

provided by the National Hormone and Pituitary Program (NIDDK, NICHHD, USDA,

Rockville, MD). Polyclonal rabbit anti-hPRL (lot# P4802-1) was from Hycor Biomedical

Inc. (Portland, ME). The AhR (Poland et al., 1991) and ARNT (Probst et al., 1993)

antibodies were kindly provided by Dr. Christopher Bradfield (Northwestern University,

Chicago, IL) and Dr. Oliver Hankinson (University of California at Los Angeles, Los

Angeles, CA), respectively. Goat anti-human IgG alkaline phosphatase conjugate, anti-

human CD19 (clone SJ25-C1), MOPC-21 IgGI, rabbit anti-rat IgG horseradish peroxidase

conjugate, and goat anti-mouse IgG fluorescein isothiocyanate conjugate (human serum

absorbed) were obtained from Sigma Chemical Co. (St. Louis, MO). Goat anti-rabbit IgG

horseradish peroxidase conjugate was from Bio-Rad Laboratories (Richmond, CA) and

horse anti-goat IgG horseradish peroxidase conjugate was from Vector Laboratories

(Burlingame, CA). Rabbit anti-rat CYPIAl was from Gentest Corp. (Woburn, MA).

Goat anti-rat IgG was from Fisher Scientific (Pittsburgh, PA), and goat anti-human IgG

was from Jackson Immunoresearch (West Grove, PA).

Oliponucleotides

Complimentary oligonucleotides were synthesized by the University of Florida

DNA Synthesis Core Laboratory.

The human CYPIAI DRE oligonucleotides were:

5'-TGCCCAGGCGTTGCGTGAGAAGGACCGGAG-3'

5'-CTCCGGTCCTTCTCACGCAACGCCTGGGCA-3'








The human CYP1A1 DREmut2 oligonucleotides were:

5'-TGCCCAGGCGTTGATTGAGAAGGACCGGAG-3

5'-CTCCGGTCCTTCTCAATCAACGCCTGGGCA-3'

The CYPIA1 8 base pair primer sequence was:

5'-TGCCCAGG-3'

The CD19-2 oligonucleotides were:

5'-GAGAATGGGGCCTGAGGCGTGACCACCGCCTTCC-3'

5'-GGAAGGCGGTGGTCACGCCTCAGGCCCCATTCTC-3

The 5'Sy2a oligonucleotides were:

5'-GGATCAGAATTGTGAAGCGTGACCATAGAAAGAA-3'

5'-TCTTTCTATGGTCACGCTTCACAATTCTGATCC-3'


Cell Lines and Culture Medium


The Nb2 1 Ic subline was generously provided by Dr. Paul Kelly, INSERM, Paris,

France. The IM-9 human B lymphoblast cell line (CCL 159), the CTLL-2 mouse cytotoxic

T cell line (214-TiB), and the human hepatoma cell line HepG2 (HB8065) were from the

American Type Culture Collection (Rockville, MD). The PJS-91 cell line was generated in

this laboratory by Epstein-Barr virus (EBV) transformation of peripheral blood

lymphocytes isolated from a healthy male donor. The IM-9 and PJS-91 cell lines were

maintained in RPMI 1640 medium supplemented with 100 gg/ml streptomycin, 100 U/ml

penicillin, 2 mM L-glutamine, and 10% FBS. The HepG2 culture medium was minimum

essential medium supplemented with 50 ig/ml gentamicin, 2 mM L-glutamine, 1 mM

sodium pyruvate, and 10% heat-inactivated FBS. The Nb2 and CTLL-2 culture medium is

described elsewhere.








Recombinant cDNA Clones


The rPRL probe was a PstI fragment of the rPRL cDNA (Cooke et al., 1980); the

human CD19 probe was an EcoRI + SacI fragment of pB4-19 (Tedder and Isaacs, 1989);

the human AhR probe was a SmaI fragment of phuAhR (Dolwick et al., 1993); the human

ARNT probe was a BamHI fragment of pBM5Neo-MI-1 (Hoffman et al., 1991); and the

human CYPIB1 probe was an EagI fragment of clone 1 (Sutter et al., 1994). The

following cDNA containing plasmids with the clone designation and corresponding ATCC

catalogue number were obtained from the American Type Culture Collection (Rockville,

MD): human 1-actin (HHCI89; 65129), human PRL (p-hPRL; 31721), human CYPIAl

(phPl-450-3'; 57259), and human c-myc (pG1-5'-c-myc; 39286).


General Methods

Cocaine Treatment of Animals


Timed pregnant or non-pregnant Sprague-Dawley rats (Harlan-Sprague Dawley,

Indiannapolis, IN) were housed in the University of Florida animal resources facility

according to University guidelines. Animals were maintained on a 12 hour light/dark cycle

with food (Purina rat chow) and tap water provided ad libitum. Body weight and food

consumption were monitored daily. Each daily dose of cocaine, 60 mg/kg calculated as the

free base, was administered as a split dose of 30 mg/kg by i.p. injection between 8 and 9

a.m. and again between 4 and 5 p.m.; control animals received saline vehicle. Cocaine

administration was initiated on gestation day 8 and continued through gestation day 19. On

day 20, the animals were anesthetized by sodium pentobarbital injection, and maternal

blood was collected by cardiac puncture into heparinized syringes. Spleens and thymuses

were collected and the animals euthanized with a second injection of sodium pentobarbital.

A total of four separate experiments with 5 to 6 animals in each group were carried out with

similar results, and the data presented herein are from one representative experiment.








Isolation of Rodent Lymphocytes


Spleens and thymuses from mice and rats were placed in lymphocyte medium

(RPMI 1640 supplemented with 100 U/ml penicillin, 100 ig/ml streptomycin, 2 mM L-

glutamine, 25 mM HEPES, 50 pm 2-mercaptoethanol, and 10% FBS). After weighing,

single cell suspensions were prepared by pressing the tissues through a stainless steel mesh

screen. Cells were then centrifuged through Histopague 1077 to obtain spleen

lymphocytes, or lysis of red blood cells with 1 mM KPO4, 150 mM NH4CI, 10 mM

EDTA, pH 7.2 to obtain thymocytes. Both spleen lymphocytes and thymocytes were then

depleted of remaining adherent cells by incubation in Petri dishes for 1 hr at 37 C.

Cell Proliferation Assay

Rodent lymphocytes. After repeated washing with lymphocyte medium, spleen

lymphocytes (2 x 105) or thymocytes (8 x 105) were added in a final volume of 200 il to

96-well microtitre plates. Lymphocytes were stimulated in triplicate with 0.25 or 2.5 tig/ml

ConA, 10 Rg/ml PHA, 0.5 or 5 pg/ml PWM, 10 units/ml IL-2, and 10 Ag/ml LPS for 72
hr. [3H]thymidine (lICi ) was added to each well for the last 18 hr, and the plates

harvested onto Whatman glass fiber filters with an automated cell harvester (Brandel Inc.,

Gaithersburg, MD). The filters were allowed to dry overnight and the cpm for individual

samples was determined on a Beckman LS7000 scintillation counter.

Human B Ivmphocvte cell lines. IM-9 and PJS-91 human B lymphocyte cell lines
were cultured overnight in medium lacking FBS to render them quiescent. On the day of

assay, the cells were washed with medium and 2 x 104 cells were added in triplicate to the
wells of 96-well microtitre plate sin a final volume of 200 p1. The plates were incubated

for 24 to 72 hr with a pulse of [3H]thymidine (llCi /well) for the last 4 hr of the culture

period, and harvested onto glass fiber filter strips with a cell harvester. The filters were

allowed to dry overnight and the cpm for individual samples was determined on a Beckman

LS7000 scintillation counter.








Measurement of Secreted IL-2


IL-2 present in spleen lymphocyte conditioned medium was measured with a bioassay

essentially as described (Coico et al., 1992). The assay used CTLL-2 cells, a mouse

cytotoxic T cell line that proliferates in response to T cell growth factor (Gillis et al., 1978)

and can thus be used to specifically assay IL-2. CTLL-2 cells were maintained in RPMI

1640 supplemented with 100 U/ml penicillin, 100 gg/ml streptomycin, 2 mM L-glutamine,

2 mM sodium pyruvate, 15 mM HEPES, 10% FBS, and 2 Units/ml recombinant human

IL-2. Cells in early-log phase were collected 3 days after feeding, washed twice, and

resuspended at lxl05/ml in the same medium without IL-2 (assay medium). Sterile-filtered

supernatants from 48 hr ConA-stimulated spleen lymphocytes were serially diluted (5-fold

dilutions starting with a 1:2 dilution) in assay medium. The samples and recombinant

human IL-2 standard were added together with lxl04cells in triplicate to the wells of a 96-

well microtitre plate. The plates were incubated for 24 hr, pulsed with 1 CCi

[3H]thymidine/well, and harvested 24 hr later. The amount of IL-2 activity in the spleen

lymphocyte culture supernatants was quantitated by converting the cpm values for each

dilution to the percentage maximal cpm relative to the recombinant human IL-2 standard.
The percentage maximal cpm were transformed to log values and plotted against the log5

dilution factor. Linear regression was used to determine the dilution factor yielding 50%

maximal cpm, and the inverse of this value was the amount of IL-2 Units/ml.

Measurement of Plasma GH. PRL. and IGF-

Duplicate plasma samples were assayed for PRL, GH, and IGF-1 by RIA. The RIAs

for PRL and GH were performed exactly as described previously (Singh et al., 1992a;

Singh et al., 1992b). Unknown values were derived from the 10 90% inhibition portion

of the standard curve and expressed as ng/ml in terms of the respective NIDDK reference

preparation (rat GH-RP-2 and rat PRL-RP-3).








The RIA for IGF-1 utilized the UB3-189 polyclonal antibody (NIDDK).

Characterization of the UB3-189 antibody has revealed that this preparation has minimal

crossreactivity with IGF-II and insulin. Since IGF-binding proteins interfere with

quantitative measurements of IGF in blood and tissue (Daughaday et al., 1987), the

acid/ethanol with cryoprecipitation procedure was used to strip IGF-I binding proteins

from plasma (Breier et al., 1991). Briefly, plasma was mixed with four volumes of 2N

HCl/ethanol (12.5%/85%) at room temperature for 30 min. The samples were centrifuged

and neutralized with Tris base prior to cryoprecipitation. Samples were centrifuged again

prior to dilution (1/70) with assay buffer. The efficiency of extraction was 54% using a

non-radioactive spike of 100 ng IGF-1 into the serum sample. Assay procedures were

followed as previously described (Daughaday et al., 1987; Breier et al., 1991) using (N-

Met) recombinant human IGF-I as the standard and label preparation. IGF-1 was

radioiodinated using chloramine-T and purified on a Sephadex G-50 column. The

sensitivity of the assay was 0.1 ng/ml with an ED50 of 1.24 ng/ml. The intra- and inter-

assay coefficients of variation were 5.8% and 9.4%, respectively.


ELISA for Secreted Human IgG and Rat Serum IgG

Ninety-six well ELISA plates (Nunc MaxisorpT) were coated with 100 pl1 of goat

anti-rat IgG or 2.5 tg/ml affinity purified goat anti-human IgG in bicarbonate coating

buffer, pH 9.6 and incubated overnight at 4 C. The wells were then washed 4 times

(between each step) with wash buffer (PBS pH 7.4, 0.05% Tween-20, 0.02% sodium

azide), followed by blocking with 5% BSA in wash buffer for 1 hr at room temperature.

After washing, culture supematants or serum samples and purified human or rat IgG were

added to the wells in triplicate and incubated for 2 hr at room temperature. The plates were
washed again and 100 tl of a 1:5000 dilution of Fc specific, affinity purified, alkaline

phosphatase-conjugated goat anti-human IgG or affinity-purified rabbit anti-rat IgG

horseradish peroxidase conjugate at 1:100,000 final dilution was added and again incubated








again 2 hr at room temperature. After washing, p-nitrophenyl phosphate or o-

phenylenediamine dihydrochloride was added and the absorbance at 405 nm and 492 nm,

respectively, was measured on an ELISA plate reader.


Sodium Dodecvl Sulfate Polvacrvlamide Gel Electrophoresis (SDS-PAGE) and Western
Immunoblotting


Conditioned medium from cell cultures was prepared by collecting, washing, and

replating cells in medium without FBS for the last 6-18 hr of the culture period. The

cultures were then centrifuged and the supernatants dialyzed extensively against 5 mM Tris-

HCI pH 8.0/0.02% sodium azide and then distilled water, while the cells were washed with

PBS and stored at -70 C until needed. After thawing on ice, the cells were lysed in NTEN

(50 mM Tris-HCI pH 8.0, 150 mM NaCI, 10 mM EDTA, 0.5% Nonidet P-40) containing

protease inhibitors (10 pg/ml aprotinin, 5 ig/ml leupeptin, and 1 mM PMSF). Whole cell

extracts were prepared by resuspension of cell pellets in IX SDS-PAGE sample buffer

(62.5 mM Tris-Cl, pH 6.8, 10% sucrose, 2% SDS, 5% 2-mercaptoethanol, .001%

bromophenol blue) and boiling for 5 min. Equal amounts of lyophilized protein,

determined by Bradford assay (Bradford, 1976), were subjected to SDS-PAGE (Laemmli,

1970), followed by electroblotting (Towbin et al., 1979) onto nitrocellulose or PVDF

membranes. For the immunostaining, the blots were blocked in TSAT (100 mM Tris-HCI

pH 8.0, 150 mM NaCI, 0.1% Tween-20, 0.02% sodium azide) containing 1% bovine

serum albumin (BSA) for 5 hr at 4 C for Protein A detection, or in TSAT containing 3%

gelatin for 1 hr at room temperature for colorimetric detection. The blots were then washed

in TSAT, and incubated with primary antibodies or normal serum from the respective host

species in TSAT-0.4% BSA overnight at 4 C, or in some experiments for 2 hr at room

temperature. Detection of antibody-protein complexes was carried out with 1 gCi

[1251]Protein A in TSAT-0.4% BSA for 5 hr at 4 C, followed by autoradiography, or goat








anti-rabbit IgG-horseradish peroxidase conjugate for 2 hr at room temperature and 3-

amino-9-ethylcarbazole as the substrate for colorimetric detection.

With Protein A detection, quantitation of individual bands on the blots was

accomplished by scanning on a Betascope 603 blot analyzer (Betagen Corp., Waltham,

MA). To account for differences in protein loading between lanes and to make the results

more reproducible between different blots, the values were normalized using a specific

counts index. This was calculated according to the formula:

[(specific counts/non-specific counts) -1] x 100

Specific counts refers to the total counts in the band of interest, while non-specific

counts is the total counts from a non-specific region of the same size in the same lane.


N-Terminal Sequence Analysis


For sequencing of the Mr 43,000 PRL-immunoreactive protein in rat spleen

lymphocytes, 200 ig of unstimulated and ConA-stimulated spleen lymphocyte lysate was

subjected to SDS-PAGE and electroblotted onto an Immobilon-P PVDF membrane

(Millipore Corp., Bedford, MA). Duplicate lanes of the membrane were Coomassie

stained or immunostained with anti-rPRL. The Mr 43,000 PRL-immunoreactive protein

was identified on the Coomassie-stained membrane by the increase in intensity in the

ConA-stimulated versus unstimulated lysate and alignment with the corresponding

immunostained band. N-terminal microsequencing was carried out by the University of

Florida ICBR Protein Chemistry Core Facility on an Applied Biosystems Model 473A Gas

Phase Protein Sequencer with an on-line HPLC system. Sequence data was analyzed

using Genbank software (GCG, University of Wisconsin, Madison, WI).


Metabolic Labeling and Immunoprecipitation

Exponentially-growing IM-9 cells were washed with serum-free medium and

cultured in methionine-deficient serum-free RPMI medium for 30 min at 37 C to allow









depletion of any residual intracellular methionine. Then 100 gCi/ml Tran[35S]LabelM was

added and the cultures continued for 3 hr. Cell lysates were prepared as described above

for SDS-PAGE. For immunoprecipitation, the cell lysates were diluted with NTEN and

precleared with normal rabbit serum + protein A-sepharose or protein A-Sepharose alone

for 1-2 hr at room temperature with constant agitation. Precipitation of specific proteins

was performed on an orbital shaker overnight at 4 C, followed by washing of the

precipitates twice with NTEN and a final wash with NTEN lacking NaCl. The precipitates

were boiled in SDS-PAGE sample buffer and analyzed by SDS-PAGE. Metabolic labeled

precipitates were visualized by autoradiography of the dried gel and unlabeled precipitates

were detected by immunoblotting.


Nb2 Lymphoma Cell Bioassay


Nb2 cells were maintained in RPMI 1640 medium supplemented with 100 pg/ml

streptomycin, 100 U/ml penicillin, 2 mM L-glutamine, 100 pM 2-mercaptoethanol, 10%

horse serum, and 10% FBS. For assay of PRL, mid-log phase Nb2 cells were collected,

washed, and resuspended in starvation medium (starvation medium is maintenance medium

without FBS) for 16-20 hr. On the day of assay, the cells were washed again with

starvation medium and 8x 104 cells in a final volume of 200 Wl were added to each well of a

96 well microtitre plate and stimulated in triplicate with rPRL or hPRL (NIDDK) in 50 pl

starvation medium. Other additions were also in 50 pl starvation medium. The cultures

were pulsed with I gCi/well [3H]thymidine for the last 4 hr of the 48 hr incubation and

harvested onto glass fiber filters using a cell harvester. The dried filters were quantitated

by scintillation counting.


Isolation of RNA


Total RNA was isolated by a modification of the single step acid guanidinium

thiocyanate-phenol-chloroform method as described (Xie and Rothblum, 1991). Poly(A)+








RNA was isolated by batch adsorption to oligo(dT)-cellulose exactly as described (Celano

et al., 1993).


Northern Blot Analysis


Analysis of rPRL mRNA. Total RNA was electrophoresed through 1.5% agarose-

formaldehyde gels, transferred to nitrocellulose membranes by capillary action, and fixed to

the membrane by baking at 80 C for 2 hr. Blots were prehybridized at 42 C overnight in

4X SSC (20X SSC is 3 M NaC1, 0.3 M Na citrate, pH 7.0), 50 mM KPO4, pH 6.5, 0.1%

SDS, 5X Denhart's solution (50X Denhart's is 1% each of Ficoll, BSA, and

polyvinylpyrrolidone), 250 tg/ml yeast RNA, and 50% formamide. Blots were hybridized

with a rPRL cDNA probe labeled with [a32P]dCTP by random priming using a Prime-a-

gene kit (Promega Corp., Madison, WI). Hybridization was performed in the same

solution as for prehybridization except containing IX Denhart's, at 42 C for 20 hr. Blots

were washed once at room temperature in 2X SSC, 0.1% SDS for 20 min, followed by

three 20 min washes in 0.1X SSC, 0.1% SDS at 42 C, 50 C, and 65 C. The blots were

then autoradiographed followed by quantitation on a Betascope blot analyzer. For

quantitation, PRL mRNA levels were normalized to p-actin levels in the same lane by

dividing each value for PRL net cpm (total cpm -background cpm) by the corresponding

values for p-actin net cpm and multiplying by 100.

Analysis of other mRNAs. Poly(A)+ RNA (4-10 tg) was electrophoresed through

1.2% agarose-formaldehyde gels, transferred to positively charged nylon membranes by

capillary action, and fixed to the membrane by UV crosslinking in a StratalinkerTM.
cDNA probes were radiolabeled with [a32P] dCTP by random priming using the Prime-It

II kit (Stratagene; La Jolla, CA). Hybrization was performed for 2 to 4 hr at 68 C with

ExpressHyb solution (Clontech; Palo Alta, CA). The blots were washed twice at room

temperature in 2X SSC, 0.1% SDS for 15 min, followed by two 30 min washes in 0.1X

SSC, 0.1% SDS at 65 C.








Preparation of Cvtosol and Nuclear Extracts


Cytosol was isolated from exponentially-growing IM-9 and PJS-91 or confluent

HepG2 cells by harvesting, and washing twice with ice-cold PBS. Washed cells were

resuspended in 3 pellet volumes of HEDG buffer (25 mM HEPES, pH 7.6, 3 mM EDTA,

1 mM DTT, 10% glycerol) and homogenized with 10-20 strokes in a Dounce homogenizer.

The homogenate was centrifuged at 100,000 x g for 1 hr at 4 C and the supernatant was

stored in small aliquots at -70 C. The cytosolic AhR complex was transformed in vitro by

incubation with 25 nM TCDD or 0.1% DMSO for 2 hr at 20 C before use in gel mobility

shift assays.

Cells were treated with TCDD or 0.1% DMSO (vehicle control) in serum free

medium for 2 hr at 37 C and crude nuclear extracts were prepared by a combination of

previously described methods (Saatcioglu et al., 1990; Ausbel et al., 1991). After

treatment, the cells were washed sequentially with ice-cold PBS and 10 mM HEPES, pH

7.6, resuspended in 3 volumes Buffer A (25 mM HEPES, pH 7.6, 1 mM DTT, 10 mM

KC1, 3 mM MgCI2) for 10 min on ice, and then homogenized with 10-20 strokes in a

Dounce homogenizer. The nuclear pellet (3,500 x g, 10 min) was washed once with

Buffer B (25 mM HEPES, pH 7.9, 0.5 mM EDTA, 0.5 mM DTT, 3 mM MgCI2, 25%

glycerol, 0.2 mM PMSF) containing 150 mM KCI, after which nuclear proteins were

extracted with Buffer B containing 500 mM KCI for 1 hr. The nuclear extract was

centrifuged at 12,000 x g and the supernatant dialyzed against 100 volumes Buffer C (20

mM HEPES, pH 7.9, 0.5 mM EDTA, 0.5 mM DTT, 20% glycerol, 150 mM KCI, 0.2

mM PMSF) for 1 hr. The dialyzed extract was clarified by centrifugation (12,000 x g, 10

min) and stored in aliquots at -70 C.

Gel Mobility Shift Assay

Complementary oligonucleotides were annealed, end-labeled with [y32P]dATP and

T4 polynucleotide kinase, and then purified on a 15% acrylamide gel. In some








experiments, the 8 base pair CYPIAl DRE primer was annealed to the complementary

strand and the 3' end filled in with the Klenow fragment of DNA polymerase I in the

presence of dATP, dTTP, dGTP, and [a32P]dCTP. Nuclear extract protein was incubated

in 15 mM HEPES, pH 7.9, 2 mM Tris-HC1, pH 7.5, 12% glycerol, 0.4 mM DTT, 0.4

mM EDTA, 80 mM KCI, 20 mM NaCI, and 0.5-1 ig poly (dI-dC):(dI-dC) for 15 min at

room temperature prior to addition of labeled probe (2-20 fmoles; 20-100,000 cpm), and

then incubated another 15 min. Unlabeled double-stranded competitor oligonucleotides

were added to the reaction mixture prior to the addition of labeled probe. In some

experiments, goat anti-mouse AhR (Poland et al., 1991) or rabbit anti-human ARNT

(Probst et al., 1993) antibody was added for 10 min at the end of the incubation with

labeled probe. The reactions were electrophoresed through non-denaturing 4% acrylamide

gels in IX TGE buffer (25 mM Tris, 380 mM glycine, 2 mM EDTA) to separate

protein:DNA complexes (Ausbel et al., 1991). Following electrophoresis, the gels were

dried down and exposed to X-ray film at -70 C with intensifying screens.


UV Cross-Linkine

The CYPIAI DRE probe used for UV cross-linking was prepared by annealing the

8 base pair CYPIAl DRE primer to the complementary strand, the 3' end filled in with the

Klenow fragment of DNA polymerase I in the presence of dATP, dGTP, 5-

bromodUTPand [a32P]dCTP, and gel-purified on a 15% acrylamide gel. Nuclear extract

protein (30-75 pg) was incubated in 15 mM HEPES, pH 7.9, 2 mM Tris-HC1, pH 7.5,

12% glycerol, 0.4 mM DTT, 0.4 mM EDTA, 80 mM KCI, 20 mM NaCI, and 5 ig poly

(dI-dC):(dI-dC) for 15 min at room temperature prior to addition of labeled probe (10-25

fmoles; 100-250,000 cpm), and then incubated another 15 min. For in situ UV cross-

linking, the binding reactions were electrophoresed through non-denaturing 4% acrylamide

gels in IX TGE buffer (25 mM Tris, 380 mM glycine, 2 mM EDTA), and the entire gel

irradiated by placing on a UV transilluminator emitting at 300 nm for 30 to 45 min at 4 C.








The gel was then exposed to film at 4 C overnight and the AhR-dependent band excised

from the gel by alignment with the resulting autoradiograph. Nuclear protein was eluted

from the gel slice in IX SDS-PAGE sample buffer (62.5 mM Tris-Cl, pH 6.8, 10%

sucrose, 2% SDS, 5% 2-mercaptoethanol, .001% bromophenol blue) containing 100 mM

NaCI overnight at room temperature, boiled for 5 min, and analyzed by SDS-PAGE on

7.5% gels. For UV cross-linking in solution, the binding reactions were irradiated on ice

at a distance of 5 cm from a UV transilluminator emitting at 300 nm for 30 to 45 nm. One

volume of 2X SDS-PAGE sample buffer was added, the samples were boiled for 5 min,

and analyzed by SDS-PAGE on 7.5% gels. In both cases, the gels were dried down and

exposed to X-ray film at -70 C.

Immunofluorescence Staining and Flow Cytometry Analysis


IM-9 and PJS-91 cells were collected by centrifugation and washed once with

diluent buffer (PBS ph 7.4, 0.1% sodium azide, 1% BSA). Cells (2 xl05) were aliquoted

to 12 x 75 mm sterile tubes and incubated with 1 gg specific antibody (mouse anti-human

CD19 or mouse IgGI isotype control) for 30 min on ice. The cells were washed three

times with diluent buffer, resuspended in 100 il of a 1:75 dilution of goat anti-mouse IgG-

FITC conjugate and incubated for 30 min on ice. The cells were again washed three times

with diluent buffer and resuspended in 0.5 ml diluent buffer. Two-color flow cytometric

analysis was performed at the University of Florida Flow Cytometry Core Facility on a

Becton Dickinson FacSortTM instrument using the LysislITM software. Dead cells were

stained with I pl of a 1 mg/ml solution of ethidium bromide immediately prior to analyzing

on the flow cytometer and subsequently excluded from data analysis. Data analysis was

performed using the Becton Dickinson PC-LYSYSTM software.








Ouantitation of Autoradiographs


Autoradiographs from Northern blot analyses and competition experiments in the

gel mobility shift assay were quantitated by densitometry. The autoradiographs were first

scanned on a desktop laser scanner. Care was taken to ensure that the intensity of the

images on the autoradiographs was within the linear range of the X-ray film. The

autoradiographic images were then digitized using NIH Image 1.52 software and the mean

density of the pixels present within the bands of interest were used for subsequent

calculations.


Statistical Analysis


In all cases where comparisons were made between control and cocaine-treated

groups or between only two treatment groups, the two-sample unpaired Student's t test

was used. Concentration-response data was analyzed using the Stat View 512+TM

analysis software by the one-way analysis of variance followed by Scheffe's F test if

significance was obtained in the analysis of variance.














CHAPTER 3
EFFECT OF TCDD ON LYMPHOCYTE PROLACTIN PRODUCTION AND
PROLACTIN-STIMULATED MITOGENESIS


Introduction

The environmental contaminant TCDD suppresses many aspects of immune

function in both in vivo and in vitro models (Holsapple, 1991). The immunosuppressive

actions of TCDD are diverse, with effects on cell-mediated, humoral, as well as innate

immunity. Additionally, the magnitude of immunosuppression varies with respect to the

components of immune function measured, the species tested, and the experimental

protocol used. Little information is available concerning the regulation of gene expression

by TCDD in lymphocytes or the molecular mechanism of immunotoxicity, but evidence

supports a role for the AhR in the immunotoxicity of TCDD (Holsapple, 1991; Kerkvliet

and Burleson, 1994).

The best described actions of the pituitary hormone PRL are on the mammary

gland, yet PRL is now accepted as also having an immunomodulatory role (Gala, 1991;

Murphy et al., 1995). Several laboratories have shown that mouse and human

lymphocytes produce PRL (Montgomery et al., 1990; Montgomery et al., 1992; Pellegrini

et al., 1992). In addition, proliferation was partially inhibited by addition of prolactin

antiserum to mitogen-stimulated lymphocyte cultures (Montgomery et al., 1987; Hartmann

et al., 1989; Sabharwal et al., 1992). PRL mRNA has also been detected in lymphocytes

(O'Neal et al., 1992; Pellegrini et al., 1992; Sabharwal et al., 1992). Lymphocytes from

mice, rats, and human express the PRL-R (O'Neal et al., 1991; Pellegrini et al., 1992;

Gagnerault et al., 1993; Viselli and Mastro, 1993; Touraine et al., 1994), a further

suggestion that PRL is likely to play a role in lymphocyte function. Thus, current evidence








indicates that lymphocytes express both components necessary for an autocrine and/or

paracrine network of PRL, which may act to promote lymphocyte growth and function.

Since there is evidence that PRL has an immunomodulatory role, factors which

regulate PRL production may therefore affect lymphocyte growth and/or function. Such

indirect mechanisms of immunomodulation involving neuroendocrine hormones are now

receiving attention, especially in relation to immunotoxicity assessment (Fuchs and

Sanders, 1994). Plasma PRL levels have been shown to be altered in rats and mice

exposed to TCDD in vivo (Jones et al., 1987; Russell et al., 1988; Moore et al., 1989; De

Krey et al., 1994), which could be responsible for some of the immunotoxic effects of

TCDD. In this Chapter, the production of PRL and proteins immunochemically related to

PRL was investigated in human B lymphocyte cell lines and rat spleen lymphocytes treated

with TCDD. Several higher molecular weight proteins were identified in cell lysates that

cross-reacted with PRL antiserum. There are also molecular weight variants of PRL in the

pituitary (Mena et al., 1992; Sinha, 1995), some with an apparent molecular weight similar

to those seen here in lymphocyte lysates.

The second part of this Chapter presents studies aimed at evaluating the effect of

TCDD on PRL-induced mitogenesis in Nb2 cells. The Nb2 cell line was established from

a lymph node tumor induced by estrogen in a male rat of the Noble strain (Gout et al.,

1980), and has been used extensively as both a bioassay for PRL and to define the

mechanism of PRL signal transduction. Nb2 cells are dependent on PRL for growth and

PRL stimulates mitogenesis in these cells which can easily be quantitated (Gout et al.,

1980; Tanaka et al., 1980). It provides a simple model for examining if, and how, a

compound may interfere with PRL action. PRL-stimulated ornithine decarboxylase activity

in spleen and thymus was impaired in TCDD-treated rats (Jones et al., 1987).

Furthermore, TCDD causes atrophy of the thymus in laboratory animals (Vos et al., 1973;

Vos and Moore, 1974; Holsapple, 1991; Kerkvliet and Burleson, 1994). The thymus is a

major site of T lymphocyte development, and since the Nb2 cell line is derived from a









lymph node tumor of T cell origin (Fleming et al., 1982) this cell line may also serve to

understand how T cells respond to PRL. Glucocorticoids also produce thymic atrophy,

and dexamethasone has been shown to alter the response of Nb2 cells to PRL (Fletcher-

Chiappini et al., 1993).




Identification of Proteins Cross-Reacting with PRL Antiserum in Human B lymphocvtes


IM-9 cells were evaluated for PRL production by SDS-PAGE and western blotting.

Fig. 3-1 A shows that several proteins in IM-9 conditioned medium and cell lysates were

recognized by a hPRL antiserum. The major cross-reacting proteins in cell lysates and

conditioned medium had apparent molecular weights of 40,000, and 27,000 and 54,000,

respectively. The cross-reactivity of these proteins was specific for the human PRL

antiserum, as no reactivity was observed when immunostained with normal rabbit serum

(data not shown). Purified pituitary hPRL migrated as a Mr 25-27,000 protein on SDS-

PAGE (Fig. 3-1 A). Although PRL is a 23 kDa protein, glycosylated variants of PRL

migrate on SDS-PAGE with slightly higher apparent molecular weight (Ellis and Picciano,

1995; Heffner et al., 1995). The Mr 27,000 protein present in IM-9 conditioned medium

migrated similar to, but slightly higher than purified human pituitary PRL; this protein (Mr

26-28,000) was considered to be PRL. The identity of the Mr 40,000 protein present in

IM-9 cell lysates and the Mr 54,000 secreted protein is unknown, although there are

structural variants of PRL in the pituitary (Sinha, 1995). These proteins that cross-reacted

with hPRL antiserum but with Mr values different from pituitary PRL will hereafter be

referred to as PRL-immunoreactive proteins. A similar set of PRL-immunoreactive

proteins was also detected in PJS-91 cells, a human B lymphocyte cell line generated by

EBV transformation of human peripheral blood lymphocytes (Fig. 3-1 B).

To determine if these PRL-immunoreactive proteins were synthesized in vitro, IM-9

cells were cultured with [35S] methionine to metabolic label newly synthesized protein.








Cell lysates containing [35S] labeled protein were then immunoprecipitated with hPRL

antiserum. In Fig. 3-2 A, several radiolabeled proteins were precipitated with hPRL

antiserum with Mr values of 68,000, 64,000, 58,000, 46,000, 42,000, 34,000, and

30,000. Only the Mr 42,000 protein was clearly specific for the hPRL antiserum, although

all bands were present at higher levels in anti-hPRL immunoprecipitates relative to normal

rabbit serum immunoprecipitates. In a similar experiment, immunoprecipitation was

performed with unlabeled IM-9 cell lysates followed by western blotting. Two PRL-

immunoreactive proteins of Mr 70,000 and 41.000 were specifically immunoprecipitated

by the hPRL antiserum (Fig. 3-2 B).

Thus, from these results, it can be concluded that in addition to secreted PRL, IM-9

cells also produced two intracellular PRL-immunoreactive proteins with Mr 68-70,000 and

Mr 40-42,000. Both of these proteins were synthesized in vitro and could be specifically

immunoprecipitated with a hPRL antiserum.


Effect of TCDD on Lymphocyte PRL-Immunoreactive Protein Production


The effect of TCDD on PRL and PRL-immunoreactive protein production by IM-9

cells was evaluated by western blot analysis. Fig. 3-3 shows PRL-immunoreactive protein

production by IM-9 cells treated with 30 nM TCDD for 72 hr. The levels of secreted PRL

(Mr 28,000), and the Mr 54,000 secreted and Mr 40,000 intracellular PRL-immunoreactive

protein were not altered by TCDD treatment. Also tested were several other agents known

to affect lymphocyte function: the synthetic glucocorticoid dexamethasone, and the

cytokines TGF-p and TNF-ac (Fig. 3-3). TNF-a treatment decreased the level of

expression of PRL (28K) and the secreted Mr 54,000 PRL-immunoreactive protein.

Dexamethasone and TGF-O1i had no effect on the level of any of these proteins in IM-9

cells.

The effect of TCDD on PRL and PRL-immunoreactive protein production was

further evaluated in concentration-response experiments in both IM-9 and PJS-91 cells.








TCDD (0.1-100 nM) did not alter the levels of secreted PRL (Mr 27,000) in either cell line

(Fig. 3-4 A). A Mr 107,000 secreted PRL-immunoreactive protein was detected in

conditioned medium that had not been previously identified, and TCDD did not affect the

level of this protein at any concentration (Fig. 3-4 A). Another PRL-immunoreactive

protein with Mr 66,000 was detected in conditioned medium in this experiment. A protein

of this size had been detected in cell lysates but had not previously been detected in

conditioned medium. The reason this protein went undetected may be due to co-migration

with a very large band in conditioned medium in earlier experiments, presumably residual

albumin left in the serum-free medium (see Fig. 3-1 and Fig. 3-3). TCDD treatment did

not alter the level of this protein in PJS-91 cells. In IM-9 cells, however, TCDD, > 1 nM,

increased the level of the secreted Mr 66,000 with a concomittant decrease in the level of

two slightly faster migrating PRL-immunoreactive bands. The level of the Mr 67,000 and

34-38,000 PRL-immunoreactive proteins in IM-9 and PJS-91 cell lysates was also

unaffected by TCDD treatment (Fig. 3-4 B).

Production of PRL and PRL-immunoreactive proteins was also evaluated by

western blotting in rat spleen lymphocytes. These cells produced a secreted Mr 25,000

(PRL) and an intracellular Mr 43,000 PRL-immunoreactive protein (Chapters 4 and 5).

Spleen lymphocytes isolated from pregnant rats administered TCDD in vivo did not show

any difference in production of secreted PRL relative to control animals. Likewise, TCDD

treatment in vtr of spleen lymphocytes from untreated rats did not affect production of

PRL or the intracellular Mr 43,000 PRL-immunoreactive protein (data not shown).


Effect of TCDD on PRL-Stimulated Proliferation in Lymphocytes


The ability of Nb2 cells to proliferate in response to PRL provides a simple, yet

reliable model for evaluating the effect of TCDD on the PRL signal transduction pathway in

lymphocytes. Fig. 3-5 shows a typical Nb2 assay measuring the proliferative response to

varying concentrations of PRL. Rat and human PRL were equally efficient at stimulating








proliferation of Nb2 cells, as the maximal stimulation (measured as thymidine

incorporation) for each hormone was similar. This preparation of hPRL was slightly more

potent than rPRL with EC50 values of 0.12 ng/ml nad 0.73 ng/ml, respectively. In the

absence of serum or PRL, Nb2 cells do not proliferate. TCDD did not significantly alter

Nb2 cell thymidine incorporation (Fig. 3-6, background) or cell viability (data not shown)

in the absence of PRL or FBS. TCDD also did not significantly affect cell proliferation

under optimal growth conditions; i.e. in the presence of FBS (Fig. 3-6). There was,

however, a slight concentration-related increase and decrease (15-20% at 100 nM TCDD)

in background and FBS-stimulated cell proliferation, respectively. The effect of TCDD on

Nb2 cell proliferation in response to increasing concentration of PRL is shown in Fig. 3-7.

These experiments used concentrations of PRL spanning several orders of magnitude to

ensure that any effect of TCDD, either enhancement or inhibition of the PRL response,

could be detected. TCDD (0.1-250 nM) did not have any effect on proliferation in

response to suboptimal or optimal concentrations of human PRL. In other experiments,

TCDD was added to the Nb2 cells at various times with respect to stimulation with PRL.

Addition of TCDD to Nb2 cells at 18 hr or 42 hr after stimulation with an optimal

concentration of PRL did not affect the response to PRL (Fig. 3-8). Further experiments

were performed treating Nb2 cells with TCDD during the period of starvation prior to the

assay (i.e., in the absence of PRL or serum). In these experiments, Nb2 cells were either

pretreated with TCDD, or TCDD was added at the same time or after stimulation with PRL.

Under these conditions, TCDD (1-100 nM) did not affect the proliferative response of Nb2

cells to a suboptimal (Fig. 3-9 A), or an optimal (Fig. 3-9 B) concentration of PRL. In

some experiments, dexamethasone was used as a positive control for inhibition of Nb2 cell

proliferation since it has been shown to inhibit both background and PRL-stimulated

proliferation of Nb2 cells (Fletcher-Chiappini et al., 1993). As expected, dexamethasone

decreased proliferation in response to a suboptimal concentration of PRL, but in the

absence of PRL (pretreatment), cell viability was greatly reduced (data not shown).








The effect of TCDD on the response of rat spleen lymphocytes to PRL was

evaluated in experiments using lymphocytes from animals that had been administered

TCDD in vivo or lymphocytes treated with TCDD in vitro. No effect of PRL on

proliferation of spleen lymphocytes from mice or rats could be reproducibly detected, either

when added alone, or in combination with the mitogens PWM or LPS. However, TCDD

treatment in vitro did suppress background and LPS-stimulated proliferation in murine

spleen lymphocytes, while PWM-stimulated proliferation was unaltered by TCDD (Fig. 3-

10). TCDD was equally effective in suppressing the proliferative response to LPS when

added before, at the same time, or after LPS. Background thymidine incorporation was

suppressed when TCDD was added at the time of assay or 6 hr after, but not when the cells

were pretreated for 18 hr with TCDD (Fig. 3-10). In this experiment, PRL did not affect

control spleen lymphocyte proliferation, nor was it able to restore proliferation in TCDD-

treated cells to normal levels (data not shown).




Although PRL is classically considered to be an anterior pituitary hormone, many

tissues have been found to produce PRL indistinguishable from pituitary PRL (Sinha,

1995). In addition to the 23K PRL form, several other forms of PRL can be detected in

biological samples from various species (Sinha, 1995). PRL variants of 16 K, 21 K, 25

K, 45 K, and >60 K have been identified, and it is likely that at least some of these variants

result from post-translational modifications.

In these studies, several PRL-immunoreactive proteins produced by the IM-9 and

PJS-91 human B lymphocyte cell lines were detected using western blot analysis. In cell

lysates, the PRL-immunoreactive proteins detected had apparent molecular weights of 40-

42,000, 34-38,000, and 66-68,000. The secreted PRL-immunoreactive proteins detected

had apparent molecular weights of 26-28,000, 54,000, 66,000, and 107,000. There are

published reports of proteins of similar molecular weight detected with PRL antiserum. In








rat pituitary homogenates, an Mr 43,000 protein was detected by western blotting (Mena et

al., 1992), and a Mr 41,000 PRL-immunoreactive protein was detected in human pituitary

homogenates (Meuris et al., 1983). Ellis metal. (1995) identified a non-glycosylated PRL-

immunoreactive protein with Mr 42,000, as well as Mr 28,000 and 32,000 glycosylated

PRL-immunoreactive proteins in human milk. Mouse thymocytes expressed a Mr 30,000

PRL-immunoreactive protein (Montgomery et al., 1990), and a Mr 45-46,000 PRL-

immunoreactive protein was identified in ConA-stimulated murine spleen lymphocytes

(Kenner et al., 1990; Shah et al., 1991). Human lymphocytes produced Mr 24,000

(Montgomery et al., 1992), Mr 27,000 (Montgomery et al., 1992), and Mr 60,000 PRL-

immunoreactive proteins (Sabharwal et al., 1992). AG-876 and Ramos, two Burkitts

lymphoma (human B lymphocyte) cell lines, expressed Mr 60,000 and 29,000 PRL-

immunoreactive proteins, respectively (Baglia et al., 1991; Sabharwal et al., 1992). PRL

is a 23 kDa protein and it is likely that the Mr 21-27,000 PRL-immunoreactive proteins in

lymphocytes identified by other laboratories (Montgomery et al., 1990; Baglia et al., 1991;

Montgomery et al., 1992; Pellegrini et al., 1992) is identical to pituitary PRL. In contrast,

there is no convincing evidence that any of the higher molecular weight PRL-

immunoreactive proteins in pituitary extracts or lymphocytes are structurally related to

PRL. It has been suggested that some of these forms result from oligomerization of PRL,

however, in some instances, reducing and denaturing agents did not affect the molecular

weight as determined by SDS-PAGE, indicating that if these variants are PRL oligomers,

then they are formed by covalent interactions. Indeed, it has been proposed that

glycosylated PRL may form covalent dimers or that PRL may become covalently linked to

binding proteins (Sinha, 1992). In line with this hypothesis, it was recently demonstrated

that Mr 75,000 and 98,000 PRL-immunoreactive proteins in human serum were covalent

complexes between PRL and the immunoglobulin G heavy chain (Walker et al., 1995).

In these studies, TCDD did not affect the production of PRL or PRL-

immunoreactive proteins by IM-9 or PJS-91 human B lymphocyte cell lines. In the TCDD








concentration-response experiment (Fig. 3-4), a Mr 66,000 PRL-immunoreactive protein in

IM-9 conditioned medium was increased by TCDD. However, this parallelled a decrease

in two closely migrating lower Mr bands. The faster migrating immunoreactive bands may

be proteolytic fragments, or dephosphorylated or deglycosylated forms of the Mr 66,000

protein since no similar bands were detected in PJS-91 conditioned medium or in either IM-

9 or PJS-91 cell lysates. Whether this effect of TCDD is due to an alteration in post-

translational modification (e.g. phosphatase activity or proteolytic processing) or

experimental artifact (sample preparation) is not known. These cell lines do express the

AhR complex capable of binding to DNA (Chapter 6), so it is unlikely the lack of effect of

TCDD on PRL production is due to an inability of these cells to respond to TCDD.

Other compounds have been shown to alter PRL production in the IM-9 cell line. It

was reported that in IM-9 cells, phorbol ester decreased PRL mRNA levels but increased

the secretion of PRL (Gellersen et al., 1989). Retinoic acid also increased PRL production

in IM-9 cells and this was demonstrated to occur at the post-transcriptional level (Gellersen

et al., 1992). On the other hand, of several regulatory factors examined, dexamethasone

was the only factor known to influence pituitary PRL production that also affected PRL

production in IM-9 cells (Gellersen et al., 1988). In the experiments presented here,

however, dexamethasone did not affect production of any of the PRL-immunoreactive

proteins. Clearly, PRL gene expression in IM-9 cells, and most likely in normal

lymphocytes, is under different regulatory control than PRL gene expression in the

pituitary. This is not surprising since the PRL mRNA 5' untranslated region in IM-9 cells

is unique from the pituitary (DiMattia et al., 1990). This difference was found to result

from a non-coding exon upstream of exon 1 that is not transcribed in the pituitary (DiMattia

et al., 1990). To this end, a lymphoid specific enhancer element has been identified in the

hPRL gene, although factors binding to this DNA region have yet to be identified (Berwaer

et al., 1994). TGF-pl1 and 32 was reported to decrease PRL mRNA levels in rat GH3

pituitary cells (Delidow et al., 1991), and this cytokine has important immunoregulatory








properties. However, in the studies presented here, there was no effect of TGF-p 1 on

PRL or PRL-immunoreactive protein production in IM-9 cells.

PRL-stimulated ornithine decarboxylase activity in the thymus and spleen was

reported to be inhibited in TCDD-t, d animals (Jones et al., 1987), although these

authors did not determine which cell types in these organs were responding to PRL. PRL

induces the transcription of omithine decarboxylase in Nb2 lymphoma cells which may be

important for cell growth (Yu-Lee. 1990). In the present study, under several different

experimental conditions, TCDD did not affect the ability of Nb2 cells to proliferate in

response to PRL. It is possible that the PRL signal transduction pathway is subtly different

in Nb2 cells compared to normal tissues, which could account for the exquisite sensitivity

of these cells to PRL and the lack of effect of TCDD on Nb2 cells. Indeed, Nb2 cells have

a mutant form of the PRL-R with an increased affinity for PRL, designated the intermediate

form because of the size relative to me previously identified alternatively spliced long and

short forms of the PRL-R (Ali et al., 1991).

TCDD also had no effect on Nb2 cell viability or cell proliferation under optimal

growth conditions. The severe thymic atrophy in TCDD-treated mice and rats is

characterized by a depletion of cortical thymocytes (Vos and Moore, 1974), although the

mechanism for this effect is unknown. Evidence suggests that one possibility for the

TCDD-induced thymic atrophy was an inability of immature T cells in the thymus of

TCDD-treated animals to differentiate (Blaylock et al., 1992; Kerkvliet and Burleson,

1994). The Nb2 cell line has previously been characterized as having an immature

CD4+CD8+ T cell phenotype (Fleming et al., 1982), suggesting that Nb2 cells might be a

good model for examining the effects of TCDD on immature T cells. Other T-cell

dependent responses are suppressed by TCDD in animal models (Vos and Moore, 1974;

Kerkvliet and Burleson, 1994), however, in some cases these results have been difficult to

reproduce using in vitro systems (Kerkvliet and Burleson, 1994). With this in mind, it is

not as surprising that TCDD had no effect on Nb2 cell growth or viability.






45

In summary, two human B lymphocyte cell lines, IM-9 and PJS-91, produced

several proteins in addition to PRL that cross-reacted with hPRL antiserum. Proteins

cross-reacting wth hPRL antiserum that have a similar molecular weight as those reported

here have been documented, however, an exact identification of these proteins has not been

made. There is still no solid evidence these proteins are structural variants of PRL. TCDD

treatment of IM-9 and PJS-91 cells did not qualitatively or quantitatively alter the

production of PRL or PRL-immunoreactive proteins in these cell lines. TCDD likewise did

not affect the ability of the PRL-dependent Nb2 lymphoma cell line to proliferate in

response to PRL. From these studies, it can be concluded that neither lymphocyte

production of PRL nor the PRL signalling pathway in lymphocytes appears to be a target

for TCDD.














A B
CM L hPRL CM L




80K 80 K




49K 49 K

r*%


33K 33K K

28^ -- ^ ^_ ^ .^
28K K _
V 28K




19K I 19K






Figure 3-1. Analysis of PRL and PRL-immunoreactive proteins in IM-9 and PJS-91
human B lymphocyte cell lines. (A) Left, PRL and PRL-immunoreactive protein
production by IM-9 cells. Right, 50 ng purified human pituitary PRL (NIDDK). (B) PRL
and PRL-immunoreactive protein production by PJS-91 cells. IM-9 and PJS-91 cells were
cultured at 5 x 105/ml for 24 and 48 hr, respectively, washed, and then related at 1 x
106/ml for an additional 4 hr in serum-free medium. Conditioned medium (CM) or whole
cell extracts (L) from 1 x 106 cells were analyzed by SDS-PAGE and western blotting.
The blots were stained with anti-hPRL-P4802-1 at a final dilution of 1:600. The major
PRL-immunoreactive proteins are denoted by arrows. The position of molecular weight
size standards are indicated on the left.
















A B
NRS antl-hPRL NRS anti-hPRL




106 6K106 K
80 K


.4- 49 K
49 K

33 K


33 K


19K
28K








Figure 3-2. Immunoprecipitation of PRL-immunoreactive proteins from IM-9 cell lysates.
(A) Autoradiograph of [35S]-labeled proteins in IM-9 cell lysates immunoprecipitated by
anti-hPRL-P4802-1. The autoradiograph shown is an overexposure to reveal the fainter
immunoprecipitated bands. Each lane contains the normal rabbit serum (NRS) or anti
hPRL immunoprecipitate from NP-40 cell lysate equivalent to 1 x 107 cells. (B)
Immunoblot of proteins in IM-9 cell lysates immunoprecipitated by anti-hPRL-P4802-1.
IM-9 cells were lysed by three rounds of freeze-thawing in hypotonic lysis buffer (10 mM
Tris-HC1, pH 7.4, 10 mM NaCI, 1.5 mM MgC12, 1 mM EDTA) containing protease
inhibitors. Each lane contains the immunoprecipitate of freeze-thaw lysate equivalent to 3 x
107 cells. The large band migrating at approximately 50 K is IgG heavy chains from the
hPRL antiserum that are recognized by the secondary antibody used in western blotting.
The weaker band migrating at approximately 25 K is most likely IgG light chains resulting
from a similar phenomenon. The major PRL-immunoreactive proteins are denoted by
arrows. The position of molecular weight size standards are indicated on the left.














CONTROL TCDD DEX TGF-pl TNF-a
CM L CM L CM L CM L CM L


Fig. 3-3. Effect ofTCDD, dexamethasone, TGF-pl, and TNF-a on PRL-immunoreactive
protein production by IM-9 cells. IM-9 cells were cultured at a density of 5 x 105/ml with
30 nM TCDD, 100 nM dexamethasone (DEX), 1 ng/ml TGF-pl, or 100 pg/ml TNF-a for
72 hr, then washed and recultured another 24 hr in serum-free medium. Conditioned
medium (CM) or cell lysates (L), 100 gg each lane, were analyzed by SDS-PAGE and
immunoblotting. Anti-hPRL-P4802-1 was used at a final dilution of 1:800. The major
PRL-immunoreactive proteins are denoted by arrows. The position of molecular weight
size standards are indicated on the left.


-t ^ p 40r r


K -- -r 4-















cU -s




45 s
0E0oo
0 0






X-


ii






a- c $.
S)i0
















2 x -a ^.

^ijgs
r ^ C~i'-




















S1 irE I

u. i tl it .
I leUI UsI*U 3t



I II
s 1 i





















100-





. 50-





0-


P-...


S0'-----



d ----0--

---0--


- rPRL

- hPRL


EC50 (ng/ml)

0.73

0.12


0.01 0.1 1 10 100

[PRL]-ng/ml


Figure 3-5. Concentration-response curves for the proliferative response of Nb2 cells to
rat and human PRL. The Nb2 assay was performed as described in Materials and
Methods. The values are the mean SE of triplicate wells from a representative
experiment. The data are expressed as the percentage of stimulation at each concentration
relative to the maximal stimulation observed with each hormone. The background values
were 684 46 cpm, and the maximal response to rPRL and hPRL was 190,367 1,931
and 184,669 3,431 cpm, respectively. EC50 values were calculated by linear regression
analysis of the linear portion of each curve.
















O background

7 FBS
T
T T L
100- -- ------------- ----
I T
1 1.


S 50-






-- 10 9 8 7

-(Iog[TCDD])


Figure 3-6. Concentration-response for the effect of TCDD on background and serum-
stimulated proliferation of Nb2 cells. Background (no FBS or PRL) or FBS-stimulated
proliferation of Nb2 cells were measured in a Nb2 assay. The values are the mean SE of
triplicate wells from a representative experiment. The data are expressed as the percentage
of control response (i.e. in the absence of TCDD). The control values for background and
FBS were 1,010 + 15 and 190,822 20,740 cpm, respectively.

















3 control

010 .1 nM 0 TCDD
10 0- .. .- ......i.n .-..T.C.DD ............ ----- ------ ...

0 I nM TCDD
-r 10 nhl TCDD

100 nM TCDD

250 nMi TCDD
50








0.01 0.1 1 10

[hPRL]-ng/ml


Figure 3-7. Concentration-response for the effect of TCDD on the proliferative response of
Nb2 cells to hPRL. TCDD and hPRL were added concurrently to Nb2 cells at the time of
assay. The values are the mean SE net cpm of triplicate wells from a representative
experiment. The data are expressed as the percent of stimulation relative to the maximal
stimulation observed with hPRL. The maximal response to hPRL was 144,963 2,837
cpm.




















100- .-. ..... ""0:



E


50-
-- --- Ohr
E
P .........-... 18 hr
---.---- 42 hr

0 1 I
-- 10 9 8 7

-(log[TCDD])


Figure 3-8. Concentration-response of the effect of TCDD added at different times on Nb2
cell proliferation. TCDD was added to Nb2 cells either concurrently with, 18 hr after, or
42 hr after stimulation with hPRL. The values are the mean SE net cpm of triplicate
wells from a representative experiment and are calculated as the percentage of stimulation
relative to the maximal stimulation observed with hPRL. The values shown are the effect
of TCDD on the proliferative response to an optimal concentration of hPRL (1 ng/ml) which
was 93% of the maximal cpm. The maximal response to hPRL in untreated cells was
92,319 6,924 cpm.





55




A
10 nM TCDD
S100 nM TCDD
S 100- --------------- ---- -- T


S 50



pretreatment only pretreatment + assay assay only

B
150-


100 -------- ---.
100


C 50

0
pretreatment only pretreatment + assay assay only
time of exposure to TCDD


Figure 3-9. Effect of TCDD pretreatment on Nb2 cell proliferation in response to
suboptimal and optimal concentrations of hPRL. TCDD was added to Nb2 cells either for
24 hr prior to the assay and washed out before the assay (pretreatment only), for 24 hr
prior to the assay and during the assay (pretreatment + assay), or only during the time of
assay (assay only). The values are the mean SE of triplicate wells from a representative
experiment. The data are expressed as the percentage of control response (i.e. in the
absence of TCDD). (A). Proliferative response to a suboptimal concentration of hPRL
(0.05 ng/ml). The control response was 6,908 182 cpm. (B). Proliferative response to
an optimal concentration of hPRL (0.5 ng/ml). The control response was 59,816 1,053
cpm.















V"u- [j 18 hr before

O same time

150- T 6 hr after





50
I

100 ----------------








background PWM LPS

mitogen

Figure 3-10. Effect of TCDD on the proliferative response of murine spleen lymphocytes.
Spleen lymphocytes isolated from pregnant mice were evaluated for background, and
PWM- and LPS-stimulated proliferation with a [3H]thymidine incorporation assay. TCDD
(30 nM) was added 18 hr before, at the same time, or 6 hr after stimulation. The values are
the mean SE of triplicate wells from a representative experiment calculated as the
percentage of control response (i.e. in the absence of TCDD). The control response was
6,017 1,123, 39,106 3,869. and 90,043 2,296 cpm for background, PWM- and
LPS-stimulated proliferation, respectively.














CHAPTER 4
EVALUATION OF IMMUNE PARAMETERS AND LYMPHOCYTE PROLACTIN-
IMMUNOREACTIVE PROTEIN PRODUCTION AFTER CHRONIC
ADMINISTRATION OF COCAINE TO PREGNANT RATS


Introduction


Cocaine is a widely abused illicit drug with an estimated 5-6 million regular users in

North America (Das, 1993). The adverse effects of cocaine are well documented and

include central nervous system and cardiovascular toxicity (Loper, 1989). Recent evidence

showing that cocaine use is associated with a variety of infections (Richter, 1993), as well

as HIV seropositivity (Anthony et al., 1991) has lead to renewed interest in the possible

immunotoxicity of cocaine. In cocaine users, alterations in peripheral blood T cell

subpopulations have been observed (Ruiz et al., 1994), and proliferation of human

lymphocytes was altered by direct exposure to cocaine in vito (Klein et al., 1993; Phillips

et al., 1995).

Animal studies have shown that cocaine affects different aspects of immune

function after in vivo administration. In mice, cocaine treatment decreased lymphoid organ

weights and the number of lymphocytes recovered from these organs (Faith and Valentine,

1983; Ou et al., 1989; Di Francesco et al., 1992; Pirozhkov et al., 1992), natural killer cell

and cytotoxic T cell activity, and splenocyte cytokine production (Di Francesco et al., 1992;

Pirozhkov et al., 1992). Primary antibody responses of splenocytes from cocaine-treated

rats (Bagasra and Forman, 1989) and mice (Faith and Valentine, 1983; Holsapple and

Munson, 1985; Havas et al., 1987; Ou et al., 1989) were increased or decreased depending

on the antigen used for immunization and the dosage regimen. In addition, cocaine directly








modulated proliferation of mouse (Klein et al., 1988) and rat (Berkeley et al., 1994) spleen

lymphocytes in culture.

Cocaine use during pregnancy has been associated with developmental toxicity

including preterm delivery (Handler et al., 1991), abruptio placentae and stillbirth (Bingol

et al., 1987), and low infant birth weight (Bingol et al., 1987; Handler et al., 1991;

Chasnoff et al., 1992). Pregnancy is accompanied by specific changes in the the lymphoid

organs of mice and rats (McLean et al., 1974; Medina et al., 1993; Clark et al., 1994),

which prompted the question of whether the pregnant animal may be more susceptible to

cocaine-induced immunotoxicity, and thereby allow the transmission of infectious diseases

to the developing fetus. In this regard, no study to date has examined the effects of cocaine

exposure on maternal immune function during pregnancy. Although the

immunomodulatory effects of in utero cocaine exposure in humans are unknown, Sobrian

et al. (1990) reported that young rats exposed to cocaine in utero had significantly altered

spleen and thymus body weight ratios.

Previous work in this laboratory established a model of chronic cocaine

administration to pregnant rats which results in significant fetal growth retardation with

minimal maternal toxicity (Salhab et al., 1994). While cocaine impaired fetal growth, no

changes were observed in placental growth or the synthesis of placental prolactin-like

proteins. The purpose of this work was to examine the effects of this cocaine dosing

regimen on the immune system of pregnant rats, including lymphoid organ weights and

lymphocyte proliferative and secretary responses in vitro. Plasma levels of prolactin (PRL)

and growth hormone (GH) were measured since both hormones have major roles in

maintaining immunocompetence (Gala, 1991), and cocaine administration alters the serum

levels of several neuroendocrine hormones in rats (Van de Kar, 1992). An additional

immune parameter studied was lymphocyte production of prolactin-immunoreactive

proteins based upon recent evidence that local autocrine/paracrine production of these

proteins by lymphocytes may have an effector role in the immune system (Hartmann et al.,








1989: Kenner et al., 1990; Montgomery et al., 1990; Shah et al., 1991). Evaluation of the

effects of cocaine on immune function in pregnant animals and women makes a significant

and relevant contribution to our understanding of the effects of cocaine on the immune

system and the relationship between pregnancy and susceptibility to immunotoxicity.


Results


Effect of Cocaine Treatment on Body and Lymphoid Organ Weights


A previous study in this laboratory showed that administration of 30 mg cocaine/kg

i.p. to pregnant rats produced a peak plasma level of 1.09 0.17 gg/ml cocaine 1 hr after

injection (Salhab et al., 1994). This blood level in rats is similar to that reported in human

saliva and plasma after i.v. administration (Cone et al., 1988). The previous study further

demonstrated that administration of 30 mg/kg twice daily from gestation days 8 to 19

significantly decreased fetal body weight without any change in placental growth (Salhab et

al., 1994). It would be difficult to separate direct effects of cocaine on the immune system

from immunomodulation as a result of inadequate diet and/or malnutrition. Therefore, food

intake was monitored closely and there was no significant difference in food consumption

between the control and cocaine-treated animals (data not shown). The body and lymphoid

organ weights and lymphocyte yields from control and cocaine-treated animals are shown

in Table 4-1. No significant differences were observed in maternal body weight, nor in

relative spleen and thymus weights between control and cocaine-treated animals. In

addition, the total number of lymphocytes isolated from spleen and thymus were

comparable between the two groups, and lymphocyte viability was >97% in all cases.

Since pregnant animals gain a considerable amount of weight during gestation,

body and lymphoid organ weights were also measured in non-pregnant rats treated with the

same dosing regimen of cocaine to determine whether cocaine treatment had any effect on








these parameters in the non-pregnant animal. As seen in the pregnant animals, body weight

and relative spleen and thymus weights were not altered by cocaine treatment (Table 4-2).


Plasma Levels of leG. PRL. GH. and IGF-1


Fig. 4-1 shows the maternal plasma levels of IgG, PRL, GH, and IGF-1 from

control and cocaine-treated animals. In the cocaine-treated dams, plasma levels of IgG

were 46% higher than in control animals (P < .05). In contrast, the plasma levels of PRL,

GH, and IGF-1 were not significantly different between the two treatment groups. These

values for PRL, GH, and IGF-1 are in the range previously reported for late gestation

pregnant rats (Carlsson et al., 1990; Davenport et al., 1990; Grattan and Averill, 1990).

The substantial variability in maternal plasma PRL levels most likely reflects the surge of

PRL secretion prior to parturition. which is in contrast to the constant levels of GH and

IGF-1.


Mitogen-Stimulated Proliferative Responses of Spleen Lymphocytes and Thymocytes


Spleen lymphocytes and thymocytes isolated from control and cocaine treated

animals were evaluated for proliferative responses in vitro to a panel of T cell mitogens

(ConA, PHA, and IL-2), the mixed B and T cell mitogen PWM, and the B cell mitogen

LPS. Basal [3H]-thymidine incorporation in unstimulated spleen lymphocytes from

control and cocaine-treated groups was not significantly different, 4,644 639 cpm and

5,616 726 cpm, respectively. Data in Fig. 4-2 A, expressed as net cpm (total cpm basal

cpm) show that stimulation with the individual mitogens increased spleen lymphocyte
[3H]-thymidine incorporation 4-40 fold, with ConA being the most potent. The levels of

thymidine incorporation by spleen lymphocytes from control and cocaine-treated groups

were comparable in the presence of both suboptimal and optimal concentrations of ConA

and PWM, as well as PHA, LPS, and IL-2 (Fig. 4-2 A). It should be noted that spleen








lymphocytes from cocaine-treated animals consistently showed a greater response to ConA

and PHA, although these differences were not significant.

The proliferative response of thymocytes was examined only after stimulation with

ConA, PWM, and IL-2 since thymocytes respond poorly to PHA and not at all to LPS

(Fig. 4-2 B). Basal thymidine incorporation by thymocytes from control and cocaine

treated animals was similar, 384 45 cpm and 239 53 cpm respectively, whereas

mitogens stimulated incorporation 50-300 fold, again with ConA being the most potent

(Fig. 4-2 B). As noted with spleen lymphocytes, thymocytes from control and cocaine

treated animals showed comparable levels of thymidine incorporation after stimulation with

the individual mitogens.


IL-2 Production by Spleen Lymphocytes


Spleen lymphocytes were stimulated for 48 hr with ConA, and the supernatants

were assayed for IL-2 activity using the CTLL-2 bioassay. The mean IL-2 production for

spleen lymphocytes from control animals (10.5 1.5 U/ml) was not significantly different

from the cocaine-treated group (13.3 2.0 U/ml) (Fig. 4-3).


Lymphocyte Production of PRL-Immunoreactive Proteins


Spleen lymphocytes were cultured in the presence of ConA for 48 hr, after which

conditioned medium and cell lysates were analyzed by SDS-PAGE and western

immunoblotting. An antiserum to rPRL specifically cross-reacted with intracellular Mr

31,000 and 44,000 proteins (Fig. 4-4 A), as well as a secreted Mr 44,000 protein (Fig. 4-4

B). The Mr 44,000 PRL-immunoreactive protein was expressed at very low levels in

unstimulated spleen lymphocytes, but was strongly induced in the presence of ConA

(Chapter 5). Quantitation of these proteins on immunoblots showed that spleen

lymphocytes from cocaine-treated animals had significantly decreased levels of the PRL-

immunoreactive proteins relative to control animals (Fig. 4-4 C). In spleen lymphocytes








from cocaine-treated animals, the level of the ConA-stimulated Mr 44,000 cross-reactive

protein was significantly decreased by 30% in cell lysates (P <.05) compared with spleen

lymphocytes from control animals (Fig. 4-4 A and 4-4 C). Similarly, the Mr 44,000

secreted immunoreactive protein was significantly decreased by 35% in conditioned

medium (P < .01) from spleen lymphocytes from cocaine-treated vs. control animals (Fig.

4-4 B and 4-4 C). The major Mr 31,000 intracellular protein was also decreased by 36%,

though not significantly, in spleen lymphocyte lysates from the cocaine-treated group (Fig.

4-4 A and 4-4 C).

It warrants note that in some experiments we observed secretion of a Mr 24,000

PRL-immunoreactive protein which is similar in size to pituitary prolactin (data not

shown). When expressed, the relative level of the Mr 24,000 cross-reactive protein

following stimulation of spleen lymphocytes with ConA, PWM, and PHA was decreased

by 69-89% in conditioned media from spleen lymphocytes from cocaine-treated animals.

Lastly, although ConA-stimulated thymocytes express a similar set of PRL-immunoreactive

proteins (data not shown), the level of protein expression was lower and highly variable,

sometimes being completely absent in thymocytes from individual animals. These findings

made it difficult to quantitate any alterations associated with in vivo cocaine exposure.


Northern Blot Analysis of PRL mRNA Expression


Northern blots of total RNA isolated from ConA-stimulated spleen lymphocytes

were hybridized sequentially with rat PRL and human 3-actin cDNA probes. The major

rPRL and P-actin RNA transcripts were 2.2 kb and 1.9 kb, respectively (Fig. 4-5). The

level of rPRL transcripts was not significantly different between spleen lymphocytes from

control and cocaine-treated animals, with mean PRL/actin ratios for the control and cocaine-

treated groups of 2.32 and 2.19, respectively. In a single sample of thymocytes from

control animals, total RNA showed a similar size PRL RNA transcript (Fig. 4-6).








Discussion

A more thorough understanding of the effects of cocaine on the immune system is

important since it is a widely abused drug and a significant percentage of the population is

exposed; furthermore, there have been a number of epidemiological studies revealing an

association between cocaine use and an increased risk of HIV seropositivity (Chaisson et

al., 1989; Anthony et al., 1991). These studies were designed to characterize the

immunological effects, as well as lymphocyte PRL production after administration of a

fetotoxic dose of cocaine (Salhab et al., 1994) to pregnant rats during gestation.

The studies presented here are unique in that to our knowledge, there are no

published reports investigating the effects of cocaine on the immune system of any

pregnant animal. Furthermore, these studies complement previous work in this laboratory

on the effects of chronic cocaine exposure on pregnancy in an animal model. The dosing

regimen used in these studies was designed to mimic the pattern of cocaine use seen in drug

users. A dose that results in plasma levels approximating that found in cocaine users (Cone

et al., 1988) was administered to the pregnant rats twice a day. The dosing was continued

through mid and late gestation as a model of chronic drug use by pregnant women. In

addition, this dosing schedule has been shown to retard rat fetal growth, without adverse

effects on placental growth (Salhab et al., 1994). In this regard, cocaine exposure during

pregnancy has been clearly associated with multiple developmental complications including

premature delivery and low birth weight in humans (Bingol et al., 1987; Handler et al.,

1991; Chasnoff et al., 1992), and intrauterine growth retardation and low birth weight in

animals (Dow-Edwards, 1991: Salhab et al., 1994). Evidence of an association between

cocaine use and infectious diseases in humans (Chaisson et al., 1989; Anthony et al., 1991;

Richter, 1993), and immunotoxicity after cocaine exposure in animals (Bagasra and

Forman, 1989; Ou et al., 1989; Sobrian et al., 1990; Di Francesco et al., 1992; Pirozhkov

et al., 1992) has raised particular concerns regarding pregnancy wherein suppression of the








maternal immune system may increase susceptibility of transmission of infectious diseases

transplacentally.

Recent evidence indicates that the hormones PRL, GH, and IGF-1 are important

regulatory factors for the immune system (Gala, 1991; Kelley et al., 1992), and cocaine

administration has been associated with significant alterations in the levels of PRL as well

as other neuroendocrine hormones (Mendelson et al., 1989; Van de Kar, 1992). In the

present study, pregnant rats chronically treated with cocaine showed no significant

differences in maternal plasma concentrations of PRL, GH, and IGF-1 (Fig. 4-1). This

finding was not expected since alterations in serum prolactin after cocaine exposure have

been described in the rat (Van de Kar, 1992); however, prolactin levels peak just prior to

birth in the rat (Grattan and Averill, 1990), which may have diminished any effect resulting

from cocaine exposure.

Immune system changes are known to occur during pregnancy including involution

of the thymus (McLean et al., 1974; Clark et al., 1994) and spleen (McLean et al., 1974),

as well as suppression of B lymphopoeisis (Medina et al., 1993), such that the pregnant

animal may be more, or possibly less susceptible to xenobiotic-induced

immunomodulation. Additionally, many immunotoxic chemicals result in changes in

lymphoid organ weights in the absence of overt toxicity. Cocaine administration had no

clear effect on the weights of the spleen and thymus (Table 4-1). This is in contrast to what

has been reported for mice, where relative thymus and spleen weights were decreased after

cocaine exposure (Faith and Valentine, 1983; Watson et al., 1983; Pirozhkov et al., 1992).

The number of cells recovered from the thymus (Ou et al., 1989; Di Francesco et al., 1992)

and spleen (Pirozhkov et al., 1992) of cocaine-exposed mice also have been reported to be

decreased; however in our studies, there was no significant difference in thymocyte yields

(Table 4-1). This disparity could be attributed to species differences between mice and

rats, or the inability of cocaine to cause further atrophy of the thymus beyond that which

occurs as a function of pregnancy (McLean et al., 1974; Clark et al., 1994). We could find








no reports in the literature comparing the effects of cocaine on mice and rats or pregnant

versus non-pregnant animals, so these two possibilities cannot be distinguished at this

time. In this regard, however, nonpregnant female rats treated with the same cocaine

dosing regimen as in the present study did not show any alterations in relative spleen and

thymus weights (Table 4-2).

It is interesting that maternal plasma IgG levels were significantly increased by

chronic cocaine treatment in pregnant rats (Fig. 4-1), as has been shown for male rats

(Tebbett and Karlix, 1994). In mice, splenocyte primary antibody responses have been

reported to be increased (Havas et al., 1987) or decreased (Faith and Valentine, 1983;

Holsapple and Munson, 1985; Ou et al., 1989), depending on the experimental protocol.

In one of the few studies with rats, male animals treated with cocaine for 10 days showed

enhanced antibody responses to both T-dependent and T-independent antigens at a low

dose, while at a higher dose the T-independent response was further increased and the T-

dependent response was inhibited (Faith and Valentine, 1983; Holsapple and Munson,

1985; Bagasra and Forman, 1989; Ou et al., 1989). Moreover, this study found a

differential change in spleen lymphocyte subpopulations with a significant increase in the

percentage of activated B cells and a concomitant decrease in the percentage of T cells.

Thus, our observation of increased maternal plasma IgG levels is consistent with the

previous report of in vivo stimulalory effects of cocaine on B cell function in rats.

Mitogen stimulated proliferation is another commonly measured endpoint in

immunotoxicological studies. In pregnant rats, the present study did not find any effect of

cocaine on the ability of spleen and thymus lymphocytes to proliferate in response to a

panel of T and B cell mitogens (Fig. 4-2). In contrast, lymphocytes from cocaine-treated

mice showed an altered proliferative response to mitogens (Faith and Valentine, 1983).

Furthermore, exposure to cocaine in vitro inhibited proliferation of mouse splenocytes and

human peripheral blood lymphocytes (Klein et al., 1988), as well as rat B and T cells

(Berkeley et al., 1994), but only at concentrations which greatly exceed those found in








serum of human cocaine users (Cone et al., 1988), or the LD50 in pregnant rats (Salhab et

al., 1994). It is noteworthy that in vitro exposure to lower, more physiological relevant

doses of cocaine (as low as 0.1 pg/ml) has been recently shown to increase human B cell

proliferation (Phillips et al., 1995), as well as human T cell proliferation, IL-2 secretion

and calcium mobilization (Matsui et al., 1992). The present study found that splenocytes

from cocaine-treated rats showed no changes in ConA-stimulated IL-2 production (Fig. 4-

3), whereas mice treated with cocaine were reported to produce significantly less of this T

cell product (Di Francesco et al., 1992). Thus, many of the immune parameters reported to

be suppressed by cocaine treatment in mice have not been found to be altered in pregnant

rats in this study.

Lymphocyte production of PRL-immunoreactive proteins was evaluated as an

additional immune parameter based upon recent evidence that PRL plays an effector role in

the immune system (Gala, 1991). The data showed that mitogen-stimulated rat spleen

lymphocytes consistently secreted a Mr 44,000 PRL-immunoreactive protein and contained

Mr 31,000 and 44,000 PRL-immunoreactive proteins in cell lysates (Fig. 4-4). These

proteins were also produced by rat thymocytes (data not shown). Proteins of similar

molecular weights in murine splenocytes and thymocytes have been reported by other

laboratories. Mouse splenocytes expressed both secreted and intracellular forms of a Mr

46-48,000 PRL-immunoreactive protein in response to mitogen stimulation (Kenner et al.,

1990; Shah et al., 1991), and cell proliferation was partially inhibited by addition of PRL

antiserum to mitogen-stimulated lymphocytes (Hartmann et al., 1989). Mouse thymocytes

were shown to synthesize Mr 33,000 and 35,000 PRL-immunoreactive proteins

(Montgomery et al., 1990). In the present study, spleen lymphocytes cultured from

cocaine-treated rats showed significantly decreased levels of a Mr 44,000 protein present in

both cell lysates and in conditioned medium following ConA stimulation. These cocaine-

related alterations in lymphocytes contrast with previous studies in this laboratory (Salhab

et al., 1994) on rat placenta in which cocaine treatment had no effect on the expression of








placental PRL-like proteins. At the present time however, the sequence identity of the

higher molecular weight PRL-immunoreactive proteins has not been established. The

functional consequences of cocaine-related decreases in splenocyte production of PRL-

immunoreactive proteins also remain unknown in view of the lack of published data

regarding the regulation of production of these proteins in primary culture. TCDD,

however, did not affect PRL-immunoreactive protein production in spleen lymphocytes

from pregnant rats or in human B lymphocyte cell lines (Chapter 3). Since dexamethasone

decreased PRL production in IM-9 cells (Gellersen et al., 1988), cocaine, by increasing

ACTH and subsequently corticosteroid levels, could decrease lymphocyte production of

PRL, but it is unlikely this is the case in our experiments since lymphocyte PRL production

was assessed after 3 days of culture. The observation that PRL mRNA transcript levels

were not changed in splenocytes cultured from cocaine-treated rats suggests that reduced

levels of PRL-immunoreactive proteins may reflect changes at the post-transcriptional level,

or alternatively, that the PRL-immunoreactive proteins are transcribed from a different

gene. Another interesting point is that although the 2.2 kb PRL-hybridizing transcript

detected in our experiments was significantly larger than the pituitary message (0.9 kb),

others have reported PRL mRNAs ranging in size from 1.0-1.6 kb in ConA-stimulated

human thymocytes (Montgomery et al., 1992; O'Neal et al.. 1992) and murine splenocytes

(Woldeyesus et al., 1989: Shah et al., 1991). In contrast, PRL message was detected only

in rat thymus, and not in other lymphoid organs, including the spleen (Touraine et al.,

1994). However, in the latter study, the analysis was performed on RNA isolated directly

from the organs examined from non-immunized animals. Previous evidence suggests

lymphocyte PRL production may be triggered by antigenic or mitogenic stimulation and

would explain why these authors did not detect PRL message in the spleen.

In this study, production of PRL-immunoreactive proteins in spleen lymphocytes

was stimulated by ConA and other T cell mitogens. This suggests that these proteins were

produced by T cells, yet cocaine exposure did not affect the production of another T cell








product. IL-2 (Fig. 4-3). Although the functional consequence of decreased lymphocyte

PRL production is unknown, because of the lack of significant effect of cocaine on other

endpoints measured, it is tempting to suggest that in rats cocaine specifically suppresses

lymphocyte prolactin production and thus may be a sensitive marker for cocaine exposure.

In summary, these studies showed that chronic administration of a fetotoxic dose of

cocaine did not alter maternal plasma PRL, GH, or IGF-1 levels, lymphoid organ weights,

lymphocyte proliferative responses or IL-2 secretion, whereas serum IgG levels were

significantly increased, and lymphocyte production of PRL-immunoeactive proteins was

significantly decreased. Down-regulation of the synthesis of these proteins may be

indicative of subtle changes in lymphocyte function not normally detected during routine

immunotoxicity assessment. The lack of a major effect of cocaine on immune parameters

in pregnant rats may also reflect a differential response to cocaine between pregnant and

non-pregnant animals, or that rats are more resistant than mice to the immunosuppressive

effects of this drug. The observation that nonpregnant female rats treated with the same

cocaine dosing schedule did not show any change in spleen or thymus weights (Table 4-2)

is consistent with the latter hypothesis. In this regard, rats are less sensitive than mice to

the acute hepatotoxic effects of cocaine (Boelsterli and Goldlin, 1991; Roberts et al.,

1992), and differ in susceptibility to the immunotoxic effects of 2-methoxyethanol and 2-

methoxyacetic acid and TCDD (Smialowicz et al., 1992; Smialowicz et al., 1994).

However, since cocaine-mediated neuroendocrine changes (Van de Kar, 1992) and

developmental toxicity (Dow-Edwards, 1991; Salhab et al., 1994) are documented in the

rat, this species is susceptible to cocaine toxicity. Variability in metabolism of cocaine may

be partially responsible for the differing susceptibility to cocaine-induced hepatotoxicity

(Boelsterli and Goldlin, 1991). A recent report described sex and strain variability in mice

in the suppression of the antibody response by cocaine that was suggested to be the result

of differences in metabolism (Holsapple et al., 1993). It will be interesting to see if further

studies determine whether metabolic variability can account for the apparent species






69

difference in the immunotoxicity of cocaine. Thus, the present study finds evidence that

chronic cocaine administration is associated with selective changes in immune parameters in

pregnant rats, but which are not of magnitude likely to compromise pregnancy

maintenance.






70


Table 4-1. Body and lymphoid organ weights of control and cocaine-treated
animals.

treatment body spleen spleen thymus thymus
group weight weight TLY weight TLY
(g) (x107)c (x107)c

control 37414 0.30.02 5.391.31 0.12.02 70.312.6

cocainea 3594 0.30+.02 6.782.10 0.12.01 59.916.7

Results are expressed as the mean SE
a n= 4 except for body weight and spleen weight (n=6)
b organ weights are expressed as the percentage of body weight
c TLY, total lymphocyte yield per organ






71




Table 4-2. Body and lymphoid organ weights of non-
pregnant control and cocaine-treated rats.

treatment body spleen thymus
group weight weight weight
(g)

control 2897 0.33.03 0.13.01

cocaine 27118 0.27.01 0.14.01

a Results are expressed as the mean SE where n=6, except where indicated
b n=5
c organ weights are expressed as the percentage of body weight
























S 300- T [ control

ci cocaine
200- T


l oo
100 =



0
IgG PRL GH IGF-1


Fig. 4-1. Maternal plasma levels of IgG, PRL, GH, and IGF-1 in control and cocaine-
treated animals. Plasma IgG was quantitated with an ELISA and plasma PRL, GH, and
IGF-I by RIA. The results are expressed as the mean SE of 5 animals for the control
group and 6 animals for the cocaine-treated group. P < .05 for cocaine-treated group vs.
control group.









A
250

. 200

S 150
E

Z 100

50

0





B
c 150






0-
E

E5 I 50.



0-


[ control

* cocaine


ConA ConA PWM PWM PHA LPS IL-2
0.25 2.5 0.5 5 10 50 10
U/ml
gg/ml


[3 control

* cocaine


ConA ConA PWM PWM IL-2
0.25 2.5 0.5 5 10
U/ml
gg/ml


Fig. 4-2. Proliferative responses of spleen lymphocytes and thymocytes from control and
cocaine-treated animals. Spleen lymphocytes (A) and thymocytes (B) were isolated,
stimulated with the indicated mitogens, and cell proliferation at 72 hr was evaluated by a
[3H]thymidine incorporation assay. The results are expressed as the mean SE of the net
cpm (cpm incorporated in the presence of mitogen background cpm incorporated) for four
animals in each treatment group.






74












20-



15 T



10-



5-




control cocaine

treatment group


Figure 4-3. ConA-stimulated IL-2 production by spleen lymphocytes from control and
cocaine treated animals. Spleen lymphocytes were stimulated for 48 hr with 2.5 tg/ml
ConA and IL-2 activity in the supematants was quantitated using the CTLL-2 bioassay. An
experiment representative of two separate groups of animals is shown. Samples from each
animal were assayed twice and the average value was taken. The data shown are the mean
+ SE of four control and four cocaine-treated animals.




























Fig. 4-4. Western blot analysis of PRL-immunoreactive proteins produced by spleen
lymphocytes from control and cocaine treated animals. Spleen lymphocytes from four
control and four cocaine-treated animals were cultured at 5x106/ml for 48 hr with 2.5 tig/ml
ConA, and lysates and conditioned medium were subjected to SDS-PAGE and western
immunoblotting. The blots were immunostained with a 1:1200 dilution of anti-rPRL-IC-5.
A representative western blot from one experiment is shown. Only the portion of the
membrane containing the PRL-immunoreactive bands of interest is shown. The position of
molecular weight markers and the major PRL-immunoreactive proteins are shown on the
left and right, respectively. (A) Western blot of PRL-immunoreactive proteins in spleen
lymphocyte cell lysates. (B) Western blot of PRL-immunoreactive proteins in spleen
lymphocyte conditioned medium. (C) Quantitation of rat PRL antiserum-specific bands.
The relative levels of PRL-immunoreactive proteins on the blots was quantitated as
described in Methods. The results are the mean SE of values pooled from two separate
western blots of the same samples (n=6 for control; n=8 for cocaine). P < .05 P <
.01.











A B
control cocaine control




I_
I-I I





106 106

80 4 8

C

5150-



33

S 28
28-



S* 19















100-
CT





150-


cocaine
71


secreted 44 K cellular 44 K cellular 31 K


"















control cocaine e

1 2 1 2


rPRL -2.2kb



P-actin -1.9kb






Figure 4-5. Northern blot analysis of rPRL mRNA expression in spleen lymphocytes and
thymocytes (control only) from control and cocaine treated animals. Cells were cultured as
described for Fig. 4-5 except with a 48 hr incubation. The northern blots were probed with
a radiolabeled rPRL cDNA, top, stripped, and reprobed with a p-actin cDNA, bottom. The
size of the hybridizing bands estimated from the position of the 28S and 18S rRNAs are
shown on the right.














CHAPTER 5
EVIDENCE THAT A MITOGEN-INDUCIBLE PROLACTIN-IMMUNOREACTIVE
PROTEIN IN RAT SPLEEN LYMPHOCYTES IS ALDOLASE A


Introduction

Several laboratories have shown that mouse and human lymphocytes produce PRL

and PRL-like proteins (Kenner et al., 1990; Montgomery et al., 1990; Shah et al., 1991;

Montgomery et al., 1992; Pellegrini et al., 1992), as well as PRL mRNA (O'Neal et al.,

1992; Pellegrini et al., 1992; Sabharwal et al., 1992). Furthermore, lymphocyte

proliferation was partially inhibited by addition of PRL antiserum to mitogen-stimulated

lymphocytes (Hartmann et al., 1989; Sabharwal et al., 1992). Clearly, mouse and human

lymphocytes are capable of producing PRL, which may act in an autocrine/paracrine

fashion to promote lymphocyte growth and/or function. In Chapters 3 and 4, several PRL-

immunoreactive proteins were detected in human B lymphocyte cell lines and ConA-

stimulated rat spleen lymphocytes. There are variants of PRL in the pituitary (Mena et al.,

1992; Sinha, 1995), some of which are of similar apparent molecular weight to those seen

in lymphocyte lysates. Although other laboratories have reported PRL-immunoreactive

proteins in mouse lymphocytes similar to those described here in rat spleen lymphocytes, it

is important to note that none of the higher molecular weight PRL-immunoreactive proteins

identified in either murine or human lymphocytes have been purified or sequenced. Thus,

the structural similarity of these proteins to to pituitary PRL is unknown. The objective of

the studies presented in this chapter was to define what, if any, relationship these proteins

have to PRL.









Results

PRL-Immunoreactive Proteins in Lymphocytes


Lymphocytes isolated from rat spleen were cultured for various times in the

presence or absence of the mitogenic lectin ConA, after which the conditioned medium and

cell lysates were analyzed by SDS-PAGE and western immunoblot procedures. A

polyclonal rabbit antiserum to rPRL specifically cross-reacted with several proteins, the

most prominent of which was an intracellular protein with Mr of 43,000 (Fig. 5-1, lanes 3

and 4). It was also determined if freshly isolated lymphocytes cultured in the absence of

mitogen expressed the Mr 43,000 protein. Since freshly isolated spleen and thymus

lymphocyte lose viability rapidly if not stimulated, the unstimulated cells were not cultured

for the same time period as the mitogen-stimulated cells. Fig. 5-1 shows that the

appearance of the Mr 43,000 protein in cell lysates occurred in mitogen-stimulated spleen

lymphocyte cultures (lanes 3 and 4), but not in freshly isolated cells (lane 1) or cells

cultured for 14 hr in the absence of mitogen (lane 2). An immunoreactive band of similar

apparent molecular weight was also detected in spleen lymphocyte conditioned medium

from ConA-stimulated cells but not unstimulated cells (Fig 5-1, compare lanes 6 and 7 to

lane 5). Additionally, cell density did not did not affect the appearance of this protein as

mitogen-stimulated spleen lymphocytes cultured at both low and high density contained

similar levels of the Mr 43,000 protein in cell lysates and conditioned medium (Fig. 5-1,

compare lanes 3 and 4, and lanes 6 and 7). A Mr 30,000 protein was also recognized by

the rPRL antiserum in some experiments (Fig. 5-1), but this protein showed variability in

the level of Con A-induced expression between individual animals. The Mr 30,000 PRL-

immunoreactive protein was present in freshly isolated lymphocytes at levels similar to

mitogen-stimulated lymphocytes, but at very low levels in unstimulated lymphocytes (Fig.

5-1, compare lanes 1 and 3 with lane 2). Similar results were obtained when identical

experiments were performed with rat thymocytes (data not shown).









Subsequent experiments focused on the intracellular Mr 43,000 PRL-

immunoreactive protein, designated rPIP-43, since it was expressed at very low levels in

freshly isolated spleen lymphocytes but strongly induced by ConA stimulation (Fig. 5-1),

suggesting a role for this protein in lymphocyte growth and/or differentiation. Various

mitogens in addition to ConA were examined for the ability to induce rPIP-43. T cell

mitogens (Con A, PHA, PWM) and B cell mitogens (PWM, LPS) all increased the amount

of this protein detected in spleen lymphocyte lysates (data not shown), although rPIP-43

levels were somewhat lower in LPS-stimulated cells.

IM-9 human B lymphoblastoid cells produced several intracellular proteins in the

Mr 20-70,000 range that were recognized by a polyclonal rabbit hPRL antiserum (Chapter

3). A Mr 40-42,000 PRL-immunoreactive protein was detected in IM-9 cells, but not rat

spleen lymphocytes with hPRL antiserum (Fig. 5-2 B). This observation suggested that

rPIP-43 was not immunologically related to human prolactin. This would be surprising

since the human prolactin antiserum cross-reacted with purified rat pituitary PRL (data not

shown). In contrast, the rPRL antiserum detected a protein in IM-9 cell lysates which

migrated identically to rPIP-43 (Fig. 5-2 A, left panel). A constitutively expressed protein

migrating identically to rPIP-43 was also detected by the rPRL antiserum in the Nb2 rat

lymphoma cell line (Fig. 5-2 A, right panel). These observations suggested that rPIP-43

may be a ubiquitously expressed protein and prompted an examination of the structural

relatedness of this protein to rat PRL.


N-Terminal Amino Acid Sequence of rPIP-43


Since the identity of any of the 40-50K PRL-immunoreactive proteins in

lymphocytes has not been established, an attempt was made to obtain the N-terminal amino

acid sequence of rPIP-43. Rat spleen lymphocyte lysate was fractionated by SDS-PAGE

and N-terminal sequencing identified a single peptide sequence in the mitogen-inducible Mr

43,000 PRL-immunoreactive band. The amino acid sequence of the 25 N-terminal








residues of rPIP-43 was obtained, and a search of the protein sequence databases showed

there was 100% identity with the rat homologue of the glycolytic enzyme aldolase A (Fig.

5-3). Further analysis showed there was no similarity between the N-terminal sequence of

rPIP-43 and rPRL, and only low (21%) overall identity between rPRL and rat aldolase A.

Aldolase A is highly conserved across species with 97% amino acid sequence identity

between the rat, mouse, rabbit, and human homologues. Native aldolase exists as a non-

covalent homo- or hetero-tetramer of 40 kDa subunits which is consistent with the apparent

molecular weight of 43,000 by SDS-PAGE.


Specificity of the rPRL Antiserum for rPRL and Rabbit Aldolase A


Further efforts were made to characterize the specificity of the rPRL antiserum.

The rPRL antiserum identified a single band of Mr 43,000 in an aldolase A preparation

from rabbit muscle (Fig. 5-4, upper panel). The staining of aldolase was not due to the

secondary antibody used in the western blot procedure (data not shown). The hPRL

antiserum, on the other hand, did not react with aldolase A (Fig. 5-4, center panel). This is

consistent with the lack of cross-reactivity of the hPRL antiserum toward rPIP-43. In this

regard, the band stained by anti-hPRL in IM-9 cell lysates appeared to migrate slightly

faster than the Mr 43,000 band in the aldolase A preparation. These observations indicated

that even though both rPRL and hPRL antiserum react with an approximate Mr 43,000

band in IM-9 cell lysates, this band is not aldolase A, and that aldolase A was not

recognized by the hPRL antiserum. Normal rabbit serum recognized aldolase A on

immunoblots (Fig. 5-4, lower panel), although in preliminary results, rPIP-43 did not

appear to be stained by normal rabbit serum. Normal rabbit serum also did not recognize

either hPRL or rPRL (data not shown). It is not presently known why normal rabbit serum

reacted with aldolase A while the hPRL antiserum (made in rabbit) is not reactive.

To further characterize the ability of the rPRL antiserum to react with rPIP-43 and

aldolase A, rPRL antiserum was preabsorbed with aldolase and rPRL before western blot








analysis. When rPRL antiserum was preabsorbed with rat PRL, the intensity of

immunostaining of purified rPRL was greatly reduced, and the intensity of staining of

rPIP-43 and aldolase A was reduced to near undetectable levels (Fig. 5-5, compare left

panel to middle panel). Thus, the antibodies specific for rPRL in this antiserum also

recognized aldolase A. When the rPRL antiserum was preabsorbed with aldolase A, the

staining intensity of rPIP-43 and aldolase A was markedly decreased (Fig. 5-5, middle

panel), although the intensity of staining was not decreased to the degree as with the rPRL-

absorbed antiserum (Fig. 5-5, compare middle panel with right panel). In contrast,

preabsorption of the rPRL antiserum with aldolase A had no effect on the intensity of

staining of rPRL. In control experiments, preabsorption of rPRL antiserum with BSA had

no effect on the intensity of staining of the rPIP-43, rPRL. or aldolase A bands (data not

shown). The rPRL antiserum did appear to have a lower affinity for aldolase A than

purified PRL as aldolase A showed considerably less intense staining than an equal amount

of purified rPRL (Fig. 5-5, left panel).


Effect of Aldolase A in the Nb2 Lymphoma Assay


Rabbit aldolase A was tested in the Nb2 assay for the ability to modulate

proliferation in Nb2 lymphoma cells. Aldolase A was added in combination with varying

concentrations of hPRL to determine if aldolase A either enhanced or interfered with the

mitogenic response of Nb2 cells to hPRL. No effect of aldolase A on Nb2 cell

proliferation was consistently detected, either when added alone or with suboptimal and

optimal concentrations of PRL (Fig. 5-6 and data not shown).


Discussion

These studies present sequence information on a Mr 43,000 protein induced by

mitogenic stimulation in rat spleen lymphocytes that was recognized by rPRL antiserum.

To our knowledge, this is the first time the sequence of any of the PRL variants or proteins









immunochemically related to PRL in the 40-50 kDa range has been determined. The Mr

43,000 PRL-immunoreactive protein characterized in these studies, designated rPIP-43, is

present at very low levels in freshly isolated or unstimulated rat spleen lymphocyte lysates

but is induced by in vitro culture with various mitogens. N-terminal sequencing of rPIP-43

showed that this protein was identical to rat aldolase A, an enzyme in the glycolytic

pathway. There is sufficient difference between the N-terminal sequence of rPIP-43 and

aldolase B (17/25 amino acids identical) and C (20/25 amino acids identical) to exclude the

possibility that rPIP-43 is aldolase B or C (Fig. 5-3). Although only one peptide sequence

was identified in the rPIP-43 band, however; it is conceivable that aldolase A migrated at

the same position as rPIP-43 on SDS-PAGE and was present in a sufficiently greater

amount so that the rPIP-43 polypeptide was not detectable during sequencing. Several

lines of evidence, however, suggested that rPIP-43 is indeed aldolase A and this

identification was not simply an artifact of the sequencing procedure. Aldolase A is highly

conserved across species and ubiquitously expressed (Lebherz and Rutter, 1969), and a

protein identical in electrophoretic mobility to rPIP-43 was found in two lymphoid cell lines

of both rat and human origin at relatively high constitutive levels. Since there was no

readily apparent similarity in amino acid sequence between rPRL and aldolase A, the two

proteins may share a common epitope which is recognized by the PRL antiserum.

Furthermore, naturally occurring rabbit anti-aldolase A antibodies could be present in the

rPRL antiserum. The results of our studies showed that the rPRL antiserum reacted with

rabbit aldolase A on western blots. The cross-reactivity of anti-rPRL towards aldolase A

suggests, but does not prove, that aldolase contamination of the rPIP-43 band on western

blots used for sequencing was not the reason for rPIP-43 being identified as aldolase A.

The fact that normal rabbit serum weakly recognized aldolase A indicated that there are

likely naturally occurring aldolase A antibodies in rabbit serum, but they are either in lower

abundance or have a reduced affinity for aldolase A, and probably could not account for the

the strong recognition of aldolase A or rPIP-43 by the rPRL antiserum on western blots.









Preabsorption of the rPRL antiserum with rPRL, but not BSA, greatly decreased the

intensity of staining of rPRL, rabbit aldolase A, and rPIP-43. Preabsorption of the rPRL

antiserum with aldolase A was not as effective in decreasing the intensity of

immunostaining of either rPRL, rPIP-43, or aldoalse A. Thus, the rPRL antiserum may

not react as strongly with native rat aldolase A as opposed to aldolase A subjected to

reducing SDS-PAGE. This seems reasonable since the native form of aldolase A is a

tetramer and the epitope recognized by the rPRL antiserum could very well be hidden.

These data suggested that the most likely reason that the PRL antiserum reacts strongly

with aldolase A was that aldolase A shared a common epitope with rPRL that was exposed

upon denaturation and recognized by the rPRL antiserum. Recently, it was reported that

two different human PRL antisera cross-reacted with purified human immunoglobulin light

chain and hPRL immunoreactivity co-purified with IgG from human serum and amniotic

fluid (Heffner et al., 1995). Together with the results presented here, this underscores the

need for testing PRL antiserum by western blotting for cross-reactivity with non-PRL

proteins before use in other quantitative immunochemical techniques, as the cross-reactivity

may lead to spurious results when measuring PRL in biological samples.

The present study confirms previous reports of the production of PRL-

immunoreactive proteins following mitogenic stimulation of lymphocytes (Montgomery et

al., 1987; Kenner et al., 1990; Montgomery et al., 1990; Shah et al., 1991; Montgomery et

al., 1992). PRL is a 23 kDa protein and it is likely that the 21-27 kDa PRL-

immunoreactive proteins identified by other laboratories (Montgomery et al., 1990; Baglia

et al., 1991; Montgomery et al., 1992: Pellegrini et al., 1992) is identical to pituitary PRL,

as PRL structural variants (deamidated, phosphorylated, glycosylated, cleaved) of this size

have been identified (Sinha, 1995). In these studies, a Mr 23,000 PRL-immunoreactive

protein in rat spleen lymphocyte conditioned medium was not consistently detected. The

present data support the conclusion that rPIP-43 is aldolase A and it is tempting to speculate

that the other PRL-immunoreactive proteins with a similar molecular weight may actually








be aldolase A as well. There is no conclusive evidence that any of the higher molecular

weight PRL-immunoreactive proteins in lymphocytes are structurally related to PRL. So

called "variants" of PRL in the 40-50 kDa range ("big" PRL) have been identified in

pituitary tissue and plasma (Meuris et al., 1983; Aston et al., 1984; Mena et al., 1992;

Sinha, 1992). Although these forms have been suggested to result from oligomerization of

PRL; in some instances, reducing and denaturing agents did not affect the molecular weight

as determined by SDS-PAGE (Meuris et al., 1983; Mena et al., 1992; Sinha, 1992),

indicating that if the 45K PRL variant is a PRL dimer, then it is formed by covalent

interactions. Indeed, there is some evidence suggesting that PRL may form covalent

dimers or that PRL may become covalently linked to binding proteins (Sinha, 1992).

Moreover, it was recently demonstrated that a 75 and 98 kDa PRL-immunoreactive protein

in human serum was a covalent complex between PRL and the immunoglobulin G heavy

chain (Walker et al., 1995).

Nb2 cells respond to lactogenic hormones (PRL, growth hormone, placental

lactogens) with increased proliferation and are commonly used to assay biological samples

for these hormones. Montgomery et al. (1987) showed that a PRL-immunoreactive protein

present in marine splenocyte culture supernatants stimulated Nb2 proliferation

(Montgomery et al., 1987). Human thymocytes produced 23K PRL which was shown to

be bioactive in the Nb2 assay (Montgomery et al., 1992). Since there is no information

regarding the lactogenic activity of the 40-50K PRL-immunoreactive proteins, and rPIP-43

was detected in spleen lymphocyte conditioned medium, aldolase A was tested for activity

in the Nb2 assay. The data showed that aldolase A had no stimulatory effect on Nb2 cells

and neither enhanced or inhibited the mitogenic response to hPRL.

The results of these studies further showed that the appearance of rPIP-43 in cell

lysates was dependent on mitogen stimulation of the cells and not simply a phenomenon

occurring during in vitro culture. It is not surprising that aldolase A production in

lymphocytes was induced by mitogenic stimulation. It is well known that mitogenic








stimulation of resting lymphocytes causes a series of biochemical changes including blast

formation and increased metabolic rate, RNA, DNA, and protein synthesis (Parker, 1987).

These processes require energy so it is logical that the levels and/or activity of glycolytic

enzymes would increase during mitogenic stimulation. Aldolase enzymatic activity has

been shown to be increased in human PBLs stimulated with PHA (Rogers et al., 1980),

and this study also reported an increase in the level of aldolase protein, confirming earlier

work using PHA and PWM-stimulated human PBLs (Kester et al., 1977). Rat thymocytes

stimulated with ConA showed a 10-15 fold increase in aldolase A mRNA and enzyme

activity (Netzker et al., 1994). Aldolase exists as a multimer, either a homo-tetramer or

hetero-tetramer with two of the A, B, and C isozymes present in various combinations.

The A, B, and C aldolase isozymes are 65% identical in amino acid sequence. Of the three

aldolase isozymes, A ("muscle-type"), B ("liver-type"), and C ("brain-type"), A is

ubiquitously expressed (Lebherz and Rutter, 1969). In the spleen, the predominant

isozyme is aldolase A, although some A-C heterotetramers were detected (Lebherz and

Rutter, 1969). Rogers et al. (1980) reported that the A homotetramer was the predominant

form in lymphocytes, but that mitogen stimulation caused the appearance of some A-C

hetero-tetramers, however; the C4 homotetramer was found only in long-term

lymphoblastoid cell lines, and there were no B subunits detected. In contrast, other reports

concluded that only aldolase A was present in lymphocytes (Steinhagen-Thiessen and

Hiltz, 1979; Netzker et al., 1994); and the reason for this discrepancy is known. Since

aldolase A, B, and C are of the same molecular weight, the data presented here suggest that

only the A isozyme is present in rat spleen lymphocytes because no sequence

corresponding to aldolase B or C was detected in the rPIP-43 band. These data do not rule

out the possibility that aldolase B or C sequence was not detected as a result of the three

isozymes having slightly different electrophoretic mobilities. A mitogen-stimulated

increase in aldolase A levels is not exclusive to lymphocytes. Aldolase A mRNA was

induced in human PBLs by serum and PHA, and by serum or phorbol ester in a human








hepatoma cell line (Gautron et al., 1991). These authors also observed that aldolase A

mRNA was expressed at very low levels in normal liver but at higher levels in hepatoma

tissue and suggested that aldolase A could serve as a marker of proliferating tissues.

Previous studies showed that expression of rPIP-43 in spleen lymphocytes from

pregnant rats was significantly reduced in animals which had been administered a fetotoxic

dose of cocaine (Chapter 4). If rPIP-43 is aldolase A, this implies an effect of cocaine on

lymphocyte aldolase A expression, either as a direct effect on gene expression or a general

effect of cocaine on the ability of lymphocytes to respond appropriately to mitogenic

stimuli. Thus, aldolase A, in addition to serving as a marker of proliferating tissues as

suggested (Gautron et al., 1991), may also represent a potential marker for toxicity. In line

with these observations, evidence has shown that lymphocyte aldolase activity was altered

in various pathological states, including glomerulonephritis and leukemia (Klinger et al.,

1983; Musolino et al., 1992).

In summary, a mitogen-inducible Mr 43,000 PRL-immunoreactive protein in rat

spleen lymphocytes has been identified as the rat homologue of the glycolytic enzyme

aldolase A. The results showed that this lymphocye Mr 43,000 PRL-immunoreactive

protein was not a structural variant of PRL, but a distinct protein. The reason for the cross-

reactivity of the rPRL antiserum toward aldolase A is unknown, but the present data

suggest that this was likely the result of a common epitope shared by rPRL and rat aldolase

A. It was recently reported that antibodies made against nucleotide cofactor-requiring

enzymes cross-reacted with other nucleotide cofactor-requiring enzymes as well as with

enzymes that bound phosphoryl compounds, including aldolase A (Tucker et al., 1993).

However, to our knowledge, there is no evidence that PRL binds either nucleotides or

phosphoryl compounds, although PRL is phosphorylated (Sinha, 1992). Since aldolase A

is a ubiquitously expressed protein and present in serum, the cross-reactivity of antiserum

with aldolase A has the potential to interfere with immunological detection of PRL in






88

biological samples. Finally, these data and others suggest that lymphocyte aldolase A may

be a useful indicator of lymphocyte responsiveness.












1 2 3 4 5 6 7


mom moo*


Figure 5-1. Analysis of PRL-immunoreactive proteins in cell lysates and conditioned
medium from rat spleen lymphocytes cultured in the absence or presence of ConA.
Lymphocyte lysates and conditioned medium were subjected to SDS-PAGE and western
blot analysis using a 1:1500 dilution of anti-rPRL-IC-5. Each lane contains equal amounts
(50 jg) of cell lysate (lanes 1-4), or conditioned medium (lanes 5-7). Lane 1, freshly
isolated lymphocyes; lanes 2 and 5, lymphocytes cultured 14 hr in the absence of mitogen;
lanes 3 and 7, lymphocytes cultured for 72 hr at low density (1 x 106/ml) with 2.5 ig/ml
ConA; lanes 4 and 6, lymphocytes cultured for 60 hr at high density (5 x 106/ml) with 2.5
jig/ml ConA. The position of molecular weight markers is indicated on the left, and the Mr
43,000 PRL-immunoreactive protein is denoted by an arrow.


106
80




49




33


-
28





19





90





A B
RSL IM-9 Nb2 RSL IM-9



106 106

80 80



49 49




33 33

2828




19 19

anti-rPRL anti-hPRL




Figure 5-2. Western blot analysis of proteins in lymphocyte lysates cross-reacting with
human and rat PRL antiserum. Lymphocyte lysates were subjected to SDS-PAGE and
western blot analysis using a 1:1500 dilution of anti-rPRL-IC-5 or a 1:1000 dilution of
anti-hPRL-P4802-1. RSL, rat spleen lymphocytes. (A) Anti-rPRL immunoblot. Left
panel, cell lysates from rat spleen lymphocytes cultured for 48 hr with 2.5 Ag/ml ConA,
and exponentially growing IM-9 cells. Right panel, cell lysate from exponentially growing
Nb2 cells. (B) Anti-hPRL immunoblot of cell lysates from 48 hr ConA-stimulated rat
spleen lymphocytes and exponentially growing IM-9 cells. The position of molecular
weight markers is indicated on the left of each panel. The Mr 40-43,000 cross-reactive
protein detected with anti-rPRL and anti-hPRL is denoted by an arrow.