Role of individual stimulator cells and mitogenic factors in human mixed lymphocyte cultures


Material Information

Role of individual stimulator cells and mitogenic factors in human mixed lymphocyte cultures
Physical Description:
ix, 150 leaves : ill. ; 29 cm.
Ruiz, Phillip, 1954-
Publication Date:


Subjects / Keywords:
Lymphokines   ( mesh )
Lymphocytes   ( mesh )
Lymphocyte Culture Test, Mixed   ( mesh )
Pathology thesis Ph.D   ( mesh )
Dissertations, Academic -- Pathology -- UF   ( mesh )
bibliography   ( marcgt )
non-fiction   ( marcgt )


Thesis (Ph.D.)--University of Florida, 1981.
Bibliography: leaves 134-149.
Statement of Responsibility:
by Phillip Ruiz, Jr.
General Note:
General Note:

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 000895934
oclc - 22838884
notis - AEK4542
System ID:

This item is only available as the following downloads:

Full Text







Dedicated to the memory of Thomas J. Ruiz


First and foremost, I would sincerely like to thank my parents, Mr.

and Mrs. Phillip Ruiz for their lifelong love, support and respect. Their

continuous encouragement has been instrumental in the completion of

this endeavor. Also, I especially wish to acknowledge my grandparents,

Margaret and Pedro Del Castillo, for their unlimited love and aid in

this and everything else I have ever undertaken. I am ever grateful to

my sister Darlene, and to her husband, Buzzy Heinrich, for their constant

love, support and presence. I am likewise deeply indebted to Melissa

Elder for her love and help during this time and during the preparation

of this dissertation.

I am very thankful to Dr. Juan Scornik for administering expert

and inspirational guidance as my advisor. His friendship and unselfish

dedication have made my stay as his graduate student a memorable and

rewarding experience. The members of my committee, Dr. Paul Klein, Dr.

Noel Maclaren and Dr. Edward Hoffmann,are also thanked for their inter-

action, assistance and availability on this project. Other members of

the Pathology faculty to whom I express special thanks for their dis-

cussions and use of equipment are Dr. Shiro Shimuzu, Dr. Michael Nor-

cross, Dr. Raul Braylan and Dr. John Cuddeback. I also thank Dr. Joost

Oppenheim of the NIH for his useful criticisms of our work. I am

grateful to the Department of Pathology for their financial support and

for providing much of the equipment used in these studies.

I wish to express my appreciation to all of the past and present

assistants of the lab, among them being Anita Wilder, Beree Darby and

Sofie Livio. Their excellent work aided in the development and com-

pletion of this project. A special thanks goes to my friend, Art

Alamo, whose time and help with the figures and other work in this dis-

sertation greatly surpassed his obligation as a lab assistant. My

thanks also goes to Joan Ireland and everyone associated with the

histocompatibility lab for their help. I would also like to thank all

the members of the clinical chemistry lab for their friendship and

assistance. In addition, I am extremely grateful to all of the persons

who have repeatedly donated blood for my experiments. The help of

Develyn Matthews and Trish Johnson in preparing this manuscript is sin-

cerely appreciated. I would also like to thank the graduate students

of the Department of Pathology for their encouragement and friendship.

Finally, these acknowledgements would not be complete without

mentioning some of the special people whose support and friendship have

facilitated everything. Among them are Lisa, Barbara and Tom Tedder,

Rube Pardo, Terry Jackson, Danny Cavallero, Tina Bear, Caroline Lepore,

Melanie Elder, Rick Zabak, Rick Marrero and Bruce Baldwin.


ACKNOWLEDGEMENTS .....................................

COMMONLY USED ABBREVIATIONS ..........................


INTRODUCTION....................... ...............

I. Cellular and Genetic Characteristics of the
II. Cytokines and the MLC ......................

SPECIFIC AIMS ........................................

MATERIALS AND METHODS ................................

RESULTS ..............................................

I. Studies of Stimulation in the Human MI.C ....
II. Investigations on a Mitogenic Factor Induced
During the MLC .............................

DISCUSSION ...........................................

I. The Stimulator Cell in MLC .................
II. MLC-Induced Mitogenic Factor ..............

REFERENCES ...........................................

BIOGRAPHICAL SKETCH ..................................

MLC ......
. .. .. .. ..

. .. .. ...

. . .

. .. .. .. ..

.. .. .. .. .

. .. .. .. ..

. .. .. ...

. .. .

. .. .. ..

. .. .

















AEF Allogeneic effect factor

0C Degrees centigrade

Ci Curie

CTL Cytotoxic T cell

CTC1 Long term human T cell line

cpm Counts per minute

ConA Concanavalin A

DMSO Dimethylsulfoxide

ESRD End stage renal disease

ESA Electronic sizing analysis

FACS Fluorescence activated cell sorter

HC Hydrocortisone

HLA Major histocompatibility complex of man

H-2 Major histocompatibility complex of the mouse

IL1 Interleukin 1 or lymphocyte activating factor (LAF)

IL2 Interleukin 2 or T cell growth factor (TCGF)

I.U. Individual units

L.D. Lymphocyte-defined

LAF Lymphocyte activating factor or Interleukin 1 (IL1)

Leu 2 Monoclonal antibody (Becton-Dickinson) which identifies human

T cytotoxic/suppressor subset

Leu 3 Monoclonal antibody (Becton-Dickinson) which identifies hu-

man T helper subset

MNL Mononuclear leukocytes (human peripheral blood)

MLC Mixed lymphocyte culture (or reaction)

MF Mitogenic factor

MLC-MF Mixed lymphocyte culture-derived mitogenic factor

MHC Major histocompatibility complex

N.D. Not done

NA Nonadherent human mononuclear leukocytes

NCS Newborn calf serum

OKT4 Monoclonal antibody (Ortho) which defines human helper T cell


OKT8 Monoclonal antibody (Ortho) which defines human cytotoxic/

suppressor T cell subset

PLT Primed lymphocyte typing

PBS Phosphate buffered saline

PHA-P Phytohemagglutinin protein fraction

PWM Pokeweed mitogen

SCID Severe combined immunodeficiency

S.D. Standard deviation

SD Serologically defined

TCGF T cell growth factor

x Irradiated

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



Phillip Ruiz, Jr.

August, 1981

Chairman: Juan C. Scornik, M.D.
Major Department: Pathology

According to the clonal selection theory, lymphocytes proliferate

and expand into specific sensitized clones when exposed to antigens. In

the case of the MLC, the belief is that clonal proliferation of spe-

cific T cells exposed to allodeterminants on the stimulator cells de-

termines the magnitude of the response, implying that the higher the

number of HLA-D antigens mismatched between the responder and stimu-

lator, the higher the MLC response.

However, it is well known that MLCs reacting to two HLA-D antigen

differences are not always higher than one HLA-D difference, implying

that other factors may play a role in an MLC.

The aim of this work was to explore the influence of those non-

HLA-D factors in MLCs. One nongenetic factor identified as a major

variable in MLCs was the intrinsic capacity of some individual cells to

stimulate in MLC. Mononuclear leukocytes (MNL) from individuals were


found to behave as "high" or "low" stimulators to panels of responder

cells in MLC. This capacity to stimulate was unrelated to HLA-D be-

cause cells from HLA-D identical individuals stimulated differently the

same responders in the same experiment. The concentration of monocytes

and B cells in a given population did not correlate with the capacity

to stimulate in MLCs. Also, individuals identified as "high" stimu-

lators induced MLC responses of equal or greater magnitude than pools

of two or ten stimulators. One factor investigated as a potential mod-

ulator of lymphocyte proliferation was MLC-derived mitogenic factor

(MLC-MF). These studies revealed that MLC-MF provides a powerful mi-

togenic signal to resting MNL, especially in the presence of irradiated,

autologous monocytes. The action of MLC-MF is not genetically restricted

or related to the presence of soluble alloantigens. Also, MLC-MF's

signal to MNL was not found to be mediated through additional production

of mitogenic substances. Column chromatography and absorption experi-

ments suggested that MLC-MF and Interleukin 2 may be closely related.

These studies have characterized stimulator cells and MLC-MF as two non-

genetic factors that influence the extent of an MLC response. Thus,

while only HLA-D antigens trigger the MLC reaction, it is a combination

of several factors that determines its magnitude.


The in vivo and in vitro immune reactions involving the major his-

tocompatibility complex (MHC) done in the context of studying allograft

rejection have now proved to be central in understanding the intrinsic

capacity of an organism to develop any immune response. The mixed

lymphocyte culture (MLC) has proved to be irreplaceable in examining

genes in the MHC and also has provided a system of exploring the basis

of lymphocyte activation by alloantigens. On the other hand, the MLC

has some practical value in establishing genetic relationships among

different individuals at a given MHC locus. Thus, the MLC has been and

continues to be a unique probe to study fundamental immunological

phenomena with a potential of practical application.

I. Cellular and Genetic Characteristics of the MLC

The MLC technique is a measurement of the recognition and resultant

immune response by lymphocytes from an individual or animal when cocul-

tured with lymphocytes from a genetically different member of the same

species. The first description of the in vitro interaction between

lymphocytes from unrelated individuals was made by Bain et al.(l)

and Hirschhorn et al.(2) when they demonstrated that such mixed lym-

phocyte cultures developed responses comparable to mitogen stimulation

of these cells. They observed in these two-way MLCs (where neither

cell population is prevented from responding) the appearance of blast

like cells which took up 3H-thymidine in autoradiographic smears (a


measurement of DNA synthesis) and which underwent mitosis. The peak

response in the cultures appeared 4-6 days after their initiation.

Cultures containing lymphocytes from only one individual did not show

any morphological change from resting lymphocytes and were not syn-

thesizing DNA at any greater rate.

Subsequently, one-way MLCs (where only one cell population is

proliferating) were developed which enabled the observation of a single

individual's responsiveness to other "stimulator" lymphocytes. The

latter were inhibited from proliferating by treatment with x-irradia-

tion (3) or mitomycin C (4). The establishment of the microculture

technique (5) enabled the use of smaller numbers of cells thus per-

mitting analysis of a greater number of variables. The development of

multisample harvesting machines (6) allowed for rapid and accurate

determinations of the amount of cell-bound 3H-thymidine that had been

incorporated by the dividing lymphocytes. The use of 3H-thymidine as a

measurement of cell proliferation or MLC response has been the standard

procedure used since its use was first proposed in 1964 (7).

Genetics of the MLC. Early studies on the mixed lymphocyte re-

action (MLR) within families indicated that no reactions occurred be-

tween monozygotic twins but reactions were sometimes found between

dizygotic twins, siblings and almost always between unrelated indi-

viduals (2,8). These initial indications of genetic influence upon the

degree of MLC reactivity prompted the proposal that the MLC might be

useful as an in vitro test of histocompatibility between individuals

(8) and thus help select for compatible donors for transplantation.

In the mouse system, the use of H-2 congenic lines enabled Dutton

(9) to determine that the major proliferative response in MLCs between

strains occurred when the animals differed at the major histocompati-

bility complex (MHC). An analogous discovery was made in the human

system in 1967 by Bach and Amos (10) who used siblings from seven large

families and found correlation between stimulation in MLCs and leu-

kocyte typings by sera from multiparous women (these sera typed a

series of leukocyte antigens in a single system then known as Hu-1 and

presently known as HLA). They proposed that Hu-1 contained the genetic

loci which controlled the degree of reactivity in the MLC and that Hu-1

was probably the MHC in man. The same authors confirmed these findings

(11) in a more in-depth analysis of the association of antigenic (sero-

logically-defined HLA-A and -B loci) similarity and MLC stimulation,

but found 3 individuals who although were serologically identical,

mutually stimulated each other in MLC. Reactions in MLC between sero-

identical individuals were also found by other investigators (12,13).

By using a HLA recombinant family, MLC reactions between serologically

identical siblings was shown to be closely associated with the HLA-B

locus (14). These investigators proposed that a locus in or near HLA

and distinct from those that were serologically identifiable could be

the primary genetic stimulating locus in the MLC. The finding that

only 10% of HLA-sero-identical unrelated individuals showed negative

MLC responses (15,16) indicated that the relative stimulation in the

MLC could not be predicted by HLA-A or HLA-B typing. When combined

with the intrafamilial MLC data, these findings pointed to a locus in

the HLA complex which strongly influenced MLC reactivity and was dis-

tinct from the serologically-defined (SD) antigens. The separate locus

which controlled the "MLC antigens" was named HLA-D by the Sixth Inter-

national Histocompatibility Workshop in 1975 to distinguish it from the

serologically-defined HLA-A,B and C antigens.

A similar situation exists in the mouse where the I region of the

H-2 complex (analogous to HLA-D) which in general has a large influence

on immune responsiveness and cellular interaction (17), also contains

the major antigenic determinants recognized in the MLC (18,19,20).

Antibodies were discovered in multiparous sera absorbed with

platelets (platelets carry HLA-A,B and C antigens) that reacted with B

lymphocytes and which were later determined to be identifying deter-

minants on molecules coded in or near the D region (21,22). These DR

(D region related) antigens identified by these antibodies are the

analogue of the Ia antigens in the mouse (23) and are found on B cells

(24), macrophages (25), and activated T cells (26,27). The first in-

dication of the association of DR antigens with the D region were the

findings of inhibition of MLC reactions by DR antisera (22). Presently,

there seems to be good correlation between the well defined DR antigens

and the D antigens as defined by MLC typing (discussed below) (28).

However, it seems that it is becoming difficult to maintain the paral-

lelism between D and DR with the new specificities (28). Evidence that

D and DR may be a separate locus includes the discovery of families

with unusual D/DR haplotypes such as DW3/DRW2 (29), the report of a

family which had a crossover between D and DR (30), and the finding of

an antiserum which inhibited MLCs after absorption with a DR positive B

cell line (31). However, as discussed by Bodmer (32), none of the

apparent exceptions to the HLA D/DR association are absolute and all

can still be explained on the basis of technical problems and incomplete

understanding of the genetics of this region of HLA. Recently, systems

of DR-associated, serologically detected antigens, namely the MB anti-

gens (33,34), MT antigens (35) and SB antigens (36) have been discovered.

Their precise relationship to the D/DR region is presently unclear.

Attempts have been made to find methods of identification of the

alleles of these "lymphocyte-defined" antigens of the HLA-D region. One

approach has been through the use of homozygous typing cells (HTC)

(37,38,39). Under the assumption that a lack of stimulation in MLC

corresponds to HLA-D identity, cells were found which were homozygous

at HLA-D, usually from offspring of first-cousin marriages. Panels of

HTC are then used as stimulators in MLCs; if a responder shows minimal

proliferation then this corresponds to compatibility between the re-

sponder and the HTC, whereas a high response indicates incompatibility.

This can be exemplified in a theoretical MLC between individual A who

has antigens 1 and 2 and individual B who is homozygous for antigen 2

(the HTC). In the MLC of B and Ax (x designating the irradiated or

stimulator cell) there should be a reaction of responder B to antigen

1, since they share antigen 2. The reverse MLC (A plus Bx),however,

should show little or no response since responder A recognizes no dis-

similar HLA-D antigen. Thus person A, if his HLA-D phenotype was un-

known, would be assumed to have antigen 2. Unfortunately, negative

typing results are rarely this clear. With unrelated individuals,

"zero reactions" (those close to the autologous control values) are

usually not found (40) and compatibility at HLA-D is instead determined

by consistently low versus high responses. In families, zero responses

tend to occur more frequently (40). Various explanations have been put

forth for this phenomenon. One is that determinants outside of HLA-D

capable of eliciting an MLC response, analogous to the mis locus in the

mouse (41,42), would not be shared between the responder and stimulator

cells. This would, in all likelihood, occur more frequently in MLCs

between unrelated individuals than in intrafamilial MLCs, thus ac-

counting for easier typings with families. In addition, in inbred

mice, F1 hybrid lymphocytes have been shown to proliferate to parental

lymphocytes even though there is apparent genetic identity (43,44).

This phenomenon has also been observed in inbred human families (21,

40). Other factors influencing the MLC could be the nonspecific pro-

liferation of lymphocytes induced by blastogenic factors produced

during the MLC (45) and varying levels of regulator or suppressor cells

(46) in the responder population.

The other cellular method used for defining HLA-D specificities is

the primed lymphocyte-defined typing or PLT (47,48) which was proposed

as an alternate method of typing based on mouse PLT data. This method

of typing is based on the concept that specific clones generated during

the course of a MLC, when harvested several days after the maximal

proliferative burst of the primary MLC, will show secondary response

kinetics when restimulated with the original sensitizing cells or with

cells sharing LD antigens of the original stimulator. Using this

method of "positive typing," PLT cells can be generated against HTC

(inducing clones specific for one HLA-D antigen) and then tested on

panels of cells to determine what HLA-D specificity the cell possesses.

HTC can also be used as the PLT cells. There are presently 12 HLA-D

specificities as defined by negative and positive HTC typing methods

presented at the Eighth International Histocompatibility Workshop. As

with the negative typing, a major problem with PLT assays are the

unusually high "background" responses which are often seen and which

inhibit easy interpretation of results. One reason for this may be the

nature of the determinants which stimulate in the primary MLC as com-

pared to determinants involved in stimulation of PLT cells in secondary

MLC. As mentioned above, the PLT test yields reagents which recognize

products encoded within the D region (47,48,49). However, it has been

put forth by several groups that the HLA-D stimulating products in the

primary MLC are distinct from the antigens recognized by MLC primed

cells (50,51). They have proposed that DR antigens are the strongest

stimuli recognized in the secondary response as demonstrated by a

higher association with DR than D in PLT responses. Other unknown loci

(PL) have also been implicated in PLT stimulation (52,53). This is

further evidence that the serologically detected DR locus is separate

from D. These differences in stimulating determinants recognized by

memory cells may have an influence on interpreting PLT results. Another

influence on the PLT response is that several clones are stimulated

even against one haplotype difference, and differential proliferative

capacity exists due to distinct determinants that are being recognized.

One approach to reduce this variable has been through the cloning of

PLT reagents (54). Their results indicate that a better discriminatory

capacity exists with cloned cells. Attempts at treating the responder

cells with nylon wool (55) have in one report aided in the interpreta-

tion of results.

Cells Involved in the MLC

Responder cells. The "thymus derived" or T lymphocyte appears to

be the predominant cell which recognizes allodeterminants on the stim-

ulator cell and which undergoes blastogenesis in the MLC. Abundant

evidence is available which shows that cells from neonatally thymec-

tomized mice (56) rats (57) and chickens (58) do not respond well in

MLCs. In the mouse system, treatment with anti-theta serum eliminates

the response in MLC (59,60). Weber (61) used chromosomally marked T

cells and B cells in chicken MLCs and found a selective proliferation

of T cells. Cantor and Boyse (62,63) subsequently demonstrated that

the T cells responding in MLC to I-region differences were of the Ly 1+

phenotype or helper T cell population. These Ly 1+ cells can also

serve as helpers for the induction of cytotoxic T cells (Ly 2+, 3+

cells) during the course of a MLC (discussed below).

In the human MLC, T cells also constitute the major fractions of

responding cells (64). Recently, monoclonal antibodies (the OKT series

and Leu series) have been developed which detect human T cell subpopu-

lations involved in the regulation of the immune response. The OKT4+

subset, which provides inducer (helper) function in T-T, T-B and T-

macrophage interactions (65,66) is the predominant population which

proliferates in the MLC (67). The other principal subset, OKT5+/8+

cells, also proliferates in MLC (67) but appears to contain the cyto-

toxic cells which develop following MLC and also suppressor cells (66).

The Leu 3 monoclonal antibody basically correlates with OKT4 and the

Leu 2 antibody with OKT5/8 (68). Differences in activities between the

two series, such as the presence of a MLC induced Leu 3+ suppressor

(69) may be due to varying methods of isolating the subsets, ("negative"

selection with antibody and complement versus "positive" selection with

the flourescence-activated cell sorter).

Stimulator Cells. The nature of the cell(s) responsible for

stimulation in the MLC is not as clear and certainly more controversial

than the nature of the responder cell. Several investigators have

shown that both T and B lymphocytes, in the mouse and human system,

stimulate well in MLC (64,70). Others have shown that B cells stim-

ulate better than T cells (59,71-74), and one group has demonstrated

that IgM-bearing B cells are the subpopulation responsible for B cell

stimulation (75). Monocytes and macrophages have also been found to be

effective stimulators in MLC (76,77,78) along with Ia-bearing, ac-

tivated T cells (79), and dendritic cells (80). Not all la-bearing

cells stimulate in MLC since B cells from leukemia patients which bear

Ia antigens were shown to stimulate poorly in MLC (81).

As mentioned before, irradiated (3) and mitomycin-C (4) treated

cells are able to stimulate and it has recently been reported that glu-

taraldehyde treated cells, while unable to proliferate, could still

stimulate in MLC (82). Stimulator cells damaged by physical or other

chemical agents have been described to lack any stimulatory ability in

MLC (83,84), despite expressing "transplantation antigen" (85).

Development Of Primary And Secondary Cytotoxic Cells During The

MLC. It has been known for sometime that allogeneic cell interaction

results in the generation of cytotoxic T lymphocytes (CTL) which spe-

cifically lyse cellular targets, independent of antibody or complement

(86). In the mouse system, primary (87,88) and secondary (89) cytotoxic

cell responses have been demonstrated following MLC. Primary (90,91)

and secondary (92) MLC-induced CTL have also been found in human cell

cultures. The effector cell in the mouse system is an Ly 2+, 3+

T cell (63) and a OKT5+/8+ (or Leu 2+) T cell (66) in humans. A recent

report (93) which phenotypically characterizes the human MLC-generated

cytotoxic cell, describes the presence of the activation antigen 4F2 on

the CTL and a lack of receptors for IgG on this cell.

Eijsvoogel et al. (90,94) showed in their studies that although

disparity at HLA-D is necessary for the generation of CTL, the antigens

recognized by the CTL are those coded by the HLA-A,B and C regions. It

has also been shown that differences at D were not absolutely necessary

for CTL induction (this is analogous to experiments in mice showing that

CTL can develop with only a K region difference)(95). Although HLA-A,B

and C determinants appear to serve as important targets for CTL, recent

work by Feighery and Stastny (96,97) demonstrates that CTL could develop

in MLCs between HLA-D mismatched, HLA-A,B matched individuals that were

directed against the specific HLA-D product. Strong cellular cyto-

toxicity to I-region determinants has also been described in the mouse

CTL system (98,99). It is presently not clear whether the products

recognized by CTL are the same as those seen by serological techniques.

The use of in vitro generated cytotoxic T-lymphocytes as typing reagents

is currently under way in several labs and a recent European CML work-

shop (100) was held to compare cellular typings of the HLA locus.

Since the techniques are presently available to clone and propagate

lymphocytes in vitro for long periods of time, the feasibility of CTL

as typing reagents is increasing.

Generation Of "Primed" Or Memory Lymphocytes During The MLC

Following the initial burst of cell division in the primary MLC,

the clones of sensitized lymphocytes begin to slow down metabolically

and revert back to cells with the morphology of small, resting

lymphocytes. This is a heterogeneous population which contains specif-

ically primed cytotoxic cells (discussed above) and other memory cells

which also show secondary response kinetics. The primed cells, as

mentioned before, display an accelerated proliferative response toward

the specific primary stimulator cell and this forms the basis of the

PLT assay. The nature of the stimulating antigens involved in the PLT

response has been previously discussed. The PLT cell appears to be

the helper or responder (63,66) cell which proliferates and aids in the

development of cytotoxic cells during the MLC.

Regulation of the MLC by Suppressor Cells

Rich and Rich (101) first described in the mouse the generation of

cells from alloantigen challenged mice which inhibited syngeneic MLC

responses in vitro. It was subsequently shown by several investigators

that the mouse MLC was itself capable of generating suppressor cells

which could inhibit other MLC and CTL responses (102,103,104). The

mechanism of the suppression appeared not to be due to killing of the

responders (101) although this claim has been disputed (105). The cell

responsible for suppression is a T cell with the Ly2+,3+ phenotype

(62,63). These negatively regulatory cells have also been described in

human MLC reactions (106,107,108,109). One group claims that the

suppressor T cell in their system is nondiscriminatory in its suppression

(106), while others find preferential suppression occurring with cells

autologous to the suppressor at HLA-D (107,108). Although the mechanism

of suppression is unclear, there are reports claiming that suppression

is mediated by soluble factors (110,111). Recently, the human MLC-

induced suppressor cell has been demonstrated to be found largely in

the fraction of T lymphocytes identified by the Leu 3 antibody (69).

This is an interesting finding since the Leu 3 subset is thought to

correspond to the "helper" T cell population in man while suppression

in other systems have been associated with the Leu 2 population.

Engleman et al.(112,113) have described the presence of suppres-

sor cells in the blood of several individuals who did not respond in

MLC to specific stimulator cells. This radiation-sensitive cell

appears to only suppress responses of autologous cells or cells which

share HLA-D with itself (thus, it is genetically restricted at HLA-D).

Additionally, the suppressor cell shows specificity in that only the

responses to specific alloantigens are suppressed. This genetically

restricted, specific suppressor cell has also been reported by other

labs (114,115). These latter descriptions of suppressors are of in-

terest since they may give some insight into the controls that the HLA

region has on normal immune function and in the role of various cell

subsets in autoimmune and other disease processes.

The Autologous MLC

In 1975 Opelz et al.(116) made the surprising discovery that when

human T lymphocytes were incubated with autologous B lymphocyte-enriched

cell populations, a significant response was measured at 6 days. These

results were corroborated in humans (117,118) and mice (119) and the

phenomenon became known as the autologous MLC. Although the responder

cell has consistently been found to be a T cell, studies by Sakane and

Green (120) have shown that the T cells responsive to autologous cells

are largely separable from T cells responsive to allogeneic cells. The

nature of the stimulator cell, as in the allogeneic MLC, is currently a

matter of controversy. While many labs claim B cells (116,121) and K

(null) cells (117) to be the major stimulating cell populations, most

have not used purified preparations. Recently, MacDermott and Stacey

(122), using highly purified cell populations, showed that B cells were

poor stimulators whereas macrophages possessed the most potent stimula-

ting capacity in the autologous MLC. Other studies have shown that the

autologous MLC, and not the allogeneic MLC can be inhibited by physio-

logic doses of hydrocortisone (118,123). Finally, the findings of

decreased autologous MLC activity in lupus patients (124,125), and the

generation of suppressor cells during the autologous MLC (126) suggest

that this phenomenon may be part of the normal autoregulating mechanisms

present in the immune response.

Clinical Significance of the MLC

Early studies on the MLC suggested that the test could serve as a

method of evaluating the degree of histocompatibility between two

individuals. Since this could be a valuable tool in transplantation

medicine (predominantly transplants of kidneys and bone marrow) numer-

ous studies have been done on the relevance between the MLC reaction

and the survival of the graft.

What has emerged is a confusing picture on the usefulness of the

MLC in regards to transplantation. The most beneficial role of the MLC

appears to be almost exclusively limited to living related donor trans-

plants, especially among HLA-identical siblings. These serologically-

identical individuals rarely react to each other in MLC (95) and also

have the greatest chance for graft survival (127). When evaluating 1-

haplotype mismatch MLCs, however, often the reactions seen are as high

as 2-haplotype MLCs. Dupont et al.(95) and others (128) have shown

tremendous overlap when analyzing the means of a considerable number of

1-haplotype versus 2-haplotype MLCs. This brings up the basic problem

that one encounters in MLCs; that is, correlating the degree of prolif-

eration in the MLC (which is often high in the 1-haplotype MLCs) with

the suitability of the prospective donor (or the histocompatibility

difference between the donor and recipient). Some labs have claimed

correlation between the MLC and graft survival by using cutoff stimu-

lation indices (129) or relative responses (130). However, several

reports exist claiming no correlation using the same means of data

analysis (131,132).

In conclusion, the emergence of the MLC as a clear-cut analytical

tool in transplantation has not yet occurred and its role in the future

in this regard certainly has to be evaluated further.

II. Cytokines and the MLC

The role of soluble cytokines in the primary proliferative phase

of the MLC is unclear, although the MLC has been shown to induce pro-

duction of many of these molecules. An attempt will be made to briefly

summarize characteristics of several lymphokines and monokines which

may have a significant influence on the MLC, although many of the

studies were performed with mitogen-stimulated cultures or in other

systems. A distinction between the human and mouse systems will be

made only when necessary.

Mitogenic Factors (MF)

Many of the presently described lymphokines and monokines can be

termed "mitogenic factors" since they stimulate cell division of lym-

phocytes. This section will describe, however, substances which induce

mitogenesis of mature, unstimulated lymphocytes, as opposed to antigen-

stimulated cultures.

Kasakura and Lowenstein (133) reported in 1965 the appearance of

soluble mitogenic factor in the supernatants of human MLCs which in-

duced resting lymphocytes to proliferate. The peak of MF production

was found 3 to 5 days after initiation of the MLC and irradiation

appeared to have little effect on the release of MF (134). T cells

proliferate in response to MF stimulation (135,136). Uotila et al.

(137) suggested that the cytotoxic T cell is the target while Kasakura

(138) has shown the development of nonspecific cytotoxic T cells after

exposure to MF. Others have claimed that B cells are also stimulated

to divide by MF (139,140). MF is produced by T cells (141) requiring

macrophages (142) to help initiate MF production. The regulation of MF

levels has been suggested to be under the influence of suppressor cells

(143,144) which when removed, allow for enhanced MF activity in the

supernatants. The need for antigen in order for MF to work is presently

unclear. Using MF derived from lymphocytes stimulated with tetanus

toxoid, Geha and Merler (145) found that MF function was only expressed

in the presence of antigen. However, others have found no genetic

restriction of MLC-derived MF, suggesting no role for solubilized

transplantation antigens (146).

The molecular weight (mw) of MF has been reported from being a

80,000 mw protein tetramer (140) to a 20,000 to 28,000 mw protein

(139). However, extensive biochemical analysis of MF is lacking. It is

currently unknown whether the activities attributed to MF and several

described human cytokines (such as IL2, see below) are performed by the

same or separate molecules. Another consideration is whether it is

wise to compare MF studies since a careful evaluation of the biochemical

and functional characteristics of MFs from different sources (MLCs,

mitogens, antigens) is not available.

Most of the studies on MF described above were done with super-

natants from human cell cultures. Few factors derived from mouse cells

have been found to stimulate mature, resting cells. One well described

mitogenic factor in the mouse system is allogeneic effect factor (AEF),

which was first reported in 1974 by Armerding and Katz (147). The

method of production of AEF is uncommon in the sense that it is the

product of a secondary in vitro MLC following a primary in vivo allo-

antigen priming by thymocytes. The original activity designated to AEF

was its ability to replace T helper cells in augmenting the in vitro

antibody responses by B cells (147). Subsequently, molecular charac-

terization of AEF revealed the presence of Ia determinants on the

molecule (148,149). The most recent and interesting biologic effects

of AEF are the stimulatory signals that AEF gives to mature resting T

cells. It has been shown that AEF has the capacity to induce the de-

velopment of cytotoxic T cells which preferentially lyse autologous

cells (150). Additionally, AEF causes the proliferation of mature T

cells which when tested after priming, react in secondary kinetics to

autologous cells (151). Thus, this product of alloantigen-activated T

cells appears to preferentially induce autoreactive effector cells. The

physiological significance of this process is presently unknown.

T Cell Growth Factor (TCGF) or Interleukin 2 (IL2)

In 1976, the discovery was made that supernatants or conditioned

media from PHA-stimulated mononuclear cell cultures could support the

growth in vitro of human T cells activated by lectin (152). This

growth factor appeared in mononuclear cell supernatants of T cells stim-

ulated with either mitogens or alloantigens (153,154). It was later

shown that T cells produced this factor (155); in particular, the

source of TCGF was helper T cells, possibly in a macrophage independent

process (156). A quantitative microassay was developed for the measure-

ment of T cell growth factor (153) using long term cytotoxic T cell

lines dependent on TCGF for their growth (157). In mice, it appears

that normal, resting T cells do not respond to TCGF (155,158,159,160),

and that the acquisition of responsiveness to TCGF is a direct result

of the T lymphocyte activation by mitogen or alloantigens. The need

for a cell to be activated in order to respond to TCGF has also been

demonstrated by the inability of resting cells to absorb out any TCGF

activity in the mouse (155) and human (161) systems, whereas activated

cells absorb TCGF very efficiently. Cytotoxic T cells are the primary

cells which respond to TCGF (155,159), and the expression of receptors

for TCGF on these cells occurs as early as 4 hours following cell-

lectin interaction (162). It has recently been shown that cytotoxic

cells may also be capable of producing TCGF when restimulated with H-2

K/D region antigens (163).

Several mouse and human tumor lines have been found that produce

large amounts of TCGF following stimulation with mitogen or phorbol

myristate acetate (164,165,166). Biochemical analysis reveals that

murine TCGF has a molecular weight of 30,000 (167) while rat and human

TCGF have a molecular weight of 15,000 (168). In addition to its

unique property of propagating cultured T cell lines, Watson et al.

(167,168) found that the TCGF column fractions had other biological

properties. These properties are the ability to augment thymocyte re-

sponses to submitogenic doses of mitogen (costimulator function),

helper T-cell replacing factor (TRF) activity in athymic spleen cell

cultures and the induction of cytotoxic lymphocytes in both thymocyte

and nude mouse spleen cell cultures. Another property of TCGF is that

steroids (169) and cyclosporin A (170) inhibit the production but not

the effect of this lymphokine. These drugs may serve as probes to in-

vestigate molecular mechanisms involved in TCGF's effect. These and

other studies on TCGF have suggested that this molecule may be the only

mitogenic signal generated during cell-lectin interaction, and that it

is intricately involved with a macrophage product known as LAF (lym-

phocyte-activating factor) or IL-1 (discussed below). Smith et al.

(171) have demonstrated that the addition of LAF containing super-

natants increases the levels of TCGF that is produced and thus in-

creases T cell proliferation. From this they have proposed that LAF

serves as a signal from the macrophage (which has bound the lectin) to

help initiate and augment TCGF production, which eventually controls

the level of the clonal T cell response.

Lymphocyte Activating Factor (LAF) or Interleukin I

Macrophages secrete a large number of products which modulate

the immune response. These substances range from small molecular weight

molecules such as prostaglandins (172) and free radicals (173), to

larger molecules (usually proteins) which are collectively known as

monokines. One monokine described by Gery et al. (174) possessed

potent mitogenic properties for thymocytes, but would minimally

stimulate peripheral T cells, and was named lymphocyte activating

factor (LAF). The source of LAF was shown to be adherent cells rather

than lymphocytes (174,175). The stimulant used by Gery to produce LAF

was bacterial lipopolysaccharide (LPS), an agent known to activate

macrophages. Other stimulants (including the MLC) found to induce LAF

from human leukocytes or murine peritoneal exudate cells, have also

been known to activate macrophages either directly or indirectly through

lymphocytes (for review, 176). Several macrophage tumor lines produce

LAF (177) when stimulated by LPS, and Mizel et a1.(178) find the

molecular weight of LAF from the P388 macrophage line around 15,000 mw.

Earlier molecular weight descriptions of LAF have been similar (174,179),

with some investigators also finding high molecular weight forms of LAF

(179). No size differences have been found between human and murine


Some biological properties of LAF include the previously mentioned

thymocyte mitogenic ability, the capacity to replace macrophages in the

induction of cytotoxic T cells (180), and the enhancement of the B cell

antibody response (181). A most interesting finding is the increased

production of TCGF by addition of LAF. As discussed previously, these

studies imply that LAF or Interleukin I is a signal generated during

lectin-macrophage interactions which aids in potentiating T cell pro-

liferation through the production of TCGF. The role that these two

interleukinss" have in the immune response, although still ill-defined,

is clearly an important one. The relationship that these products have

to other cytokines is currently an area of intense interest.


According to the clonal selection theory (182), lymphocytes pro-

liferate and expand into specific sensitized clones when exposed to anti-

gens. In the case of the MLC, the belief is that the clonal prolifera-

tion of specific T cells exposed to allodeterminants on the stimulator

cells determines the magnitude of the response, implying that the higher

the number of antigens mismatched between the responder and stimulator,

the higher the MLC response.

However, comparison of many 1-haplotype and 2-haplotype MLCs shows

that although the overall mean of the responses in 1-haplotype MLCs is

lower, the overlap between the two groups is substantial. Since these

results do not sustain the concept that greater HLA-D disparity equates

with higher MLC reactivity, a series of studies were initiated to iden-

tify which factors other than HLA-D might affect the magnitude of the

proliferative response in MLCs.

These studies addressed the following areas.

1. Examination of the overall or "general" stimulating ability and re-

sponsiveness of MNL.

a. Peripheral blood MNL from HLA-identical siblings were used

to stimulate different responder cells to determine whether

there was any difference in their stimulatory ability. These

potential differences would not be related to HLA-D.

b. Individual stimulator cell populations were compared with

pools of stimulators for their capacity to stimulate in MLC.

These studies were done to determine the stimulation of

one stimulator cell population relative to a pool of stimu-

lators with multiple HLA-D disparities present.

c. Individuals who were used in several MLCs were analyzed for

their "relative" stimulating ability to determine whether

some persons were consistently better stimulators than others

in MLCs.

d. Mitogen assays were simultaneously performed with MLCs to

determine whether an association of responsiveness could be

established in different assay systems.

2. A major effort was dedicated to better define MLC-derived soluble

mitogenic factors (MLC-MF) and to explore the possibility that they

may be an important source of nonspecific proliferation in the MLC.

a. Studies were done to optimize conditions for the procurement

and production of sufficient quantities of MLC-MF with high

levels of activity.

b. The levels of MLC-MF in MLC supernatants were measured and

compared to the response of the original MLC (where the super-

natant was taken) in an attempt to correlate MLC-MF production

with the MLC response.

c. The effect of supernatants from MLCs containing MLC-MF ac-

tivity was tested in autologous and allogeneic MLCs.

d. Studies were performed to determine the nature of the cell

which was providing a "helper" effect to cells responding to


e. Studies were done to find out whether the proliferative

effect of MLC-MF was mediated through further production of

mitogenic factors or whether the effect on MNL was direct on

responder cells.

f. Experiments were also designed to explore whether soluble

alloantigens could be involved in the effect of MLC-MF.

g. The cells responding to MLC-MF were also analyzed using mono-

clonal antibodies to T cell subsets.

h. In column chromatography and absorption experiments the rela-

tionship between MLC-MF and IL2 was investigated.

These investigations led to the identification of a number of vari-

ables that can significantly influence the outcome of a MLC. In addition,

they definitively established stimulatory properties of cells independent

of HLA-D and a clarification of the way in which MLC-induced mitogenic

factors exert their nonspecific effects on lymphocytes.


Individuals. All the studies were performed using cells obtained

from normal individuals and in some cases with end stage renal disease

(ESRD) patients. The behavior in MLC of the cells from ESRD patients

as a group was similar to the cells from normal individuals.

Cell preparations. All of the cell preparations and assays were

performed sterilely.

Human mononuclear leucocytes (MNL). Human peripheral blood was

drawn in heparin (Upjohn, Kalamazoo, MI, 10 I.U./ml). The buffy coats

were removed, layered over Ficoll-Hypaque and MNL isolated by centri-

fugation (183). The MNL were washed three times (1600 rpm, 10 minutes,

room temperature) with phosphate buffered saline (PBS) and resuspended

in RPM1-1640 (GIBCO, Grand Island, NY) supplemented with 50 ug/ml

gentamycin (Schering, Kenilworth, NJ) and 20% pooled human AB serum

(Bio-bee, Boston, MA). This medium (complete) was used for most assays

unless otherwise noted.

Non-adherent (NA) human MNL. Human peripheral blood MNL (isolated

as described above) were run through a series of steps in order to

obtain the population of cells referred to as nonadherent cells. The

first step consisted of incubation of 10 X 106 MNL in 100 X 15 mm plas-

tic petri dishes (Falcon, No. 1029) in 5 ml of complete medium for 1

hour at 370C. Duplicates were run according to the number of cells

needed. Following the incubation, the plates were examined microscop-

ically to ensure that cells were attaching, and the non-adherent cells

were rinsed off, using 3 washes per plate. These non-adherent cells

were then pooled and isolated by centrifugation (1600 rpm, 10 minutes,

room temperature). The supernatant was decanted and the cells were

resuspended in .5 ml of complete medium at 370C. The cells were then

added to nylon wool columns which had been equilibrated to a tempera-

ture of 370C. The columns were removed, and medium was run through to

elute the non-adherent cells. Subsequently, the cells were centrifuged,

resuspended and counted.

The proportion of lymphocytes and monocytes before and after sep-

aration was determined by electronic sizing in a Coulter channelizer

and Coulter Counter Model 2 ZI. Approximately 105 cells were placed in

10 ml of Isoton Plus and 0.77 ml of Lyse S Plus (both from Coulter

Diagnostic, Hialeah, FL) and aspirated into the Coulter Counter. The

size distribution was recorded into a Hewlitt-Packard 9845T computer

and the proportion of the large and small cells was calculated according

to the areas of the corresponding peaks. The large cells monocytess)

are practically eliminated by sequential attachment to plastic surfaces

and nylon wool. Monocyte concentration was estimated by myeloperoxidase

staining (184).

MLC-primed cells (PLT). These cells were obtained by incubating

10 X 106 responder MNL from one individual plus 10 X 106 irradiated MNL

from another individual in a volume of 12 ml of complete medium in 75

cm2 flasks (Corning, New York, NY) for 10 days at 370C. Following this

period, the cells were collected, washed twice, and viable cells counted.

Murine thymocytes. Thymuses were removed from 4 to 6 week old

C57BL/6 mice, cut and pressed through a nylon mesh screen, and run

through a series of varying gauge (19,21,23 and 25) needles to obtain a

single cell suspension. The cell suspension was then overlayed on

Ficoll Hypaque, and centrifuged at 1200-1300 rpm for 15 minutes at room

temperature. The pelleted cells (or high density thymocytes) were then

removed, washed 3 times with PBS and resuspended in medium used for the

costimulator assay (see below), and counted. Phytohemagglutinin (PHA)-

stimulated MNL were prepared by incubating MNL at a concentration of 1

X 106 cells/ml in complete medium with .1% PHA-P (Difco, Detroit, MI)

for 3 days at 370C, 5% CO2, in 75 cm2 tissue culture flasks (Corning).

These cells were collected, washed 3 times with PBS and counted.

Cryopreservation and thawing of human MNL. Human MNL were isola-

ted from peripheral blood as described above with the exception that

the MNL were resuspended after the final wash in RPM1-1640 supplemented

with 20% newborn calf serum (NCS) (GIBCO) and antibiotics. After

counting the cells were adjusted to a concentration of 6.0 X 106

cells/ml with the same medium. An equal volume of RPM1-1640 containing

20% DMSO and 20% NCS was then added to the cell mixture at the approxi-

mate rate of 1 ml/minute. This step was done on ice and the final cell

concentration was now 3.0 X 106 cells/ml. The cell suspension was then

added in 1.0 ml aliquots to freezing ampules (Nunc, Intermed, Roskilde,

Denmark) kept on ice. The freezing of the cells was done according to

a method described by Dr. Rene Duquesnoy in a personal communication.

Briefly, the ampules were transferred to a styrofoam box which had

multiple holes in it. The covered box was then put in a -700C freezer

for 24 hours (the holes served the purpose of allowing the temperature

within the box to gradually decrease). After 24 hours, the ampules

were transferred to liquid nitrogen, where they were kept until used in


Frozen cells were thawed in the following manner. The ampules

were removed from the liquid nitrogen, placed in a 400C water bath for

60-90 seconds until thawed, and then immediately placed on ice. The

1.0 ml of cells was then transferred to a larger tube and 9.0 ml of

RPM1-1640 with 20% NCS was added at a rate of 1 ml per minute. The

cells were then spun down, resuspended, and viable cells enumerated.

In Vitro Assay Systems

Mixed lymphocyte cultures (MLCs). (Figure 1) MLCs were performed

in round-bottom 96 well plates (Dynatech, Alexandria, VA), each well

containing 0.2 ml of RPM1-1640 supplemented with 20% human AB serum and

gentamycin (complete medium). The wells also contained 30 X 10'

responder MNL and, unless noted, 120 X 103 stimulator MNL irradiated

with 2500 rads from a '37Cs source (Gammator Model M). The plates were

then incubated at 370C, 5% CO2, for a total of 6 days. On day 5, the

cultures were pulsed with 1.0 uCi of 3H-thymidine (Schwarz/Mann, specif-

ic activity of 6 Ci/mM) and harvested 18 hours later onto filter papers

with a 24-line harvester (Otto Hiller, Madison, WI). The filters were

air dried, placed into vials with an appropriate volume of scintilla-

tion fluid and counted in a LKB Model 81000 liquid scintillation counter.

MLC-MF assay. (Figure 2) Supernatants from MLCs were routinely

tested for their mitogenic activity on resting human MNL in 96 well

Individual A (Responder)

Peripheral blood MNL (30 x 103 cells)

"0 96 well roundbottom
3000 Rods microtiter plate
30OO Rbed

Peripheral blood MNL(120 x 103 cells)

Individual B (Stimulator)

5 days
370C, 5% C02

--- H-thymidine
18 hours

Quantitate Cell-bound cpm

Human mixed lymphocyte culture (MLC).

Figure 1.

round-bottom plates. 30 X 103 responder MNL in 0.1 ml of complete

medium were incubated with 0.1 ml of MLC supernatants in each well.

The cultures were incubated for 6 days at 370C in 5% CO2. On day 5,

cultures were pulsed with 1.0 uCi/well of 3H-thymidine and harvested 18

hours later with a 24 line harvester. Filters were collected and

counted as described in the MLC assay. When testing for helper cell

activity, 120 X 103 autologous, irradiated (3000 rads) unseparated or

NA MNL were added with the responder cells at the beginning of the

incubation period. In some experiments, hydrocortisone sodium succinate

(HC) (Upjohn) was added initially to some wells in 10 ul volumes at

different concentrations and allowed to remain for the duration of the


On several occasions PLT cells were also used as responder cells

to MLC supernatants. The assay consisted of 30 X 103 PLT cells in 100

ul of complete medium in the presence of 100 ul of supernatant. Pulsing

was done after 48 hours at 370C and harvesting 72 hours after initiation

of the cultures.

Costimulator assay. MLC supernatants and column fractions were

tested for their ability to enhance mitogen (at suboptimal dose) in-

duced responses in mouse thymocytes (185). 5 X 105 C57BL/6 thymocytes

in RPM1-1640 supplemented with 1.0% fresh autologous mouse serum,

mercaptoethanol (5 X 10-5 M), 10 mM HEPES buffer, 100 I.U./ml penicillin,

100 ug/ml streptomycin and .25 ug/ml Fungizone (GIBCO) Were added in a

total volume of 100 ul to 100 ul of MLC supernatant or column fraction,

with or without 0.3 ug/ml of Conconavalin A (ConA) (Miles-Yeda, Re-

hovot, Israel). After 48 hours, cultures were pulsed with 0.5 uCi/well

of 3H-thymidine and harvested 24 hours later.

T cell growth factor (TCGF) assay. Levels of TCGF in supernatant

or fraction samples were determined by the stimulatory activity of

these samples on a long term, cultured human T cell line (CTC1). Thirty

thousand (30 X 103) CTC1 cells in 100 ul were incubated at 370C with 100

ul of the sample being tested. The cultures were pulsed with 1.0 uCi of

3H-thymidine at 24 hours and harvested 48 hours following the initiation

of the incubation.

The CTC1 cell line was derived from a 10 day old MLC. Following

the initial MLC, the cells have been propagated in vitro by the addi-

tion of unrelated, irradiated MNL to the culture flasks in the presence

of mitogenic doses (.1%) of PHA-P. Six days prior to the TCGF assays,

the cells are maintained only by the addition of supernatants from MNL

stimulated by PHA. At this time, the cells are usually exclusively

responsive to TCGF and not to PHA. Assays were not considered in which

the CTC1 cells responded to PHA.

Mitogen assays. MNL from patients and controls were also used in

proliferation assays using three well known mitogens, PHA, ConA, and

pokeweed mitogen (PWM) (Sigma, St. Louis, MO). Briefly, the responding

MNL were used at 30 X 103 cells per well (in round-bottom microtiter

plates) in a volume of 100 ul. The mitogens were also added in 100 ul

with the optimal concentrations being .1% for PHA, 0.6 ug/ml for ConA

and 12 ug/ml for PWM. After 48 hours incubation at 370C, the cultures

were pulsed with 1.0 uCi/well of 3H-thymidine and harvested 18 hours

later as described in the assays above.

Production of MLC-MF (Figure 2). Mixed leucocyte cultures were

set up in 17 X 100 mm tubes (Falcon, Oxnard, CA) using 3.0 X 106


S3.0 106 cells) [ (3 x 106 cells)

3000 Rods


5mI of Medium/tuoe(16 i 00 m))

2-3 Days at 37*C

Soin fown cells. collect suoernatant
filter Sterilize
!0 .(3 H,00 3H-thymidine
ceils Quaonittated
^~~~~ / 5cell- bound
5 days IShrs cOm

rr~ata 96 we l
MNL found boom
20 3 'nucronter
ellss plate

Figure 2. MLC-MF: Production and assay.

responder MNL and 3.0 X 106 irradiated stimulator MNL (2500 rads) in a

total volume of 5 ml of complete medium (RPM1-1640 with 20% human AB

serum). After 48-72 hours, the tubes were mixed, centrifuged (2000 rpm,

10C, 10 minutes) and the cell-free supernatants removed. All super-

natants were pooled, filtered through a .45 um filter (Sybron/Nalge,

Rochester, NY), aliquoted and kept frozen at -700C until used. In some

experiments, hydrocortisone, prepared in PBS, was added initially to

cultures at several concentrations and allowed to remain in the super-

natant when collected. On several occasions, PLT cells were specif-

ically restimulated on day 10 with the original stimulator cell and

supernatants were collected 48-72 hours later.

Absorption of MLC supernatants. Human and murine tumor lines,

resting MNL and PHA-stimulated MNL were used for absorption at concen-

trations from 20 X 106 per ml of MLC supernatant. The human tumor

line was a T cell lymphoma that is terminal deoxyribotransferase and

leu 1 (Becton Dickinson, Oxnard, CA) positive (kindly provided by Dr.

Raul Braylan). The mouse tumor line used was M3, a methylcholanthrene-

induced fibrosarcoma derived from C57BL/6 mice (kindly provided by Dr.

Paul Klein). PHA-stimulated and resting MNL were prepared as previously

described. Absorptions were performed at 4C for 30 minutes (to prevent

the release of factors from the absorbing cells), mixing every 10

minutes. After the incubation, the tubes were spun down in a refrig-

erated centrifuge, and the MLC supernatants were removed, sterilized

through a .45 um filter and kept frozen at -700C until used.

Treatment of MNL with OKT antibodies. Unseparated and NA MNL were

treated with the monoclonal antibodies OKT4 and OKT8 (Ortho, Raritan,

NJ), and rabbit complement (Pel-Freeze). Thirty microliters of each

antibody was added to 3.0 X 106 MNL which were in a volume of 170 ul of

RPM1 alone. The reactions were done sterilely in 12 X 75 mm tubes at

40C for thirty minutes, agitating every 10 minutes. Following this

incubation, 200 ul of complement was added to each tube and the reaction

allowed to proceed for 60 minutes at 370C. After this step, the tubes

were spun down (1400 rpm, 100C, 10 minutes) and washed twice with PBS.

Finally, the cells were resuspended in complete medium and the viable

cells counted.

Column chromatography. MLC-MF fractionation was accomplished by

Sephadex G100 chromatography (100 X 2.6 cm column; pH 7.4 PBS, constant

flow elution of 15 ml/hour) of the MLC supernatants after 50-70 fold

concentrations (Diaflo ultrafilter PM10). Standards for molecular

weight estimation included bovine serum albumin, lysozyme and ovalbumin


HLA typing. Typing for the HLA-A and B locus antigens was done

with the microcytotoxicity technique (186) with antisera obtained from

the National Institutes of Health and from local sources by screening

with a panel of 30 cells of known phenotypes. DR typing was performed

with the double immunofluorescence technique (187). DR antisera was

obtained from Dr. Terasaki's laboratory at the University of California

at Los Angeles.


I. Studies of Stimulation in the Human MLC

Establishment of the Optimal Conditions for the MLC

Several studies were done to determine optimal and practical

parameters of the MLC. An analysis of the responder cell in the MLC

indicated that 30 X 103 MNL consistently provided responses which were

as good as higher numbers of cells (data not shown). In studies of the

stimulator cell concentrations, numerous cell titrations were used with

30 X 103 responders of which several are shown in Figure 3. It was

determined that 120 X 103 MNL induced an excellent response in most

MLCs. The use of this responder/stimulator cell ratio of 30 X 103/ 120

X 103 also had the advantage of being a small number of cells, which

would allow for a greater number of variables to be analyzed with the

limited cell numbers often obtained from some patients and normal

individuals. Optimal MLC responses were also found when pooled human

AB serum was used at 10-20% concentration.

Experiments were also performed to determine whether preincubation

of responders and stimulators significantly affected their respective

activities. As shown in Figure 4, the ability of cells to respond

slightly increased when they were preincubated for 4 days. Cells left

in culture for 2 to 4 days and then irradiated, showed an enhanced abil-

ity to stimulate in comparison with fresh cells. On the other hand,

Figure 3. Varying numbers of stimulator cells in MLCs. Each line
represents a different MLC combination in which the stimulator cells
were used at four cell concentrations (30, 60, 120 and 240 X 103 cells).
Each point represents the mean of triplicate values.








1 -

30 60 120 240

Stimulator Cells ( x18-3 )



o o0- 1 4-)
C. A -C 0a) C
U) S- U) S.- *r
a0 0 0 C
4- (1- 0-+ a) 4 .

> .E -C *r- U
o to
c 0o "0
4-) a)

CU MO 4- 0
-0 E c: 4-
4-) *- 0 o-
4r- 4- =

r- o- 0)
a) 4- 04-) C
)> 4-)0 t -
I- (A fu V) -0 I-)
= (a =
S-. -0 S- 0CL
S*r- 0) S-
C) 4- > U *r- 0
... *r- S-

3E 20 >,=

lC -. >C

C0 -) -40
4-)4- S- La)
(o 0) ro a) -
-r- O -0 4-)
3 a) E
*- a1) C 4- C 0

4-) (o M.
a) S-0 C S- -o

*C = 4-
C 4-S- a) 4 -)
U) *C- r- to 0) *
0) *- U S-

01CCUa0 tCl
> C oa cC -c >)

0. 4- + 1

S- U S- a o)
in TI 0 3 3
C )1 4- 4- 1 -
O 0- .C 4- C ro
a) V) a) 1->

0 I- a)
*-- 4- L.C= C -O 4--
4- 0 E *- 'to
ta 0 4- U
.- > U >) 4-)4*r-
34- 0 4 0 --
S*- T 3a) -0 a-
C 4-' *-
*r- *r- in ) 0) -
0) .10 m j .4-)
S. Co 3 0 I-
CL 4- r- 4-
a) n 0) 0
4- .= *- -0 u *
0 4- .C 0 VC

a)S. 0 '
4-e u, a)o
- o aW a)
SQ.I- i- C 0 .C

u u00 0 3 n
S- -0 4-)
C- t- -= a) *r- In C
n 3 r- s- 0
Wcua) (i oWa)
S- S- s- C UL -0 a)
= = 4-- C S-
o7) 0M 1 D 0 CL
*- *r- 4- a) .C L 0
LL. L- 0 U IF- V) LS




c t6 ( c
CV () cm cu

( E-0TX ) Wdo








irradiated cells left in culture had less ability to stimulate than

non-irradiated cells left in culture for the same time period.

Studies of Individuals (Patients and Controls) with Plant Mitogens and
Mixed Lymphocyte Cultures

Experiments were initiated to attempt to evaluate the general

immune responsiveness of patients and normal individuals. Peripheral

blood MNL were isolated and used simultaneously in both MLCs (against a

panel of ten stimulators) and in assays with the mitogens, PHA, ConA

and PWM. In all cases, patients were analyzed with normal individuals

as controls and often the same patient-control pairs would be studied

on several occasions. Figure 5 shows a representative study of the

responses of a patient and control in three separate experiments using

the 10 stimulator cell-panel MLCs and the mitogen assay systems. It

can be noted that a significant degree of variability exists from one

experiment to the next when using the mitogens, ConA and PHA, while the

smaller responses seen with PWM were less variable. The mean of the 5

highest MLC responses was used as the method to represent the general

alloresponsive state of the individual. In the study shown and in

others performed, the MLC was consistently less variable than the

mitogen responses. For example, by calculating the mean of the 5

highest MLC values in 3 experiments, Figure 5 shows that the patient

(open bars) was always a lower responder than the control. PWM did in-

duce a lower response by the patient in all 3 experiments, however this

consistency was not found in other studies. ALso apparent is the lack

of correlation between the MLC and the mitogen response. In many situ-

ations, the highest responses to mitogens by an individual could not

be associated with a similar high MLC response by his cells.

tL C 0- WU C i
4- 4- C *r- ro i -
C O 3 0 Q -

S- w) C 4-) u 4-) 0

QL.- aL c W3 4- 4- )
x S- (A M.0 0 (A cc
I c) a *- a)
CL (A c C cn 3
-P-' W WO0 U W 0*-
1E ( E 0.--'4
S- -) 4- ((A (n
0 0 J CU- C S 4-0

SS- C -0(A 4-' C
an (o a) ( aA
L. (l 4- 1o 4-) 1 S-
0U 0) 0 0 S *

W (l S.- a- ) 4-

=c a a -c a W -
4-J a) S- (0- S- CU "
0L- 0 n CL
c 1 C *- -C
A) n a)0 V) 0
4-) c) c>-l S.-.
r000 CC

( C: -L (0 ( A

-C- (U 4- > O ) *4-
B a ro-

C- O 4-+ C 0(-C )
*- O0 U) 4- 4-
> 4- a) E En
*- c: c *C C 4- W
-o o _C) W &- cu 0 -
C F 0- -0 Q) M
*- V))C4-L'C
-0 (n X 4-) C &-
O C Lsu M( a
S e0o^ (U E (D 3
+4-1 E S- E
4-3 E 0 a) -J
4- C 0- 4- -C a)
O0 C .CE:

V 4- ) S- S.-
> ro -S 3 0 >i W

(/) to r- C
mI 1) (a 0
C C L 3 0
4- 0 WU W 4- (nA
0 L)- .CE & 0 U U
0I .... -1- 3
OZ < C "C a a

EW .c .c
.C = 0 WI- I- =
C u (a. -P
-0 0 C)- C -I-
C -1 +-1 M: cD
4-C 4- M *

UC -I .
0 4- *r- C

0. 0 C: E -
E Ut >- 0W) -- a 4-

> o0 i. a. (
- )- U ) 4- 0 (A
4-) C: 3 S- (U O-

a) S-+ s- n a 3 o
+ ( 0 C r- = -
>) a)- 3 :3 C- (0 M
U-j E >

M- iW .C *-

S) > aU (n c 3
x a) m = a) s. -
*r- (A 4-3 -C a *C

S- C 4-) >,+-2
= 3 L(0 0 Q
Lo 0) C: a) -c- + l

Q) 4- S- c* L) C
S- ) ( 01 a S) a) a)
:3 v)4--- 0j u3
OT *-4- -) CL C -
*i- -C to X .- 4-
LL- F- 0 o ()o > o

-7.- 0

MLC Stimulation by a Panel of Unrelated MNL

ESRD patients and controls were evaluated for their ability to

respond in MLCs to panels of stimulator cells (frozen). Figure 6 shows

that responder 1 (open bars) was clearly a lower responder than re-

sponder 2 (striped bars) to every stimulator cell tested. Analysis of

these and other experiments allowed the observation that populations of

MNL from many individuals induced either high or low stimulation with

most of the other responder cells used in that MLC. This is exempli-

fied in Figure 6 in which cells which induced high stimulation in

responder 1 were also a good stimulator for the other responder.

Statistical analysis of the responses of these two individuals (Figure

7) in 4 independent MLCs to the same panel of cells indicated that in

fact there is a correlation of both responses to a given stimulator

cell. Since both responders were reactive to each other in MLC, oc-

casional similarities at the D locus with a given cell of the panel

would be expected to produce less stimulation in one responder but not

in the other. The results shown in Figure 7 suggest that this is not

the case and also that the degree of stimulation was unrelated to the

concentration of monocytes or B cells. These results, which were

representative of a trend seen in three other similar experiments,

suggested that cells display a high or low general stimulatory activity

to unrelated cells.

MLC Stimulation by HLA-Identical Cells to a Panel of Responder Cells

The following experiments confirmed the previous suggestion and

established that the different capacity to stimulate in MLC was unre-

lated to different degrees of HLA-D disparity.

Figure 6. Mean of 4 separate MLC responses to a panel of cryopreserved
stimulator cells and a pool of those cells by an ESRD patient (open bars)
and a control (striped bars). Each letter designates a stimulator cell
(frozen) used in each experiment, and each bar represents the mean +
S.D. of the 4 MLC responses to each stimulator cell.






a -
X= 50







Figure 7. Correlation between the responses of an ESRD patient (Re-
sponder 1) and a normal control (Responder 2) induced by a panel of
cryopreserved normal stimulator cells. The correlation observed (r =
.62) was at the limit of statistical significance (p = .051). Each
letter refers to each individual stimulator cell, X being a pool of all
of them. The values that they represent in the graph are means of 4
independent experiments performed with the same participants at dif-
ferent times. The coefficient of variation was less than 20% except for
X versus responder 1 (25%), E versus responder 2 (26%) and F versus re-
sponder 2 (28%). Autologous controls were under 2500 cpm. The re-
sponse of responder 1 stimulated by responder 2 was 20,000 to 38,000 cpm
in the different MLCs and responder 2 versus 1 was 31,000 to 49,000 cpm.


Monocytes % Nd. 28 22 26 26 36 20
B Cells % Nd. 8 9 4 5 4 7

45 50

The first experiment shown (Figure 8) compared the ability of

cells from 2 normal HLA-identical siblings to behave as stimulators to

a panel of ten unrelated individuals as responders. Neither of the

siblings responded to each other in MLC, indicating identity at HLA-D.

It can be seen that for every responder tested, one of the siblings was

a consistent low stimulator whereas the other was a reasonably good

stimulator. In the same experiment, cells from both siblings were

added together as stimulators with results consistent with summation of

individual effects rather than suppression by the low stimulator. Both

siblings were used in another MLC with two unrelated responders and

similar results were obtained, the same sibling stimulating low re-

sponses on both occasions.

The next experiment (Figure 9) shows the response of four unre-

lated individuals to stimulator cells from identical twins. Again, one

of the siblings stimulated all responders significantly better than the

other. In this experiment, monocytes and B cells were enumerated. The

proportion of monocytes, as estimated by myeloperoxidase staining and

Coulter sizing, was similar in both twins. The proportion of B cells

was slightly higher in the sibling that produced lower stimulation.

HLA and red cell antigens were identical for both twins (see legend for

Figure 9). Figure 9 also shows that the response of each sibling to

different stimulators did not correlate with their capacity to stimu-

late in MLC, a fact that we observed repeatedly in other MLCs.

Finally, Figure 10 shows the same phenomenon with two normal HLA-

identical siblings stimulating other members of the family. One of the

siblings was capable of inducing higher responses from all the family

members than the other sibling. Identity at HLA-D between the siblings

Figure 8. Stimulation by two HLA identical individuals to a panel of un-
related responders. Each letter designates a different responder cell.
Stimulator and responder cells were all obtained the same day of the
experiment. Each pair of bars corresponds to the stimulation produced
by the 1st (striped) and 2nd (open) sibling to a given responder cell.
Lines in each bar indicate standard deviation of triplicate cultures.
Both siblings were HLA-A3,29, HLA-B22,44, HLA-DR7. Autologous control
of 1st sibling was 743 cpm + 121 and the response to her sibling was
381 + 13. Autologous control of the 2nd sibling was 1061 + 218 and the
response to her sibling was 613 + 135.



- 16
CL 12



t I II*



Figure 9. Stimulation and response of two normal identical twins versus
unrelated cells. Top: Each pair of bars indicates the stimulation pro-
duced by the cells from the 1st (striped) and the 2nd (open) twin to
the corresponding responder. Bottom: Each pair of bars indicates the
response of each twin to the corresponding stimulator. All cells were
obtained the same day of the experiment. Lines in each bar indicate
standard deviation of triplicate cultures. Both twins, in addition to
obvious physical and behavioral similarities, shared HLA-A11,26, B14,22
DR2,3 antigens as well as red cell antigens (ABO, Rh complex, MNS, P,
K, Fy and Jk). Autologous control of the 1st twin was 2400 cpm + 1122
and the response to her sibling was 3015 + 991. Autologous control of
the 2nd twin was 1521 + 166 and the response to her sibling was 1874 +
738. Myeloperoxidase positive MNL in the 1st and 2nd twins were 6 and
9% respectively, monocytes quantitated by Coulter sizing were 9 and 11%
and B cells were 9 and 5% respectively.



1 I28

x 96


A a






0 60


Figure 10. Stimulation induced by two HLA-identical siblings on other
members of the family. A is an ESRD patient, B is a patient's sister,
C, a sister, and the control is an unrelated individual. Sib 1, a
sister, and Sib 2, a brother, are the two HLA-identical siblings also
used as stimulators. The solid bars are stimulation by sibling 1,
striped bars are stimulation by sibling 2 and open bars are stimulation
by the unrelated control. Lines in the bars are standard deviation of
triplicate cultures. The haplotypes are: a:A1,B8, b:Aw31,Bw35, c:A2,B7,
d:Aw24,B7. DR typings were not done.




S 32


A B C Control Sib I Sib2
a/c b/d a/d b/c b/c

used as stimulators was established due to a lack of reaction in MLC

between them. Thus, the different stimulatory activity of HLA-identical

individuals is seen not only with unrelated responders, but also with

one and two haplotype mismatch intrafamilial combinations.

Individuals with "High" and "Low" Stimulatory Ability

The sibling which induced low stimulation in Figure 8 is a labora-

tory technologist that participated in a number of MLCs as a normal

control for ESRD patients and relatives. The relative response induced

by this person's cells on the responders of each MLC (it is actually

"relative stimulation") is shown in Table 1 (Individual 2). It can be

seen that in six of the seven MLCs the "relative stimulation" was less

than 1.00, suggesting that this individual is a low stimulator. In

contrast, the same analysis performed in another individual that was

also studied in several MLCs as a control showed that the relative

stimulation was higher than 1.00 in six out of nine MLCs (Table 1,

Individual I). A third individual analyzed in the same manner produced

results intermediate between the other two.

Responses to Stimulator Pools

The existence of high stimulators was also shown in experiments

comparing the response to single stimulators with the response to pools

of two stimulators. The response of cells (30 X 10') from one individ-

ual was tested with a panel of unrelated stimulator cells (60 and 120 X

103). The same responder was also tested with several pools of two

stimulator cells (60 X 103 each) in a way that if the stimulators are A

through K, the pools were formed by A-B, A-C, A-D, etc. Table 2 shows


Individual 1 Individual 2
Experiment #(a) Relative Stimulation Experiment # Relative Stimulation

1 2.09 + .48(b) 10 .83 + .21

2 2.29 + 1.20 11 .45 + .01

3 .84 + .21 12 .52 + .27

4 1.14 + .09 13 1.67 + .80

5 2.95 + 1.30 14 .32 + .21

6 .53 + .10 15 .77 + .04

7 1.30 + .38 16 .82 + .25

8 .81 + .08

9 2.26 + .24

Average(c) 1.57 + .83 .76 + .44

Experiments with Individuals 1 and 2 were not performed simultaneously.

This value was obtained in the following way. The MLC was done with a
patient (A), relative (B), control (C) and individual 1 as a second con-
trol (D). Results were as follows:

Response of A to DX: 56,604 cpm (a)
Response of A to either B,C or
pool, whichever was higher: 29,756 cpm (b) Ratio (a/b) = 1.90

Response of B to Dx: 67,044 cpm
Highest response of B to A, C or pool: 38,452 cpm Ratio = 1.74

Response of C to Dx: 54,658 cpm
Highest response of C to A, B or pool: 20,585 cpm Ratio = 2.65

Avg + S.D. = 2.09 + .48
The relative stimulation produced by Individual 1 and Individual 2 was
significantly different (p = 0.037) in a two sided t test for identity of
population means.



- CO 00
D 0 CM
i- CF L-
cn in i










c (

I- -





00 -o
CM CM 0)

00 .0 a-

S u LA

- r ~

a) -) CD Li

CM 00



m cr

CM oC -r -C
- CM c\j

tD r-. C)
4- c


L.. CD =I -

00 r m n C C Ln

Cnm(V~IDCO~-CO r--

SCO C\JN O 't 0 o r- CO o inO
c\j Lo oD oji r" m m e- m -
o kO Lt r O )- O O0 LD p

to Ul ko to t.0 r~ ko r 0o ko





Ex 0
4- U



0) S0-


S S-

C a)

0 <

4 r-
C2 3

r- 4


-0 >-


C) ,

cn S-
0 C


(A c c

SF o

(A -
~ 4-)

(CU *0
3 .C)

i n O 4 0
u -0 -0

1/1 i-
34- 10

- (A







x --

C0 D 0
- 0
T 0m Lt)
o0 m0 L.O)
0C) CM m







one representative experiment out of three performed. It can be seen

that while most of the pools stimulated more than the individual stim-

ulators alone (presumably as a consequence of additional HLA-D dis-

parities) there was one individual (I in Table 2) which produced a

stimulation higher than most of the pools studied. These results show

that one good stimulator, with a maximum of two HLA-D disparities with

the responder, can induce a response higher than that produced by three

or four disparities. Individual high stimulators were also more effec-

tive than pools of 10 stimulators. Table 3 shows twenty MLCs in which

the best of 10 individual stimulators was compared to a pool of all

them. In each MLC there was at least one stimulator that stimulated

equal or better than the pool.

II. Investigations on a Mitogenic Factor Induced During the MLC

The above results indicated that the ability to stimulate in the

MLC was possibly influenced by non-genetic factors present during the

course of an MLC. One of these non-genetic variables that was investi-

gated was a blastogenic or mitogenic factor generated during the MLC


Effect of MLC Supernatants (MLC-MF) on Normal Resting MNL

The initial studies on MLC-MF utilized supernatants collected on

day 5 from MLC microtiter plates. These supernatants were added to MLC

(e.g. ABx) combinations to see if there.was an increase in the response,

especially in the case of low responders. The results shown in Figure

11 are typical of what occurs when MLC supernatants are added to allo-

geneic MLC combinations. In most cases there is no effect when


Pooled Stimulators Highest Response to Pooled/Highest
Experiment Responder (cpm) One Stimulator (cpm) Ratio

1 A(a) 37807 51548 .73

2 A 42976 46238 .93

3 B 54620 72215 .76

4 B 46915 58660 .80

5 B 77662 83961 .92

6 B 48870 84811 .58

7 C 40019 45234 .88

8 C 56157 59020 .95

9 D 43265 43496 .99

10 E 12979 12325 1.05

11 E 21910 32194 .68

12 F 24027 55254 .43

13 F 23456 43089 .54

14 F 31292 51524 .61

15 F 22326 38035 .59

16 G 92948 99434 .93

17 H 80176 75416 1.06

18 H 49738 56262 .88

19 I 74932 73097 1.03

20 I 26663 38166 .70

in different oppor-

(a) The same letter refers to the same individual studied
tunities. Stimulator cells were cryopreserved.

Figure 11. Effect of MLC supernatants on autologous and allogeneic MLCs.
MLC supernatants added at 50% concentration to MLC combinations (ACx
and BCx) and to the autologous controls (AAx and BBx). The open
bars represent the values when no MLC supernatant was added and the
striped bars the values seen when MLC supernatant was added. Each bar
represents the mean of triplicate values + the S.D.


Responders / Stimulators

measured at 6 days and rarely is the response augmented. However,

there was a greatly increased response at 6 days in the autologous or

control combination (AAx) when the MLC supernatant was added (Figure

11). For example, the response of AAx is above 20,000 cpm in the

presence of MLC-MF and approximately 500 cpm without.

These MLC supernatants were also tested for their capacity to

stimulate MNL from a patient who had Severe Combined Immunodeficiency

(SCID) and MNL from a patient with hypogammaglobinemia. The results in

Table 4 indicate that MNL from the SCID patient failed to respond MLC-

MF and also in the MLC. In contrast, there was no deficiency in the

ability of the SCID patient's cells to stimulate in MLC. The MNL from

the patient with hypogammaglobinemia responded well in MLC and to MLC-

MF (as did the controls) and also served as good stimulators in MLC.

The effects of the MLC supernatants on normal resting cells was

the impetus for the work described below.

Establishment of Conditions for the Production of MLC-MF

The MLC-MF preparations made in plates used in the initial experi-

ments had the disadvantages of having low activity, of small volumes,

of being tedious to procure, and of taking 5 days to obtain. Studies

were thus planned on improving the preparatory conditions, volume and

activity of MLC-MF. In a time course experiment (described below,

Figure 16) it was found that the peak of MLC-MF appeared from 2 to 3

days following initiation of the MLC. Using this information, "macro"

MLCs were set up in 16 X 100 mm tubes using variable numbers of re-

sponders to stimulators in different volumes and each combination was

run as either a 0, 1 or 2 way MLC (0 way = both cell populations




A = SCID Patient
B = Patient with Hypogammaglobinemia
C = Normal Control
D = Normal Control

Combinations cpm + S.D.(a)

A(30) + Ax 420 + 147
+ Ax + MLC-MF 940 + 655
+ Cx 397 + 98
+ Cx + MLC-MF 401 + 15
+ Dx 251 + 83
+ Dx + MLC-MF 242 + 120

B(30) + Bx 804 + 59
+ Bx + MLC-MF 3,335 + 38
+ Cx 6,281 + 3,340
+ Dx 18,294 + 173

C(30) + Cx 985 + 323
+ Cx + MLC-MF 7,999 + 624
+ Ax 6,789 + 1,591
+ Bx 1,710 F 1,029
+ Dx 19,309 + 7,911

D(30) + Dx 600 + 17
+ Dx + MLC-MF 3,368 + 1,259
+ Ax 3,388 + 1,114
+ Bx 4,333 + 2,740
+ Cx 12,961 +2,785

(a)cpm = counts per minute + standard deviation
MLC-MF was added at 50% concentration

irradiated; 1 way = one cell population irradiated; 2 way = neither

population irradiated). Table 5 shows that a significant increase in

the MLC-MF activity of MLC supernatants was obtained in most of the

tube combinations when compared to the plate supernatants. Of special

significance were the 5 ml volume tubes with 3.0 X 106 cells of each

cell population (culture no. 8, Table 5). The high activity obtained

in the one-way MLC combination showed that a large volume of MLC-MF

could be prepared by setting up multiple tubes in the same combinations

with an accessible number of cells. Also of interest is the high MLC-

MF activity seen in 0-way combinations which indicates that irradiated

cells are very capable of producing MLC-MF. Two-way MLC combinations

actually contained the least activity. This was presumably due to

consumption of MLC-MF during the proliferation occurring in some cells

early in the MLC.

The supernatants were also tested for their mitogenic activity on

cells which had been stimulated with PHA for 6 days. These activated

cells did not further respond to PHA but would to TCGF-containing

supernatants. As shown (Table 5), many of the supernatants induced

proliferation of these cells. An analysis of the association between

the actions of these supernatants on resting cells and on the PHA-

stimulated cells indicated a significant correlation between the two

activities (Figure 12).

Comparison of MLC Response with MLC-MF Activity of Supernatant

A study was performed to determine whether an association existed

between the degree of proliferation measured in the MLC and the amount

of MLC-MF that could be measured in the supernatant. Table 6 shows the


o ln cr,

Ln 0n m

in n r

to0 (O ) rn LA m
- co Ln C.0 CO
cr- i-*- d- rm CM


- C) CJ

+ 1+ 1+ 1

- C0 "



















' Ln co
CL n cO

en n

CM 0) L
Cn cn r-


M Lco


0- M I -
CN) C\j 1 00 (O

.+1+1 *+1+1 *+I+I
z z z
0 rr 3 LO
r- m rO m cn
C C- o Ln M
CJi :I- C4 C4 LA
c~j 'd- en m



Ln C\




* +1+1

S r-

CM en


0 0 0 0 0
00== ^0== c0 = 0= l= -=

-= = CLO= c\J= = =


ko 00
en -I
ft f

C\j CO


mO C


o00 C co
000 O~
o CM in

o V. i

i- C Ce) d LA 0 r


C= = -=

i- CM

CM= = d

i- C

CM o n oC en .

.0 CO r OC r- -

+ 1+1+1 ++ 1+ 1
m cn CMi t.0 r- 0
C\j m 0 [CO r-
L -. CM\ Cl C

CO o







+ I *




co r,

X CO x C x co
<.0 <



0= = 0== -= = CMJ

M o-



w ~ 0

E- 0
cu (D



0 CM -

r0 -0
3) 4l C

cu3 c u
r (o

3 I
W0 *- ri-

0 0

o a- re=

C- 4-c (

C-3 3- a
.0 4,- 1

4-' .
o-o E

= C
CO 0.-

0 =W
S0 *r-

3 -) S -

c c- 0 -
J s- a (U

3 (

C r(D E
4-' 4,-)

SrC) C-
Cr.S- C 0a
4- a .)1


(A +1
LL- >) -
)< M (a C 00
I -a = CM
CYn -j S. a
C- S- -s
o 000
(A 4- (
C>, .03

0 10 4-0) Q->

O C .)
1o tn w- -

4 3 CA

4-' 0m ) E
C C W 0 1.
.90 c Co

0 0- a3 0
Ss- E 1- E

W0 4

(A 0 rC 0.0
+r> 3 I

4 en 01 E C

) C0XL

4- 0( M- C1
0) 0a 4-

4-3 U (A
4-) '- W (C
- ( 0 -r C)
)D a) ) 0 CO

0,. 3 +E> <:
-7>r O C".
-E u00

+ zJ a C *'1
So0 C -C C0

4O 4- CL)

C0 I- 0 0 (04+-
S00 S-

0 S- C*U 0C

W> 4-+ 10 2 0-

u 0i S- 4 r- U
c CL 0c

-0 L- a- 0U
U CI S)I o- C
S 0 C w 0a
>.- C0CC

0- 0 l. S.- a0
4-- 1- c

OW L 0

~ (4- 'r C '-r


Figure 12. Statistical analysis of correlation between the activity of
MLC supernatants on resting cells and on PHA-activated cells (values ob-
tained from Table 5). The data show that there was correlation since
the linear correlation coefficient was significantly different from 0
with P = .00000021. Values on both the x and y axis are measuring cpm.

Correl. coaff. (r) .31
Sample size

t. statistic
Degr. freedom

p "alue
confidence leel ('%)

7 i*

one-sided test


two-sided test


x x


5 10 15 20 25 30 35 40 45 50

Activity on Resting Calls ( x18-3 )



Combination # MLC Response MLC-MF Activity of MLC Supernatants

Responder 1
1,902 + 1,086
14,904 + 3,582
6,884 + 1,150
10,392 + 4,541
1,009 + 142
4,222 + 2,715
7,428 + 2,804
1,689 + 1,036
1,197 + 688
12,870 + 1,135
13,529 + 739
11,840 + 902
1,808 + 727
8,382 + 3,672
4,215 + 1,449
1,709 + 508
519 + 250
8,370 + 1,713
8,763 + 1,834
5,386 + 1,085
895 + 261
2,487 + 218
3,269 + 1,902
3,837 + 1,737
3,597 + 403
6,763 + 2,269
9,663 + 1,416
7,114 + 2,005
4,146 + 2,538
5,167 + 1,051
3,496 + 1,616
2,062 + 315

Responder 2
899 + 360
11,942 + 3,004
10,012 + 2,254
5,256 + 234
1,550 + 797
4,228 + 503
1,644 + 940
1,387 + 71
777 + 168
11,276 + 1,171
12,652 + 3,011
7,386 + 3,052
420 + 189
3,827 + 1,551
4,360 j 3,076
1,804 + 770
1,031 + 663
6,036 + 926
6,986 + 1,727
5,327 + 2,928
1,210 + 450
1,310 + 582
1,649 + 663
2,252 + 1,998
1,183 + 695
5,062 + 1,184
4,707 + 1,489
6,479 j 1,208
1,387 + 927
6,344 + 3,678
3,929 + 1,962
1,457 T 644


were performed between family members and controls in muleiple

combinations of responders to stimulators. In addition to measuring
the response at 6 days ("MLC response" column), supernatants were
removed from each combination, sterilized and assayed for MLC-MF
activity on resting cells ("MLC-MF assay" column) from two individ-
uals responderss 1 and 2).



responses measured in 32 different MLC combinations (MLC response

column) and the MLC-MF activity of the supernatants from those MLCs, as

tested on resting MNL from two responders. A pattern of associations

appears such that high MLC cultures possessed high MLC-MF activity in

the supernatant and conversely, low MLCs had low MLC-MF present.

Their association was directly tested for statistical significance

for both responders (Figures 13,14). In both cases, correlation was

established between the MLC response and MLC-MF activity. This indi-

cates that the degree of blastogenesis seen in the MLC may be the

result of the level of MLC-MF produced, which may differ from one

person to the next, or alternately, the level of MLC-MF may be the

result of the degree of blastogenesis. In any case, an association

between the two appears to exist.

Titration of MLC-MF Activity

A MLC supernatant known to possess good MLC-MF activity was seri-

ally diluted to determine at what dilution no response would occur. As

indicated in Figure 15, this particular MLC supernatant induced signif-

icant mitogenic doses to a 1:32 dilution. In general, most MLC-MF

assays were run at a 1:2 dilution.

Time Course Response of Resting and PLT Cells to MLC-MF

A series of experiments were designed to determine when the peak

response occurred to MLC-MF by resting cells and 10 day old MLC-primed

(PLT) cells. Figure 16 shows that both cell populations respond to

MLC-MF with the same kinetics as they do to alloantigen stimulation.

Resting cells, in the example shown, had the peak proliferation to MLC-

MF at 6 days, the same peak as seen in the MLC. An analogous situation

Figure 13. Statistical analysis of the correlation between MLC responses
and the MLC-MF activity of the supernatants obtained from those MLCs (all
values obtained from Table 6). Figure 13 represents the MLC cpm versus
MLC-MF cpm as detected on responder 1. Each number on the line represents
the combination number described in Table 6. There was significant cor-
relation since the linear coefficient was significantly different from
0 with P = .0000027.

3. 7;

one-sided test

?9.9a 14

two-sided test


Correlatlon BIetwmn MLC Reponm and MLC-WF Rotivtty (Reeponw 1)

25 59 75

MLC cpa ( xl8-3 )

.3e3r. r' io-3m

z n' i aden.:* 1 .: 1 4 ':





o~oi ii

. 7a

3A. ad

on*-Siaa tall.

d.a 6643

two-111*. to%



25 -

2B 0

2 15 i-

MLtC c ( x18-3 )

Figure 14. Statistical analysis of the correlation between MLC responses
and the MLC-MF activity of the supernatants obtained from those MLCs
(all values obtained from Table 6). Figure 14 represents the MLC cpm
versus MLC-MF cpm as detected on responder 2. Each number on the line
represents the combination number described in Table 6. There was
significant correlation since the linear coefficient was significantly
different from 0 with P = .0000086.

la o a 1
35 a
1a 14
.-^ -'^" ^ Ia *
rr^ .* f
I---- ----- ____________________

Figure 15. Titration of MLC supernatant. MLC-MF activity (as detected
on resting MNL) of serial dilutions of an MLC supernatant. Each point
is the mean of triplicate values + the S.D.

1:2 1:4 1:9 1:16 1:32 1:64 1:129 1:256
Dilution of MLC Supernatant

F- 00o

M3 E

C L0 .

c I
4- 1
10 L)

s- 4-

CL "0

0 L...J

3) -I


---J L' L
a 0 4-'
S..4 0 >

S: Sro

c o >, a)

-I -I ',
Z Za C0 4-)


C C 10 e-
4*- r- E 0Q
4- 4.. *- *.-
l 4-' S...

In Ln Ln -C
a a) cu 4-
S- a0


44- > a
0 -Q -

u O tyl CL

C in C)
4a) 4-)

SF- S-- C
S.0 4)-)

a) 0-, Da

S 4-~a)

o *v- U
E (A 4-'


*r- Sr- 0.



Lai cur

_i /

w /

o y
a- >=


o U,

O u,



C-0Tx ) WdO

exists with PLT cells. These memory cells respond with secondary kine-

tics (2-3 days) if exposed to the same alloantigen which was used for

the primary stimulation. Similarly, culture supernatants with MLC-MF

activity induced a maximal proliferation 2-3 days following the initia-

tion of the incubation.

Enhancement of the Response to MLC-MF by Irradiated Autologous Cells

As mentioned before, the MLC-MF activity was found to enhance the

autologous MLC (AAx) combinations but not the allogeneic MLC. The

role of the irradiated cells appeared important since the response of A

alone to MLC-MF was much lower than the AAx combination (an example is

shown in Figure 19). A study was done to determine the optimal number

of irradiated cells which was needed by incubating 30 X 103 MNL with

varying concentrations of autologous irradiated cells (Figure 17). In

addition, the same number of non-irradiated, autologous cells were

added to other wells to compare the responses. As indicated in Figure

17, wells which were given non-irradiated autologous cells produced

slightly higher responses than those given irradiated cells at the

lower concentrations. However, when 120 X 103 irradiated cells were

added to the 30 X 103 MNL the response was equal to that of 150 X 10'

MNL. Since the irradiated cells themselves were not proliferating to

the MLC-MF, it was clear that they were augmenting or "helping" the

response of the non-irradiated MNL.

Determination of the Optimal Radiation Dose for the "Helper" Effect by
Autologous Cells

The observations that irradiated cells were enhancing the response

to MLC-MF prompted experiments designed to determine whether the dose

C )

4- 4-) C CL

o -

-0 0
vn ..- -o3 L
L 0)S

.1 V)- -
E L S- -

c -o 1/

0 0 *

*O *( 0-

S *S 0)

0 E N

S- ca- c+
&- 0 I -
aC 0 I >
0 S- cV)

0 5..C (0

0 C>

Lo 4-0 0
0- -
-10 S- -

4- O 4- 0

*rO -OO

s- -o E .

_j.. C(A
r 34-' 3

L L(n C:

(0 U Uo

*.- C L. .S- VI
1 : 'w L -

=0 Ca 4-J

S- C I -
s-- 5-. s-

S-- L2 0

0 -'*- 0) 0

4 r- *, I-
3 O 1a0 rE

4- LS 0 A 3
D *i- (J -0
0 -I
*r- -0 L. .C -

C -C0




B 0


H c

( E-BIX ) Wd3


of irradiation used until then (3000 rads) was the optimal dose for the

helper effect. Cells were exposed to a series of radiation doses and

then added at a concentration of 120 X 103 cells to 30 X 103 auto-

logous, non-irradiated cells responderss) in the presence of 50% MLC-

MF. The response measured at 6 days (Figure 18) indicates that the

optimal radiation dosage for helper activity occurred from 2600 to 3200

rads. These cells alone did not respond to MLC-MF.

Using these experiments as a basis, all MLC-MF assays done in

which autologous irradiated cells were added to the 30 X 103 responders,

used a "helper" concentration of 120 X 103 cells given 3000 rads.

Adherent Cells Provide the "Helper" Effect in the Irradiated Cell

The observation that 120 X 103 unseparated, irradiated MNL aug-

mented the response of 30 X 103 autologous MNL allowed us to study the

role that monocytes played in the response to MLC-MF. The first pair

of bars in Figure 19 represents the response with and without MLC-MF of

30 X 103 nonadherent MNL. This cell population contained a very small

percentage (3% by electronic sizing analysis [ESA] and 0% by peroxidase)

(Figure 20) of monocytes, in comparison to the unseparated MNL popula-

tion (23% by ESA and 10% by peroxidase). The addition of 120 X 103

irradiated, unseparated MNL significantly enhanced the response to MLC-

MF. In contrast, the addition of the same number of irradiated non-

adherent cells did not affect the response. A repeat of this experi-

ment is shown in Figure 21 (Coulter analysis in Figure 22) with simi-

lar results.


C\J =
- I

0) .C

s- CC

t- 4-

t1 WI

.CO E=


*r- -0
> c +I

*- o

u 0 0

40 L) C
Qs i-

(U cv >
.- O
cM a)v

S- -
m- 0 vi

C0 4-)
04 C

c o L
ai C

-a E
-o In

.*- v C
( 0 .)

0t C
4- *- L)
0 +- S-

C- *.- 4-'
U -0 C

L4. Qr

.- U

LL 0) r




S ^
*-^-i- ts

\ [D +>
\ M 3

\n cr,
\ C3 c
( ---T --- W 5

Figure 19. Determination of the cells providing the "help" in the
iradiated MNL population. Open bars represent wells with no MLC-MF
and striped bars are wells with 50% MLC-MF. Ax(120) corresponds to
120 X 10' autologous responders. Anax(120) corresponds to nonadherent,
irradiated MNL which were added to responders. The first pair of
bars had only responders, with no autologous helpers added. The bars
represent the mean of triplicate values + the S.D.



m 15


E 10


Ax(12B) Rnax(128)
Cells Added

Figure 20. Coulter analysis of irradiated unseparated and NA MNL that
were used as helpers in Figure 19. Other data on these cells included:
Unseparated MNL had 8% B cells (Surface IgM), 10% peroxidase positive
cells and 23% monocytess" by Coulter analysis (top of Fig. 20). NA MNL
had 1% B cells (Surface IgM), 0% peroxidase positive cells and 3%
monocytess" by Coulter analysis (bottom of Fig. 20).


88 -

8.7 43.4 SS 78 1 5 5 112.1 13.2 147.8 4. 182.3
VOLUME (cubic microns)


99 -

se -

51 -

-c -

8.7 2G 1 43.4 6a.9 79 1 95.5 112.1 132.2 147.6 164.9 182.3
VOLUME (cubic microns)

Figure 21. Determination of the cells providing the "help" in the
irradiated MNL population. Open bars represent wells with no MLC-MF
and striped bars are wells with 50% MLC-MF. Ax(120) corresponds to
120 X 10' autologous responders. Anax(120) corresponds to nonadherent,
irradiated MNL which were added to responders. The first pair of bars
had only responders, with no autologous helpers added. The bars
represent the mean of triplicate values + the S.D.


108 -


B0 L



Ax(120) Rnax(128)
CELLS RDDED ( x1000 )

Figure 22. Coulter analysis of irradiated unseparated and NA MNL
used as helpers in Figure 21. Unseparated MNL contained 7% monocytes
by Coulter analysis and NA MNL contained 1% monocytes by Coulter








.7' 6a8 73.L1 a.5 112.1 131.2 t47.S 164.3 182.
VOLUME tcubic microns)






.743.4 6 8 a 73.1 t 5.5 112.1 l38. 2 147.6 164.3 182.3
VOLUME (cubic microns)

These results suggest that monocytes amplify the response to MLC-

MF although they may not be required to induce it. The possibility is

minimal that B cells were providing the help since in data not shown,

nonadherent cells from petri dishes, which contain many B cells and few

monocytes, provided little help when added to nonadherent responders.

Proliferation to MLC-MF is not Due to Additional MF Production

Since IL1, a monocyte product, can induce lymphocytes to produce

IL2 (171), it was next investigated whether the MF activity of MLC

supernatants was due to a factor directly mitogenic to MNL or to a

factor only capable of inducing MF production by the responding cells,

but not being mitogenic itself.

Two approaches were utilized to answer this question. The first

approach involved the use of adrenal steroids, which have been reported

to inhibit IL2 production but not its effect on activated lymphocytes

(169). A representative experiment is described in Table 7. MLC-MF

production was induced in the presence and in the absence of hydrocor-

tisone. A significant inhibition of MLC-MF production was seen at con-

centrations of 10-5 and 10-6 M. However, when hydrocortisone was added

after the MF was produced, the effect on responder cells was minimal.

Thus, since production of MLC-MF does not occur in the presence of

hydrocortisone, the mitogenic activity observed when hydrocortisone is

added to MLC-MF cannot be explained by the production of additional MF

by the responder or irradiated cells.

A second, more direct approach was also utilized to explore the

same problem. MLC-MF was incubated for 2 days at 370C alone and in the

presence of unseparated MNL (Table 8). These supernatants, when tested


HC Present During(a) HC Added to(b) cpm(c)
Generation of MLC-MF MLC-MF Exp 1 Exp 2

0 0 12,285 + 1,771(d) 14,538 + 3,109
10-5M 0 1,226 F 120 2,259 T 665
10-6M 0 4,707 4,415 + 537
0 10-5M 9,268 + 2,568 7,772 + 3,525
0 10-6M 12,601 + 4,588 10,722 + 3,420
No MLC-MF Added 2,937 + 1,705 2,044 + 788

HC was added at concentrations of 10'5M and 10^6M
the production of MLC-MF in several tubes.

at the start of

\Uj HC was added at the start of the MLC-MF assay to MLC-MF produced
in absence of HC.
(c) Counts per minute (mean of triplicate values + S.D.).
(d) All supernatants were tested in two MLC-MF assays (described in
Materials and Methods).

Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd