Defining guinea pig monocyte heterogeneity using cells separated by counter-flow centrifugation elutriation


Material Information

Defining guinea pig monocyte heterogeneity using cells separated by counter-flow centrifugation elutriation
Counter-flow centrifugation elutriation
Physical Description:
xi, 145 leaves : ill. ; 29 cm.
Noga, Stephen Joseph, 1954-
Publication Date:


Subjects / Keywords:
Centrifugation   ( mesh )
Macrophages   ( mesh )
Monocytes   ( mesh )
Phagocytes   ( mesh )
Reticuloendothelial System   ( mesh )
Pathology thesis Ph.D   ( mesh )
Dissertations, Academic -- Pathology -- UF   ( mesh )
bibliography   ( marcgt )
non-fiction   ( marcgt )


Thesis (Ph.D.)--University of Florida.
Bibliography: leaves 122-144.
Statement of Responsibility:
by Stephen Joseph Noga.
General Note:
Photocopy of typescript.
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 - 000373252
oclc - 10196709
notis - ACB2426
System ID:

This item is only available as the following downloads:

Full Text







Copyright 1983


Stephen Joseph Noga

Dedicated to Sigurd Johns Normann,

my mentor and friend


I owe a debt of gratitude to the many people who have

helped me throughout my graduate career.

I would like to acknowledge the sacrifices made by my

parents, Stephen and Betty Noga, who provided me with a

strong educational foundation from which to continue my

academic pursuits. I am deepy indebted to my wife, Bonnie,

for her constant love, support and friendship throughout my

graduate career. I have relied heavily upon her resilient

quality of inner strength as well as her wise counsel to

overcome many anxious periods incurred during the

completion of this dissertation. I would also like to

thank Lonny Seitz, my eighth grade biology instructor, who

nurtured my interests in biology. His genuine desire to

broaden everyone's perspective of the biological sciences

should serve as a role model for students and teachers


This dissertation would not have been possible without

the guidance of my supervisory committee which I sincerely

thank for their help. I especially wish to show my

appreciation to my supervisory committee chairman, Dr.

Sigurd J. Normann, for instructing me in the scientific

method, and in so doing has rightfully become my mentor.

His keen scientific intellect and objectivity, coupled with

his unquestionable integrity, have provided me with the

tools necessary to pursue a fulfilling investigative

career. More importantly, he is truly a friend whom I can

trust and depend on in times of need.

I would like to thank Dr. Roy Weiner for his

scientific critique, friendship, and his unlimited use of

the resources of the Division of Hematology/Oncology at

the University of Florida. In addition, he has played an

instrumental role in the attainment of my career goals and

has served over and above his duties as a member of my

supervisory committee.

I am grateful to Dr. Ammon B. Peck and Dr. Raul C.

Braylan for all their constructive criticism and direction

as members of my supervisory committee. Additionally, I

wish to thank Dr. Peck for his outstanding role in

constructing a viable and strongly competitive graduate

department in pathology and for his firm support of student

representation on the graduate curriculum committee.

The experimental data in this dissertation could not

have been completed in its degree of thoroughness and

accuracy without the technical assistance of Lisa Anderson

and Pat Tarantula. Not only was the work exceptional, but

the many hours they unselfishly contributed were done out

of friendship rather than obligation. I am also grateful

to Janet Cornelius for contributing her technical expertise

and more importantly for being my friend throughout my

graduate experience. Al Cockrell has been most helpful in

preparing the photographic materials for this dissertation.

I am indebted to Crystal Grimes for the many hours she

spent in the preparation and typing of this dissertation.

This document attests to her superior technical skills and

her attention to detail, which again were a reflection of

the friendship between us.

There are many friends who have contributed to the

completeness of my graduate career. I offer my sincere

thanks for their help and understanding. Cindy Bevis, in

particular, has unhesitantly contributed the time and

effort necessary to develop a long-lasting friendship.

This kind of friendship can never be attained passively but

is rewarded in its vitality. I also wish to thank Melissa

Elder for her warm friendship and guidance during our time

together as graduate students. Viren Shah deserves special

mention for being a loyal friend and technical advisor

throughout my graduate career. I am grateful for the

friendship of Dr. Raymond Hackett who has provided both

professional and personal counsel on inumerable occasions.

He has also taught me the value of brewing a good pot of

coffee. I would also like to thank Dr. Saeed Khan, Dr.

Noel Maclaren, Dr. Gerald Elfenbein, Dr. John Gudat and Dr.

Herman Baer for being my friends and for taking an interest

in helping me fulfill my career goals.

Finally, I wish to thank the Department of Pathology

for the use of their facilities and the grant support of

the Public Health Service (CA 29266 from the National

Cancer Institute) for making the research in this

dissertation possible.







Origin of the Mononuclear Phagocytes
Techniques for Obtaining and Isolating
Macrophages . .
Techniques for Isolating Monocytes .
Functions of Macrophages .
Macrophage Heterogeneity .
Monocyte Heterogeneity .
Theories on Macrophage Heterogeneity



Mononuclear Cell Preparation .
Counter-Flow Centrifugation Elutriation
Cell Volume Analysis . .
Histochemical Stains . .
Adherence . .
Percoll Gradients . .
Phagocytosis . .

* 1

* 23
* 25
* 26

. 30

. .
* .
. .
. .
. .
* .

Guinea Pig Anti-Sheep Erythrocyte Antibody .
Fc Receptor Assay . .
Antibody-Dependent Cellular Cytotoxicity
(ADCC) . . .
Tumoricidal Assay . .


Isolation of Mononuclear Cells .
Counter-Flow Centrifugation Elutriation (CCE)
Histochemical Staining . .
Adherence Characteristics of Guinea Pig
Monocytes . . .
Percoll Gradient Separation . .
Phagocytosis . .
Fc Receptors . .
Development of Fc Receptors in Culture .
Antibody-Dependent Cellular Cytotoxicity .
Tumoricidal Activity . .



Separation Procedures for Isolating Guinea
Pig Monocytes . . 103
Functional Properties of Guinea Pig Monocytes 107
Monocyte Heterogeneity . 114
Synthesis . . 118



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



Stephen Joseph Noga

August 1983

Chairman: Sigurd J. Normann, M.D., Ph.D.
Major Department: Pathology

Counter-flow centrifugation elutriation (CCE)

separates human blood monocytes into 2 subpopulations that

differ in modal volume, Fc receptors and tumoricidal

activity. Further studies would be facilitated if an

animal model could be found which displayed a similar

heterogeneity. The guinea pig was selected for examination

because its hematopoietic system is similar to man.

Greater than 95% of guinea pig peripheral blood monocytes

were recovered by centrifugation over Ficoll-Hypaque (F/H)

of specific gravity 1.101. Post-F/H mononuclear cells were

injected into a Beckman elutriator chamber at a rotor speed

of 3,000 rpm and a loading flow rate of 10.0 ml/min. Cell

fractions were collected at 2 predetermined flow rates of

24 ml/min (400 ml collection volume) and 28 ml/min (200 ml

collection volume). Further, the 24 ml/min fraction was

collected in two separate aliquots (200 ml each) and

designated 24A and 24B. Residual (larger) cells still in

the chamber were collected by turning the elutriator rotor-

off (R/O) and continuing medium flow. The smaller guinea

pig monocytes were found in the 24A fraction and the

largest in the R/O fraction. Further functional character-

ization was performed on the small (24A), intermediate (28

ml/min) and large monocytes (R/O).

Small monocytes had a modal volume (MV) of 283p3 and

were weakly adherent, non-specific esterase and Fc receptor

positive, moderately phagocytic, but negative in acid

phosphatase and ADCC activity. This fraction contained the

highest level of native tumoricidal activity when tested

against the xenogeneic P-815 mastocytoma. Intermediate-

sized monocytes (MV 317p3) demonstrated moderate

adherence, non-specific esterase, acid phosphatase, ADCC,

and phagocytic activity but were negative for Fc receptors

and tumoricidal activity. Large monocytes (MV 354p3)

were strongly adherent, esterase positive, highly

phagocytic, Fc receptor positive and active in ADCC. They

showed moderate reactivity for acid phosphatase and weak

tumoricidal activity.

We concluded that heterogeneity does exist in the

guinea pig monocytes and that their reproducible separation

by CCE provides a model system with which to further

characterize the biology of circulating monocytes.


A century ago, Eli Metchnikoff introduced the word

macrophage to describe those cells in fixed connective

tissue which were mononuclear and had the properties of

ameboid motion and phagocytosis (136-138). Subsequently,

Landau (111) and Aschoff (10-11) noted that non-granulo-

cytic phagocytes were widely distributed in the body based

on their staining affinity for vital dyes such as lithium

carmine and trypan blue. In 1924, Aschoff proposed that

these cells in different body cavities were related not

only by morphology and staining characteristics but also by

function; he called these cells the reticuloendothelial

system (11).

In present day usage, the term mononuclear phagocyte

is often used to describe the major cell group originally

ascribed to the reticuloendothelial system. The modern

concept of mononuclear phagocytes includes cells from all

organs and cavities of the body. Fixed tissue macrophages

or histiocytes are relatively immobile phagocytes found in

connective tissue, bone marrow and all organ systems. In

some tissues, they constitute a significant percentage of

the total cells. For instance, the Kupffer cells are fixed

macrophages constituting one-third of the cells of the

liver (159). In addition, a population of mobile macro-

phages are widely distributed in the body as exemplified by

the circulating monocytes as well as pleural and peritoneal

macrophages. Macrophages are a major cell component at

sites of inflammation. These macrophages are derived from

circulating monocytes (76,216,218) of bone marrow origin

(227) and at the inflammatory site they may undergo

differentiation to a variety of recognized cell descendents

(199,200). In a chronic focus of inflammation, the cells

derived from circulating monocytes include macrophages,

epithelioid cells and multi-nucleated giant cells (200).

Although controversy still exists among investigators,

other cell types are strong candidates for inclusion into

the mononuclear phagocyte system. One such cell is the

Langerhans cell, a sparsely distributed, supra-basal

dendritic cell found interdigitating between keratinocytes

in the epidermis (66,98). The cells responsible for the

remodeling and reutilization of bone, the osteoclasts, have

been included also in the mononuclear phagocyte system

(12,32,54,157) as well as the microglial cells of the brain

and mesangial cells of the kidney (106,175,207). Other

investigators include the reticular cells of lymph node and

thymus as constituents of this system (68,87,224,225).

Indigenous or inducible properties which are shared by all

these cell types include adherence, phagocytosis, non-

specific esterase activity, the presence of Fc receptors

and lysozyme secretion (217).

Origin of the Mononuclear Phagocytes

It is currently accepted that the blood monocyte is a

direct precursor of the macrophage. Van Furth has

eloquently demonstrated with in vivo labeling studies that

monocytes leave the peripheral circulation and mature into

the various macrophages found throughout the tissues

(218). Monocytes have relatively long circulation times

with half-lives of 17 hours in mouse and 71 hours in man

(220). The majority of monocytes do not divide under

normal conditions and thus they represent the penultimate

form of the mature phagocyte (218, 221). Some investi-

gators believe that a small percentage of monocytes still

retain the potential for division after leaving the

circulation (215). At present, there is considerable

debate as to whether these cells represent a more immature

precursor cell, a specific subpopulation of monocytes or

derive from a reactivation of otherwise mature monocytes

due to factors in the microenvironment.

The earliest cell types which can be clearly

identified as precursors of the monocyte are the monoblasts

and promonocytes of the bone marrow. In mice, there are

estimated to be only 2.5 x 105 monoblasts and 5 x 105

promonocytes (219,221). Under steady state conditions, the

monoblast has a cell cycle time of 11.9 hours and gives

rise to two promonocytes (77). Promonocytes have a cell

cycle time of 16.2 hours and give rise to two monocytes

(219). When inflammatory states alter normal homeostasis,

both monoblast and promonocyte cell cycle times are

considerably reduced, leading to an enhanced efflux of

monocytes from the bone marrow and into the circulation.

It is difficult to identify the earliest progenitor

cells of the macrophage with present-day technology,

although indirect assays for their characterization and

enumeration have been developed. These progenitor cells

are identified by the colonies they form when plated in

semi-solid agar and are quantitated in colony forming units

(CFU) (76,77). The first step in the generation of the

macrophage is thought to be the division of an uncommitted

pluripotent stem cell (CFU-S) that gives rise to a stem

cell committed to the formation of either granulocytes or

macrophages. These colony forming cells are designated GM-

CFC. Metcalf believes that two types of progenitor cells

exist in bone marrow that give rise to the tissue

macrophage pool (134). The first progenitor, GM-CFC,

produces mixed colonies of macrophages and granulocytes,

granulocytes alone or macrophages alone under the

appropriate stimulation by colony stimulating factors

(CSF). The second precursor is committed solely to the

production of macrophage colonies (M-CFC) (30,114,116,117).

The latter progenitor cells are found not only in bone

marrow but also in peritoneal exudates, blood and other

organs. The significance of these macrophage-colony

forming cells (M-CFC) to the total macrophage pool is

unclear. However, this self-replicating pool of

macrophages is being investigated to determine if local

production of macrophages is a normal event. Volkman

argues for the existence of a self-replicating pool of

macrophages within the tissues which are not of bone marrow

origin (226). This recently described M-CFC could account

for this extramedullary source of macrophage precursors.

Two types of CSF have been identified (134). The

GM-CSF (syn.CSF II) cause differentiation of the formerly

described progenitor cells (GM-CFC) to either pure colonies

of macrophages or granulocytes or to mixed colonies as in

the majority of cases (20,27,135). Changing the

concentration of GM-CSF in the medium will alter the

granulocyte/macrophage colony ratio (133). The latter

progenitor cell, M-CFC, will differentiate into macrophage

colonies under the influence of GM-CSF (116). The second

CSF that has been isolated, M-CSF (syn.CSF I), interacts

exclusively with the GM-CFC precursors to produce

macrophage colonies (201,202).

Techniques for Obtaining and Isolating Macrophages

Macrophages are found in association with a myriad of

cell types in sundry anatomic locations. Isolating them in

high yield and purity is prerequisite to an effective study

of their functions and properties. Although a variety of

methods have been developed to isolate and purify macro-

phages, only two of these techniques will be described


A. Selective Increase in the Macrophage Population

A common method used to obtain workable amounts of

macrophages involves their selective induction at a

particular site by a variety of stimuli. Intraperitoneal

injection of thioglycollate, proteose-peptone, BCG, or

Listeria monocytogenes causes an increase in both the

absolute number and relative percentage of macrophages

within the exudate population (9,34,50,51,57,58). Intra-

peritoneal injection of Listeria also will cause an

increase in the absolute number and relative percentage of

monocytes in the peripheral blood without altering other

blood cell numbers (5,17). The major disadvantage of such

techniques lies in correlating results obtained with such

cells to cells obtained under steady state conditions.

Such elicited cells may not be indicative of macrophages

found under homeostatic conditions. Nevertheless,

considerable information on macrophage physiology has been

determined using such enrichment techniques. Very

commonly, exudates produced in this manner are subjected to

additional techniques to purify the macrophages.

B. Adherence

One of the earliest and still popular methods for

macrophage enrichment is based on the ability of

mononuclear phagocytes to adhere to glass or plastic

surfaces (33,145,161). Macrophage-enriched populations

obtained by inflammatory induction are often the starting

population of cells. This technique was first described by

Mosier who noted that a macrophage-rich, adherent popula-

tion of cells could be obtained after incubation of a mixed

leukocyte population on glass plates (145). The cells that

were washed off the plates were found to be relatively

depleted in macrophages and rich in lymphocytes. These

cells were designated as the non-adherent population.

Unfortunately, the macrophages had to be used as a

monolayer in the same vessel from which they were isolated

since they were firmly attached to the glass surface.

Later, it was discovered that trypsin or physical scraping

of the plate could remove cells for further study (161).

However, this resulted in poor recovery and low viability.

Subsequently, chelators such as EDTA or the anesthetic,

lidocaine, were used to detach adherent macrophages

(43,69,168,242). Macrophages were obtained in modest

numbers and acceptable viability for subsequent analysis.

Nonetheless, this approach was still not ideal as

functional damage and a biased selection of macrophages

often occurred. Further, cell types other than macrophages

reside in the adherent as well as detached population

(161). This prompted a search for better methods of

purification and recovery.

Newer methodologies take advantage of the ability of

macrophages to bind specifically to fibronectin-coated

surfaces through interaction with the fibronectin receptor

on the macrophage surface (19,167,178). Higher purity of

adherent macrophage populations is possible with this

technique compared to glass or plastic surfaces. The

adherent cells are easily removed with low concentrations

of EDTA as the fibronectin-macrophage interaction is

calcium-dependent (1,123). Although cells obtained in this

manner are ideal for studying some properties of phago-

cytes, recovery is neither ideal nor consistent. Further-

more, recent evidence points to the selection of particular

groups of phagocytes, leaving the less adherent macrophages

in the non-adherent fraction (91,206).

Techniques for Isolating Monocytes

Monocyte studies have not progressed as rapidly as

those using macrophages due to problems associated with

recovering monocytes in large enough quantities and

sufficient purity for analysis. Monocytes constitute only

2-8% of the white cells in the peripheral blood which also

contains vast numbers of erythrocytes and platelets (244).

The following methods have been developed to enrich for

monocytes and such techniques had to be developed before

analysis of this cell type was possible.

A. Density Gradient Separation Using Ficoll-Hypaque

The density gradient technique developed by Boyum was

a significant achievement that formed the basis of modern

methods for isolating the mononuclear cell population from

peripheral blood (22). Boyum used density gradients formed

by the high molecular weight polysaccharide polymer,

Ficoll, and the radiocontrast preparation, Hypaque. Human

peripheral blood, when centrifuged over a 1.079 g/ml

density gradient, accumulated lymphocytes and monocytes at

the interface between the plasma and the density gradient.

Ficoll caused the erythrocytes to stick to each other with

the resulting rouleaux formation increasing their specific

gravity. Consequently, the erythrocytes sedimented through

the gradient and pelleted at the bottom of the tube. The

hypertonicity of the Hypaque solution caused granulocyte

shrinkage presumably due to the inability of this cell to

appropriately regulate its water volume. The increased

density and decreased drag force of the granulocytes caused

them to be sedimented towards the bottom of the tube.

However, the lymphocytes and monocytes were found at the

gradient interface. These cells have a lighter density

than granulocytes and they are more efficient in

controlling their smaller volume of cellular water.

This technique was first developed as a means of

isolating lymphocytes from peripheral blood and the

monocytes were considered a contaminant. Later, however,

this technique was viewed as a method for the recovery of

monocytes. By eliminating granulocytes, a 5-10 fold

increase was achieved in monocyte purity. Even under the

most ideal conditions, however, monocyte concentration is

rarely greater than 30% (164,237).

B. Density Gradient Separation Using Percoll

A more recently developed technique for isolating

monocytes from peripheral blood utilizes an isoosmotic

suspension of non-toxic polyvinyl-pyrollidine-coated silica

called Percoll. It is used either as a continuous gradient

of increasing density (24,79) or as selected densities of

Percoll layered successively to form a step gradient

(64,208). Isopycnic centrifugation causes cell mixtures of

different densities to be separated as each cell type

localizes at the specific gravity of the Percoll equivalent

to its own density.

It is possible to isolate monocytes in high purity by

this technique using either whole blood (64,162) or the

mononuclear preparation obtained after Ficoll-Hypaque

separation (79). Use of whole blood drastically limits the

number and purity of monocytes obtained as the gradients

often are overloaded with cells. Mononuclear cell

preparations allow the recovery of more monocytes although

two separation steps are required with a resultant variable


C. Fibronectin-Mediated Adherence

Some, but not all monocytes adhere to glass and

plastic surfaces (161). Consequently, problems with

differential adherence and low yield result. Microexudate

and serum-coated plates both contain high levels of

fibronectin and have been used to isolate monocytes in

greater than 90% purity and viability (1,67,110,122,123).

Again, monocyte recovery is not optimal. Preparative steps

often include use of a mononuclear cell preparation

obtained by Ficoll-Hypaque separation.

D. Counter-Flow Centrifugation Elutriation

The recently developed procedure of counter-flow

centrifugation elutriation (CCE) has been used to isolate

monocytes (35,62,65,170,187,203,237). Essentially, cells

are subjected to centrifugation within the rotor chamber of

a continuous flow centrifuge. Centrifugal force, which

tends to pellet cells towards the rotor periphery, is

counter-balanced by medium flow in the opposite direction

towards the rotor center. Under exact predetermined

conditions of rotor speed and medium flow, cells remain

suspended and reorient within the chamber according to

their sedimentation coefficients (118,186). An increase in

flow or decrease in rotor speed will cause cells of low

sedimentation velocity to leave the chamber while retaining

the faster sedimenting larger cells. This results in a

clean separation of cells on the basis of size and to a

lesser degree, density. There are many advantages in using

this technique of separation over the other methodologies

described. First, large numbers of mononuclear cells can

be introduced into the elutriator for separation

(62,65,187,237). Second, most investigators report

monocyte recovery and viability in excess of 90%

(62,65,187,237). Third, functional studies conducted on

elutriated cells demonstrated no apparent alteration due to

the elutriation procedure (62,65,187,237). Lastly, large

numbers of monocytes are obtained by a selection method

which does not cause major alterations in monocyte

physiology due to the selection procedure itself.

Functions of Macrophages

The macrophage has experienced a renaissance in the

last decade with over 50 properties and functions now being

ascribed to this particular cell (143). These properties

can be grouped into several broad categories which relate

to functional roles.

A. Antigen Presentation

That macrophages present antigen and help regulate the

immune response were recent key discoveries that redefined

our concept of immunobiology (209). It is now clear that

an interaction between macrophages and T cells is necessary

for the normal generation of an immune response. The

sequence of events appears to be initiated by macrophage

uptake of antigen followed by antigen catabolism that

eventuates in some partially degraded material being

associated with the macrophage membrane (17,210). This

processed antigen becomes closely associated with the I

region membrane determinants. Only Ta positive macrophages

can participate in presenting antigen and macrophage-T cell

interaction is restricted to cells with histocompatible I

regions (18,179,196,209). Specific lymphokines, such as

macrophage activating factor (MAF) and macrophage Ia

recruiting factor (MIRF), have the ability to increase the

number of macrophages possessing Ia and antigen presenting

ability (210), while the prostaglandin PGE2 reduces the

number of Ia positive macrophages with resultant down-

regulation of the immune response (210).

B. Secretory Functions of Macrophages

Macrophages secrete an incredible array of enzymes,

plasma proteins and biologic modifiers (143). The

bacterial compound, lysozyme, is considered a constitutive

product of all macrophages (75). Further, macrophages

synthesize the first five components of the complement

system (40,143,158). The secretion of neutral proteanases

such as plasminogen activator, elastase and collagenase by

macrophages is important to the role these tissue cells may

play in inflammation (39,183). In addition, major biologic

modifiers such as the prostaglandins (25,90), interferon

(120,121) and colony stimulating factor (31,72,142) are

produced by macrophages.

C. Anti-Microbial Functions of Macrophages

Macrophages have the ability to destroy a wide

spectrum of procaryotic and eucaryotic parasites (127).

Killing usually takes place after fusion of the phagosome

with primary and secondary lysosomes (127). It is

speculated that several pathways exist for intracellular

microbial killing which may vary depending upon the type of

organism and the type of macrophage. A hexose monophos-

phate shunt (HMS) dependent oxygen burst precedes most

intracellular killing events (105,127,146,243). A variety

of oxygen metabolites are produced in this manner. Super-

oxide anion, hydrogen peroxide and other oxygen intermedi-

ates are major products generated in phagolysosomes and are

thought to be the principal microbicidal agents (56,238).

Mechanisms other than those involving oxygen may also

function in microbicidal activity. Lysozyme, cationic

proteins, acid hydrolases and other lysosomal constituents

have all been postulated to play an anti-microbial role.

Under certain conditions, microbicidal activity within

macrophages is increased and such macrophages are referred

to as activated for microbial killing (97). A variety of

lymphocyte-derived factors from immune-mediated reactions

to antigens (26) and obligate intracellular parasites have

the ability to activate the macrophage (155). Although the

events leading to activation are not well understood, it is

well established that the result is a cell with both

heightened oxygen metabolic machinery and microbicidal

capacity (127).

D. Anti-Tumor Functions of Macrophages

It was a fortuitous discovery that macrophages can

kill tumor cells, for the original objective of these

studies was concerned with the enhanced ability of

activated macrophages to destroy microbes. However, such

fully activated macrophages were found capable of

destroying tumor cells and virus-infected neoplastic cell

lines in vitro and in vivo. Astonishingly, normal cells

present in the same culture vessel were spared from attack


Most of the early data on macrophage tumoricidal

activity was elucidated by Hibbs and by Keller. Hibbs

first demonstrated an increased resistance to both

autochthonous and transplantable neoplasms in mice

chronically infected with obligate, intracellular parasites

(84). Activated macrophages from chronically infected

animals were able to destroy tumor cells but not normal

cells in vitro. At the same time, Keller had shown that

activated macrophages could suppress tumor growth in vivo

(101). He demonstrated that adoptive transfer of

irradiated peritoneal exudate cells which were functionally

composed of activated macrophages were able to suppress

tumor growth in total body irradiated syngeneic

recipients. Keller later contended that the major effect

of activated macrophages was on the proliferative capacity

of the tumor and that cytostasis was of varying degrees

depending on the target cell tested (103). Both Hibbs and

Keller concluded independently that tumor cell killing by

activated macrophages was by a non-phagocytic, non-specific

means involving macrophage/tumor cell contact (85,102).

Data from the laboratory of Fidler support this hypothesis

in as much as activated macrophages destroyed B16 melanoma

cells in culture, but again, did not kill normal cells

(59). He then confirmed this observation in vivo by

demonstrating that IV injection of activated macrophages

into B16 melanoma-bearing mice significantly reduced the

number of established pulmonary metastases. Tumor growth

was also inhibited at primary sites of innoculation by

activated macrophages (45). Furthermore, macrophages from

primary sites of immunogenic tumors were tumoricidal in

vitro (122,184).

The ability of activated macrophages to recognize and

destroy tumor cells has been postulated to depend upon a

specific recognition event (60). However, species-specific

antigens, tumor-specific or associated antigens, and

histocompatibility antigens are not responsible for the

selective destruction of neoplastic cells, although

macrophage-tumor cell contact has been shown to be

necessary for tumor cytolysis to occur (61). Some

investigators have demonstrated highly specific binding of

activated macrophages to tumor cells (124,125). Such

macrophage/tumor cell aggregates can be disrupted by the

addition of excess unlabeled target cells. Addition of

partially purified tumor cell membranes also suppresses

tumor binding and cytolysis. Normal unstimulated

macrophages do not form macrophage-tumor cell aggregates

and are not tumoricidal. At present, a specific tumor

binding receptor has not been identified.

The mechanisms involved in the actual cytolytic event

are poorly understood. Substances such as oxygen

intermediates, arginase, and prostaglandins (all of which

are made by macrophages) can damage tumor cells (2,4,5).

However, high levels of these products seem necessary to

affect tumor killing. For example, the amount of H202

necessary to effect killing of some strains of tumor cells

in vitro is far in excess of that which would be synthe-

sized by a quantity of macrophages causing comparable

damage (147). A recently discovered neutral proteinase,

which is excreted exclusively by activated macrophages and

is inhibited by serine proteinase inhibitors, kills tumor

cells in low concentrations, e.g., the LD50 for tumor

targets is 1 x 10~9 M (4). It is postulated that the

tight adherence of macrophages to tumor targets provides a

channel for protecting neutral proteinase from the

inhibitory effects of plasma. Neutral proteinase has also

been found to act synergistically with other macrophage

secretary products such as hydrogen peroxide in affecting

tumor cell lysis (3).

Another mechanism by which macrophages may bind and

kill tumor cells involves specific anti-tumor antibody and

interaction with the Fc receptors on macrophages.

Antibody-dependent cellular cytotoxicity (ADCC) is a well

known property of macrophages (193,197,233). Originally

assayed by measuring radiolabeled chromium release from

antibody sensitized erythrocyte targets (141), ADCC has

been shown to be a mechanism of lysing nucleated cells as

well (109). Essentially all macrophages can participate in

ADCC by virtue of their Fc receptors. Direct correlation

with expression of Fc receptors and cytolysis has been

demonstrated (92,108). Activated macrophages are more

effective in mediating ADCC than resting macrophages

(107). The mechanism involved in the actual cytolysis of

tumor cells is again unclear at this time, but essentially

all mechanisms described previously for activated

macrophages have been implicated.

Hibbs et al. proposed that macrophages may function in

tumor surveillance. He found that activated macrophages

were considerably more cytopathic for transformed cell

lines in vitro than for normal cell lines which exhibited

contact inhibition, anchorage dependence, etc. (86). His

work supported the contention that macrophages were a part

of a primitive non-immunologic surveillance mechanism

capable of detecting and destroying cells manifesting

aberrant growth, thus preventing cancer development.

The activated macrophage may not be the only cell

capable of immunologic non-specific tumoricidal activity,

however. Current work focuses on destruction of nascent

tumors by a class of circulating cells called natural

killer (NK) cells. By broad definition, these cells have

in common the ability to destroy tumor cells of hemato-

poietic origin, across species barriers (80,82,241).

Current data do not unequivocally link NK activity to one

particular cell. Herberman has isolated a large granulated

lymphocyte (LGL) in high purity from human and murine blood

by isolation on Percoll gradients (80). This cell has the

properties of a T lymphocyte (surface markers, theta

antigen, morphology). Curiously, other investigators have

determined that the NK cell is a promonocyte possessing

macrophage surface markers but lacking the typical charac-

teristics of the macrophage phagocytosiss, adherence,

esterase) (119). These LGL's and promonocytes are capable

of destroying tumor cells of hematopoietic origin in

accordance with the definition of NK cell activity.

Augmentation of NK activity results in vivo from the

addition of immune modulators such as interferon (46,70),

poly I:C (153,154), or BCG (163). Infection with viral

agents also raises NK activity (47). Eremin has implicated

the Kurloff cell of the guinea pig as responsible for NK

activity (52). This cell, with its large eosinophilic,

eccentrically located inclusion body, has both properties

of lymphocytes and macrophages, thus making its origin

uncertain (104,171). The techniques used to establish the

NK cell activity of Kurloff cells were not definitive,

relying upon negative depletion techniques to show loss of

NK activity (52).

Macrophages and their precursors have also been

implicated as possessing an immune surveillance role.

Unlike NK activity which is manifested against a rather

restricted tumor target range, the native tumoricidal

activity of mononuclear phagocytes covers a broad range of

tumor targets, many not of hematopoietic origin (94).

Native tumoricidal activity is usually measured over a

longer assay time (18-72 hours) (131,205) than NK activity

(4 hours) (82). Again, augmentation of tumoricidal

activity occurs with the addition of immune modulators


Accumulating data on NK activity and native

tumoricidal activity have led some investigators to

speculate that more than one cell type is responsible in

vivo for affording protection against neoplastic cell

growth (37). Mice whose immune systems have been destroyed

by high dose irradiation or nude mice lacking a functional

T cell system manifest the same rates of spontaneous

hematopoietic cancers as normal animals with functionally

intact immune systems. Interestingly, no disease states

have been described wherein the host lacks a mononuclear

phagocyte system. This condition might be incompatible

with life for a variety of reasons, including protection

against the aberrant growth of neoplastic cells.

Therefore, cells of both macrophage and lymphocyte origin

may play strategic roles in immune surveillance.

E. Role of Macrophages in Inflammation and Repair

That macrophages phagocytize foreign matter has been

known since the first observations of Metchnikoff (136-

138). Thus, it comes as no surprise that macrophages play

a crucial role in the inflammatory process.

The temporal events of cell migration and accumulation

at an inflammatory site are well established. After the

initial wave of neutophils, monocytes become the principal

cell entering the lesion (228). Monocytes are a chemotac-

tically responsive cell and in vitro migrate directionally

in response to gradients of C5a or f-met-leu-phe (188,198).

In vivo at the site of inflammation, they first adhere to

post-capillary venular endothelium, then emigrate between

endothelial cells. Once present, macrophages phagocytize

dead tissue, senescent granulocytes and particulate debris.

It is believed that monocyte secretion of plasminogen

activator may be involved in a localized lysis of the

basement membrane allowing the monocyte to enter the area

of injury (240). Indeed, macrophages are considered the

principal cell responsible for wound debridement and for

normal wound repair. They appear to secret a factors)

involved in fibroblast migration and collagen synthesis

(228) as well as neovascularization (165,166). Macrophages

in various states of maturation and differentiation are

found at sites of inflammation. The chronicity of the

lesion, the composition or antigenicity of the phlogistic

agent and the immune status of the host determine the

predominant types of mononuclear phagocytes that reside at

the inflammatory site.

Once the monocytes have arrived at the site of

inflammation, an impressive array of biological inter-

actions and biochemical events takes place that lead to the

transformation of monocytes into macrophages. In immune-

mediated inflammation an increased modulation of macrophage

functions may occur via lymphokine production by T cells

(42). Alternatively, immune complexes alone can stimulate

macrophage functional activity (41). Phagocytosis seems to

be the event that triggers macrophages to secrete a variety

of enzymes into the surrounding area (190). Once

initiated, secretion becomes a function divorced from the

initial phagocytic trigger. The three major groups of

enzymes are lysozyme, lysosomal hydrolases and neutral

proteinases. Receptor complexes such as C3b-C3b, cause the

selective secretion of acid hydrolases (191) whereas C5a

stimulates both acid hydrolase and neutral proteinase

secretion (129,211). The latter group of enzymes together

with reactive oxygen intermediates (56,238) is now thought

to be responsible for most of the tissue damage present at

sites of chronic inflammation (13). Indeed, macrophages

play a key role in perpetuating and actually accentuating

the inflammatory process in some disease states (44).

Alternatively, the production of prostaglandins by

macrophages at inflammatory sites could down-regulate the

local immune response thus leading to a more controlled

inflammatory process (14,21,73).

Very few products secreted by macrophages have been

analyzed in relation to their role in inflammatory

physiology. Nevertheless, considerable progress has been

made in understanding the complex biochemical events

modulated by macrophages in inflammation, a cell that two

decades ago was thought to have the singular function of

removing foreign matter and necrotic debris from a nidus of


Macrophage Heterogeneity

Early work on macrophage function concentrated on

characteristics expressed by the cell population as a

whole. In analyzing some of the classical properties which

characterize elicited macrophage populations, it emerged

that considerable heterogeneity exists at the individual

cell level with some macrophages even lacking properties of

adherence (113,204), non-specific esterase (63), high

phagocytic rate (100,181,229), Fc receptors (99,230) and

ADCC activity (194). Additionally, considerable

differences exist among peritoneal macrophages elicited by

various means (74,148,189). Major differences also exist

among unfractionated macrophage populations obtained from

different sites. For instance, alveolar macrophages rely

primarily upon oxidative metabolism while peritoneal

macrophages use a glycolytic pathway (156). The lysosomal

content of alveolar macrophages is higher than peritoneal

macrophages but alveolar macrophages are considered less

efficient in phagocytosis and lysis of bacteria (112,239).

The Fc receptor avidity is lower in alveolar macrophages

compared to peritoneal macrophages (172) and the former are

less responsive to chemotactic agents (48,49). Another

example of heterogeneity in macrophages obtained from

different sites derives from work on Ia expression. The

percentage of Ia positive macrophages within a body

compartment is constant in normal animals although

significant differences exist between compartments

(15,36,51). Fifty percent of thymic (16), hepatic

(174,176) and splenic macrophages (36) are Ia positive

whereas 8-30% of peritoneal macrophages (192,246) and

peripheral blood monocytes (6) possess this marker. The

ability of these macrophages to present antigen also

correlates with the percentage of Ia positive cells (51).

Resident macrophages, which have low microbicidal

activities, do not produce appreciable amounts of neutral

proteinases and have negligible activity against tumor

cells (2,55,211). Peritoneal cells elicited by

intraperitoneal injection of thioglycholate (211),

Corynebacterium parvum (143,158), phorbol myristate

acetate (223), lymphokines (149), or phagocytosable

particles (190) result in activated macrophages possessing

high levels of antimicrobial and antitumor activity and

high neutral proteinase levels.

The techniques previously described for isolating

macrophages and monocytes have also been used to

demonstrate heterogeneity among macrophage populations.

Both density (173,194) and size (231) have been used to

physically separate macrophage populations and demonstrate

differences in functionality. Albumin, Ficoll and Percoll

gradients have been used to fractionate macrophages and

show a correlation between size and density with antigen-

presenting capability and tumoricidal activity (160,173,

230,231,235). Counter-flow centrifugation elutriation has

been used to demonstrate that large macrophages are more

tumoricidal than smaller-sized phagocytes within the same

macrophage population (140).

Monocyte Heterogeneity

The use of adherence for monocyte purification first

indicated that monocytes were heterogeneous inasmuch as

some, but not all, monocytes adhered to glass or plastic

surfaces (89). The use of velocity sedimentation (8) and

CCE (63,152,237,247) have demonstrated size and phenotypic

differences in human monocytes. As first isolated by

Norris et al., small human monocytes were reported to be Fc

receptor negative and inactive in ADCC while the large

monocytes were Fc receptor positive and active in ADCC

(152). Subsequently, other investigators demonstrated

differences between CCE isolated monocytes in lysosomal

enzyme content (63), hexose monophosphate shunt activity

(195) and native tumoricidal activity (151). Of particular

note, small monocytes possessed native tumoricidal activity

while large monocytes did not and the former were more

capable of being activated to higher tumoricidal levels in

vitro than were large monocytes. Thus, heterogeneity

appears to be present not only among the mature macrophages

but also among the circulating monocyte precursors.

Theories on Macrophage Heterogeneity

Accumulating data provide evidence for both intra-

population and inter-population heterogeneity in the

mononuclear phagocyte system. Current research has focused

on the origin of this heterogeneity and its significance.

At present two main concepts have been proposed to explain

macrophage heterogeneity.

A. Vertical Heterogeneity: Maturation-Activation

The concept of vertical heterogeneity states that

macrophage maturation with ensuing differentiation accounts

for differences between and within macrophage populations.

As the macrophage matures, different properties would be

ascribed to different stages of differentiation. Under

appropriate stimulation, mature macrophages would differen-

tiate into the highly specialized activated macrophage

capable of efficient microbicidal and tumoricidal

functions. At a particular site, the macrophage pool would

be expected to be in dynamic flux with its monocyte

precursors, thus providing a steady supply of unmaturated


Evidence to support this concept includes the

following: First, monocytes derived from bone marrow

continually renew macrophage populations at various body

sites. Van Furth used tritiated thymidine labelled cells

to demonstrate such monocyte traffic under steady state

conditions (214). A different approach has been to use

various marker enzymes. The enzyme, peroxidase, is present

in monocytes and disappears when the cells mature (38).

Peroxidase positive cells constitute a small but

significant percentage of macrophages at various locations

(23,221). Their number increases upon stimulation by

various agents indicative of the increased influx of

monocytes. The opposite situation pertains to the

ectoenzyme 5'-nucleotidase (28). Second, monocytes are

continuously differentiating into macrophages and a

spectrum of functional characteristics can be expressed by

any given macrophage (232).

Central to the argument for vertical heterogeneity is

the requirement for only one type of committed macrophage

stem cell in the bone marrow. That this may be so is

supported by the bone marrow culture studies of Calamai et

al. (29). They have shown that Ia positive and Ia

negative macrophage colonies can be derived from a single

bone marrow precursor. Depending upon the type of

stimulation used, all colonies can be converted to the Ia

phenotype. This suggests that the committed macrophage

stem cell is originally Ia negative and is induced to

differentiate and express the Ia phenotype under

appropriate immune stimulation.

B. Horizontal Heterogeneity: Population Diversity

The concept of horizontal heterogeneity states that

different stem cells exist that give rise to monocytes with

restricted functions and characteristics. Most studies in

support of this concept have been carried out on macrophage

colonies derived from bone marrow culture. In contrast to

the data presented above supporting vertical heterogeneity,

these studies have reported homogeneity within macrophages

from discrete colonies but heterogeneity among the colonies

themselves (71,126,134,182,234). The use of highly

specific monoclonal antibodies has demonstrated major

antigenic differences in macrophage colonies (182). The

work of Metcalf also supports multiple stem cells in bone

marrow as some CFU are capable of forming only macrophages

while others can differentiate into either granulocytes or

macrophages (134). The detection of a small but

significant percentage of tissue macrophages not of

monocyte origin with the capacity to replicate has been

used as evidence for the existence of multiple macrophage

stem cell precursors. This postulates the existence of at

least 2 independent replicating progenitors as renewing the

macrophage pool (114,116,117). In addition to this work

with macrophage colonies, the heterogeneous nature of

circulating human monocytes has been used as evidence for

macrophage stem cell heterogeneity. These proponents argue

that macrophage precursors would use a common pathway, the

circulation, to reach their destinations in the tissues and

should be homogeneous. However, such homogeneity is not in

fact observed, neither with respect to the properties

already mentioned nor with monoclonal antibodies that have

identified different populations of monocytes in human

peripheral blood (88,169). Finally, it is apparent that

the vast array of properties and functional diversity

ascribed to the macrophage is unlikely to be possessed by

every cell in the mononuclear phagocyte system. Indeed,

the amount of cellular machinery necessary to produce such

varied products in significant amounts argues strongly for

functional stratification and therefore a heterogeneous

macrophage population.


At the present time, there is equal support for both

the vertical and horizontal hypotheses of macrophage

heterogeneity. Important to the interpretation of both

hypotheses is an understanding of the characteristics and

functions of the macrophage precursor cells: monoblasts,

promonocytes and monocytes. Although culturing of mono-

blasts and promonocytes obtained from bone marrow has been

successful in both human and mouse, it is well accepted

that in vitro culture causes the maturation of these pre-

cursors into macrophages, thus distorting the differenti-

ation-maturation steps which may occur. An alternative

approach is to procure macrophage precursors in large

numbers in an unaltered state from which characterization

studies can be conducted. Human monocytes have been

obtained in large numbers and high purity by the technique

of counter-flow centrifugation elutriation. These mono-

cytes displayed heterogeneity in physical characteristics

and function. It is difficult to access the significance

of these findings in only one animal species. Furthermore,

investigation of monocyte heterogeneity in human subjects

is limited since most studies could be conducted only in

vitro with little or no in vivo manipulation. The use of

an animal model would circumvent these problems and con-

currently demonstrate monocyte heterogeneity in a species

other than man.

The objectives of the experiments to be reported are

to establish a method for isolating monocytes from a small

laboratory animal, to demonstrate the functional integrity

of the isolated monocytes and to investigate the

possibility that monocytes of species other than man exist

in different physical and functional states.

The following criteria should pertain to any method of

isolating monocytes:

1. The starting monocyte preparation before separation

should include all the monocytes found in the peripheral


2. The separation method should yield cells in large

enough quantities for characterization studies.

3. The separation method should be reproducible, showing

consistency in the fractionation of the starting


4. The methods involved in separation should not cause

alteration in the physical and functional properties of the


The guinea pig was chosen for study because of the

similarity of its hematopoietic system to that of man in

both total and differential white blood cell counts

(7,244). In addition, moderate quantities of guinea pig

peripheral blood can be obtained without causing harm to


the animal or significant alteration in hemodynamics.

Counter-flow centrifugation elutriation was chosen to

separate guinea pig monocytes based upon the prior

investigation of human monocyte heterogeneity using this



Mononuclear Cell Preparation

Seven male Hartley strain guinea pigs (750g) were

anesthetized with a subcutaneous injection of 30 mg/kg

Ketamine and 2 mg/kg Xylazine (78) after which a total of

150 ml of acid citrate dextrose (ACD) anticoagulated blood

were obtained by cardiac puncture. The blood was diluted

1:1 with calcium and magnesium free Hank's balanced salt

solution (HBSS) containing 100 mg EDTA/L. Twenty-one

milliliters of diluted blood were placed in each of a

series of 50 ml siliconized glass centrifuge tubes and 19

ml of Ficoll-Hypaque were introduced below the blood using

an 18 gauge canula. After centrifugation at 400 x g for 40

minutes at 180C (4), the interface layer was removed,

diluted in HBSS, and the cells collected by centrifugation

at 750 x g for 15 minutes at 4C. The cells were washed

twice and collected by centrifugation (150 x g for 10

minutes). Contaminating red blood cells were removed by

exposure to buffered ammonium chloride (177) for 3 minutes

after which the mononuclear cells were washed twice with

the final washing being performed in elutriation medium

consisting of calcium and magnesium free HBSS (pH 7.20)

with 100 mg EDTA/L, 50 mg BSA/L and osmolality of 290 m

Osmos/L. Total cell counts were made by both a

hemocytometer and a Coulter Counter Model ZF (Coulter

Electronics, Hialeah, Fla.). Viability was determined by

trypan blue dye exclusion.

Counter-Flow Centrifugation Elutriation (CCE)

Mononuclear cells (2x108) suspended in 5-10 ml of

elutriation medium were injected into the inlet stream

leading into the Beckman J-6B centrifuge equipped with a

standard JE-6B elutriator rotor and chamber. A Masterflex

peristaltic pump (Cole-Palmer Instruments, Chicago, IL) was

equipped with a vernier potentiometer to provide precisely

metered flow. The cells were loaded into the chamber at a

flow rate of 10 ml/min, rotor speed of 3,000 RPM and a

temperature of 40C. Rotor speed was held constant and the

cells eluted by changing the flow rate. In the first

method, the flow rate was raised to where small cells began

to exit the chamber and then sequential fractions of 100 ml

each were collected at 1 ml/min increments in flow. In the

second method, cells were eluted to exhaustion at each of 2

predetermined flow rates. In both techniques, cells

remaining in the elutriator chamber were collected by

continuing medium flow after stopping the rotor. Each cell

fraction was collected by centrifugation, washed twice in

RPMI 1640 and counted, after which viability was

determined. Slides were prepared using a cytocentrifuge

(Shandon Southern Instruments, Sewickley, Pa.) at 5 x 104

cells per slide in 50% fetal calf serum. Cell differen-

tials were done on 500 cells stained with Wright-Giemsa

(Camco Quick Stain, Am. Scientific Products, Ocala, FL). A

Leitz Dialux 20 microscope with orthomat camera was used to

photograph the stained slides.

Cell Volume Analysis

Cell volume was determined with a Particle Data System

80 XY Electrozone Celloscope integrated with the REX 604

software program (Particle Data Systems, Chicago, IL).

This system provides a computer smoothed volume plot based

upon total cells in the peak channel.

Histochemical Stains

The non-specific esterase stain was modified from that

described by Yam et al. (245). Formalin-acetone fixed

cytocentrifuge slides were reacted for 50 minutes at 37C

with alpha napthyl acetate, then washed in tap water, and

counter-stained for 3 minutes in 1% methyl green. Guinea

pig monocytes unlike human monocytes stained weakly at room

temperature but reacted well at 37C to yield a reddish-

brown reaction product. Acid phosphatase staining was done

using the fast garnett GBC method with Napthol AS-BI

phosphoric acid salt (Sigma Chemical Co., St. Louis, MO) as

substrate and scored for the intensity of orange-red

reaction product found within the cytoplasm (115).

Peroxidase staining was performed either as described by

Kaplow (95,96) at pH 6.0 with maximal staining after 1

minute incubation or as modified by Meltzer (130) using a 2

minute reaction at pH 7.0. The slides were counter-stained

with Giemsa and the reaction product scored. Guinea pig

peripheral blood smears were always run as controls for the

above histochemical stains. Staining intensity was graded

on a 0-4+ scale of intensity and monocytes having less than

a 1+ reaction were considered negative.


Cells suspended in alpha-MEM containing 10% heat

inactivated guinea pig serum were tested for adherence

using four methods. First, cells were allowed to adhere to

untreated 75 cm2 plastic tissue culture flasks (Costar,

Cambridge, MA.) or second, to flasks coated with a micro-

exudate from a previous culture of a BHK-21 cell line as

described by Ackermann and Douglas (1). In both methods,

non-adherent cells were removed by repetitive washing as

judged by inverted phase microscopy after which the

adherent cells were recovered by vigorous agitation

following a 15 minute exposure at 370C to 10 ml of medium

containing 5mM/L EDTA. Third, 2 ml of 1% type I bovine

gelatin (Sigma Chemical Co., St. Louis, MO) were added to

35 X 10 mm tissue culture plates (Falcon Plastics, Oxnard,

CA) and refrigerated overnight. After removing excess

gelatin, the plates were overlaid with 2 ml of fresh

heparinized guinea pig plasma or purified human fibronectin

(Sigma Chemical Co., St. Louis, MO) at 37*C for 30 minutes

(67). After washing three times, the various cell

fractions were added for 1, 4, or 8 hours. Then non-

adherent cells were washed away and adherent cells detached

by overnight incubation and mild agitation. Fourth, up to

ixl08 cells were passed over nylon wool columns prepared

as described by Weinblatt et al. (236) using 3 ml columns

containing 0.2 grams of sterile, acid-washed nylon wool.

After incubation, non-adherent cells were eluted with 100

ml of medium and adherent cells recovered by exposing the

column for 15 minutes to medium containing 5mM/L EDTA and

mechanically expressing the cells from the column. All

adherence procedures were performed at 370C in a humidified

atmosphere containing 5% CO2. In some experiments,

lipopolysaccharide derived from Salmonella typhimurium

(Difco Laboratories, Detroit, MI) was used to activate the

cells prior to adherence. The percentage of adherence was

calculated from the ratio of the number of monocytes

recovered from the adherent fraction or the number depleted

from the non-adherent fraction to the total number of

applied monocytes. The reported percentage of monocyte

adherence is the higher of these two calculations although

in most instances the two calculations were identical.

Percoll Gradients

Percoll (Pharmacia Fine Chemicals, Piscataway, NJ)

was adjusted to a specific gravity of 1.070, pH of 7.20 and

12 ml used to generate a continuous gradient in a 13 x 100

mm polycarbonate tube by centrifuging at 30,000 x g in a

fixed angle rotor for 20 minutes at 180C in a Sorval RC-2B

centrifuge. Cells from the 24A (50 x 106), 24B (6 x

106) and 28 ml/min (10 x 106) were mixed with 1 ml of

Percoll removed from the bottom of the gradient and then

underlaid in the same gradient tube. These gradients along

with a control gradient containing density marker beads

(Pharmacia Fine Chemicals, Piscataway, NJ) were centri-

fuged at 400 x g for 20 minutes at 180C. Mean cell

densities were determined by comparing the distance

migrated by cells to that of the marker beads. The various

cell fractions were removed and washed twice before

analysis for cell number, viability and size.


Non-immune phagocytosis was performed using carbon

particles and albumin-coated polystyrene latex spheres.

Carbon particles (suspension lot C-11-1431a, from Gunther-

Wagner, Hanover, West Germany) were diluted to 25 mg/100 ml

in saline and 15 ul added to 2 x 106 mononuclear cells in

a total volume of 1.0 ml of RPMI 1640 with 10% fetal bovine

serum. Fluorescent labeled latex particles of 0.7 micron

diameter (Covaspheres FX Particles) were obtained from

Covalent Technology Corp., Ann Arbor, MI. Covalent bonding

of protein to the activated bead surface was performed by

adding 50 ul of Covaspheres to 0.5 ml of HBSS containing

0.1 mg BSA/ml. Bonding was nearly instantaneous and the

albumin coated beads were collected by centrifugation,

washed twice in HBSS and resuspended in 7 ml of HBSS

containing 10% fetal bovine serum. The phagocytosis assay

was performed in 12 x 75 mm polypropylene tubes at 370C by

exposing 2 x 106 mononuclear cells in 0.5 ml to an equal

volume of bead suspension. In both assays, non-ingested

particles were removed by gently centrifuging the cells

(100 x g, 8 minutes) through 1 ml of fetal bovine serum.

Cytocentrifuge slides were prepared from the cell pellet

and stained with Wright-Giemsa. The intensity of carbon

phagocytosis per 100 phagocytes was scored on a 0 to 4+

scale using the oil immersion objective of a light

microscope. The fluorescent labeled beads were visualized

using a Zeiss epifluorescent microscope and the number of

beads counted in 100 phagocytes. In both assays, cell

suspensions also were placed directly on glass slides and

viewed using a combination of tungsten and ultraviolet

light with a Zeiss epifluorescent microscope. This method

helped distinguish those cells which had beads bound to

their surfaces (rosettes) as opposed to those which had

engulfed them.

The ability of monocytes to participate in antibody-

dependent phagocytosis was assessed with antibody coated

sheep erythrocytes. Sheep erythrocytes sensitized with

rabbit anti-sheep red blood cell immunoglobulin (anti-SRBC

IgG from Cordis Laboratories, Miami, FL) were added to

guinea pig mononuclear cells suspended in RPMI 1640 with

10% fetal bovine serum at a concentration of 100 erythro-

cytes per monocyte. Following light centrifugation (150 x

g for 3 minutes) and incubation at 370C in a 5% CO2

incubator, the uningested erythrocytes were lysed by

exposure to ice cold buffered ammonium chloride for 5

minutes. The remaining mononuclear cells were washed once

in RPMI 1640 and cytocentrifuge slides prepared after

resuspension in RPMI 1640 containing 50% fetal bovine

serum. The percentage of monocytes phagocytizing erythro-

cytes as well as the number of RBC's phagocytized per 100

monocytes was scored by the use of oil immersion light

microscopy on Wright-Giemsa stained preparations.

Candida albicans (human blood isolate) was cultured

overnight in Saboraud's broth on a 180 RPM shaker at 37*C.

For opsonization, 5 x 107 organisms in 1 ml of RPMI 1640

were incubated on ice with 1 ml of fresh guinea pig serum

for 15 minutes. Opsonized and non-opsonized Candida

albicans were added to the various mononuclear fractions

at a 10:1 ratio of organisms to phagocytes for various time

periods at 370C. Phagocytosis was evaluated using

cytocentrifuge preparations stained with Wright-Giemsa and

reported as the number of Candida ingested per 100


For some experiments, the C3 component of guinea pig

sera was neutralized by incubating rabbit anti-guinea pig

C3 (Cappell Laboratories, Cochranville, PA) with fresh

guinea pig sera for 30 minutes at 370C. Immune precipitate

was removed by ultracentrifugation (Beckman airfuge,

Beckman Instruments, Palo Alto, CA). The C3 component was

no longer demonstrable when this sera was reacted against

anti-C3 by Oucterlony analysis.

Guinea Pig Anti-Sheep Erythocyte Antibody

Autologous IgG against sheep red blood cells (SRBC)

was prepared in 3 guinea pigs by injecting intramuscularly

and subcutaneously 0.1 ml of 20% SRBC's in complete

Freund's adjuvant (Difco Laboratories, Detroit, MI). An

additional 0.1 ml of cell suspension was administered

subcutaneously 14 days later. After 5 weeks, blood was

drawn by cardiac puncture from which serum was obtained and

heat-inactivated for 30 minutes at 560C. Guinea pig IgG

was purified using ammonium sulfate fractionation. Then,

serial dilutions of this antibody were incubated for 30

minutes at 370C in a 1:1 volume ratio with a 5% suspension

of washed SRBC's in 0.01 M EDTA gelatin-veronal buffer

(EDTA-GVB from Cordis Laboratories, Miami, FL). After

washing three times, the pellets were resuspended and

evaluated for agglutination under an inverted phase

microscope (Swift Instruments, Inc., San Jose, CA) using

the 40 x phase objective. The smallest dilution of

antibody which did not produce detectable clumping was


Fc Receptor Assay

Fresh sheep red blood cells (SRBC) were collected in

Alsever's solution and washed three times in cold phosphate

buffered saline. Rabbit anti-sheep red blood cell

immunoglobulin (anti-SRBC IgG from Cordis Laboratories,

Miami, FL) was determined to have a subagglutinating titer

of 1/2000. To 1 ml of 5% SRBC in EDTA-GVB was added an

equal volume of a subagglutinating dilution of anti-SRBC

IgG and the mixture incubated at 370C for 30 minutes.

Fresh ox (bovine), guinea pig and human red blood cells

were sensitized with anti-erythrocyte immunoglobulin in a

similar manner. Rabbit anti-ox (bovine) erythrocyte

immunoglobulin (subagglutinating titer 1:500) and rabbit

anti-guinea pig erythrocyte immunoglobulin (subagglutina-

ting titer 1:5000) were obtained from Cappel Laboratories,

Cochranville, PA. Human anti-RhoD immunoglobulin (Dade

Diagnostics, Miami, FL) reacted optimally at a titer of

1:50 against 0 positive human red cells. After reacting

with antibody, the sensitized erythrocytes were washed

three times in EDTA-GVB buffer, resuspended in HBSS

containing 2% BSA, counted and adjusted to 4 x 107


The assay for Fc receptors was based upon the capacity

of the monocytes to form rosettes with antibody sensitized

erythrocytes of guinea pig, ox and human. In 12 x 75 mm

polypropylene tubes, 100 pl of sensitized erythrocytes were

added to 100 Vl of mononuclear cells (4 x 106 cells/ml)

to yield a 10:1 ratio of erythrocytes to monocytes. The

suspension was vortexed lightly, centrifuged at 100 x g for

3 minutes and incubated for 30 minutes at 370C. Then the

pellet was gently resuspended and 15 ~l placed on a glass

slide under a coverslip. The percentage of monocytes

binding 0-2, 3-4 or 5 or more erythrocytes was determined

in a 200 monocyte count using oil immersion phase contrast

with greater than 2 erythrocytes bound per monocyte

considered an Fc receptor positive cell. Phase contrast

highlighted the granulated cytoplasm found in monocytes and

allowed positive identification of these cells. In

addition, cytocentrifuge slides were made, stained with

Wright-Giemsa and the percentage of rosette-positive

monocytes determined. In all assays, a positive control

was run concurrently consisting of proteose-peptone

elicited peritoneal guinea pig macrophages which were

routinely greater than 90% Fc receptor positive.

Previously, Fc receptors on guinea pig monocytes were

determined on cells after 48 hours of culture (23).

Accordingly, mononuclear cell preparations in some

experiments were incubated at 370C and 5% CO2 for various

time periods before determination of Fc receptors. For

culture, mononuclear cells were diluted to a concentration

of 1 x 106 cells/ml with RPMI containing 10% guinea pig

serum, 10% horse serum, 10 mM HEPES buffer, 50 pg/ml

gentamycin, 10,000 U penicillin, and 5,000 U streptomycin.

Ten milliliters of cell suspension were cultured in 30 ml

capacity Teflon FEP culture bottles (Nalge Corp.,

Rochester, NY) and Fc receptor analysis performed as


Antibody-Dependent Cellular Cytotoxicity (ADCC)

Fresh sheep erythrocytes were sensitized with a

subagglutinating titer of rabbit anti-SRBC immunoglobulin

as described for the determination of Fc receptors. When

used as target cells in ADCC, they were labeled with

radioactive chromium by exposing 4 x 107 cells in the

bottom of a centrifuge tube to 200 pCi of radioactive

chromium (250-500 mCi/mg chromium, Amersham, Arlington

Heights, IL) for 1 hour at 370C. After washing twice in

HBSS and once in RPMI, the labeled cells were resuspended

in ADCC medium (RPMI 1640 with 10% guinea pig serum, 10 mM

HEPES buffer, 50 pg/ml gentamycin, 10,000 U penicillin and

5,000 U streptomycin). Labeled cells averaged 10,000

CPM/106 SRBC using a Beckman 300 Gamma Counter (Beckman

Instruments, Fullerton, CA). Effector cells in 1 ml of

ADCC media were dispensed into 12 x 75 mm polypropylene

tubes and 100 pl of target cells added (2 x 105 SRBC).

Non-specific isotope release was determined by adding 100

Pl of target cell suspension to 1 ml of assay medium. All

tests were done in triplicate. The assay was performed by

gently centrifuging the cell suspension (100 x g for 3

minutes) and incubating the pellets at 370C in a 5% CO2

incubator. At the conclusion of the assay (8 or 16 h), the

tubes were gently agitated, the cells centrifuged at 400 x

g for 5 minutes and 550 pl of the supernatant removed and

counted in a Beckman 300 Gamma Counter. The percent 51Cr

release was determined using the formula:

% 51Cr release = Experimental CPM Control CPM x 100

Total CPM Control CPM

Tumoricidal Assay

The P-815 mastocytoma cells were maintained in ascites

form by serial IP transplantation of 2 x 106 cells in

syngeneic DBA/2 mice. Tumor cells were harvested 3 to 5

days after transplantation and cultured overnight at 370C

and 5% CO2. Culture medium consisted of RPMI 1640 with

10% guinea pig serum, 10mM HEPES buffer, 50 yg/ml genta-

mycin, 10,000 U penicillin and 5,000 U streptomycin. Tumor

cells were labeled for 4 hours with 0.5 pCi/ml 125_

iododeoxyuridine (1251udr from Amersham Corp., Arlington

Hts, IL), washed three times in culture medium and counted.

Viability was determined by trypan-blue dye exclusion.

Labeled target cells averaged 4,000 CPM/104 cells with

greater than 95% viability by trypan blue dye exclusion.

Effector cells were washed twice in RPMI 1640, resuspended

in culture medium and counted. Differential cell counts

were made on cytocentrifuge slides stained with Wright-

Giemsa. A constant number of tumor cells (1 x 104 in 100

pl) was added to 1 ml of effector cells in 12 x 75 mm

polypropylene tubes at varying effector:target (E:T) ratios


based on monocyte number. Controls consisted of radio-

labeled tumor cells added to 1 ml of culture medium. All

tests were done in triplicate. At the conclusion of the

assay, the tubes were gently mixed, centrifuged at 400 x g

for 5 minutes and 550pi of the supernatant removed.

Percent isotope release was calculated using the formula:

% 125Iudr release = Experimental CPM Control CPM x 100

Total CPM Control CPM


Isolation of Mononuclear Cells

Ficoll-Hypaque of specific gravity 1.079 is used

routinely to isolate human monocytes in high yield but the

same preparation collected less than 25% of guinea pig

monocytes at the interface layer. Table 1 presents data on

the recovery of guinea pig monocytes using Ficoll-Hypaque

of increasing specific gravity. From this analysis, we

concluded that a high specific gravity of 1.101 was

required to routinely recover greater than 95% of the

monocytes applied to the gradient. Although red cell

contamination increased with the higher density Ficoll-

Hypaque, these cells were easily removed by ammonium

chloride lysis. In addition, granulocytes contaminated the

mononuclear cell fraction when high density Ficoll-Hypaque

gradients were prepared in polypropylene tubes. This

problem was corrected by using siliconized glass tubes in

place of polypropylene tubes. As shown in Table 2, the

absolute number of granulocytes was reduced by using

siliconized tubes whereas mononuclear cell numbers remained











U 41
00 C
0 0

C >

o -


u u























W E-

a u

0 E-

z 9

U z
0 0

>U) C

a4 a4
< a4



0' 0
r- C

CM 0





C) -

I 0

0 0


Counter-Flow Centrifugation Elutriation (CCE)

At a loading flow rate of 10 ml/minute and rotor speed

of 3,000 RPM, the mononuclear cells remained suspended in

the chamber while platelets and residual red cells were

purged. No mononuclear cells were collected until the flow

rate reached 24 ml/minute. Table 3 presents the profile of

cells eluted at sequential 1 ml/minute increases in flow

above 24 ml/minute showing a progressive decrease in

lymphocytes and an increase in monocytes. The 28 ml/minute

fraction had the highest monocyte purity (70%) and yield

(54%). The residual cells purged from the chamber after

stopping centrifugation were designated rotor-off (R/0)

cells. Monocytes accounted for 46% of the R/O cells along

with large lymphocytes, Kurloff cells and some


Volume determinations showed that the 24 ml/minute

fraction resembled the major post-Ficoll-Hypaque peak

having an identical modal volume of 153p3. Analysis of

the fractions collected at each flow increment demonstrated

a gradual decrease in the 153p3 peak and the appearance

of a 317p3 peak starting at 25 ml/minute. This latter

peak corresponded to the shoulder of the original post-

Ficoll-Hypaque preparation. The 28 ml/minute fraction,

which had the highest monocyte percentage, showed the

lowest percentage of cells in the 153p3 range and the

highest percentage in the 317p3 range. The R/O fraction

had a modal volume slightly greater than the 28 ml/minute

fraction (354p3). We concluded that the size profile of

the cells was a useful indicator of the differential cell

composition with a peak at 153V3 being characteristic of

lymphocytes and 317p3 of monocytes.

In the next series of experiments, separation was

achieved using only two flow rates (Table 4). As in the

previous method, the first collection was made at 24

ml/minute but this flow rate was now held constant until

400 ml had been collected. For purposes of analysis, this

collection was divided into two fractions of 200 ml each.

The first 200 ml (24A fraction) contained 92% lymphocytes,

5% monocytes and 3% Kurloff cells. It displayed a volume

peak at 153i3 and had a low shoulder representing cells

with volumes greater than 300p3 (Figure 1). After 200

ml had been collected, very few cells exited the chamber so

that the second 200 ml fraction contained less than 1 x

107 total cells. This fraction, designated 24B,

contained cells which displayed a bimodal size profile

consistent with the numbers of lymphocytes (28%) and

monocytes (64%) found by differential cell counting (Figure

1). A second collection was made at a flow rate of 28

ml/minute and all cells capable of eluting were collected

in a total volume of 200 ml. This fraction had an average

monocyte purity of 81% (Table 4). Modal volume analysis of

this fraction displayed a single narrow peak at 317p3 and

failed to detect any cells with a peak volume of 153u3

(Figure 1). The R/O cells were similar in differential

composition and volume analysis to those obtained when

elutriation was carried out at 1 ml/minute increases in

flow rate.

The morphology of the monocytes obtained by the two

techniques of elutriation was identical (Figures 2-8).

Monocytes collected at 24 and 28 ml/minute possessed an

eccentric, reniform nucleus, a glassy sky-blue cytoplasm

when stained with Wright-Giemsa and a slight cytoplasmic

vacuolization. The R/O monocytes were larger, possessed a

centrally located large nucleus and had a pyroninophilic

cytoplasm with vacuolization. To further delineate the

significance of the cells contained in the 24B fraction,

24A and 28 ml/min cells were reconstituted and re-eluted

using the two flow rate technique (Table 5). A 24B

fraction of identical total cell number was again

recovered. This fraction had a higher percentage of

monocytes than the original 24B fraction with a modal

volume identical to that obtained with the 28 ml/min (317


In comparison to incremental flow rates, the two flow

rate procedure yielded greater purity of the intermediate

sized monocytes and was easier to perform and control. For

these reasons, the two flow rate procedure was utilized in

subsequent studies.



Volume analysis of guinea pig mononuclear cells
separated by counter-flow centrifugation
elutriation. Drawing reproduced from an actual
computer smoothed volume plot obtained from a
Particle Data System pulse height analyzer.

'-4 cn-.
lv .-~D

C -H
4-) 4-'



Ln La
O- -


0 '


H) -1

*4 rr4










8 I:









n 23

- o

U) -H



o Ln
+1 +1

+ +1
o rz

+1 +1

+1 +1




0 0


.dJ -H



U <)

) 8



(0 .8-

C( 4Ji






O1 v) 0

0 r -1

- :j' (N
1~~- -4

Y %0m m
(N r-4

l^- 0 rl

In i


u q
a) 0









Oc w








-H M

44 -
B- u


;8 l




0 a)
4 r-O
o -H

U f0

-4 .4

( ra
(U m


z g

0 *t



01 f
'B (

c: ra
0 *
*H 3+
0 CN
.i-4 a) -rH

4J 4
Vo a)

8 0)

0 ,c

4-' CP+U
8 -Ia)


Morphology of guinea pig monocytes (peripheral
blood smear x 2500). The monocyte in this
peripheral blood smear is indicative of the
morphology of the majority of guinea pig mono-
cytes. The nucleus is convoluted and classi-
cally reniform and the cytoplasm is finely
granulated with well-defined cell borders.

* I
* e

~ a

t 4

1 7


1~ i



Morphology of guinea pig mononuclear cells
separated by counter-flow centrifugation
elutriation (x 2500).
Post Ficoll-Hypaque mononuclear cell
preparation before elutriation. The majority
of cells are lymphocytes but monocytes and
Kurloff cells are evident. Large numbers of
platelets contaminate the preparation before






Morphology of guinea pig mononuclear cells
separated by counter-flow centrifugation
elutriation (x 2500).
The 24A fraction (24 ml/minute, 0-200 ml
collection). This fraction is composed
largely of lymphocytes. In the center is a
small monocyte with its typical reniform
nucleus and scant cytoplasm.




Morphology of guinea pig mononuclear cells
separated by counter-flow centrifugation
elutriation (x 2500).
The 24B fraction (24 ml/minute, 200-400 ml
collection). This fraction has predominantly
monocytes with lesser numbers of lymphocytes
and Kurloff cells.


Morphology of guinea pig mononuclear cells
separated by counter-flow centrifugation
elutriation (x 2500).
The 28 ml/minute fraction. This fraction
contains intermediate-sized monocytes
characterized by a reniform nucleus and slight


Morphology of guinea pig mononuclear cells
separated by counter-flow centrifugation
elutriation (x 2500).
The rotor-off (R/O) fraction. This fraction
contains large monocytes which have a
spherical nucleus, abundant cytoplasm and
heavy vacuolization. Although not shown, an
occasional monocyte is binucleated.


Morphology of rotor-off (R/O) monocytes of
guinea pig peripheral blood (x 2500).
R/O monocytes constitute an extremely small
percentage of the total blood leukocytes in
the guinea pig (0.14%). As depicted, this
cell may be binucleated with the nucleus more
spherical than those of the majority of guinea
pig monocytes. The cytoplasm is also more
abundant and more heavily granulated with
irregular cell borders.

Histochemical Staining

Post-elutriation monocytes were analyzed for non-

specific esterase, acid phosphatase, and peroxidase. All

guinea pig monocytes were non-specific esterase positive

with monocytes of the 24A fraction having a 1+ localized

reaction, the 24B fraction a 1+ to 2+ localized reaction,

the 28 ml/minute fraction a 2+ diffuse reaction and R/O

monocytes a 4+ mixed reaction. Acid phosphatase was graded

2+ in R/O monocytes and in monocytes of the 24B and 28

ml/minute fractions but zero in monocytes of the 24A

fraction. Peroxidase was negative in monocytes from all

fractions using the standard Kaplow staining technique

(95,96), although a few cells of the 28 ml/minute and R/O

fractions showed a + perinuclear reaction. However, 40% of

post-Ficoll Hypaque guinea pig monocytes were positive (1+

to 2+) for peroxidase using Meltzer's modified peroxidase

method (130). With this technique, 41% of the 24A, 45% of

the 24B and 33% of the 28 ml/minute monocytes were positive

displaying a 1+ to 2+ reaction. Only 4% of the R/O

monocytes were positive.

Adherence Characteristics of Guinea Pig Monocytes

Adherence of human and murine monocytes to plastic,

nylon wool, microexudate or fibronectin (plasma)-coated

plates has been used either to purify monocytes (1,67,110,

236) or to deplete them from preparations of lymphocytes

(93,145) as well as natural killer (NK) cells (81).

Accordingly, the capacity of guinea pig monocytes to adhere

to similar substrates was investigated, searching not only

for a method to increase monocyte purity, but also for

differences between the monocyte fractions.

Post-Ficoll-Hypaque guinea pig monocytes or monocytes

contained within the 24A, 24B and 28 ml/minute collections

did not adhere to plastic tissue culture flasks or to BHK-

21 microexudate coated plates (Table 6). Addition of gram

negative bacteria (Pseudomonas fluorescens) or 5 pg/ml

lipopolysaccharide caused monocyte adherence within two

hours demonstrating that adherence in a serum containing

medium was an acquired property of these cells. In serum

free medium, both monocytes and lymphocytes adhered but

detachment of the monocytes did not permit adequate

recovery. The R/O monocytes suspended in serum did not

adhere readily to plastic but did so to microexudate-coated


Gelatin-coated plates incubated with plasma absorb the

cell adherence-promoting protein fibronectin (19,67). When

added to such plates, monocytes of the 24A fraction

demonstrated relatively low adherence at all incubation

times up to 8 hours. In contrast, a moderate number of 24B

monocytes and nearly all 28 ml/minute monocytes adhered

after 1 hour although longer incubation times caused the

majority of monocytes to detach (Table 6). R/O monocytes

adhered to fibronectin-coated plates readily and so

strongly that they could not be detached.

Nylon wool columns often are used to separate lympho-

cytes and NK cells from monocytes using a 1 hour incubation

in a serum containing medium (81). Under such conditions,

however, only 18% of the 24A monocytes and 40% of the 24B

monocytes adhered although the 28 ml/minute and R/O mono-

cytes were adherent (74 and 77% respectively). After 4

hours incubation on the column, greater than 70% of the 24

ml/minute monocytes were retained.

Percoll Gradient Separation

Cells isolated by CCE were subfractionated by density

using a continuous Percoll gradient (Table 7). Cells of

the 24A fraction formed one diffuse and two distinct

bands. One distinct band corresponded to a mean density of

1.062 g/ml and contained 85% of the Kurloff cells applied

to the gradient with a purity of 91%. Kurloff cells had a

single volume peak at 230p3. A second distinct band was

identified at a mean density of 1.075 g/ml which contained

86% of the applied monocytes with a purity of 35%. This

band which also contained 65% lymphocytes, had a bimodal

volume peak with maximums at 153p3 lymphocytess) and 283

P monocytess). The morphology of the Kurloff cells and

small monocytes isolated from the 24A fraction by Percoll

gradient centrifugation is presented in Figures 9 and 10.

Cells of the 24B fraction yielded the same two density

bands as the 24A fraction. The 1.062 g/ml band had a

Kurloff cell purity of 60% (247p3) while the 1.075 g/ml

1- mN III Ci c r c

0 OtD Mnon aso
N o 0 com rm

(N o0 w OO r- co-i
1-1 C1 (n 1-4 r-' W


to rCO







8 c
o U

4 ,


r- C)(






-s i




-P w





CN in r-
cc 1



0 cc

m cc

m CY








0 (D

4J) OJ
O (0

4 .i

T4~J C)

8 8
C .6 C

0 '0 0

4-1 P
$4 Z
P- : 0

mm a
4 a)

-P r n

ro 0)
0 t

r-i3 B
C0 O w
r. 0

Fg c

ON- r- d
ti( 0 ? <

*H 41 e(

ru{s u

P-4 r-

cc o








0 >




4 14
tl (
Q 0


o 0
4-) C 4-)
0 0^ rr
.H su
S4. La

4) 4


Morphology of Kurloff cells and small
monocytes of the 24A fraction after Percoll
density gradient centrifugation (x 2500).
Mean density of 1.062 g/ml. This band is
composed of 91% Kurloff cells. Note the
eccentric inclusion bodies which characterize
these cells.


Morphology of Kurloff cells and small
monocytes of the 24A fraction after Percoll
density gradient centrifugation (x 2500).
Mean density of 1.075 g/ml. This band is
enriched in small monocytes (35%) although
lymphocytes are still the predominant cell

band had a monocyte concentration of 85% (30013) with an

80% monocyte recovery. The majority of lymphocytes from

both fractions formed a diffuse band below a density of

1.080 g/ml. Percoll gradient separation of the 28

ml/minute fraction yielded one distinct band at a mean

density of 1.075 g/ml. This band was composed of 85%

monocytes (70% yield) and had a single volume peak

identical to that of the original preparation (317p3).

It was concluded that a combination of CCE and Percoll

gradient separation resulted in a method to isolate (a)

Kurloff cells in high yield and purity, (b) a population of

small monocytes (283p3) with a purity of 35% enriched

from 5% of the 24A fraction and (c) a population of

intermediate-sized monocytes (300p3) of high purity (85%)

from the 24B fraction. While Percoll gradient separation

did not improve the purity of the large monocytes isolated

by CCE at 28 ml/minute, it did demonstrate that these

monocytes had a mean density similar to that of small and

intermediate-sized monocytes.


A. Antibody-Independent Phagocytosis

Non-immune phagocytosis was performed using two sub-

strates: carbon particles and polystyrene latex spheres.

Both particles are frequently used to demonstrate

phagocytosis and are readily engulfed by macrophages of a

variety of animal species including guinea pig.

As shown in Table 8, virtually no carbon was ingested

by 24A, 24B or 28 ml/minute monocytes within a 4 hour time

period. In contrast, R/O monocytes readily engulfed this

material within 30 minutes. The amount of carbon ingested

by R/O monocytes did not increase from 30 minutes to 120

minutes but almost doubled over the ensuing 2 hours of

incubation. Post-Ficoll-Hypaque monocytes demonstrated a

very low amount of carbon ingestion consistent with the

small numbers of R/O monocytes found in this preparation.

Phagocytosis of albumin-coated fluorescent polystyrene

spheres (CovaspheresR) was next examined using both

cytocentrifuge and wet-mount preparations under a Zeiss

epifluorescent microscope. After 15 minutes of incubation,

monocytes of the 24A, 24B and 28 ml/minute fractions had

beads bound to their surfaces but very few beads had been

ingested (Table 9). Concurrently, the R/O monocytes had

very few surface bound beads but did contain numerous

intracellular beads demonstrating that these cells were

avidly phagocytic. By 60 minutes of incubation, surface

bound beads appeared to have been ingested, as beads no

longer were attached to the cell surface of any monocyte

fraction (Figures 11-13). Monocytes from the 24A fraction

had the lowest phagocytic activity while the 24B and 28

ml/minute monocytes showed moderate bead ingestion. The

R/O monocytes had the highest phagocytic activity, although

the number of engulfed beads had not appreciably changed

between 15 and 60 minute periods. A few R/O monocytes had

ingested so many beads that accurate counting was not

always possible and some appeared to have ruptured.

Examination of the phagocytic index revealed that fewer

monocytes of the 24A fraction were phagocytic than in the

other monocyte fractions. Since the avidity index

indicated that approximately the same number of beads were

ingested by cells actually engaged in phagocytosis in all

fractions except the R/O cells, the low phagocytic activity

in the 24A fraction appeared to be due to fewer monocytes

actually engaging in phagocytosis.

B. Antibody-Dependent Phagocytosis

Both elutriated and unfractionated monocytes were

evaluated for differences in their ability to phagocytize

sheep erythrocytes opsonized with rabbit anti-SRBC IgG.

After thirty minutes, only 28 ml/minute and R/O monocytes

demonstrated a moderate degree of phagocytosis (Table 10).

The small monocytes of the 24A fraction showed a low rate

of engulfment even after 4 hours of incubation whereas the

24B monocytes were moderately phagocytic after 2 hours.

The monocytes of the 28 ml/minute and R/O fractions

displayed a continual increase in the phagocytic rate with

increased time. After 4 hours, it was difficult to count

the erythrocytes in some R/O monocytes due to the large

quantities ingested as well as their resultant fusion

within the phagocytic vacuoles. The phagocytic index

revealed a gradation in the percentage of monocytes which

were phagocytic with the 24A monocytes having the lowest

percentage and the R/O having the greatest number of actual

phagocytes. In contrast to the phagocytic avidity for

latex beads, a difference in phagocytic avidity was

observed between the monocyte fractions. Monocytes of the

24A fraction had low phagocytic avidity, monocytes of the

24B and 28 ml/minute fractions intermediate avidity, and

R/O monocytes high phagocytic avidity for opsonized

erythrocytes. Accordingly, the very low phagocytic

activity of the 24A fraction for opsonized erythrocytes was

due to low phagocytic avidity as well as very few monocytes

actually engaging in phagocytosis.

C. C3-Dependent Phagocytosis

Whole cells and cell wall components derived from

Candida albicans are known to activate the alternative

complement pathway. Further, Morrison and Cutler (144)

have shown that phagocytosis of this fungus by murine

macrophages is dependent upon C3 but not upon antibody

specific for Candida albicans.

We first determined whether or not phagocytosis of

Candida albicans by guinea pig monocytes was similar to

murine macrophages in being dependent upon a heat labile

serum opsonin. Figure 14 presents the kinetics of yeast

uptake by post-Ficoll-Hypaque monocytes in the absence of

serum or in the presence of fresh serum or serum

inactivated by heating at 560C for 60 minutes. Over a 4

hour time period, phagocytosis of Candida albicans in the

presence of heat inactivated serum was identical to the

uptake of the yeast in the absence of serum. These results

demonstrated not only that there was no heat stabile

opsonin in guinea pig serum which supported Candida

albicans phagocytosis but also that fresh guinea pig serum

contained a heat labile opsonin that markedly increased

phagocytosis within 60 minutes although some organisms

continued to be engulfed over the ensuing 2 hours. This

decreased rate of uptake after 60 minutes suggested that

the monocytes were reaching their maximum phagocytic

capacity. The dramatic increase in ingestion in the

presence of fresh serum compared to heat inactivated serum

supports the contention of Morrison and Cutler (144) that a

heat labile opsonin, presumably C3, is involved in the

phagocytosis of Candida albicans. When Candida albicans

was opsonized with fresh guinea pig sera rendered C3

deficient by incubation with anti-C3, less than 2

organisms/100 phagocytes were ingested by post-Ficoll-

Hypaque monocytes in 60 minutes. Thus C3 must play an

integral role in the phagocytosis of Candida albicans by

guinea pig monocytes.

Having demonstrated that a heat labile factor in serum

promotes phagocytosis of Candida albicans by post-Ficoll-

Hypaque quinea pig monocytes, the next experiment examined

the complement dependent phagocytosis of the monocyte

fractions isolated by CCE. As observed in Table 11, all

monocyte fractions were similar to the post-Ficoll-Hypaque

monocytes in that each monocyte fraction required a heat

labile component of serum for optimal phagocytosis of

Candida albicans. While heat inactivated serum promoted a

low rate of phagocytosis, opsonized yeast were ingested

readily by all monocyte fractions. However, phagocytosis

by the 24A monocytes appeared to take longer than that of

the R/O or 28 ml/minute monocytes. After 120 minutes,

however, the 24A monocytes had ingested only 27% fewer

yeast than the 28 ml/minute monocytes. This modest

reduction in phagocytic activity was due entirely to a

reduced percentage of active phagocytes and not to a

reduced phagocytic avidity. Thus, ingestion of Candida

albicans in the presence of a heat labile opsonin was

similar to non-immune phagocytosis of polystyrene beads in

that both particles were readily ingested by all 4 monocyte

fractions. The modest decrease in phagocytic activity by

monocytes of the 24A fraction was due entirely to a

decrease in phagocytic index rather than a decrease in

phagocytic avidity. In contrast, the antibody-dependent

phagocytosis of opsonized erythrocytes and the non-immune

ingestion of carbon particles was markedly reduced in the

24A fraction compared to the other monocyte fractions.

This reduced phagocytic activity was due not only to a

reduced percentage of active phagocytes but also to a

reduced phagocytic avidity for these particles.












CN 0 H- (N

U Q)

> 0 0(
- Q00'
>14J 4JU

u 10 0 U
.0 .,O r-,

0 1

> 0 0

0 --

x 0

S h

O O 0 0

0 0

0 0



O *

u u

+ 0

0 >

to 4
0 (0

~4 E

- H
C )

44 ,
0 c

Mc >
0) a)



4-f 54


0 re


40 M

04 -
W 4

U) 0

C >

U) .,.
W 0

. CO 0
a- (0 >*i

- > 0

* .



C>- 34



0 r r-i r-1 1 -1

CN cN -N (N -

+1 +1 +1 +1 +1
0 0 0 0 0

+1 +1 +1 +1 +1
o 0 0 0 0
4 -)

0 0


h i C

.6 )


rti 0
> ^


*P O

V rI24



4 -4


U) U


S tQ

4+1 0



0 0


a) U)
a -
(0 ">

4 V8y (

*- 1




0H a





Phagocytosis of albumin-coated fluorescent
polystyrene beads by guinea pig monocytes
isolated by CCE. (Photographed by using
epifluorescence microscopy x 2500).
The 24A fraction (24 ml/minute, 0-200 ml
collection) after incubation for 60 minutes at
370C. 60% of small monocytes ingest a
moderate number of beads whereas the remainder
of the monocytes are not phagocytic.



Phagocytosis of albumin-coated fluorescent
polystyrene beads by guinea pig monocytes
isolated by CCE. (Photographed using
epifluorescence microscopy x 2500).
28 ml/min fraction after incubation for 60
minutes at 370C. A greater percentage of
intermediate-sized monocytes (76%) are phago-
cytic compared to small monocytes (60%).

b ""



Phagocytosis of albumin-coated fluorescent
polystyrene beads by guinea pig monocytes
isolated by CCE. (Photographed using
epifluorescence microscopy x 2500).
The rotor-off (R/0) fraction after incubation
for 60 minutes at 370C. This fraction
contains the highest percentage of phagocytic
monocytes (88%). Unlike the smaller monocyte
fractions, these cells demonstrate greater
avidity for the albumin-coated beads as shown
in this photomicrograph.

H +1 I
4 -) X .4 1
Ha 0 0 m 0 +

rc) ro
0 C 0 O

C r5

+1 +1 +1 +1 +1
8 5n) i y

Sr 0 5

0)- +-4 4 4(

N + + 1+1 + +1 0 *
IO *

~ C *
S. .
~ "n (UI Xq r
4 04 +1 + I + I + 1 +1 ^ *

|- m| %V N 0

0N M) -4 ON
1 2 +1 +1 +1 +1 +1 5

4. L'I S) q
|~1 -^8 w'f

icn +s +' + +1 + Is l 6
c ~ m c nm 4j --IS 4
(N ror'' i CN FN rq ^ '

a 4J .
4-) *

8 0I 0 fL

s-4 0 g 1 L *

L) 2P IS CN c0



Kinetics of yeast uptake by guinea pig post-
Ficoll-Hypaque monocytes demonstrating
enhanced phagocytosis in the presence of a
heat labile opsonin.
Candida albicans was incubated in medium
alone, fresh guinea pig serum or heat
inactivated serum (560C, 30 minutes) for 15
minutes at 40C at a 10:1 ratio of organisms to
phagocytes. Wright-Giemsa stained
preparations were used to evaluate the number
of yeast ingested per 100 monocytes.

1-t r- >-i cc) in
C. cl P Ce' C

in O LA c
co LO 1 0% m



H u



5M !




i X
ry <
V -I

c o v 0 mA
Ln r-4 -4 () m o

















(0D '

Us .

'0 *,"I









x .

4 C


0 1
0 <
(0 (
10 S


u *H

*i N

C (0


Fc Receptors

Guinea pig monocytes were evaluated for the presence

of Fc receptors by a rosette assay in which three sources

of red blood cells were used. The results are summarized

in Table 12.

In the first assay, sheep red blood cells were

sensitized with either rabbit or autologous guinea pig anti-

sheep erythrocyte immunoglobulin. Although the customary

assay for Fc receptors on monocytes uses rabbit anti-sheep

immunoglobulin (180), autologous antisera was also

investigated to determine if more rosettes would form when

both Fc receptor and antibody were derived from the same

species. In general, the percentage of monocytes forming

rosettes with autologous antisera was equivalent to or

slightly greater than the number obtained with rabbit anti-

sheep immunoglobulin. Using sheep erythrocytes sensitized

with rabbit anti-sheep immunoglobulin, the post-Ficoll-

Hypaque monocytes were 35% Fc receptor positive when

examined on cytocentrifuge slides, while direct analysis of

the cells in suspension on glass slides under coverslips

yielded 45% Fc receptor positive monocytes. Since this

difference was consistently observed for all monocyte

preparations examined regardless of the indicator

erythrocyte or source of antibody, the results of the two

methods of observation were averaged. Thus, post-Ficoll-

Hypaque monocytes averaged 40% Fc receptor positive

monocytes. Monocytes of the 24A fraction were 99% Fc

receptor positive, the 24B monocytes 35% and the 28

ml/minute monocytes only 12%. The Fc receptor positive

monocytes of these fractions bound an average of 3.5

erythrocytes per monocyte. The R/O monocytes were 85% Fc

receptor positive and averaged greater than 10 erythrocytes

per rosette. Other Fc receptor positive cells in the 24A

fraction (average of 3 erythrocytes/cell) were identified

as lymphocytes (presumably B lymphocytes). These non-

monocyte Fc receptor positive cells usually comprised 5 to

7% of the 24A fraction. In the 24B and 28 ml/minute

fractions, Fc receptor positive lymphocytes constituted 2%

of the cells of each fraction. A variable number of Fc

receptor positive cells in the R/O fraction were identified

as lymphocytes (2-10%) and granulocytes (8-15%).

Monocytes of the 24A and 24B fractions did not ingest

sheep erythrocytes opsonized with rabbit immunoglobulin

during the 30 minute incubation period used to determine Fc

receptors. During this same interval, 12% of the monocytes

of the 28 ml/minute fraction ingested erythrocytes for a

total of 28 erythrocytes/100 monocytes (phagocytic avidity

of 2.3 erythrocytes/phagocyte). The lower rate of

ingestion of sensitized erythrocytes observed with the Fc

receptor assay compared to the phagocytosis assay (Table

10) at comparable incubation times was accounted for by the

difference in the ratio of erythrocytes to monocytes used

phagocytosiss assay 100:1; Fc receptor assay 10:1).

In the second assay, the source of erythrocytes was

varied and erythrocytes from ox (bovine) and human were

used. Sensitized ox erythrocytes usually are more reactive

with weak Fc receptors. However, this did not prove to be

the case for guinea pig monocytes. Sensitized ox

erythrocytes consistently revealed fewer Fc receptor

positive monocytes than did sheep erythrocytes sensitized

with either rabbit or guinea pig anti-sheep immunoglobulin

(Table 12). This was most dramatically revealed with the

24A fraction where 99% of the monocytes reacted with

sensitized sheep erythrocytes whereas only 19% reacted with

sensitized ox erythrocytes. Human erythrocytes sensitized

with human anti RhOD immunoglobulin were even less

reactive than sensitized ox erythrocytes. Only 1 to 2% of

the monocytes in the post-Ficoll Hypaque preparation formed

rosettes with sensitized human erythrocytes and no Fc

recptor positive monocytes were found in the 24A, 24B or 28

ml/minute fractions. However, the R/O fraction contained

20% Fc receptor positive monocytes which averaged 5 human

red blood cells per rosetting monocyte.

Development of Fc Receptors in Culture

In a previous study of guinea pig monocyte Fc

receptors, Brade and co-workers reported that nearly all

monocytes expressed Fc receptors (23). However, their

study was performed after placing post-Ficoll-Hypaque

monocytes in culture for 48 hours. Since current results

showed only 40% of post-Ficoll-Hypaque monocytes possessed

Fc receptors when tested immediately after isolation, it

was hypothesized that monocytes might be induced to express

Fc receptors in culture.

Post-Ficoll-Hypaque monocytes and monocytes isolated

by CCE were tested for Fc receptors immediately after

isolation and after 48 hours of culture. Teflon bottles

were used for culture to prevent adherence and to allow

recovery of all monocytes. Analysis for Fc receptors was

performed using sheep erythrocytes sensitized with rabbit

anti-sheep erythrocyte immunoglobulin. As seen in Table

13, expression of Fc receptors on post-Ficoll Hypaque

monocytes increased from 41% at two hours after separation

to 97% after 48 hours of culture. Examination of the

individual CCE fractions revealed that 24A monocytes

remained Fc receptor positive in culture while the 28

ml/minute fraction increased from 12% to 98% Fc receptor

positive. In fact, 70% of 28 ml/min monocytes were Fc

receptor positive after 4 hours in culture. Rosette

formation was completely blocked by the addition of 100 mM

of 2-deoxyglucose suggesting that the expression of Fc

receptors was an energy dependent process. Similarly, Fc

receptor expression increased in the 24B fraction and the

R/O monocytes so that Fc receptors expression on these

cells averaged 81% and 99% respectively.


N N (N
0 0 C 4-

(N 3f 3- (1 4-i W)
C* m C* N w a
0 0 0 z
.s. ~o o o c'

H +)

0 I +5 +5 +1 +1 + ? 0
4J Pi
0 W + *4 d
U w

a4- m fmB
4 U 4j
en u 4J 3(

m CN c q CN ^ |o g w
W45- 5.Jl 1 (0

S+ + + + +1
(N O2N 0N B(N

8 e a)
a to
C1 -^ a ) M .
g 0 S VO+1 +1 +1 +1 +1 *5
%D/ CN Ln H ^M

(N, U) O w w Q Min 4-iQ+3+
0) 4) 4J 4J

la .-H

3) 4J -- W
Q) iA ti co r

44 ~ 4Z.
0 o 8 4-i 0 4j
I( iH

to i
I 12 I 2z^l

Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EP03J3QCD_3XCP00 INGEST_TIME 2012-02-20T21:20:05Z PACKAGE AA00009107_00001