Identification of bacterial Fc receptors and characterization of a group C streptococcal Fc receptor

MISSING IMAGE

Material Information

Title:
Identification of bacterial Fc receptors and characterization of a group C streptococcal Fc receptor
Physical Description:
x, 107 leaves : ill. ; 29 cm.
Language:
English
Creator:
Reis, Kathleen J., 1952-
Publication Date:

Subjects

Subjects / Keywords:
Receptors, Fc   ( mesh )
Streptococcaceae   ( mesh )
Genre:
bibliography   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1983.
Bibliography:
Includes bibliographical references (leaves 99-106).
Statement of Responsibility:
by Kathleen J. Reis.
General Note:
Typescript.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
oclc - 12039485
ocm12039485
System ID:
AA00011120:00001

Full Text
















IDENTIFICATION OF BACTERIAL Fc RECEPTORS
AND CHARACTERIZATION OF A GROUP C STREPTOCOCCAL Fc RECEPTOR/


KATHLEEN J.


t
REIS a-


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


UNIVERSITY OF FLORIDA


1983
















ACKNOWLEDGEMENTS



I would like to acknowledge the following people for their

assistance in helping me complete this study.

I would like to thank and express my sincere appreciation to my

advisor, Dr. Michael Boyle, for his guidance, friendship and patience

during my stay in his laboratory.

I would also like to thank Dr. Elia Ayoub for his many helpful

discussions and suggestions throughout this study; and also the

remaining members of my committee, Drs. Kenneth Berns, Parker Small and

Paul Klein for their comments and suggestions.

I would like to offer a special thanks to Dr. Vincent Fischetti,

my outside examiner, for sending me the bacteria and bacteriophage

mentioned in this study and for taking the time to review and discuss

my work.

I would also like to acknowledge Dr. Alan Keitt who encouraged and

supported me during my decision to give up blood banking and enter

graduate school and Dr. Adrian Gee for his advice and lessons in

grammer.

I would like to thank Dr. Michael Lawman whose encouragement and

friendship helped make this past year more rewarding and enjoyable.

Finally, I would like to thank Mrs. Sandy Ostrofsky for helping me

to prepare this manuscript.


















TABLE OF CONTENTS

Page

ACKNOWLEDGEMENTS ............... ........ ..................... ii

LIST OF TABLES ..... ........................................... v

LIST OF FIGURES ................. .......................... .... vi

KEY TO ABBREVIATIONS ......................................... viii

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

CHAPTER

ONE INTRODUCTION ...................... ................. 1

Staphylococcal Protein A ............................. 1
Distribution, Isolation and Properties ............. 1
Immunochemical Applications ........................ 2
Streptococcal Immunoglobulin Fc Receptors (FcR) ...... 3
Distribution of Fc Receptors Among Streptococci .... 3
Immunoglobulin Species and Subclass Reactivities ... 6
Pathogenicity ................. ....................... 9
Isolaton and Properties ............................ 10
Summary .............................................. 14

TWO METHODS FOR SCREENING BACTERIA FOR Fc RECEPTORS AND
FOR QUANTITATING THESE RECEPTORS IN SOLUBLE FORM ..... 18

Introduction ......................................... 18
Materials and Methods ....................... .......... 19
Results .............................................. 23
Discussion ........................................... 36

THREE DETERMINATION OF THE STRUCTURAL RELATIONSHIP OF
STAPHYLOCOCCAL PROTEIN A AND STREPTOCOCCAL Fc
RECEPTORS ............................................ 39

Introduction ......................................... 39
Materials and Methods ............................... 39
Results .............................................. 41
Discussion ........................................... 52











Page

FOUR ISOLATION AND PARTIAL CHARACTERIZATION OF THE Fc
RECEPTOR FROM A GROUP C STREPTOCOCCUS ................ 54

Introduction ......................................... 54
Materials and Methods ................................ 55
Results .............................................. 60
Discussion ........................................... 74

FIVE COMPARISON OF THE FUNCTIONAL AND ANTIGENIC
RELATIONSHIP OF A GROUP C STREPTOCOCCAL Fc
RECEPTOR WITH STAPHYLOCOCCAL PROTEIN A ............... 79

Introduction ..... .................................... 79
Materials and Methods ................................ 79
Results .............................................. 82
Discussion ........................................... 91

SIX CONCLUSION ... ...................................... 95

REFERENCES ............................................................ 99

BIOGRAPHICAL SKETCH .......................................... 107
















LIST OF TABLES


TABLE Page

1-1. Distribution of IgG Fc Receptors on Streptococci ...... 4

1-2. Species and Subclass IgG Reactivities of Bacterial
Fc Receptors .......................................... 7

1-3. Types of Bacterial Fc Receptors ....................... 8

1-4. Characteristics of Fc Receptors Isolated from
Streptococci .......................................... 13

2-1. PA-like Activity in Bacterial Extracts ................ 35

3-1. Effects of Specific Chicken Anti-Protein A Antibody
on the Ability of Bacteria to Adsorb IgG .............. 51

4-1. Partial Purification of a Group C Streptococcal
Fc Receptor ........................................... 66

5-1. Inhibition of Binding of 1251-PA or 1251-FcRc to
Immobilized Human IgG by IgG from Different Species ... 88

5-2. Inhibition of Binding of 1251-PA or 1251-FcRc to
Immobilized Human IgG by Human Myelomas ............... 90
















LIST OF FIGURES


FIGURE Page

2-1. Adsorption of human IgG by 1 x 1010 streptococci ..... 25

2-2. Adsorption of human IgG in the presence and absence
of the corresponding human IgG F(ab')2 fragments ..... 27

2-3. Adsorption of human IgG by varying numbers of
bacteria ............................................. 28

2-4. Adsorption of rabbit IgG in the presence and absence
of the corresponding rabbit IgG F(ab')2 fragments .... 29

2-5. Two-stage assay selective for PA ..................... 32

2-6. Standard Curve generated using purified PA in the
two-stage PA assay ................................... 33

3-1. Competitive binding assay for quantitation of chicken
anti-protein A antibody .............................. 43

3-2. Ability of Staphylococcus aureus Cowan I and various
streptococci to selectively interact with the Fc
region of human IgG or with the Fab2 region of chicken
anti-protein A antibody .............................. 46

3-3. Binding of human IgG to bacteria pretreated with
chicken anti-protein A anti-serum or normal chicken
serum ................................................ 48

4-1. Cellulose phosphate chromatography of 50% (NH4)2
preparation of crude phage lysate .................... 64

4-2. Ion exchange chromatography of phage lysate recovered
from the pass through of a cellulose phosphate column 65

4-3. Nondenaturing polyacrylamide gel electrophoresis of
affinity purified FcRc and crude phage lysate
containing FcRc ...................................... 68

4-4. Nondenaturing polyacrylamide gel electrophoresis of
affinity purified unlabeled FcRc and 1251-labeled
FcRc ................................................. 69

4-5. SDS polyacrylamide gel electrophoresis of 20 ig of
unlabeled affinity purified FcRc ..................... 70










FIGURE Page

4-6. Inhibition of binding of 1251-affinity purified
FcRc and its components to immobilized human IgG by
unlabeled, unfractionated FcRc ....................... 72

4-7. Inhibition of binding of affinity purified
unfractionated 1251-FcRc to immobilized human IgG
by chicken antibody prepared against the major charge
species (peak II) in the FcRc preparation ............. 73

5-1. Inhibition of binding of 1251-PA or 125I-FcRc to
immobilized human IgG by unlabeled PA, affinity purified
unfractionated FcRc or the major FcRc charge species .. 83

5-2. Inhibition of binding of 125I-FcRc or 1251-PA to
immobilized human IgG by antibody against the major
charge species of FcRc or against PA .................. 85

5-3. Inhibition of binding of 1251-FcRc or 125I-PA to
immobilized human IgG by sheep (A), cow (B), or
goat (C) IgG .......................................... 87

5-4. Inhibition of binding of 1251-PA or 125I-FcRc to
immobilized human IgG by human IgG subclass standards
IgG1 (A), IgG2 (B), IgG3 (C), or IgG4 (D) ............. 89


















KEY TO ABBREVIATIONS


EA Erythrocytes sensitized with subagglutinating doses of
antibody

EDTA-gel 0.15 M Veronal buffered saline, pH 7.4, containing 0.01 M
trisodium ethylenediamine-tetraacetate and 0.1% gelatin

FcR Fc receptor contained on or obtained from any bacteria

FcRc Fc receptor contained or obtained from the group C
streptococcal strain 26RP66

I.M. Intramuscularly

PA Protein A, the Staphylococcus aureus Fc receptor

PBS 0.15 M Phosphate buffered saline, pH 7.4
w-
SRBC Sheep red blood cells

VBS-gel 0.15 M Veronal buffered saline, pH 7.4, containing 0.001 M
Mg2+ and 0.00015 M Ca2+, and 0.1% gelatin


viii

















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


IDENTIFICATION OF BACTERIAL Fc RECEPTORS
AND CHARACTERIZATION OF A GROUP C STREPTOCOCCAL Fc RECEPTOR

By

Kathleen J. Reis

December 1983

Chairman: Michael D.P. Boyle, Ph.D.
Major Department: Immunology and Medical Microbiology

The ability of bacteria to remove IgG in the presence of an

equimolar amount of F(ab')2 fragments was used to identify

streptococci with Fc receptors on their surface. A quantitative

competitive binding assay was developed for the measurement of soluble

Fc receptors during extraction and purification procedures. An

antibody prepared against protein A, the staphylococcal Fc receptor,

was used to demonstrate that selected streptococcal Fc receptors and

protein A were antigenically distinct. Using these assays, a group C

streptococcus with high levels of Fc receptor activity on its surface

was selected for further study. The Fc receptor was extracted in high

yield by lysis of the bacteria following infection with bacteriophage

and purified by sequential chromatography on cellulose phosphate, DEAE

and selective elution from a column of immobilized human IgG. Four

hundred micrograms of the purified protein was obtained per gram (wet

weight) of bacteria extracted. The affinity purified receptor was









functionally homogeneous in binding to the Fc region of human IgG,

however, 4 major protein bands were observed on both nondenaturing and

SDS polyacrylamide gels. Antibody prepared against the major 64,000

dalton protein was capable of reacting with all the fractions and

competitive binding studies suggested that the purified Fc receptor is

a single receptor and that the differences in charge and size were due

to covalently bound cell wall constituents. Comparison of the group C

streptococcal Fc receptor (FcRc) and protein A indicated that although

they can compete with each other for binding to the Fc region of human

IgG, the receptors are antigenically unrelated. Differences in species

and subclasses reactivity showed that FcRc bound more efficiently to

goat, sheep, and cow IgG, while protein A bound more efficiently to dog

IgG. Differences were also observed in the reactivity of the two

bacterial Fc receptors towards human IgG suclasses. The reactivity of

the soluble FcRc corresponds to a type III streptococcal Fc receptor

classified by the reactivity of intact bacteria. This is the first

report of the isolation in high yield of a functionally homogeneous

type III receptor.
















CHAPTER ONE
INTRODUCTION



Bacterial receptors capable of reacting with the Fc region of

various classes and subclasses of immunoglobulins from mammals have

been reported on a number of strains of staphylococci (Jensen, 1958;

Forsgren and Sjoquist, 1966) and streptococci (Kronvall, 1973). The

most extensively studied of these is the Fc receptor isolated from

Staphylococcus aureus and designated protein A.

Staphylococcal Protein A

Distribution, Isolation and Properties

Protein A is produced by most Staphylococcus aureus strains as a

cell surface and/or secreted product. Studies of the distribution of

protein A among staphylococci indicate it is present on the surface of,

or secreted by, approximately 90% of all staphylococci studied

(Langone, 1982a; Sperber, 1976). There are, however, marked quantita-

tive differences between the levels of protein A on different staphylo-

cocci. The most widely studied Staphylococcus aureus, Cowan I strain,

has large quantities of surface protein A while the Staphylococcus

aureus Wood strain expresses very low levels (Freimer et al., 1979).

Initial attempts to extract protein A by heating or using lysozyme

resulted in a heterogeneous product (Forsgren, 1969; Forsgren and

Sjoquist, 1969). A more homogeneous form of protein A has been

described following treatment of Staphylococcus aureus Cowan strain










with lysostaphin. This receptor has been shown to be an elongated

42,000 dalton protein (Sjoquist et al., 1972; Bj6rk et al., 1972),

which binds selectively to the Fc region of certain immunoglobulin

species, classes and subclasses (Kronvall et al., 1970; Richman et al.,

1982). A number of biological properties have also been attributed to

protein A. Soluble and insoluble forms of protein A stimulate

proliferation of human B cells in in vitro culture systems (Romagnani

et al., 1981) and complexes of protein A and IgG are capable of

activating the classical complement pathway (StAhlenheim et al., 1973).

Additionally, injection of protein A into guinea pigs is capable of

causing Arthus and anaphylactoid type hypersensitivity reactions

(Gustafson et al., 1968).

Immunochemical Applications

The ability of protein A to react selectively with the Fc portion

of immunoglobulin molecules has been utilized in a variety of

immunologic procedures primarily for purifying and quantifying reactive

classes and subclasses of IgG (for review see Goding, 1978; Langone,

1982a). Protein A has also been reported to bind to certain species of

IgA (Patrick et al., 1977) and IgM (Lind et al., 1975). These

properties have been useful in establishing the existence of different

subclasses within isotypes, enabling them to be isolated and

characterized with ease (Lind et al., 1975; Patrick et al., 1977).

Protein A can be efficiently used to remove IgG from human serum for

use in a variety of assays to detect non-IgG antibodies (Langone et

al., 1979; Boyle and Langone, 1979) and protein A-bearing staphylococci

have been used extensively in place of second or precipitating antibody

in a variety of radioimmunoassays (Goding, 1978). More recently










radiolabeled or enzyme-linked protein A has been used as a universal

tracer in a number of assays for cell surface and soluble antigens

(Langone 1978, 1980b; Langone et al., 1979a,b; Boyle and Langone, 1979;

Gee and Langone, 1981, for recent review see Langone 1982b). The

usefulness of protein A for immunochemical studies is only limited by

the range of species, isotypes, and subclasses with which it reacts.

Streptococcal Immunoglobulin Fc Receptors (FcR)

Distribution of Fc Receptors Among Streptococci

In 1973, Kronvall (1973) found that groups A, C and G

streptococcal strains were capable of agglutinating red blood cells

sensitized with subagglutinating doses of antibody (EA), indicating the

presence of an IgG Fc receptor on the surface of these bacteria. The

distribution of immunoglobulin G-Fc receptors among various

streptococcal strains has been studied by a variety of methods.

Methods for detection of IgG FcR on the surface of streptococci

include: 1) agglutination of EA (Kronvall, 1973; Christensen and

Kronvall, 1974), 2) binding of radiolabeled polyclonal or myeloma IgG

(Christensen and Oxelius, 1974; Myhre and Kronvall, 1977, 1980a; Myhre

et al., 1979; Ericson et al., 1979) and 3) mixed reverse passive

antiglobulin haemagglutination assay (Freimer et al., 1979). In this

test bacterial Fc receptors are detected by preincubating bacteria with

purified Fc fragments of IgG. These bound fragments are subsequently

detected by agglutination of red blood cells that have antibodies to Fc

regions of IgG coupled to their surface.

The results from these investigations are summarized in Table 1-1,

and suggest that Fc receptors can be found with a high frequency on










TABLE 1-1

Distribution of IgG Fc Receptors on Streptococci


Lancefield No. FcR Positive Method of
Source Group Species No. Tested Detection Reference


Human


S. pyogenes
ND
--


Human


Human


Human


Human
Human
Human &
Non-Human
Non-Human


pyogenes
agalactiae
ND


pyogenes
agalactiae
ND
ND


pyogenes
agalactiae
ND
ND


equisimilis
dysgalactiae
zooepidemicus


C S. equii


Human G
Bovine G
(B-hemolytic)
Bovine G
(a-hemolytic)


Human
Oral
Isolates


mutans
sanguis
milior
salivarius
miller


9/32
5/10
1/15

3/3
1/3
3/3
2/3

19/30
0/40
25/30
0/30
25/30

4/7
0/2
2/2
0/3
2/2

10/10
12/12
18/18


(28%)
(50%)
(6%)

(100%)
(33%)
(100%)
(63%)

(63%)
(0%)
(83%)
(0%)
(83%)

(57%)
(0%)
(100%)
(0%)
(100%)

(100%)
(100%)
(100%)


Kronvall
(1973)


a,b




a





d





b


7/7 (100%)

17/20 (85%)
13/16 (81%)


Christensen
and Oxelius
(1974)


Myhre and
Kronvall
(1977)



Freimer
et al.,
(1979)



Myhre and
Kronvall
(1980a)


Myhre et al.
(1979)


0/18 (0%)


0/11
0/6
0/5
0/5
0/4


(0%)
(0%)
(0%)
(0%)
(0%


Ericson
et al.
T97-9)


a = uptake of radiolabeled polyclonal IgG
b = uptake of radiolabeled human myeloma IgG
c = agglutination of sheep red cells (SRBC)
d = mixed reverse passive hemagglutination


ND = Not Determined
NG = Not Grouped










human isolates of group A (35/72, 49%), group C (45/55, 82%) and group

G (47/70, 67%) streptococci. Although no direct correlation between M

proteins found on group A strains and FcR has been found, it has been

noted that some strains are more likely to produce Fc receptors than

others, e.g., serotype M8 (5 of 6 strains tested) and serotype M22 (23

of 23 strains tested) were positive for either surface FcR or hot-acid

extractable FcR, while none of the 54 serotype M12 strains had either

surface or hot-acid extractable FcR (Havlicek, 1978; Burova et al.,

1981). Recent reports indicate that IgG Fc receptors can be detected

in hot-acid extracts of serotype M12 positive Fc receptor negative

strains following passage in mice. Appearance of an extractable FcR,

is accompanied by a loss of the surface M type 12 antigen (Burova et

al., 1980, 1981; Christensen et al., 1979a).

The stability of FcR production by a given bacterial strain during

subculture or storage is an important consideration when selecting

strains for study of these receptors. There are conflicting reports

concerning the stability of Fc receptors on the bacterial cell surface.

Christensen and Kronvall (1974) reported that repeated subculturing of

groups A, B, C, D, and G streptococci did not suppress the ability of
'V
these bacteria to agglutinate EA. Havllcek (1978) looked at Fc

receptor activity in hot-acid extracts of a number of fresh isolates,

collection strains and freeze-dried strains of group A streptococci and

reported that Fc receptors are found more frequently on fresh than on

collection or freeze-dried strains. Extracts of 38 of 175 (22%)

collection strains and 10 of 32 (31%) freeze-dried strains had Fc

receptors while 49 or 49 (100%) fresh isolates had detectable levels of

Fc receptor activity as measured by the ability to agglutinate EA.










Havlicek does not indicate whether these fresh isolates were followed,

after subculturing or storage, for loss of FcR activity. Schalen et

al. (1983) found that 10 of 20 (50%) reference strains of group A

streptococci of various M types had Fc receptors as detected by binding

of radiolabeled polyclonal and monoclonal IgG. Additionally, fresh

isolates of group A streptococci have been reported to have a greater

capacity to adsorb radiolabeled myeloma IgA than do collection strains

(Christensen and Oxelius, 1975; Schalen, 1980). Although the frequency

of FcR positive group A strains appears to be higher in fresh isolates

than in older cultures, the stability of Fc receptors appears to follow

no precise pattern during subculturing and storage.

Immunoglobulin Species and Subclass Reactivities

Characterization of the Fc-reactivity of streptococci is based on

the ability of serum or IgG fractions from different animal species to

inhibit the binding of radiolabeled human IgG to a variety of bacteria

(Myhre and Kronvall, 1977, 1980a; Myhre et al., 1979). Inhibition

experiments with isolated Fc fragments demonstrated that the binding

site for the immunoglobulin Fc receptor is located in the CH2 domain

of the Fc fragment of the immunoglobulin molecule (Christensen et al.,

1976; Myhre and Kronvall, 1980b). Subclass reactivity has been

measured by the ability of a variety of radiolabeled myeloma proteins

to bind directly to the bacteria (Kronvall et al., 1979a; Myhre and

Kronvall, 1981b). Using these assays Myhre and Kronvall have described

five different Fc receptor types (see Tables 1-2 and 1-3).

These results are semi-quantitative and only reflect major

differences in reactivities. Similarily, known differences in affini-

ties of both staphylococcal protein A (Langone, 1978) and streptococcal











TABLE 1-2

Species and Subclass IgG Reactivities of Bacterial Fc Receptorsa

IgG Fc Receptor Typeb
I II III IV V


Human








Mouse





Cow



Sheep



Goat



Horse




Rabbit

Guinea Pig

Rat

Dog

Cat

Pig

Chicken


IgG1
IgG2
IgG3
IgG4
IgAc
IgM

IgG1
IgG2a
IgG2b
IgG3

IgGI
IgG2

IgGl
IgG2

IgG1
IgG2

IgG(ab)
IgG(c)
IgG(T)

IgG

IgG

IgG

IgG

IgG

IgG

IgG


+++
+++
+++
+++




4-
+++
+++
+++


- +

- +






- +++
- +



- +

(+) (+)

NT +++

NT NT

NT

NT


+++


NT +++

NT NT


indicates strong reactivity
weak, atypical reactivity


+ = indicates low reactivity


a = Summarized from Kronvall, 1973; Myhre and Kronvall, 1977, 1980a,b, 1981b
b = see TABLE 1-3
c = IgA Fc receptor is found on certain group A streptococci, but is
distinct from the IgG Fc receptor (Christensen and Oxelius,
1975; Kronvall et al., 1979a; Schalen et al., 1980).
d = NT, not tested


+++ =
(+) =










TABLE 1-3

Types of Bacterial Fc Receptorsa


Type of IgG Fc Receptor Bacterial Species


Type I Staphylococcus aureus (Protein A)

Type II Group A streptococci

Type III Strep. equisimilis (group C)b
Strep. dysgalactiae (group C)
Human group C streptococci

Type IV Bovine B-hemolytic group G streptococci

Type V Strep. zooepidemicus (group C)


a = From Myhre and Kronvall, 1981b.
b = The group C strain Streptococcus equii binds negligible levels of
human IgG (Myhre and Kronvall, 1980a).









Fc receptors are not reflected in the results presented in Tables 1-2

and 1-3 (Kronvall, 1973; Myhre and Kronvall, 1977, 1980b, 1981a). More

sensitive assays using purified protein A have been very useful in

defining its relative species and subclass reactivity. For example,

using a quantitative competitive binding assay Langone (1978) has shown

that 50% inhibition of binding of 1251-PA to immobilized rabbit IgG

required 60 ng of rabbit, human or guinea pig IgG, 135 ng of pig IgG,

4,500 ng of mouse IgG and 40,000 ng of sheep IgG. Similar assays with

streptococcal Fc receptors have not been carried out since purified

receptors and hence the radiolabeled tracers are not available.

Pathogenicity

No one factor has been shown to be solely responsible for the

pathogenicity of streptococci, but factors such as M proteins (Bisno,

1979; Lancefield, 1954; Todd and Lancefield, 1928), opacity factor

(Maxted and Widdowson, 1972), hyaluronic acid (Kass and Seastone,

1944), neuraminidase (Davis et al., 1979) and the M-associated proteins

(Maxted and Widdowson, 1972) have all been implicated. There is little

information available on the relationship of streptococcal Fc receptors

and pathogenicity.

Passage of streptococci in mice has been used to increase mouse

virulence and passage in fresh human blood has been used to select

phagocytic resistant strains, both of these methods are currently in

use for the study of virulence factors. Several studies of FcR produc-

tion and virulence have suggested a possible role for Fc receptors in

pathogenesis. Mouse passage of a number of group A streptococci (M

serotypes 3, 12 and 46) has resulted in increased virulence in mice,

loss of M protein and a concurrent production of an extractable Fc









receptor (Christensen et al., 1979a; Burova et al., 1980). Burova et

al. (1981) extracted and semipurified an Fc receptor from a group A

serotype M15 streptococcus that was highly virulent in mice. When this

receptor was mixed with a serotype M12 strain of low virulence, an

increase in virulence was observed both in vivo and in vitro.

There has been recent evidence to suggest that the synthesis of at

least two pathogenic factors of group A streptococci, M protein and

opacity factor is under extrachromosomal regulation (Cleary et al.,

1975; Spanier and Cleary, 1980). Burova et al. (1983) have looked at

the role of plasmids in the expression of antiphagocytic activity,

opacity factor, IgG, and IgA receptors and suggest that expression of

these factors may be triggered by insertion of plasmid DNA into the

bacterial chromosome.

These findings show that Fc receptor activity can be induced, or

expression enhanced, along with other factors that have been associated

with virulence. Before the importance of Fc receptors in virulence can

be critically assessed, purified Fc receptors will have to be isolated

and studied in more detail.

Isolation and Properties

The earliest attempts to extract a soluble streptococcal Fc

receptor employed the hot-acid or Lancefield technique (Lancefield,

1928). This method involves adjusting the pH of a bacterial suspension

to pH 2.0, heating in a boiling water bath for 10 mins and neutralizing

the pH. A modification of this procedure, the hot-alkaline extraction,

is carried out in the same manner except that the pH is adjusted to

10.0. These methods are commonly used for the extraction of the

streptococcal group specific carbohydrate and streptococcal proteins,









including the M protein from group A streptococci. Extracts obtained

by these methods from groups A, C, and G streptococci were able to

agglutinate antibody sensitized red blood cells (Christensen and

Kronvall, 1974; Christensen and Holm, 1976; Havlcek, 1978; Schalen et

al., 1978, 1980; Christensen et al., 1979a,b), inhibit the binding of

radiolabeled IgG to intact streptococci (Christensen and Oxelius, 1974;

Christensen and Holm, 1976) and precipitate human serum or IgG

components (Schaln et al., 1978, 1980; Christensen et al., 1979b;

Grubb et al., 1982). Extraction of a group C streptococcus by phage

lysis, and a group A strain by autoclaving at 1200C for 30 minutes, was

used by Christensen and Holm (1976) to obtain soluble Fc receptors.

The same investigators reported only limited success when hot-acid or

hot-alkaline methods, ultrasonic treatment, X-pressing or Mickle

disintegration was used on a limited number of strains (see Table 1-4).

It is difficult to compare extraction methods, purification yields, or

molecular weight determinations of the products obtained since four

different strains were extracted, each by a different method.

Additionally, each investigator used a different assay to quantitate

the solubilized Fc receptors and for determining the molecular weight.

Only Grubb et al. (1982) using a group A serotypee M15) streptococcal

strain was able to obtain a functionally homogeneous product. The Fc

receptor was solubilized by hot-alkaline extraction and purified by

DEAE ion exchange chromatography and immunoadsorption on an IgG column

and subsequent elution with 0.1 M sodium acetate, pH 3.5, containing

0.5 M NaCI. All steps in the extraction and purification of this

receptor required the presence of protease inhibitors. The authors

report that when protease inhibitors were not included during the










extraction and isolation, results were not reproducible and Fc receptor

activity often decayed rapidly. Six hundred pg of purified Fc receptor

could be obtained from 60 g bacteria (wet weight), and a yield of 11%

Fc receptor present in the crude hot-alkaline extract was achieved.

The authors also report that some charge and size heterogeneity on

SDS-polyacrylamide and agarose gel electrophoresis was evident in the

purified material, even though it was determined to be functionally

pure based on its ability to bind immobilized monoclonal or polyclonal

IgG.

The Fc receptor(s) of group A streptococci have been shown to be
4v
distinct from the M proteins (Havlicek, 1978; Christensen et al.,

1979b; Schalen et al., 1980), the group carbohydrate and peptidoglycan

(Christensen et al., 1979b; Schalen et al., 1980) as well as

lipoteichoic acid (Schalen et al., 1980). A group C Fc receptor

extracted by phage lysis and a group A Fc receptor extracted by heating

to 120C, were both shown to be sensitive to trypsin and heat treatment

at 95C for 10 mins at pH 2.0 (Christensen and Holm, 1976). The group

A streptococcal Fc receptor extracted and purified by Grubb et al.

(1982) (see Table 1-4) was found to have a molecular weight of 29,500

when determined by SDS-polyacrylamide gel electrophoresis. However,

gel chromatography run under nondenaturing conditions revealed a

hydrodynamic volume between that of IgG and IgA, indicating that the

native molecule is either elongated or forms oligomers. Grubb et al.

(1982) have also determined the amino acid composition of their group A

streptococcal Fc receptor and shown that it is distinct from that of

staphylococcal protein A.



















Sco


o CBw -


0
4.J
O
,-1


*-- C
'4-1 *i



0. 0


-4
a) -,

0


--
O



0-'
O -
0 -4

0


0

0
a ,o

(C


C:
0
*-
4-
ca

U0 C
-4
- 4-4
.L. -.4
w w
ca$Z










P-

-4




0


-o4 o
S4-J Z "W



'0 Z C

I 0 C I
us oa
S1-41 <0 CM1 41U

' b I
IC' I


U



> 0

cc
-4'-<

0 -





u
1-i
0


4.J r-
S-4 u


)a







0




0h




--4


0 0
-4




.0 -'.0
-4 10
a 4 oo



E aca
0 C10
0 0 00


-C


0 o
C-4 C)


ca

ou
N
(0
'0
t1o
(0

N
C0

0
N
0. (8I
C3


' h





0 E-
o _


--4




0











4 .




0 CL


MI


.0 01

0


ca

0 r-
C 0)

O1
0 0

0
=C


.,-1


$ o
0

*-- C 4-'
3 -4 C0

0) *-
$4 4-4


W









a C a) 4- I
0. N 0





C 0 A 2


S0.

O -
C) ^-


O
0



C


0 a
0 X













-4
O a
o

0 0





o d








0
.-i -
0. >
















N 3
a o



,-4







00
O

















-o a U-
0 0























O O




c ,
u 0












































I U
-4 c
0o





























Coa 0
'o u
a) '0












C C
o--4 44 4

















C 00
*^ 0 *
II ~ II I
t0XI<-


PSX

u


- >
XI --
3 4-1
- U
0<
Cil





C
0
-4
4-J
u

1-1
x
X


C:
--4
cc

4-1
U)









The role of Fc receptors in the pathogenesis of streptococcal

infections and their potential as immunochemical reagents has not been

fully assessed. Before these types of studies can be undertaken,

methods of obtaining reasonable quantities of purified Fc receptors

must be developed.


Summary

The existence of immunoglobulin Fc receptors on streptococci was

established by Kronvall in 1973. Inhibition of binding of radiolabeled

IgG to streptococci by various species, classes and subclasses of

immunoglobulins has been used to determine the specificity of

immunoglobulin reactivity of a variety of streptococcal strains. Based

on the pattern of binding of differing species of immunoglobulins to

streptococci, four types of streptococcal Fc receptors have been

described that are distinct from the staphylococcal protein A

receptor.

Soluble streptococcal Fc receptors have not been as extensively

studied as staphylococcal protein A. Attempts to extract and purify

streptococcal Fc receptors using phage lysis, heat and hot-acid or

hot-alkaline extraction procedures have met with limited success. The

use of protease inhibitors during the extraction procedure has recently

been reported to give increased yields of a stable product extracted by

the hot-alkaline method. The extracted material is capable of

agglutinating antibody sensitized red blood cells, inhibiting the

uptake of radiolabeled IgG by intact streptococci and precipitating

human IgG in agarose gels containing dextran. The molecular weights of

the streptococcal Fc receptors extracted to date have a range from









30,000 to 100,000 daltons, depending on the organism extracted, the

method of extraction and the method used for molecular weight

determinations, i.e., gel filtration or SDS polyacrylamide gel

electrophoresis.

The biological properties of streptococcal Fc receptors have not

been extensively characterized. Streptococcal Fc receptors are clearly

distinct from the streptococcal virulence factor, the M proteins found

on group A streptococci, but may be associated with them. Recent

reports indicate that production of streptococcal Fc receptors may be

induced or enhanced by mouse passage of selected bacterial strains, or

by the insertion of plasmid DNA into the bacterial chromosome.

Additionally, partially purified Fc receptors have been shown to alter

pathogenicity and phagocytosis of selected group A strains in in vivo

and in vitro experiments.

The sparsity of information on the physicochemical and biological

properties of streptococcal Fc receptors is due, in large part, to the

absence of reproducible sensitive assays to detect these proteins. A

number of problems have been encountered with the methods currently in

use for the detection and quantitation of streptococcal Fc receptors.

Hemagglutination reactions are frequently weak and best results are

achieved when human red cells sensitized with the Ripley antibody, a

human anti-Rh antibody, are used. Hemolysis of red blood cells by

bacterial hemolysins has been reported, but may be overcome by heat

treatment of culture supernatants or extracts prior to testing. When

using 1251 labeled IgG to detect Fc receptors, high levels of

adherence of 1251 IgG to reaction vessels and entrapment by

bacteria have been observed, despite the presence of nonspecific









protein sources and low levels of detergent. Self-aggregating

streptococci can cause entrapment of unsensitized, as well as

sensitized red blood cells, and entrapment of 1251-labeled IgG

making interpretation of results difficult.

The ability to identify the bacterial groups responsible for

selective reaction with the Fc region of IgG and to isolate them in

high yields in a form that maintains functional activity would be

clearly useful. In particular, bacterial groups which react with a

wider range of immunoglobulin species, classes, and subclasses than

protein A would enable the immunochemical techniques currently

employing protein A to be expanded. For example, some group A, C and G

streptococci have been reported to react with human IgG3 and group C

streptococci have been found that react with sheep and cow IgG. These

activities are not associated with protein A. Isolation of bacterial

groups with fewer species or subclass reactivities than protein A would

enable techniques to be developed using both bacterial reactivities to

focus on a narrower range of specificities. The usefulness of

bacterial Fc receptors in immunochemical studies is at present limited

only by the range of species, class and subclass reactivities with

which these receptors will react. The ability to isolate Fc receptors

in a functionally homogenous, soluble form with reactivities different

from those of protein A would allow the highly successful approaches

using staphylococcal protein A to be extended. Additionally, purified

bacterial Fc receptors from different types of streptococci will be

required to study IgG-FcR interactions and to define their role in

vivo.









Consequently, the purpose of this study was to identify strepto-

cocci with surface receptors capable of reacting selectively with the

Fc region of immunoglobulin molecules. Once methods for the detection

of such molecules were developed, suitable Fc receptor-positive

streptococcal strains were used to develop methods for the extraction

and purification of these receptors. Emphasis was focused on

streptococcal Fc receptors with IgG isotype species or subclass

reactivities different from those of staphylococcal protein A. Once Fc

receptors were isolated, their physicochemical and biological

properties were characterized and their usefulness in a variety of

immunoassays were evaluated. The specific aims of this study were to:

(1) Develop methods for screening bacteria for Fc receptors and

for quantitating these receptors in a soluble form

(Chapter 2).

(2) Determine if the streptococcal Fc receptors were distinct

from staphylococcal protein A (Chapter 3).

(3) Isolate and physicochemically characterize one of these Fc

receptor(s) (Chapter 4).

(4) Characterize the functional and biological activities of the

isolated Fc receptor(s), e.g., species immunoglobulin

reactivities (Chapter 5).
















CHAPTER TWO
METHODS FOR SCREENING BACTERIA FOR Fc RECEPTORS
AND FOR QUANTITATING THESE RECEPTORS IN SOLUBLE FORM



Introduction

The initial aim of this study was to develop methods for detecting

cell surface and soluble Fc receptors. Currently, two basic types of

assays have been used to measure Fc receptors on or secreted by

streptococci. The first utilizes hemagglutination which measures the

ability of intact bacteria or soluble extracts to cause agglutination

of red cells sensitized with subagglutinating concentrations of anti-

red cell antibodies (Kronvall, 1973; Christensen and Kronvall, 1974;

Havllcek, 1978). The second assay measures the ability of bacteria to

bind 1251 labeled IgG (Christensen and Oxelius, 1974; Christensen

et al., 1976; Myhre and Kronvall, 1977).

Both of these assays suffer from a number of technical limita-

tions. For example, a large number of strains of streptococci self

associate and cause apparent agglutination of unsensitized red cells

(Christensen and Kronvall, 1974; Freimer et al., 1979). In studies

utilizing 1251 labeled IgG high background levels of adherence of

label to the reaction vessel have made it difficult to measure

accurately Fc-reactivity (Christensen and Oxelius, 1974; Myhre and

Kronvall, 1977).

This chapter describes a method for the selective, semiquantita-

tive measurement of Fc receptors on the streptococcal cell surface.










The method described is rapid, reproducible, not subject to high

background values and can be used with auto-agglutinating bacteria.

The second part of this chapter describes the development of a competi-

tive binding assay for the quantitation of soluble Fc receptors.

Materials and Methods

Bacteria Strains, Media and Growth Conditions

Laboratory strains and fresh isolates of B-hemolytic streptococci

and Staphylocococcus aureus strains were used in these assays. All

strains were grown in Todd-Hewitt broth (Difco) for 18-24 hrs at 37"C.

They were heat killed at 80*C for 5 mins (Kronvall et al., 1979b),

harvested by centrifugation and washed twice in phosphate buffered

saline (PBS), pH 7.2, containing 0.02% Na azide. The optical density

at 550 nm was determined to standardize the concentration of organisms

used in subsequent tests. Sodium azide was added to a final

concentration of 0.02% to culture supernatants which were stored at 4C

until testing.

Staphylococcus aureus Cowan I served as a protein A (PA)

producing positive control.

The B-hemolytic streptococci were grouped by the Phadebact

Streptococcus Test, Pharmacia Diagnostics.

Immunoglobulin

Stock human and rabbit IgG were prepared by chromatography of

normal human or rabbit serum on DEAE cellulose (Boyle and Langone,

1980). Aliquots were stored at -700C until use.

Purified Protein A

Purified PA was obtained from Pharmacia Fine Chemicals,

Piscataway, New Jersey.










Iodination of PA and IgG

Purified PA (Pharmacia) was radioiodinated by the mild lactoper-

oxidase method using enzyme beads (Bio-Rad) (Morrison et al., 1971).

The labeled protein was separated from free iodine by passage over a

G25 column (PL-10, Pharmacia) and collected in veronal buffered saline,

pH 7.4 containing 0.001 M Mg2+ and 0.00015 M Ca2+, and 0.1%

gelatin (VBS-gel). The labeled material routinely had a specific

activity of approximately 0.3 mCi/mg.

F(ab')2 Preparation

Human and rabbit IgG F(ab')2 fragments were prepared by pepsin

digestion of the stock IgG preparation by a modification of the method

described by Nisonoff et al., (1960). Essentially, 20 mgs of IgG was

incubated in 0.1 M Na acetate, 0.05 M NaCI pH 4.0 with 5% w/w of pepsin

(Sigma) for 18 hrs at 370C. The resulting F(ab')2 fpagments were

separated from undigested IgG and Fc fragments by chromatography on a

Sephadex G-200 column in veronal buffered saline, pH 7.35. The peak

corresponding to the F(ab')2 fragments was pooled and any contami-

nating undigested IgG was removed by adsorption with PA-Sepharose

(Pharmacia). The pooled PA adsorbed F(ab')2 fragments were concen-

trated by negative pressure dialysis, aliquoted and stored at -700C.

IgG Assay

Human IgG in solution was quantified by the procedure of Langone

et al. (1977). In this assay, 0.2 mis of a test sample or buffer was

mixed with 0.1 ml of a standard suspension of rabbit IgG covalently

coupled to agarose beads (Immunobead R-l Bio Rad Laboratories) and 0.1

ml of 1251 PA (approximately 20,000 cpm). After incubation at 37C

for 90 mins, 2 mis of veronal buffered saline containing 0.01 M










trisodium ethylenediaminetetraacetate and 0.1% gelatin (EDTA-gel) was

added to each tube and centrifuged at 1,000 g for 5 mins and the

supernatant fluid decanted. After an additional wash, the

radioactivity associated with the beads was determined in a gamma

counter (either Packard or LKB). The number of counts bound in the

absence of fluid phase IgG was compared to the number of counts bound

to the beads in the presence of known amounts of fluid phase IgG. The

degree of inhibition was determined and a standard curve relating

quantity of IgG to percent inhibition was generated. The quantity of

IgG in the test sample can be determined by comparing the percent

inhibition with the standard curve. Fifty percent inhibition of

binding of 1251-PA to the Immunobeads is consistently achieved with

30-50 ng of IgG.

Adsorption of IgG by Bacteria

The detection of Fc receptors on the surface of bacteria was

determined by the ability of bacteria to adsorb IgG from solution. A

standardized number of organisms, as indicated in the text, was added

to 1 ml of VBS-gel containing 1 pg of IgG and incubated at ambient

temperature for 1 hr. The bacteria were removed by centrifugation and

the residual IgG in an aliquot of the supernatant fluid was measured as

described above. Results are expressed as the percent of IgG

adsorbed.

Protein A (PA) Assay

A two-stage assay for the detection of soluble PA or PA-like

molecules is described in the text. This method is a modification of

the competitive binding assay for the detection of soluble PA described

by Langone et al., (1977).










Hemagglutination Assay

Soluble streptococcal and staphylococcal Fc-receptors were tested

for their ability to agglutinate human red blood cells sensitized with

subagglutinating doses of IgG (Sjiquist and StAlenheim, 1969; Kronvall,

1973; Winblad and Ericson, 1973). Human red cells were drawn in

heparin, washed twice in EDTA-gel, once in VBS-gel, and resuspended to

3% in VBS-gel. Cells were sensitized to subagglutinating doses with

commercial anti-Rh typing antiserum (Gamma Biologicals, Inc.) by

incubating 0.2 mls of two-fold dilutions of anti-Rh antisera with 2 mis

of 3% Rh positive red blood cells (RBCs) for 1 hr at 37C. The cells

were washed three times in VBS-gel to remove unbound antibody and

resuspended to 0.3% in the same buffer.

The ability of bacterial culture supernatants to agglutinate

sensitized erythrocytes was determined in V bottom micro-titer wells.

Fifty pl of three-fold serial dilutions of streptococcal or

staphylococcal culture supernatants was mixed with an equal volume of

sensitized or unsensitized erythrocytes. Hemagglutination was scored

following incubation at ambient temperature for 2 hrs and again after

overnight incubation.

Lancefield Extraction

Bacteria for extraction were obtained after overnight growth at

37"C in Todd Hewitt broth. The bacteria were collected by centrifuga-

tion and washed once in PBS, pH 7.2. Lancefield extracts (Lancefield,

1928) were performed on approximately 0.1 g (wet weight) of bacteria

suspended in 3 mls of 0.15 M NaC1. The pH was adjusted to 2.0 with 1 N

HC1. The tubes containing the bacteria were placed in a boiling water

bath for 10 mins, cooled on ice and the pH adjusted to 7.0 with 1 N










NaOH. The extracts were recovered after removal of bacteria by

centrifugation and filtration through a 0.2 pm filter.

Lysostaphin Extraction

Lysostaphin extracts were performed according to the method of

Sjoquist et al., 1972. Bacteria obtained as described above were

extracted in 3 mis of 0.05 M Tris-HCl, 0.015 M NaC1, pH 7.5 containing

0.2 mgs lysostaphin (Sigma) and 10 pg DNase (Sigma). Following 4 hrs

incubation at 37*C the extracts were recovered by centrifugation

followed by filtration through a 0.2 pm filter.

Results

Quantitative Adsorption of Human IgG by Streptococci

The purpose of the experiments described in this chapter was to

develop a rapid and selective assay to measure Fc receptors on

streptococci, including autoagglutinating strains. A method was sought

that would enable binding via the Fc region to be readily distinguished

from binding through the F(ab')2 region and nonspecific entrapment.

The initial approach was to measure the ability of .streptococci to

adsorb IgG from solution.

In this procedure 1 ml of human IgG at a concentration of 1 pg/ml

was added to a pellet containing approximately 1010 streptococci.

The mixture was incubated for 1 hr at ambient temperature and the

bacteria removed by centrifugation. The residual IgG in solution was

measured in duplicate 0.2 ml aliquots of the supernatant using the

competitive binding radio immunoassay described in the methods. This

assay is based on the inhibition of 1251 protein A binding to

immobilized IgG and is consequently a selective measure of IgG










Fc regions. Using this assay there was no detectable change in the

concentration of IgG added to a blank tube and carried through the

procedure.

The IgG adsorption assay was used to screen a number of fresh

isolates of B-hemolytic streptococci. The results presented in

Figure 2-1 indicate that different levels of IgG binding capacities

could be detected by this method. The Cowan strain Staphylococcus

aureus was included as a reference positive control and the Wood 46

strain as a reference low level positive control (Freimer et al.,

1979).

In addition to the binding of the Fc region of IgG, streptococci

may also bind IgG via the antigen combining sites. The human IgG used

for adsorption had been isolated from a single donor and it would be

unlikely that greater than 10% of the total immunoglobulin would be

directed against bacterial antigens. However, to determine that the

observed adsorption of IgG was Fc specific, the ability of the

adsorbing streptococci (Figure 2-1) to remove IgG in the presence of an

equimolar quantity of F(ab')2 fragments was tested. The F(ab')2

fragments were prepared from the same IgG source used in the adsorption

assay and would contain the same distribution of IgG antigen reactivi-

ties. IgG F(ab')2 fragments in the reaction mixture would compete

with any binding of specific IgG antibody, resulting in decreased

adsorption of, the IgG. However, if binding was through the Fc region,

no decrease would be observed when F(ab')2 fragments were present.

These studies could only be done because the radioimmunoassay for IgG I

am using is based on the competitive binding of 1251 labeled































4i


Hu 5 Hu6 Hu 7 Hu 8 Hu 9 HulO


FIJI


I _____ L -


Hu 12 Hu 14 COWAN WOOD


1010 BACTERIA


Figure 2-1.


Adsorption of human IgG by 1 x 1010 streptococci. One
jg of IgG was mixed with the number of bacteria indicated
and the IgG remaining in an aliquot of the supernatant was
determined. Results are expressed as the percent of IgG
adsorbed by the bacteria, + the standard deviation. Hu6
is a group C streptococcus, the remainder belong to group
A. Staph. aureus strains Cowan and Wood are shown for
comparison.


-II]


-L L










protein A to the Fc region of immobilized IgG and F(ab')2 fragments

are not measured in this assay.

The results presented in Figure 2-2 indicate that the adsorbtion

of IgG was not markedly altered when carried out in the presence of

F(ab')2 fragments. These results indicate that these group A and

group C streptococci bind IgG through the Fc region of the immuno-

globulin molecule. In all cases the capacity of individual strains to

adsorb human IgG was related to the number of bacteria used in the

assay. This is demonstrated in Figure 2-3 for two group A strepto-

coccal strains. These results indicate that certain clinical isolates

of streptococci have adsorbtive capacity for human IgG of approximately

5-35% of that of Staphylococcus aureus Cowan strain, under these assay

conditions.

This approach can also be used to determine the species reactivity

of bacterial Fc receptors by changing the source of IgG used for

adsorption. A representative example of this approach, utilizing

rabbit IgG and the corresponding F(ab')2 fragments, is presented in

Figure 2-4. A comparison of Figures 2-2, 2-3, and 2-4 demonstrates

that the patterns of adsorption of human IgG and rabbit IgG were

similar, with one exception. The group A strain, 529, adsorbed human

IgG but failed to adsorb rabbit IgG. Functional differences between

Fc-reactivity towards different species of IgG has previously been

reported (Myhre and Kronvall, 1977). The assay being used to measure

functional Fc receptors could be readily extended to other species of

IgG classes or subclasses.



















] Human IgG
SHuman IgG
Human F(ab')2


i T
| rf r| | .-,,


J xj l lJx


Hu7 Hu9 Hull Hul2 Hul3 Hul4 Hu26 Hu27
lx 100 STREPTOCOCCI


Figure 2-2.


Adsorption of human IgG in the presence and absence of the
corresponding human IgG F(ab')2 fragments.
Streptococci, 1 x 1010, were mixed with 1 yg of human
IgG and 1 ig of human IgG plus an equimolar concentration
of the corresponding F(ab')2 fragment. The amount of
whole IgG remaining after adsorption by the bacteria was
determined. Results are expressed as the percent of IgG
adsorbed, + the standard deviation. Hu6 is a group C
..streptococcus, the remainder belong to group A.
Adsorption by 1 x 109 Staph. aureus Cowan I is shown for
comparison.


x109
COWAN


-if^.





























































106 101 5xl09 10
NUMBER OF BACTERIA


Figure 2-3. Adsorption of human IgG by varying numbers of bacteria.
Results are expressed as the percent of IgG adsorbed, +
the standard deviation.


0
cn
O

m
- 0

n0





































O
O
Ir
0
V)



























Figure 2-4.


64 529 Hu

Ix 1010 BACTERIA


Adsorption of rabbit IgG in the presence and absence of
the corresponding rabbit IgG F(ab')2 fragments. Staph.
a.ureus strains, Cowan I and Wood 46; group A streptococci
strains 64, 529 and Hull and Group C streptococcus strain
Hu6 were adsorbed with 1 p; rabbit IgG as described in
Figure 2-2 and text.


*









Measurement of Fc-Receptors Released by Streptococci

To assess the role of Fc receptors in the pathogenesis of

bacterial infections it may be valuable to distinguish between cell

bound Fc-reactive material and that secreted by the bacteria during

growth. For this reason a rapid, reproducible method for detecting

Fc receptors in bacterial culture supernatants was sought. Langone et

al. (1977), have previously described a competitive binding assay that

could be used to measure nanogram quantities of IgG or protein A. This

assay was adapted to measure selectively protein A and related Fc

receptors. This has been achieved by carrying out the assay procedure

in two stages as outlined in Figure 2-5.

In the first stage 1 ml of test sample or buffer was mixed with

0.1 ml of a standard sample of agarose beads with covalently bound

rabbit IgG (Immunobead R-l Bio-Rad Laboratories). The mixture is

incubated for 60 mins at 370C, and then 2 mls of 0.01 M EDTA buffer

containing 0.1% gelatin, pH 7.4 is added and the Immunobeads pelleted

by centrifugation at 1,000 g for 5 mins. The supernatant is discarded

and the IgG beads, with any completed Fc-reactive material, are washed

with 2 mls of the 0.01 M EDTA-gel buffer. One tenth of a milliliter of

125I labeled protein A (approximately 20,000 cpm) and 1.0 ml of

VBS-gel are added to the washed beads and incubated an additional

60 mins at 370C. At this time the beads are washed twice with 0.01 M

EDTA-gel buffer as described above and the quantity of 125I protein

A associated with the beads measured.

These assay conditions were found to be optimal from preliminary

kinetic studies. Under these assay conditions, maximal binding of

1251 protein A to the immunobeads was approximately 7,000 cpm. The










number of counts recovered when the assay was performed without

Immunobeads present (i.e., the background) was approximately 200 cpm.

By comparing the number of counts bound in the absence of cold protein

A to the number bound to the beads in the presence of known amounts of

unlabeled protein A (Pharmacia), the degree of inhibition can be

calculated and a standard curve obtained, Figure 2-6. The quantity of

Fc-reactive material in unknown samples can be measured by comparing

the observed percentage inhibition with the standard inhibition curve

and the results expressed in protein A equivalent units. For purified

protein A, 50% inhibition of binding was consistently observed with

solutions containing 10-20 ng/ml (see Figure 2-6). Similar results

were obtained when the assay was run in Todd Hewitt broth or in

Trypticase Soy Broth (data not shown). No inhibition was observed in

this two-stage assay with human IgG samples containing 100 pg/ml,

Figure 2-6. Non-specific interference from constituents of bacterial

culture media was observed when a one-stage assay for protein A or

protein A-like Fc receptors was used. This method was applied to

measuring secreted protein A-like material in the overnight culture

supernatant fluids of various streptococcal cultures. Of 15

a-hemolytic streptococci tested only two strains secreted detectable

levels of Fc-reactive material by the two-stage PA assay. The superna-

tant from an overnight culture of one group C strain (Hu6) contained

the equivalent of 10.2 + 1.7 ng of protein A/ml. The supernatant from

an overnight culture of a group A strain (#64) contained the equivalent

of 5 + 0.1 ng of protein A/ml. For comparison, overnight cultures of

Staphylococcus aureus Cowan strain and Wood 46 strain were cultured

under identical conditions as representative of high and low protein A























2 STAGE BINDING ASSAY

SELECTIVE FOR PA

SIg- G immobilized
on beads
u o PA in sample

I- incubate
I, .Ohr 370c
wash X 2
discard supernatant

I I


Add buffer +
125 I-PA




incubate
1.0hr. 37c
n centrifuge

Sdiscard
z supernatant
0
0o
cn | wash X 2
Discard supernatant





cpm bound
to immobilized IgG


Two-stage assay selective for PA.


Figure 2-5.



































Quantification of
by the 2- Stage


10 10'


PA and IgG
Binding Assay


10-
ng/ml PAorIgG


Figure 2-6.


Standard Curve generated using purified PA in the
two-stage PA assay. Known quantities of cold purified PA
were tested for their ability to inhibit the binding of
r251 PA to IgG immunobeads as described in the text.
100 ug of human IgG did not significantly interfere in
this assay.


I
0,
m
(M

c:
t'-4





c
0
.0
-c
5
0


I II 1 11111 I I1 1 111111 I I I I I 11I I I 1 11111 I I I ll

-- PA
50% Inhibition by PA 15.0 ng /ml





Ig G


I I 1-I I I I l I i l l -- I 11 1 1 1 1 i I i l ; 1 1 II I


IO"


10-


A










producers. The overnight culture supernatant from the Cowan strain

contained 1850 + 200 ng protein A/ml and the Wood strain supernatant

contained 162 + 7 ng PA/ml.

Any Fc receptors detected by this assay must bind to the IgG Fc

region at or near the PA-binding site. To ensure that my results using

streptococcal culture supernatants reflected absence of Fc receptors,

rather than failure to detect the material due to binding at a site

remote from the PA binding site, the ability of streptococcal culture

supernatants to mediate agglutination of red cells with

subagglutinating concentrations of antibody on their surface was tested

as described in Materials and Methods. The culture supernatants were

heated to 800C prior to testing to destroy hemolysins that made this

assay unworkable. [This method has previously been used to demonstrate

soluble Fc-reactive material in streptococcal culture supernatants and

extracts (Kronvall, 1973 and Havllcek, 1978).] I found that only the

two streptococcal strains that contained detectable Fc receptors by the

two-stage PA assay were capable of mediating agglutination of the

sensitized red cells. The culture supernatants of the streptococcal

strains Hu6 and 64 gave hemagglutination titers of 243 and 27

respectively and the Staphylococcus aureus strains Cowan I and Wood 46

had hemagglutination titers of 4,500 and 2,187 respectively.

Detection of Fc Receptors in Lancefield and Lysostaphin Extracts of
Streptococci.

The results in the previous section suggest that despite the

presence of relatively high levels of Fc receptors on the surface of

certain streptococci these receptors were secreted only at low levels

during culture. In order to determine if this assay could be used for

streptococcal Fc receptors once solubilized or extracted from the









TABLE 2-1

PA-like Activity in Bacterial Extracts


PA-like Material Extracteda
Streptococcal ng PA Equivalent
Group
Strain Buffer Lancefield Lysostaphin
Control Extract Extract

529 A 0 40.3 0.0
64 A 0 70.3 50.2
3706-T A 0 0.0 0.0
H-1-JP NGb 0 10.3 0.0
C 691 C 0 29.6 25.3


a = For details see text.
b = Not Grouped.










bacteria I tested for the presence of Fc receptors in Lancefield and

lysostaphin extracts of streptococci which had surface Fc receptors as

demonstrated by the IgG adsorption assay. The results presented in

Table 2-1 demonstrate that Fc-reactivity could be detected in these

extracts. This would suggest that the competitive binding assay would

detect certain Fc receptors from streptococci if they had been

secreted.

Discussion

In this chapter, methods for screening streptococci for the

presence of surface Fc receptors and for measuring Fc-reactivities in

bacterial culture fluids and extracts are described. The assay for

measuring surface Fc receptors is based on the ability of differing

numbers of streptococci to adsorb IgG in the presence or absence of

corresponding F(ab')2 fragments. This assay did not have any of the

problems associated with assays involving 1251 labeled IgG. There

was no significant nonspecific binding of IgG to the reaction vessel or

difficulty distinguishing IgG bound through F(ab')2 sites from that

bound via the Fc region. Although the presence or absence of Fc

receptors on a particular strain of bacteria was a consistent finding,

the amount of IgG that could be adsorbed by Fc receptor positive

strains varied from experiment to experiment. Since the reproduci-

bility of adsorption was high for replicates within a culture, it is

likely that this variation is due to differences in surface properties

of bacteria obtained from different cultures. The method described was

superior to the hemagglutination assay in sensitivity, objectivity, and

reproducibility and was applicable to a wider range of streptococcal










strains, e.g., autoagglutinating bacteria. These studies demonstrated

that certain clinical isolates of group A streptococci had 5-35% the

level of Fc receptors of the protein A rich, Cowan strain

Staphylococcus aureus, measured under identical conditions. This

finding suggests that these strains would be a reasonable starting

source for the extraction and characterization of a streptococcal

protein A-like molecule.

Additionally, this assay can be used to characterize the species

of IgG with which streptococcal Fc-reactive material would react (see

Figures 2-3 and 2-4). For example, streptococcus 529 effectively

adsorbed human IgG but did not adsorb rabbit IgG (compare Figures 2-3

and 2-4). Langone (1978) has shown that the competitive binding assay

for IgG can be used to quantify both protein A-reactive and nonreac-

tive species of IgG, the methods described in this study could be

expanded to examine the reactivity of bacteria with any species, class

or subclass of immunoglobulin.

The assay for soluble Fc-reactive material described is an exten-

sion of the competitive binding assay for IgG and protein A previously

described by Langone et al. (1977). By carrying out the assay in two

stages, it was made selective for PA-like Fc-reactive material and

could be carried out in culture media without loss of sensitivity.

The sensitivity of this procedure was not affected by the presence of

bacterial hemolysins. In agreement with previous reports, secreted

protein A from staphylococci could be readily detected (Winblad and

Ericson, 1973) but only a few streptococci secreted Fc receptors. In

those strains the levels of Fc-reactive material secreted were close to

the limit of detection of the assay system. Significant levels of










Fc-reactive material could be detected in Lancefield and lysostaphin

extracts of certain group A streptococci using this assay. Although

this method is based on direct competition of 125I protein A for

its receptor site on the Fc region of IgG, I was unable to find a

streptococcal culture fluid that could mediate hemagglutination which

was not positive in the PA binding assay.

The assays for cell-bound and soluble Fc receptors represent

improved methods for screening bacteria for the presence or production

of these biologically active molecules. These methods are used as the

basis for the remainder of my work to detect and characterize both cell

surface and soluble Fc receptors.

















CHAPTER THREE
DETERMINATION OF THE STRUCTURAL RELATIONSHIP BETWEEN
STAPHYLOCOCCAL PROTEIN A AND STREPTOCOCCAL Fc RECEPTORS



Introduction

In Chapter 2 methods for identifying streptococci with surface Fc

receptors were described. In this chapter I have used a monospecific

polyclonal antibody to protein A prepared in chickens to determine

whether the Fc receptor(s) on a number of strains of Staphylococcus

aureus and group A and group C streptococci are related antigenically

to protein A, or represent distinct surface moieties with a common

functional activity, namely the ability to bind to the Fc region of

IgG. This approach has proved valuable in establishing total or

partial identity between bacterial surface antigens that share a common

functional activity, e.g., the anti-phagocytic M protein of group A

streptococci that is known to exist in a variety of antigenic forms

(Ferrieri, 1975). The purpose of this study was to ensure that any

streptococcal Fc receptor I attempted to isolate was distinct from

protein A.

Materials and Methods

Bacteria Strains, Media and Growth Conditions

Staphylococcus aureus Cowan I strain and human isolates of

Staphylococcus aureus and a-hemolytic streptococci of groups A and C










were used. Media and growth conditions were as described in

Chapter 2.

iodination of Protein A

Purified protein A (Pharmacia) was iodinated by the mild lactoper-

oxidase method using enzyme beads (Bio-Rad) (Morrison et al., 1971), as

described in Chapter 2.

Chicken Anti-Protein A

Monospecific antiserum to staphylococcal protein A was a gift from

Dr. John Langone, National Cancer Institute, Bethesda, Maryland. This

antiserum was prepared by immunizing roosters using the following

schedule. Preimmunization blood samples were obtained five days prior

to the initial injection. Roosters were injected with 50 yg of protein

A (Pharmacia) intramuscularly (I.M.) in complete Freund's adjuvant.

Blood samples were obtained one week later to test for antibody

production. Four weeks after the initial injection, the roosters were

boosted I.M. with 50 pg of protein A in incomplete Freund's adjuvant

and serum collected one week later. The resulting antiserum gave a

precipitin line against purified protein A in a double immunodiffusion

assay. No line was observed when preimmunization serum was used,

confirming that the reaction was not mediated via the Fc region of

chicken immunoglobulins. All of the anti-protein A activity could be

removed by passage over protein A immobilized on Sepharose. The

ability of the antiserum to precipitate 1251 protein A could be

completely inhibited by cold protein A, either in the purified form

(Pharmacia), or as extracts of Staphylococcus aureus Cowan I strain.










Human IgG and IgG Assay

Human IgG was prepared by chromatography of normal human serum on

DEAE cellulose (Boyle and Langone, 1980) and quantified by the

competitive binding assay of Langone et al. (1977) as described in

Chapter 2.

Selection of Bacteria with Surface Fc receptors

Fc receptors on the surface of bacteria were detected by the

ability of bacteria to adsorb human IgG from solution. The details of

this assay have been described fully in Chapter 2.

Using this assay four clinical streptococcal isolates with surface

Fc receptors were selected. Three of these strains are group A (9,

64/14, and 11) and one Group C (Hu6). The ability of these bacterial

strains to adsorb IgG was described in Chapter 2.

Results

Competitive Binding Assay for the Quantitation of Antibodies
Specifically Directed Against Staphylococcal Protein A

A competitive binding assay was developed to quantify antibody to

protein A. In this assay the antigen combining sites on chicken

anti-protein A antibody compete with the Fc region of IgG immobilized

on agarose beads (Bio-Rad) for radio iodinated protein A. Two tenths

of a milliliter of anti-protein A diluted in veronal buffered saline,

pH 7.4, containing 0.1% gelatin (VBS-gel) is mixed with 0.1 ml of a

standard suspension of agarose beads containing immobilized rabbit IgG

(Bio-Rad) and 0.1 ml of iodinated PA containing approximately 20,000

cpm. Following a 1.5 hr incubation at 37C, 2 mis of EDTA-gel are

added to each tube. The tubes are centrifuged at 1,000 g for 5 mins

and the supernatant fluid is decanted. Following a second wash with an









additional 2 mis of EDTA-gel as described, the amount of 1251

protein A adhering to the beads is determined in an LKB Gamma Counter.

Maximal binding of 1251 protein A to the Immunobeads was

approximately 6,000 cpm. The number of counts recovered when the assay

was performed without Immunobeads present (i.e., the background) was

approximately 200 cpm. By comparing the number of counts bound in the

absence of chicken anti-protein A to the number of counts bound to the

beads in the presence of various dilutions of chicken anti-protein A,

the degree of inhibition of protein A binding can be calculated and a

standard curve relating antibody concentration to inhibition can be

obtained. The assay is summarized in Figure 3-1A and a typical

inhibition curve is shown in Figure 3-1B.

Under the conditions of the assay a 1:10,000 dilution of the anti

protein A antiserum routinely inhibited the binding of 1251 protein

A to Immunobeads by 50%. No inhibition of 1251 protein A binding

to Immunobeads was observed in the presence of a 1:10 dilution of

preimmunization chicken serum, confirming that protein A is not

reactive with the Fc region of chicken immunoglobulins (Langone,

1978).

Detection of Protein A on the Surface of Bacteria

In these studies the protein A positive Staphylococcus aureus

Cowan I strain and four Fc receptor positive streptococcal strains were

tested for their ability to react with anti-protein A antibody via the

(Fab')2 region, by adsorbing a standard dilution of the anti-protein

A antibody with various numbers of bacteria (Figure 3-2). In this

adsorption assay, differing numbers of bacteria were mixed with 1 ml of

a 1:2,000 dilution of chicken anti-protein A for 90 mins at 37'C and


















COMPETITIVE BINDING ASSAY

FOR THE DETECTION OF ANTI PROTEIN A


Rabbit IgG
immobilized on beads
Chicken anti- protein A
^ 125I-protein A


incubate
1.5 hr 37C
centrifuge


discard
supernatant


wash x 2
discard supernatant


cpm bound to
immobilized IgG


Figure 3-1. Competitive binding assay for quantitation of chicken
anti-protein A antibody.
A. Flow chart of assay.






























80-
CL
1
60


Z 40-


Z20



F I










Figure 3-1.


2 4 8 16 32 64

RELATIVE [CHICKEN SERUM].


(continued)
B. Typical standard curve obtained using the chicken
anti-protein A antiserum. A relative concentration of 1
represents a 1 to 64,000 dilution of the chicken
anti-protein A antiserum. The results are presented as
percent inhibition of 1251 protein A binding + the
standard deviation. For precise experimental details see
text.










centrifuged. Duplicate samples of an aliquot of the supernatant fluids

were tested for residual anti-protein A using the assay described

above. The percent anti-protein A adsorbed was determined by comparing

the amount of anti-protein A in the adsorbed supernatant fluids to

control tubes containing anti-protein A and no bacteria. Results are

expressed as the percent of anti-protein A adsorbed by various numbers

of bacteria (Figure 3-2). Adsorption of anti-protein A by Staphylo-

coccus aureus Cowan I strain was not affected by the presence of a 1:10

dilution of preimmunization chicken serum (data not shown), confirming

that normal chicken immunoglobulin is nonreactive with protein A.

These results demonstrate that chicken anti-protein A is able to

bind antigenically with a protein A-bearing-bacteria strain and the

amount adsorbed is dependent on the number of bacteria used for

adsorption (Figure 3-2A). None of the IgG Fc-reactive streptococcal

strains were capable of adsorbing anti-protein A, even though these

streptococci were able to adsorb equivalent amounts of human IgG at the

concentrations of bacteria tested. These findings indicate that the Fc

receptors on streptococci are not antigenically related to protein A.

By contrast the results with Staphylococcus aureus Cowan strain would

be consistent with the Fc receptor on these bacteria being protein A or

being co-expressed along with protein A.

The following experiment was designed to examine whether pretreat-

ment of the Cowan strain bacteria with anti-protein A antibody would

block its ability to adsorb human IgG via the Fc region.
























STAPH. AUREUS COWAN


5X108 108


5X107 107


Hu6 Hu9 Hull 64/14

I X 1010


Figure 3-2.


Ability of Staphylococcus aureus Cowan I (A) and various
streptococci (B) to selectively interact with the
Fc-region of human IgG (F -) or with the Fab2 region of
chicken anti-protein A antibody (F; Results are
presented as percent IgG or of antibody to protein A
absorbed + standard deviation. For precise experimental
details see text.


A B


STREPTOCOCCI










Blocking of Fc receptor Activity on Staphylococcus aureus Cowan
Strain by Pretreatment with anti protein A

Aliquots of a standard suspension of Staphylococcus aureus Cowan

strain that would adsorb approximately 350-400 ng of human IgG from

solution were tested for their ability to remove IgG from solution

following treatment of the bacteria with anti protein A antibody. A

standard number of bacteria was incubated with 1 ml of various

dilutions (1:125 to 1:4,000) of chicken anti-protein A or normal

chicken serum as described above for the anti-protein A adsorption

assay. Following an incubation of 1 hr at 37C, the bacteria were

washed twice with 2 mis of 0.01 M EDTA-gel to remove any unbound

anti-protein A. One milliliter of VBS-gel containing 500 ng of human

IgG was added to each bacteria pellet and to control tubes containing

no bacteria. Following a 1 hr incubation at 37C, all tubes were

centrifuged and aliquots of the supernatants were tested for the amount

of IgG adsorbed as described earlier. The Fc receptor positive group C

streptococcus (Hu6) which failed to react with the anti-protein A

antibody was included as a negative control, (see Figure 3-2). The

results in Figure 3-3A show that the adsorption of IgG was completely

inhibited when the Cowan I strain was preincubated with high

concentration of anti-protein A antiserum, but not by equivalent

concentrations of normal chicken serum. Pretreatment of the group C

streptococcus with anti-protein A did not affect its ability to

subsequently Adsorb human IgG (see Figure 3-3B).

Comparison of Fc-receptors on the Clinical Staphylococcus aureus
Isolates to those of the Laboratory Cowan Strain

The presence of protein A on other strains of staphylococci has

been suggested by their ability to react with the Fc region of IgG.

Since certain streptococci share this property using a receptor















































RELATIVE [CHICKEN SERUM]


Figure 3-3.


Binding of human IgG to bacteria pretreated with chicken
anti-protein A anti-serum or normal chicken serum.
A. Results using Staphylococcus aureus Cowan I.

























80-


60-


40-


20


RELATIVE [CHICKEN SERUM]


Figure 3-3.


(continued)
B. Results using a representative streptococcal strain.
In this case a human clinical group C isolate designated
Hu6.
For both A and B a relative concentration of 1 for
chicken anti-protein A anti-serum or the preimmunization
chicken serum (normal) represents a 1 to 4,000 dilution.
Results are expressed as the percent IgG absorbed +
standard deviation. For precise experimental details see
text.


B



Anti-Protein A
S~ -- -" T Normal








Grp C Strep Hu6




I I I I I I













antigenically distinct from protein A, in the next series of

experiments I determined whether all staphylococcal Fc reactivity was

mediated by protein A-like molecules or if differing Fc receptors could

be defined on the surface of various staphylococcal strains other than

the Cowan strain.

Clinical isolates of Staphylococcus aureus were screened for their

ability to adsorb human IgG from solution. As expected, a range of

Fc-reactivities was observed with certain of the strains being capable

of adsorbing equivalent quantities of IgG to the Cowan strain on a per

bacterium basis, while other strains were less effective. Of the nine

isolates screened, all were capable of adsorbing significant quantities

of IgG under the standard assay conditions (data not shown). From

these preliminary screening experiments, the staphylococcal isolates

were adjusted to yield approxiamtely equivalent levels of IgG adsorbing

capacity by varying the number of bacteria, Table 3-1. The ability of

these bacteria to remove IgG when preincubated with buffer, normal

chicken serum or chicken anti-protein A was tested. For these studies

a single concentration of antibody (1:250) was selected, based on the

results obtained in Figure 3-3. In each case the ability of clinical

isolates of Staphylococcus aureus to adsorb human IgG could be entirely

blocked by preincubation in chicken anti protein A, Table 3-1. No

inhibition was observed when the bacteria were pretreated with normal

chicken serum, Table 3-1. These observations indicate that despite the

obvious quantitative differences in IgG binding capacity of the

staphylococcal isolates tested, the Fc receptor was a protein A-like

molecule in all cases. Four streptococcal strains, two group A and two

group C, which demonstrated the ability to adsorb IgG were included.









TABLE 3-1

'Effects of Specific Chicken Anti-Protein A Antibody
on the Ability of Bacteria to Adsorb IgGa


Number % IgG adsorbed
Genera Strain
Bacteria Normal Chicken Chicken Anti-
Buffer Serum 1:250 Protein A 1:250

Staph.
aureus Cowan 5x108 81 + 8 88 + 2 <5
Hul6 109 67 + 5 83 + 1 <5
Hul7 2.5x108 48 + 7 46 + 4 <5
Hul8 109 69 + 4 55 + 4 <5
Hul9 5x108 79 + 1 71 + 1 <5
Hu20 1010 62 + 6 69 + 4 <5
Hu21 109 74 + 3 79 + 1 <5
Hu22 2x108 66 + 1 50 + 3 <5
Hu23 2.5x108 48 + 1 48 + 7 <5
Hu25 1010 88 + 1 89 + 1 <5


Strep.
Grp. C 26RP66 109 88 + 2 89 + 1 89 + 1
Hu6 1010 92 + 1 94 + 1 88 + 1


Strep.
Grp. A Hu9 1010 39 + 8 39 + 8 33 + 1
64/14 1010 90 + 2 96 + 4 83 + 4


a = Bacteria were preincubated in buffer,


in normal chicken serum, or in


chicken anti-protein A antiserum, for 1 hr at ambient temperature.
The bacteria were washed twice and their ability to remove human IgG
from solution was tested.









Preincubation of these streptococcal strains with either anti-protein A

or normal chicken serum did not inhibit the adsorption of IgG

(Table 3-1).

Discussion

The purpose of this part of the study was to determine whether all

bacterial Fc receptors were structurally related to protein A, or if

distinct Fc receptors were present alone or together with protein A on

the surface of other bacteria. The results presented in this chapter

clearly demonstrate that the Fc receptor(s) on four different

streptococcal strains were not antigenically related to staphylococcal

protein A. By contrast, the reactivity of all Fc receptors on

Staphylococcus aureus Cowan strain could be totally inhibited by

pretreating the bacteria with anti-protein A antibody. Studies with

nine fresh clinical isolates of Staphylococcus aureus demonstrated

differences in the quantity of Fc receptors expressed on their surface

(see Table 3-1). In all cases, a close correlation was observed

between their ability to bind anti-protein A antibody and to adsorb IgG

via the Fc region. Pretreatment of any of these bacterial strains

with anti-protein A antibody totally inhibited their ability to react

with the Fc region of human IgG. These results would suggest that all

staphylococcal Fc receptors were protein A-like and that a subgroup of

non-protein A Fc receptors could not be detected.

The results presented in this chapter support the idea that there

are different classes or types of Fc receptors on staphylococci and

streptococci (Myhre and Kronvall, 1981b). They do not exclude the

possibility that the region of the receptor molecule that directly










interacts with the Fc region of IgG may not be immunogenic or is weakly

immunogenic. Such a region would be equivalent to the antigenic

combining region of an immunoglobulin and would only be detected by the

equivalent of an anti-idiotypic antibody. Although the exact nature of

the combining site in the Fc receptor molecule that binds to the Fc

region of IgG has not been elucidated, I have shown that the Fc

reactivity on streptococci is not mediated by protein A or a closely

related molecule. By contrast, all Fc binding activity associated with

staphylococci was shown to be protein A-like in nature.

Using the methods described in Chapters 2 and 3, the next part of

this study was to attempt to solubilize and isolate a non-protein

A-like streptococcal Fc receptor.

















CHAPTER FOUR
ISOLATION AND PARTIAL CHARACTERIZATION
OF THE Fc RECEPTOR FROM A GROUP C STREPTOCOCCUS



Introduction

Attempts to extract and purify streptococcal Fc receptors have met

with limited success. Unlike protein A none of the streptococcal Fc

receptors are secreted in significant quantities during culture

(Kronvall, 1973; Schalen, 1982). A variety of extraction procedures

have been tested including phage lysis (Christensen and Holm, 1976),

heat (Christensen and Holm, 1976; Christensen and Kronvall, 1974) or

treatment with hot acid or hot alkali (Havlfcek, 1978; Schalen et al.,

1978; Christensen et al., 1979b). In most reports the yield of soluble

Fc receptor was low. The most highly characterized streptococcal Fc

receptor was isolated by Grubb et al. (1982) from a group A strepto-

coccus following alkaline extraction. This receptor was heterogeneous

in size with the predominant activity having a molecular weight of

29,500 daltons and was only obtained when protease inhibitors were

included during purification.

Using the techniques outlined in Chapter 2, I was able to identify

a group C streptococcus with Fc binding capacity approximately equiva-

lent to the adsorbing capacity of the protein A-rich Staphylococcus

aureus Cowan I strain. The receptor on this group C streptococcus was

shown to be antigenically distinct from protein A (Chapter 3). This

chapter describes the isolation in high yield of a functionally










homogenous Fc receptor from this group C streptococcus, which has been

designated FcRc.

Materials and Methods

Bacteria and Bacteriophage

The B-hemolytic group C streptococcal strain designated 26RP66 and

the Cl bacteriophage were a gift from Dr. Vincent Fischetti of the

Rockefeller University, New York, New York. This strain was selected

based on its high surface Fc receptor activity as determined by

immunoassay as described in (Chapter 2). For all of the studies

bacteria were grown in Todd Hewitt broth and phage lysis was carried

out using a modification of the procedure of Fischetti et al. (1971).

Phage-associated Lysin Activity

Phage-associated lysin activity was detected by the lysis of a

group A streptococcal strain as described by Fischetti et al. (1971).

Extraction of Fc receptors

The streptococcal strain 26RP66 was grown overnight in 3 liters of

Todd Hewitt broth. A bacterial pellet was recovered by centrifugation

and washed once in phosphate buffer saline (PBS) pH 7.4. Aliquots

containing approximately 0.25 g of bacteria (wet weight) were enzyme

extracted into 6 ml of appropriate buffer containing 100 ,g DNAse under

the following conditions:

1) Mutanolysin (Miles) extraction was carried out using 2,000

units of enzyme in 0.05 M KH2P04, pH 6.5 (Siegel et al.,

1981).

2) Pepsin (Sigma) extraction was carried out using 1,750 units in

0.05 M KH2P04, pH 5.8 (Manjula and Fischetti, 1980).









3) Lysozyme (Sigma) extraction was carried out using 24,000 units

in 0.05 M KH2P04, pH 6.3 (Forsgren, 1969).

4) Lysostaphin (Sigma) extraction was carried out using 175 units

in 0.05 M Tris-HCl, 0.15 M NaCI, pH 7.5 (Sjoquist et al.,

1972).

These conditions were previously used to isolate either

streptococcal cell wall constituents or staphylococcal protein A

(Forsgren, 1969; Sj5quist et al., 1972; Manjula and Fischetti, 1980;

Siegel et al., 1981). All extractions were carried out for 4 hrs at

37C. The extracts were then centrifuged at 10,000 g for 15 mins and

the supernatants recovered, dialyzed against PBS, and stored at 4C

until tested for functional Fc receptor activity by the method

described below. Detergent extraction was carried out in a similar way

using 1% Tween-20 in 0.15 M PBS, pH 7.4.

Hot Acid/Hot Alkaline Extracts

Hot acid/hot alkaline extracts were carried out according to the

method of Lancefield (1928). Bacteria, 0.25 g, were suspended in 3 mls

of 0.15 M PBS and the pH was adjusted to 2.0 (or 10) with 0.5 M HC1 (or

0.5 M NaOH). The bacterial suspension was boiled for 10 mins and the

pH was neutralized. The final volume was adjusted to 6.0 mis and the

supernatants recovered as described above.

IgG, IgG Fragments and Immobilized IgG Preparations

Stock human IgG was prepared by chromatography of normal human or

rabbit serum on DEAE cellulose (Boyle and Langone, 1980). Aliquots

were stored at -70'C until use. The F(ab)2 fragments of human IgG

were prepared by pepsin digestion as described in (Chapter 2).

Immunoglobulin G was immobilized to Immunobeads (Bio Rad, Richmond,










California) for use in the competitive binding assay as described by

Langone et al., 1979a.

Immobilized IgG for affinity purification of the streptococcal Fc

receptor was prepared by covalently coupling human IgG to the high

capacity Affi-gel 15 activated beads (Bio-Rad, Richmond, California).

Ten milligrams of gel washed with 3 volumes of isopropanol and 3

volumes of deionized water was mixed with 10 ml of human IgG containing

7.3 mgs IgG/ml. The coupling reaction was carried out in 0.1 M HEPES,

pH 7.5 at 4"C overnight with gentle rocking. Unreacted sites were

blocked by the addition of 0.1 ml of 1 M ethanolamine HC1, pH 8.0 for

each ml of gel. One hour was allowed for complete blocking and the IgG

coupled gel was extensively washed in VBS gel and stored at 40C in VBS

gel containing 0.02% sodium azide. Prior to use the immobilized IgG

was washed with 10 volumes of glycine-HCl, pH 2.0 and reequilibrated in

phosphate buffered saline, pH 7.4.

lodination of PA and the Streptococcal Fc-receptor

Purified PA (Pharmacia) and the streptococcal Fc receptor (FcRc)

were iodinated by the mild lactoperoxidase method using enzyme beads

(Bio-Rad) (Morrison et al., 1971), as described in Chapter 2.

Detection of Soluble Fc-receptors

The method developed for the detection of soluble Fc receptors in

extracts is described in (Chapter 2). Essentially, this competitive

binding assay measures the ability of Fc receptors to inhibit binding

of either 125I labeled protein A or 1251 labeled Fc receptor to

immobilized human or rabbit IgG. In the initial part of this study,

125I protein A was used and one unit of Fc receptor activity was

defined as the concentration of material that would inhibit its binding









by 30% under standard assay conditions. Once streptococcal Fc

receptors had been isolated and labeled, they were used for the assay

and an absolute value in ng/ml was assigned based on the inhibition of

affinity purified standards included in each assay. The absolute

protein concentration of the standard was determined using the Bio-Rad

protein assay (Richmond, California), which is a modification of

Bradford's method (Bradford, 1976).

Polyacrylamide Gel Electrophoresis

FcRc preparations containing 15-60 pg of unlabeled material or

1.7 x 105 cpm of iodinated material were applied to 7% polyacrylamide

disc gels. Samples were electrophoresed at 1.5 mamp per gel in 0.025 M

Tris, 0.2 M glycine, pH 8.3 until the tracking dye was 1 cm from the

bottom of the gel. Gels were either fixed and stained with Coomassie

Brilliant Blue or frozen and sliced into 1 mm sections. Gel slices

were either counted for radioactivity on an LKB gamma counter or eluted

for 72 hrs into VBS-gel and tested for functional activity. Samples

were also run on 0.1% SDS, 7% polyacrylamide gels as described above

with two exceptions. First, all samples were boiled for 1 min in 2%

SDS prior to electrophoresis. Second, the electrophoresis buffer was

0.1% SDS, 0.025 M Tris, 0.2 M glycine pH 8.3. Molecular weight

standards (Sigma, St. Louis) were included in each SDS polyacrylamide

gel assay. Myosin (200,000 daltons), B-galactosidase (116,000

daltons), phosphorylase b (94,000 daltons), bovine serum albumin

(68,000 daltons), egg albumin (43,000 daltons), carbonic anyhydrase

(30,000 daltons), 6-lactoglobin (18,400 daltons).









Preparation and Measurement of Chicken Anti-FcRc

Antibody to the streptococcal FcRc was prepared by immunizing

white leghorn hens using the following schedule. Preimmunization

samples were obtained two days prior to the initial injection.

Chickens were injected with 25 yg of FcRc intramuscularly (I.M.) in

complete Freund's adjuvant. Blood samples were obtained two weeks

later to test for antibody production. Two weeks and five weeks after

the initial injection, the chickens were boosted I.M. with 25 ug of

FcRc in incomplete Freund's adjuvant. Serum was collected one week

after the final injection.

Competitive Binding Assay for the Quantitation of Antibodies

Antibodies to streptococcal FcRc were measured by a modification

of the competitive binding assay described in (Chapter 3). In this

assay the antigen combining sites on chicken anti-FcRc antibody compete

with the Fc region of human IgG immobilized on agarose beads (Bio-Rad)

for radio iodinated FcRc. Antiserum to FcRc was diluted in veronal

buffered saline, pH 7.4, containing 0.1% gelatin (VBS-gel) and 0.2 ml

of the dilution was mixed with 0.1 ml of a standard suspension of

immobilized human IgG beads and 0.1 ml of iodinated FcRc containing

approximately 30,000 cpm. Following a 1.5 hr incubation at 37C, 2 ml

of EDTA-gel are added to each tube. The tubes were centrifuged at

1,000 g for 5 mins and the supernatant fluid decanted. Following a

second wash with an additional 2 ml of EDTA-gel, the amount of 1251

FcRc adhering to the beads was determined in an LKB Gamma Counter.

Maximal binding of 1251 FcRc to the Immunobeads was approximately

6,000 cpm. The number of counts recovered when the assay was performed

without Immunobeads present (i.e., the background) was approximately









200 cpm. By comparing number of counts bound in the absence of chicken

anti-FcRc to the number of counts bound to the beads in the presence of

various dilutions of chicken anti-FcRc, the degree of inhibition of

FcRc binding can be calculated and a standard curve relating antibody

concentration to inhibition can be obtained. No inhibition of binding

of 1251 FcRc was observed in the presence of preimmunization

chicken serum.


Results

Solubilization of Fc Receptor

The group C streptococcal strain designated 26RP66 was selected

because of its high level of surface Fc receptors. A variety of

extraction procedures were tested including phage lysis, alkaline

extraction, acid extraction, detergent extraction, enzyme treatment

with pepsin, lysostaphin, lysozyme or mutanolysin. The resulting cell

free lysates were tested for soluble Fc receptors using the competitive

binding assay described in the Methods. The only treatments that

resulted in significant quantities of soluble Fc receptor activity were

phage lysis (approximately 5 x 104 units/g bacteria extracted),

mutanolysin treatment (approximately 3 x 104 units/g bacteria

extracted) and treatment with hot acid (approximately 2 x 103 units/g

bacteria extracted). Extraction of the bacteria with detergent, hot

alkali, lysozyme or lysostaphin did not solubilize detectable

quantities of a functionally active Fc receptor. The extracts were

compared by three criteria: 1) the total yield of Fc receptor

recovered/unit weight of bacteria, 2) the specific activity calculated

as the soluble FcR activity divided by the OD280 of the extract










and 3) the charge heterogeneity of functional activity. This was

measured following elution from non-denaturing polyacrylamide gels.

The material obtained by phage lysis of bacteria demonstrated the

highest yield, highest specific activity and was among the least

heterogeneous of the extracts. Consequently I chose to pursue the

solubilization of Fc receptors using phage lysis.

Isolation of Fc Receptors from the Supernatant of a Phage-Lysed Group C
Streptococcal Culture

The group C streptococcus was grown to an OD650 of 0.3 in

Todd Hewitt broth. To this culture was added approximately 3 x 1012 pfu

of Cl bacteriophage/liter of culture and the bacteria allowed to lyse.

After lysis was complete EDTA was added to a final concentration of

0.05 M and DNase to a final concentration of 0.5 Ug/ml. The resulting

supernatant which was filtered through a sinter glass filter was shown

to contain two soluble activities: 1) an Fc receptor activity and 2) a

bacteriolytic enzyme activity the phage associated lysin originally

described by Fischetti et al. (1971). The supernatant contained no

detectable protease activity and the Fc receptor activity was found to

be stable for over a month at 40C or in excess of six months at -70C.

The crude supernatant was concentrated 30 fold using a Millipore

Pellicon concentrator with a molecular weight cut off of 10,000

daltons. Residual cellular debris was removed by centrifugation at

27,000 g for 2 hrs and the resulting supernatant was precipitated by

adjusting to 50% saturation with (NH4)2SO4. The precipitate was

recovered by centrifugation at 27,000 g for 1 hr at 4*C and then

resuspended in a minimal volume of 0.5 M phosphate buffer pH 6.1

containing 0.005 M EDTA. This material was dialysed against the same










buffer and then ultra-centrifuged at 90,000 g for 5 hrs at 4C. The

soluble supernatant contained both the Fc receptor activity and the

phage associated lysin activity and was subjected to further

purification.

Previously Fischetti et al. (1971) had defined conditions under

which that the phage-associated lysin binds to cellulose phosphate.

Using these conditions, 20 ml of the crude phage lysate was applied to

a 1.5 x 16 cm cellulose phosphate (Whatman P11) column which was equil-

ibrated with 0.1 M KH2PO4, pH 6.1, containing 0.005 M EDTA and 10%

glycerol. Once the OD280 had returned to base line, the column

was eluted with the same buffer containing 0.4 NaC1. In agreement

with the previous report (Fischetti et al., 1971), the phage-associated

enzyme was eluted from the cellulose phosphate under these conditions

(Fig. 4-1). Aliquots of the fractions collected tested for Fc receptor

activity showed that 98% of the total recovered activity and 95% of the

total recovered OD280 passed directly through the column.

The phage associated lysin can be stabilized and stored as

described by Fischetti et al. (1971). The cellulose phosphate step

does not result in any significant purification of the Fc receptor;

however, I felt that the ability to isolate the crude phage associated

lysin represented a significant biproduct of the purification.

In the next step of the purification procedure all fractions

containing Fc receptor activity from the cellulose phosphate flow

through material were dialyzed against 0.015 M NaCL and applied to a

DEAE column equilibrated with 0.015 M NaCl, pH 7.4, and the unbound

material was eluted in 0.015 M NaC1. Once the OD280 had returned

to base line values, a linear gradient of NaCl from 0.05-0.5 M was









applied and finally, the column was eluted with 1.5 M NaCl. The NaCI

concentration was followed in the collected fractions by measuring

conductivity and the soluble FcRc activity was monitored using the

competitive binding assay described in Chapter 2. The majority of

the Fc receptor activity was recovered in a single peak (peak I) which

was eluted between a NaCl concentration of 0.12 and 0.18 M (see

Fig. 4-2). A second peak (peak II) containing approximately 5% of the

recovered activity was obtained at a NaCI concentration close to

0.24 M.

Fractions containing Fc receptor activity from the DEAE peak I

were pooled and concentrated by Amicon ultrafiltration using a PM10

(molecular weight cut off of 10,000) and further purified by applying

to a column of human IgG immobilized on Affi-gel 15. The column was

washed with 0.15 M PBS, pH 7.4 to remove unbound material and the bound

Fc receptors was eluted from the column using 0.1 M glycine-HC1,

pH 2.0. The eluted fractions were dialyzed against PBS, pH 7.4 and

tested for functional Fc receptor activity and protein content. The

resulting product contained 5,334 Fc receptor units/ml and 58 Ug/ml of

protein. There was no detectable sugar as measured by the phenol

sulphuric acid method (Dubois et al., 1956). The overall purification

achieved by this procedure is summarized in Table 4-1. The affinity

purified Fc receptor was then tested for functional and chemical

homogeneity.

Functional Activity and Properties of the Isolated Fc Receptor

The affinity purified Fc receptor was concentrated 10 fold by

Amion Ultrafiltration using a PM-10 membrane. One hundred microliters

of this material, containing approximately 5 x 103 Fc receptor units

























i
I
I '



I 0













o .
0I


Elute
0.4 m NaCI









SEnzyme
Activity


10 20 30 40 50 60 70 80


FRACTION NUMBER


Figure 4-1.


Cellulose phosphate chromatography of 50% (NH4)2
preparation of crude phage lysate. Twenty milliliters of
the crude phage lysate was applied to a 1.5 x 16 cm column
of cellulose phosphate pre-equilibrated in 0.1 M KH2PO4,
pH 6.1, containing 0.005 M EDTA and 10% glycerol. The
unbound material was eluted in the same buffer and the
phage lysin eluted in the same buffer containing 0.4 M
'NaCI. Four milliliter fractions were collected. Fc
receptor activity (o---o), OD280 (9--*).


8000

7000

6000
0
5000

4000 S

3000
0
2000
- W
1000 W


UL.


5-


-- I-~~ (I-LCIILL~LLY 1CL


-C L-U-~~L_--_ 1_


I











































20 40 60 80 100 120 140 160


FRACTION NUMBER


Figure 4-2.


Ion exchange chromatography of phage lysate recovered from
the pass through of a cellulose phosphate column. Fifty
milliliters containing Fc receptor activity was applied to
a 2.5 x 20 cm DEAE column, equilibrated in 0.015 M NaC1,
pH 7.2 and eluted with a linear salt gradient from 0.05 M
'to 0.5 M. Five milliliter fractions were collected. Fc
receptor activity (--o), OD280 ( )--),
conductivity (---).










TABLE 4-1

Partial Purification of a Group C Streptococcal Fc Receptora


FcRu OD280 Specific
Fraction Vol FcRu Recovery OD280 Recovery Activity Purifi-
ml /ml Percent /ml Percent FcRu/OD280 cation


Crudeb
Lysate 115 9,720 100 30.40 100.0 320 1.0

Cellulose
Phosphate 560 1,302 65 4.30 69.0 302 0.9

DEAE 170 3,876 60 2.00 10.0 1,967 6.0

Immobilized
IgG 66 5,334 32 0.10 0.2 53,340 167.0


a = One unit of Fc
required to in


receptor
hibit the


activity (FcRu)
binding of 1251


is the concentration
protein A by 30%.


of material


b = The crude lysate refers to the cell free 50% ammonium sulfate cut as
described in the Results.









and 58 ig of protein, was iodinated using the Immunobead reagent. When

an aliquot of the iodinated, affinity-purified Fc receptor was mixed

with immobilized human IgG, 96% of the radioactivity could be removed

by two adsorptions. This adsorption was not inhibited by the addition

of human IgG F(ab')2 fragments derived from the same isolated IgG

pool used to prepare the immobilized human IgG beads. These results

demonstrated that the recovered affinity purified Fc receptor was

functionally active and that binding was through the Fc region of IgG.

Treatment of the labeled Fc receptor with trypsin resulted in a time

dependent loss of binding to immobilized human IgG further indicating

the protein nature of the receptor.

Fifty-eight micrograms of unlabeled affinity purified Fc receptor

was applied to duplicate 7% non-denaturing polyacrylamide disc gels.

One gel was stained with Coomassie blue while the second gel was sliced

and eluted into 0.15 M VBS-gel pH 7.4 for 72 hrs. The functional

activity in the eluted samples was measured using the competitive

binding assay described in the Methods. The results presented in

Figure 4-3B demonstrate that four major protein bands were detected by

staining and these bands corresponded to the functional Fc receptor

activity. A similar pattern was observed when radiolabeled Fc receptor

was applied to gels and the distribution of 1251 monitored

(Fig. 4-4). The distribution of counts indicated that band I contained

22%, band II (the major stained band) contained 26%, band III contained

11% and band IV contained 8% of the labeled Fc receptor material

respectively. The remainder of the counts were dispersed at low levels

throughout the gel (Fig. 4-4). The crude lysate electrophoresed under

similar conditions demonstrated a similar pattern of functional






68







I II III IV












o>
c
1000- B



-~ 800-
E


S 600-



a 400-
u
nA-

200-





A
E
N 300-
I-
c
> 200-



a: 100-




10 20 30 40 50 60

GEL SLICE NUMBER





Figure 4-3. Nondenaturing polyacrylamide gel electrophoresis of
affinity purified FcRc and crude phage lysate containing
Fc Rc .
A. Functional FcRc activity in crude phage lysate
following electrophoresis and elution of gel slices into
VBS-gel for 72 hrs.
B. Affinity purified FcRc, 30 pg, was applied to parallel
gels. The functional activity of eluted gel slices is
compared with a gel stained with Coomassie blue.



















I 11 III IV


10 20 30 40 50 57


SLICE NUMBER


Figure 4-4.


Nondenaturing polyacrylamide gel electrophoresis of
affinity purified unlabeled FcRc (30 pg) and 125i-
labeled FcRc (2 x 106 cpm).


rO
O
x


80

60


20














200


116
94


110
90

64

48


C"u
I
0D


68


43


E


30
18


Figure 4-5.


SDS polyacrylamide gel electrophoresis of 20 pg of
unlabeled affinity purified FcRc (a). Molecular weight
standards were included for reference (b).









activity indicating that the four peaks did not develop during the

purification procedure (Fig. 4-3A).

Four major diffuse bands were also observed on SDS gels with

molecular weights of 110,000, 90,000, 64,000, and 48,000 respectively.

The predominant stained protein species was the 64,000 molecular weight

protein (Fig. 4-5).

To determine whether the observed heterogeneity of Fc receptor

activity represented distinct receptors or a common receptor with

differing cell wall constituents covalently linked, two approaches were

used. In the first, each of the active fractions of radiolabeled Fc

receptor recovered by elution from nondenaturing polyacrylamide gels

was tested for its ability to be inhibited from binding to immobilized

human IgG by various concentrations of unfractionated unlabeled Fc

receptor in the competitive binding assay. The results presented in

Figure 4-6 demonstrated superimposable inhibition curves for each

fraction and would suggest that the Fc receptor activity in each peak

was directed against a similar site on the Fc region of the immobilized

human IgG and that each receptor demonstrated a similar affinity.

The second approach to study the interrelationship of the four

charged species of functionally active Fc receptor was to prepare

antibody to the major Fc receptor activity. The affinity purified Fc

receptor was separated by electrophoresis on a series of nondenaturing

polyacrylamide gels. Each gel was stained and the region of the gel

containing the major stained protein band was cut out, emulsified in

complete Freund's adjuvant and injected into chickens following the

immunization schedule detailed in the Methods. The production of

antibody was followed by the ability of the immune chicken serum to

























I I I I I l il I I I I I I I| I I I I i II

100
Lobeled 50% Inhibition
Tracer ng /ml
Unfractionated 17.5--
C 80 Peak I 15.0 Unfroction ed
U- Peak n" 16.5 Pk --
E Peak Im 18.0 -

0 Pk M





Pe 10 100 1000
L 40-



20-



10 100 1OO I000
ng FcRc/ml








Figure 4-6. Inhibition of binding of 1251-affinity purified FcRc
and its components to immobilized human IgG by unlabeled,
unfractionated FcRc. Individual peaks of 1251-FcRc
correspond to the four major changes species eluted from
nondenaturing polyacrylamide gels. 125I-FcRc,
unfractionated (o----) and fractions eluted from
dondenaturing gels, 12SI-FcRc peak I (--A),
125I-FcRc peak II (i---), 125I-FcRc peak III
(0--.-), 125I-FcRc peak IV (o---o).




























o 80
LL

I1)
C- 60-
(9
z





20-
0





1:72









Figure 4-7.


,900 1:24,300


SERUM DILUTION


Inhibition of binding of affinity purified unfractionated
1251-FcRc to immobilized human IgG by chicken antibody
prepared against the major charge species (peak II) in the
FcRc preparation. Chicken anti-FcRc (0---*), pre-immune
chicken serum (---o).










inhibit binding of 1251 Fc receptor to immobilized human IgG beads.

The results presented in Figure 4-7 demonstrate that the resulting

antibody could completely inhibit binding of the 1251 labeled

unfractionated Fc receptor to immobilized human IgG. Chicken serum

obtained prior to immunization was without effect. The labeled tracer

contains all four major charge species of Fc receptor (see Fig. 4-4)

and the antibody was prepared only against the second peak which

contains 26% of the total Fc receptor activity. These findings suggest

that each of the four peaks in the affinity purified Fc receptor

preparation contains antigenically related structures. Taken together

the results in Figures 4-6 and 4-7 would suggest that the group C

streptococcus has a single functional Fc receptor that is extracted

with differing covalently bound fragments that account for the

heterogeneity observed on nondenaturing and SDS polyacrylamide gels.

Attempts to establish conditions to convert the four peaks to a

single functionally active molecular weight form by treatment with a

variety of enzymes have not been successful.

Discussion

In this chapter the purification and partial characterization of a

group C streptococcal Fc recptor is described. A number of extraction

procedures were tested including phage lysis, hot acid and alkali

extraction and treatment with a variety of enzymes including pepsin,

lysostaphin, lysozyme and mutanolysin. Soluble Fc receptor activity

was observed following hot acid extraction, phage lysis or treatment of

the group C streptococcus with mutanolysin. The most favorable

starting material for further purification was found to result from










phage lysis. This extraction procedure resulted in the highest yield

of soluble Fc receptor activity with the least amount of charge and

size heterogeneity.

The Fc receptor solubilized following phage lysis was stable for

at least one month at 4C and for a minumum of six months at -70'C and

at no time was protease activity detectable in any extract from this

bacteria. The Fc receptor activity was stable to hot acid, destroyed

by hot alkali and destroyed by trypsin.

The Fc receptor activity could be isolated to functional

homogeneity by sequential (NH4)2S04 precipitation, cellulose

phosphate chromatography, DEAE ion exchange chromatography and by

binding to and selective elution from a column of immobilized human

IgG. [The cellulose phosphate step is not essential and similar

results were obtained when this step is omitted. The advantage of

including this step is that the phage associated lysin can be recovered

in a reasonably pure form and can be further purified as described

previously by Fischetti et al. (1971).]

All of the material recovered from the immobilized IgG column was

functionally active as judged by its ability, following radioiodina-

tion, to bind to immobilized IgG. This binding was unaffected by the

presence of F(ab')2 fragments, indicating the receptor was binding

a site on the Fc region of human IgG. The functionally active Fc

receptor was physicochemically heterogeneous being resolved into four

major bands on non-denaturing polyacrylamide gels. A similar pattern

with four major diffuse bands was also observed on SDS gels. The major

protein bands had molecular weights of 110,000, 90,000, 64,000 and

48,000 daltons.









Despite the obvious heterogeneity in the size and charge of the

solubilized Fc receptor it appeared to demonstrate remarkable unifor-

mity in its binding to the Fc region of human IgG. Similarly, when the

cold unfractionated material was used to compete with individual

labeled peaks eluted from gels, superimposable inhibition curves were

observed with all combinations (see Fig. 4-6). Similarly, when each of

the individual unlabeled peaks eluted from non-denaturing polyacryla-

mide gels was tested for its ability to compete with the radiolabeled

unfractionated receptor, superimposable inhibition curves were also

observed (Chapter 5). These results suggested that for all practical

purposes the affinity purified Fc receptor preparation contained a

single functional activity, i.e., it bound to the same site on the Fc

region of IgG with a constant affinity. An antibody prepared against

the major charge species of the solubilized Fc receptor preparation was

found to be capable of totally inhibiting the functional activity of

the unfractionated Fc receptor (see Fig. 4-7). These results would

suggest that the size heterogeneity and apparent functional homogeneity

most probably results from the solubilization of a single receptor

molecule covalently linked to various other cell wall constituents.

Heterogeneity of this type has been observed in earlier studies

attempting to isolate the M protein from streptococcal cell walls

(Fox and Wittner, 1969; Fischetti et al., 1976; Kuhnemund et al.,

1981). More recently, extraction conditions have been established that

allow a single minimal molecular weight form of the M protein to be

isolated (Manjula and Fischetti, 1980).

In the initial attempts to convert the heterogeneous soluble Fe

receptor preparation to a single species I have tested pepsin, trypsin










and lysozyme treatments under a variety of optimal and suboptimal

conditions for enzyme action. To date, none of these treatments has

been successful in reducing the number of protein bands, and pepsin and

trypsin both lead to a dose-dependent loss of functional activity.

The Fc receptor isolated here was recovered in a higher yield than

previously reported. I am able to recover 400 pg of affinity purified

FcRc/g wet weight bacteria extracted, compared to the maximum yield

previously reported of 10 ig/g wet weight of bacteria extracted (Grubb

et al., 1982). The heterogeneity observed was similar to that

described by others. The most homogeneous form of a streptococcal Fc

receptor reported was the one isolated by Grubb et al. (1982) that

resulted from alkaline extraction of a group A streptococcus. This

receptor was isolated in a predominant 29,500 molecular weight form

only when protease inhibitors were present. This receptor differed

markedly from the receptors described here. The smallest of these

group C Fc receptors was 48,000 daltons and the functional activity was

totally destroyed by treatment with hot alkali, the condition used by

Grubb et al. (1982) for their initial extraction. In addition, with

the group C streptococcus used here there was no evidence of protease

contamination, degradation or change in heterogeneity of my soluble Fc

receptor during purification (Fig. 4-3).

The ability to isolate, in high yield, a functionally active

streptococcal Fc receptor with apparent homogeneity in binding to the

Fc region of IgG represents a potentially useful immunochemical

reagent. Staphylococcal protein A, by virtue of its selective Fc

binding activity, has proved to be extremely valuable when radio or

enzyme labeled as a tracer in immunoassays (Langone, 1978, 1982b; Gee









and Langone, 1981), or once immobilized for isolation of various

classes and subclasses of IgG (Ey et al., 1978; Patrick and Virella,

1978), separation of antigen-antibody complexes (Kessler, 1976;

MacSween and Eastwood, 1978) or for selective removal of IgG from serum

(Boyle and Langone, 1980; Langone et al., 1979a; Goding, 1978). If the

Fc receptor isolated from this group C streptococcus has species or

subclass reactivities different from staphylococcal protein A it should

be valuable for expanding the immunochemical approaches currently using

protein A. A comparison of the reactivity of staphylococcal protein A

and the streptococcal Fc receptor is presented in the following

chapter.
















CHAPTER FIVE
COMPARISON OF THE FUNCTIONAL AND ANTIGENIC
RELATIONSHIP OF A GROUP C STREPTOCOCCAL Fc RECEPTOR
WITH STAPHYLOCOCCAL PROTEIN A



Introduction

Studies of bacterial Fc receptors on streptococci by Myhre and

Kronvall (1981b) have suggested that there are five bacterial Fc

receptors with differing ranges of species and subclass reactivities.

These receptors, if they could be obtained in a solubilized

functionally homogeneous form might then be anticipated to extend the

usefulness of bacterial Fc receptors beyond those already described for

protein A. In the preceding chapter I described a method for the

isolation of a functionally homogeneous Fc receptor from a group C

streptococcus and in this chapter I will compare the functional

activities of this receptor to those of staphylococcal protein A. The

results presented suggest that this isolated streptococcal Fc receptor

(FcRc) has the binding characteristics of the type III receptor

described by Myhre and Kronvall (1981b) based on the Fc-reactivities of

heat-killed streptococci.

Materials and Methods

Purified Streptococcal Fc Receptor (FcRc)

The soluble streptococcal Fc receptor (FcRc) was isolated and

purified to functional homogeneity from a group C strain designated

26RP66 as described in Chapter 4.









Polyacrylamide Gel Electrophoresis

The affinity purified FcRc was separated by electrophoresis into

four functionally active fractions on 7% non-denaturing polyacrylamide

gels as described in Chapter 4.

iodination of PA and FcRc

Purified protein A (Pharmacia) and the affinity purified FcRc were

iodinated by the mild lactoperoxidase method using enzyme beads

(Bio-Rad) as described previously Chapter 2.

Competitive Binding Assay for Functional PA and FcRc Activity

Protein A and FcRc were quantified using a modification of the

competitive binding assay of Langone et al. (1977). In this assay

0.2 ml of a test sample or buffer is mixed with 0.2 ml of a standard

suspension of agarose beads with covalently coupled human, rabbit or

goat IgG (Bio-Rad Laboratories, Richmond, California), and 0.1 ml of

1251 protein A or 1251 FcRc (approximately 20,000 cpm) and

incubated at 370C for 90 mins. Two milliliters of EDTA-gel was added

to each tube and centrifuged at 1,000 g for 5 mins and the supernatant

fluid decanted. After an additional wash, the radioactivity associated

with the beads was determined in an LKB Gamma Counter. The number of

counts bound in the absence of fluid phase PA or FcRc was compared to

the number of counts bound to the beads in the presence of known

amounts of fluid phase PA or FcRc and the degree of inhibition

determined. The functional activity of these two receptors was

compared by competing unlabeled PA or FcRc with either 1251-PA or

1251-FcRc.










Immobilized Human, Rabbit and Goat IgG

Human IgG was coupled to Immunobeads (Bio-Rad) for use in the

competitive binding assays as described by Langone et al., 1979a.

Rabbit and goat IgG covalently coupled to Immunobeads (Immunobead R-l

and Immunobead G-l, respectively) were obtained from Bio-Rad, Richmond,

California.

Chicken Antibodies to Protein A and FcRc

Monospecific antiserum to staphylococcal protein A was a gift from

Dr. John Langone, National Cancer Institute, Bethesda, Maryland. The

antiserum was prepared as described in Chapter 3. Monospecific anti-

serum to streptococcal FcRc was prepared as described in Chapter 4.

Competitive Binding Assay for the Quantitation of IgG

IgG from a variety of species was quantitated by a competitive

binding assay developed by Langone et al. (1977) and modified as

described in Chapter 2. For this study the ability of different

species IgGs to inhibit the binding of either 1251-PA or

1251-FcRc to immobilized human IgG was compared.

Immunoglobulins

Stock human IgG was prepared by chromatography of normal human

serum on DEAE cellulose (Boyle and Langone, 1980). Aliquots were

stored at -700C until use. Purified rabbit, cow, sheep, goat, rat,

dog, and pig IgG were purchased from Cappel (Cappel Laboratories, Inc.,

Cochranville, PA).

Human IgG Subclasses

Human IgG subclasses were provided by the WHO/IUIS Immunoglobulin

Subcommittee. Two samples of each subclass were tested:

IgG1 (k) lot #0781 and IgG1 (X) lot #0180

IgG2 (k) lot #0380 and IgG2 (x) lot #0981










IgG3 (k) lot #0282 and IgG3 (x) lot #0381

IgG4 (k) lot #0981 and IgG4 (W) lot #0880

Results

Inhibition of Binding of 1251 PA or 1251 FcRc to Immobilized Human
IgG by Unlabeled PA or FcRc

The isolated functionally active FcRc has previously been demon-

strated to be composed of four major charged species, that can be

readily separated and recovered following electrophoresis and elution

from non-denaturing polyacrylamide gels (Chapter 4). Each fraction

eluted from the gel has been shown to bind the Fc region of IgG and all

are antigenically related (Chapter 4).

In the initial experiments described here I compare inhibition of

binding of labeled tracer to immobilized human IgG by 1) unlabeled

protein A, 2) unlabeled affinity purified FcRc or, 3) affinity purified

FcRc that was further fractionated on polyacrylamide gels. The results

presented in Figure 5-1A demonstrate that binding of 1251 PA could

be inhibited by any of the FcRc fractions tested and each FcRc fraction

showed a superimposable inhibition curve. These results indicated that

the binding site on the Fc region of human IgG for protein A and FcRc

are either identical or in close proximity. When the experiment was

repeated using 1251 FcRc as tracer similar results were obtained

(see Fig. 5-1B). As expected the FcRc was more effective in inhibiting

binding of 1251 FcRc to the immobilized human IgG than in

inhibiting 1251 PA. By contrast, protein A demonstrates equivalent

inhibition with both tracers, suggesting that its affinity for the Fc

region of human IgG is higher than that of the FcRc. In similar

comparative binding assays using immobilized rabbit or goat IgG

























100-


100-


ng FcRc or PA/tube














Figure 5-1. Inhibition of binding of (A) 1251-PA or (B)
125I-FcRc to immobilized human IgG by unlabeled PA,

affinity purified unfractionated FcRc or the major FcRc
charge species. For precise experimental details, see
text.


A


unfroclionoted
PA FcRc

bond I

bond IM
bd bond I


bond M








PA unfracfionaed
bond 1ZE FCRC
b ond M530
bond 1I

bond I






o10 100 1000










in place of human IgG, no heterogeneity in binding was observed within

any of the affinity purified FcRc fractions. These findings would

support my previous conclusion that the charge and size heterogeneity

of the affinity purified FcRc preparation could be attributed to

covalently linked cell wall constituents attached to a single type of

receptor (Chapter 4). Consequently in the remaining experiments

presented here I have compared the activity of the total FcRc

preparation to protein A.

Antigenic Relationship of Protein A and FcRc

Polyclonal antibodies to protein A or to the major charge species

of FcRc were prepared in chickens as described in Chapters 3 and 4.

Each antibody was tested for its ability to prevent binding of labeled

1251 PA or 1251 FcRc to immobilized human IgG. In this assay

labeled tracer and immobilized human IgG were incubated for one hour at

37"C with dilutions of serum containing antibody to protein A, serum

containing antibody to FcRc or normal chicken serum. The quantity of

radiolabel associated with the immobilized IgG was quantified after

washing to remove soluble antigen-antibody complexes containing the

labeled tracer. Inhibition detected in this assay requires that the

antibody will combine with a site on the Fc receptor that will

sterically inhibit its interaction with the corresponding binding site

on IgG. The results presented in Figure 5-2 indicated that the binding

of protein A or FcRc was only inhibited when the corresponding antibody

was used. There was no evidence of any antigenic crossreactivity

between these two bacterial Fc receptors. Neither protein A nor the

FcRc reacted with any component in normal chicken serum.

























LL,
80-
to
ANTI FcRc
0 60-
z
z
m 40-


Z
z
20- /ANTI PA

SPRE-IMM
lI __- SERUM
1:72,900 1.24,300 1:8600 1:2700 1.900 1.300

SERUM DILUTION



100- B


80-
/ ANTI-PA

z 60-
0

40-

i 0 ANTI-FcRc
20 PRE-IMM
SERUM /

----- i| --.,..^ .- _.^ ^ ---
1:21870 17290 1.2430 1810 1270 1:90 130

SERUM DILUTION








Figure 5-2. Inhibition of binding of (A) 1251-FcRc or (B)
125I-PA to immobilized human IgG by antibody against
the major charge species of FcRc or against PA. (A)
anti-FcRc ( -- ), anti-PA (0---o), pre-immune chicken
serum (*---); (B) anti-PA ( -- ), anti-FcRc (o---o),
pre-immune chicken serum (-- ).










Comparison of Species Reactivity of Protein A and FcRc

Using the competitive binding assay described in the Methods the

ability of different species of IgG to inhibit binding of 1251 PA

or 125- FcRc to immobilized human IgG were compared. The results

presented in Figure 5-3 and Table 5-1 demonstrate a number of clear

differences in binding of the two Fc receptors. In particular, sheep,

cow and goat IgG were much more reactive with the FcRc than with

protein A (Figure 5-3). Under the assay conditions used similar

inhibition was observed using rabbit IgG, however protein A was more

efficient in its reactivity with human IgG than the FcRc. An absolute

comparison of reactivities of protein A and FcRc can not be made since

the FcRc preparation is heterogeneous and accurate estimates of the

specific activity of the 1251 labeled tracer cannot be made.

The reactivities of human IgG subclasses were also compared in

similar experiments. The results presented in Figure 5-4 demonstrate a

number of interesting reactivities. The labeled FcRc reacted with all

four human subclasses with IgG3 and IgG1 showing approximately

equivalent reactivity while IgG2 and IgG4 demonstrated lower

reactivity. There was considerable variability between the two myeloma

proteins of each subclass tested. It is not clear whether these

differences relate to unique receptors on immunoglobulins from

different individuals, e.g., allotypic sites (Haake et al., 1982;

Schalen, 1982), reactions with Fab regions (Inganas, 1981; Erntell et

al., 1982) or differences in amino acid composition of myeloma proteins

within the site where the bacterial receptor binds. Examples of each

of these types of reactivity of bacterial Fc receptors have been




































q o0 --------



0 80-
In 100- a COW




S 0- (405 n ) q"





Z (21,000 ng)
D 20-


i 0o----------'_______
M 100- C GOAT


80-//


60- / 25 FRc
(180 ng)

40- I, .,-
/ / 1251 PA
/ (13,000 ng)
20- -

![-- ---np---------- 4^_______
10' 102 103 14 05

ng IgG ADDED







Figure 5-3. Inhibition of binding of 1251-FcRc or 1251-PA to
immobilized human IgG by sheep (A), cow (B), or goat (C)
IgG. 1251-FcRc (*---), 1251-PA (o---o); numbers
in parenthesis represent the concentration of IgG required
to inhibit by 50% the binding of the labeled tracer to
immobilized human IgG. For precise experimental details,
see text.









TABLE 5-1

Inhibition of Binding of 1251-PA or 1251-FcRc to
Immobilized Human IgG by IgG from Different Species


Nanograms IgG Required to Inhibit by 50%
Species
1251-FcRc 1251-PA


Rabbit 125 130
Human 44 13
Pig 70 118
Goat 180 13,000
Sheep 240 40,000
Cow 405 21,000
Dog 13,000 100
Rat >105 >105






89
























100- A -00

80- -
a-
i- 60- /7 60 i
t O







O00 100- 0 O-.0 --
x -
a-







2 0- o" 0 b s20/ / .
1.U z










/ ,/ / /
00ng gG A E ng gG A E



































1251-PA (o---o); reactivity of individual subclass
m



































standards, A light chains, with 125i-FcRc (m----); and
reactivity of individual subclass standards, \ light
chains, with 1251-PA (D---).
20 I -










TABLE 5-2


Inhibition of Binding of 1251-PA or
Immobilized Human IgG by Human


125I-FcRc to
Myelomas


Nanograms IgG Required to Inhibit by 50%
IgG Subclass
1251-FcRc 1251-PA


IgG1 (k) 50 30
IgG1 (x) 165 60

IgG2 (k) 190 265
IgG2 (X) 465 960

IgG3 (k) 120 666
IgG3 (A) 70 none detected

IgG4 (k) 80 39
IgG4 (A) 580 90