Citation
Regulation of serum protease activity : action of C1-inhibitor on the activities of serum C1 and nerve growth factor

Material Information

Title:
Regulation of serum protease activity : action of C1-inhibitor on the activities of serum C1 and nerve growth factor
Creator:
Faulmann, Ervin L. W., 1949-
Publication Date:
Language:
English
Physical Description:
v, 75 leaves : ill. ; 29 cm.

Subjects

Subjects / Keywords:
Antibodies ( jstor )
Enzymes ( jstor )
Hereditary angioedemas ( jstor )
Incubation ( jstor )
Molecular weight ( jstor )
Molecules ( jstor )
Nerves ( jstor )
Protease inhibitors ( jstor )
Teeth ( jstor )
Zymogens ( jstor )
Blood Proteins ( mesh )
Complement 1 Inactivators -- analysis ( mesh )
Complement 1 Inactivators -- biosynthesis ( mesh )
Dissertations, Academic -- Immunology and Medical Microbiology -- UF ( mesh )
Immunology and Medical Microbiology thesis Ph.D ( mesh )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1985.
Bibliography:
Includes bibliographical references (leaves 69-74).
Additional Physical Form:
Also available online.
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Ervin L.W. Faulmann, Jr.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright [name of dissertation author]. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Resource Identifier:
020892695 ( ALEPH )
16896114 ( OCLC )

Downloads

This item has the following downloads:


Full Text












REGULATION OF SERUM PROTEASE ACTIVITY:
ACTION OF Cl-INHIBITOR ON THE ACTIVITIES
OF SERUM Cl AND NERVE GROWTH FACTOR






By

ERVIN L.W. FAULMANN, JR.


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


1985










ACKNOWLEDGEMENTS


I wish to express my thanks to Dr. Michael D.P. Boyle

for his guidance, support, and understanding during my

training under him. It has been a privilege to work with

Mike, with his wit, enthusiasism, intelligence and constant

pursuit of "What's new and exciting" he enhances the world

around him.

I would also like to express my appreciation to the

graduate students (both past and present), technicians, and

faculty with whom I have worked the last three years. Their

able assistance, personal sharing, and willingness to put up

with my "helping behaviors" are deeply appreciated.

My parents, Leithel and Ervin Faulmann continue to be a

supporting factor in my life. Though time and distance do

not allow direct contact, I know that their hopes and

prayers are for what I feel is best for me. I find that

virtually all of my personal strengths stem directly from

the foundations that they provided me with many years ago.

I thank them.

Finally, I would like to thank my companion in life,

Ms. Phoebe Castle Faulmann. Her support of my endeavors was

essential for all my work in pursuing my goals. The

strength of our relationship remains not in looking at each

other, but in looking in the same direction. To her, my

wife, and my friend, I would like to dedicate this and all

endeavors.

















TABLE OF CONTENTS


Page


ACKNOWLEDGEMENTS . . . . . . . . ..

ABSTRACT . . . . . . . . . . .

CHAPTERS

I INTRODUCTION . . . . . .. .. . .

II SPONTANEOUS ACTIVATION OF PARTIALLY PURIFIED
SERUM C1. . . . .. .. . . . .

Introduction . . . . . . . .
Materials and Methods . . . . .. ..
Results . . . . . . . . .
Discussion . . . . . . . . .

III REGULATION OF Cl ACTIVATION . . . . .

Introduction . . . . . . . .
Materials and Methods . . . . . .
Results . . . . . . . . . .
Discussion . . . . . . . ..

IV INACTIVATION OF THE PROTEOLYTIC ACTIVITY OF
NGF BY Cl-INH . . . . . . . .

Introduction . . . . . . . .. .
Materials and Methods . . .. . .....
Results . . . . . . . ... . ..
Discussion . . . . . . . . .


V SUMMARY AND CONCLUSIONS

BIBLIOGRAPHY .. ... . . ..

BIOGRAPHICAL SKETCH . . .


. . . . . . . 63

. . . . . . . 69

. . . . . . . 75














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



REGULATION OF SERUM PROTEASE ACTIVITY:
ACTION OF Cl-INHIBITOR ON THE ACTIVITIES
OF SERUM Cl AND NERVE GROWTH FACTOR

By

ERVIN L.W. FAULMANN, JR.

MAY 1985

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

The interactions between serum proteases (serpins) and

serum protease inhibitors are the basis of the regulation of

many homeostatic processes. Evidence from a number of

researchers suggests that protease inhibitors may have

physiological roles in addition to inactivating proteolytic

enzymes.

The first component of the classical complement pathway

(Cl) exists in serum in a precursor zymogen form. During

purification Cl undergoes spontaneous activation yielding

activated Cl (aCl); however the mechanism of this

autoactivation process is unclear.

In this study, by means of the Cl hemolytic assay,

partially purified serum Cl was shown to spontaneously

activate in a time-and-concentration-dependent manner.

Addition of aCl did not alter the level of Cl activation.

iv










These results suggest a model of serum Cl activation that

requires intermolecular interactions between the Cl

molecules independent of the enzymatic activity of aCl

produced. However, addition of purified Cl esterase

inhibitor protein (Cl-Inh), a serpin that inhibits the

enzymatic activity of aCl and a variety of cascade

proteases, inhibited the spontaneous activation of serum Cl

in a concentration-dependent manner. The effect of Cl-Inh

regulating Cl activation was independent of its activity in

inactivating aCl. These results document a new role of

Cl-Inh in which it reversibly inhibits the activation of

zymogen Cl as compared to its irreversible inhibition of

aCl.

The role of Cl-Inh in inhibiting another classical

complement pathway activator, the gamma subunit of nerve

growth factor (gamma NGF), was also examined. The enzymatic

activity of gamma NGF was inhibited by Cl-Inh in a

time-dependent manner with 1:1 stoichiometry. The

interaction of Cl-Inh with gamma NGF produced a sodium

dodecyl sulfate (SDS) stable complex that was susceptible to

cleavage by hydroxylamine, thus suggesting a covalent bond

between the protease and protease inhibitor. This reaction

between gamma NGF and Cl-Inh is similar to that between

gamma NGF and a newly described class of cell secreted

protease inhibitors, called nexins, and provides new insight

as to the possible physiological roles of both Cl-Inh and

gamma 'JGF.
















CHAPTER I
INTRODUCTION


The activation and control of specific serum proteases

is the basis of many homeostatic processes. These processes

include the clotting system, the fibrinolytic system, the

kinin generating system, and the complement system, all of

which have important roles in inflammation and in the normal

response to tissue damage. Each of these processes entails

the serial interaction of specific serum proteins, in which

the activation of one component can initiate a chain

reaction (cascade reaction) culminating in the generation of

a specific effector molecule, e.g., the activation of

Hageman factor initiates a series of reactions that can

result in the formation of a fibrin clot. Many of the

interactions of these cascades are mediated through the

activation of a proteolytic enzyme from an inactive

precursor (zymogen) molecule. At various points along these

cascade reactions there are specific amplification steps in

which one activated molecule can stimulate the activation of

a large number of subsequent components in the cascade.

During the course of these cascade reactions biologically

active split products are produced, e.g., the anaphylatoxins

of the complement system. It is clear that if such

pharmacologically active split products are produced







2

unregulated there would be serious physiological

consequences. Serum protease inhibitors provide an

important level of regulation by inactivating specific

proteolytic enzymes associated with key amplifying roles in

these cascade systems.

Recent evidence suggests that protease inhibitors may

have physiological roles in addition to the inactivation of

proteolytic enzymes. In 1980, Baker et al. described the

secretion of a protein from normal human fibroblasts that

formed a covalent acyl linkage with urokinase and thrombin.

The resulting proteolytically inactive complexes were

internalized and degraded (Knauer et al., 1983). These

regulating anti-proteases were found to be secreted by a

wide range of normal and transformed cell lines and were

named nexins. Detailed studies using a variety of sources

of nexins and a number of serine class proteases have

indicated the existence of three distinct protease nexins

(PNs). These regulating proteins designated PNl, PN2, and

PN3 share many of the functional characteristics of the

serpins, but appear to be both physicochemically and

antigenically distinct proteins (Knauer et al., 1983). The

three groups, PNl, PN2, and PN3, differ in the enzymes which

they inhibit. The first, PNl, is a protein of molecular

weight of 38,000 Daltons that binds thrombin and urokinase.

Complexes with either of these enzymes bind to the surface

receptors on a variety of cell types and are rapidly

internalized and degraded. In the presence of heparin, cell







3

binding is enhanced (Knauer et al., 1983). The second, PN2,

is a protein of 93,000 Daltons with selective reactivity for

the epidermal growth factor (EGF)-binding protein (Knauer et

al., 1983). These complexes are also cleared by a variety

of cell types but their uptake is not augmented by addition

of heparin. The last, PN3, is a 31,000 Dalton protein with

selective reactivity for the gamma subunit of nerve growth

factor (gamma NGF) and the resulting gamma NGF-PN3 complexes

are rapidly cleared by a variety of cells. As in

PN2-containing complexes, clearance is not augmented by

heparin. The binding specificities of the PNs are not

absolute and some reactivity of gamma NGF and PNl and PN2

has also been reported (Knauer et al., 1983). The

physiological function of the nexins is not clear. The

current findings suggest that they act to regulate the

biological activities of certain serine proteases by forming

complexes that are rapidly cleared and degraded. At

present, no stimulatory biological activity has been

reported for any PN-protease complex.

Of particular interest to this study is the Cl esterase

inhibitor protein (Cl-Inh) that regulates a variety of

cascade systems associated with the host response to tissue

damage and inflammation (see Figure 1). Human Cl-Inh is a

single-chain glycoprotein with a molecular weight of 116,000

Daltons (Harrison, 1984) (97,000 Daltons in SDS-PAGE (Harpel

and Cooper, 1975)), containing about 35% carbohydrate

(Harrison, 1984). It was first recognized for its ability













Complement
System


Clotting
System


Kinin
Generating
System


C4 C2
aCl # >Y
classical
pathway


factor IX
PTA (factor Xl) -70-> /
activation


/ I-


kallikrein


kininogen
>kinin
kinin


Fibrinolytic
System


plasmin L


fibrin split products
CI activation


Figure 1. INHIBITORY FUNCTIONS OF Cl-INH.







5

to regulate the enzyme activities of the first component of

the classical pathway of complement (Cl) (Levy and Lepow,

1959). Serum Cl exists as a zymogen that can be activated

to produce a proteolytically active focm (aCl) which cleaves

C4 and C2 (Patrick et al., 1970). The Cl-Inh has been shown

to form a covalent bond with the activated subunits of aCl,

activated Clr (aClr) and Cls (aCls), leading to the

inactivation of the enzyme and the dissociation of Cl into

its subunits (Harpel and Cooper, 1975; Minta and Aziz,

1981).

Hereditary angioedema (HAE) is a genetic deficiency of

functional Cl-Inh, which is associated in patients with

episodic bouts of swelling. Donaldson has attributed the

edema to uncontrolled production of a kinin produced by the

cleavage of C2 (i.e., C2b) (Donaldson, 1970). Others have

suggested the swelling results from Hageman factor

activation (Schapira et al., 1983). As a serpin, Cl-Inh is

thought to operate by being first cleaved by the specific

protease and then forming a covalent bond with an

appropriate amino acid within the active site of the serine

esterase. The possibility that a variety of functionally

distinct Cl-Inh may exist is suggested by the observation

that Cl-Inh from sera of certain HAE patients fail to block

aCl activity, but continue to inhibit kallikrein and plasmin

effectively (Donaldson, 1982; Travis and Salvesen, 1983).

Similar variations of the range of activity for other

serpins have been observed (Travis and Salvesen, 1983). The







6

altered activity has been shown to be associated with

mutations in the amino acid sequence of the target site on

the serpin recognized by the specific enzyme (Carrell,

1984). During a disease episode, the serum Cl of HAE

patients is fully activated and there is no measurable C4

and C2 (Gelfand et al., 1979). Asymptomatic HAE patients,

however, exhibit low but measurable levels of C4 and C2

(Gelfand et al., 1979), hence their Cl is not fully

activated. The low levels of functional Cl-Inh and the

anomalous episodic activation of the serum Cl of HAE

patients appear to be causally related. Clinical

administration of Cl-Inh reverses the disease process and

the patient's C4 and C2 levels are restored (Gadek et al.,

1980).

The main focus of this study was to examine the

activation of serum Cl and determine the role of Cl-Inh in

regulating the activation process. Previous studies have

reported the autocatalytic activation of partially purified

serum Cl, in which aCl reacts enzymatically with zymogen Cl

resulting in the conversion of zymogen Cl to aCl (see Figure

2) (Lepow et al., 1965). In this case, Cl-Inh would

regulate the activation of Cl by inhibiting the enzymatic

activity of aCl. A recent report has shown that purified

zymogen Cl spontaneously activates by a intramolecular event

(concentration independent) that was not autocatalytic

(Ziccardi, 1982a). Furthermore, Cl-Inh was shown to inhibit

this spontaneous activation, though the mechanism of this

























z C I slow



zCI,


aCl


CI INA

aoCI


Figure 2. SCHEMATIC OF THE CLASSICAL AUTOCATALYTIC
ACTIVATION MODEL.







8

inhibition remains unclear (Ziccardi, 1982b). This study

was initiated to resolve the apparent controversy regarding

the spontaneous activation of zymogen Cl and the role Cl-Inh

may play in it. Partially purified serum Cl was shown to

undergo spontaneous activation in a

time-and-concentration-dependent manner. This spontaneous

activation was independent of the enzymatic activity of

exogenous aCl. Serum fractions depleted of Cl and purified

Cl-Inh were shown to inhibit the spontaneous activation of

serum Cl. These studies suggest a model of Cl activation

that is dependent on an intermolecular interaction between

zymogen Cl molecules that is independent of the enzymatic

activity of either the Clr or Cls subunits.

Recently, it has been reported that the high molecular

weight form of NGF can substitute for activated Cl in

activating the classical complement pathway cleaving C4 and

C2 (Boyle and Young, 1982). High molecular weight NGF is a

multisubunit protein composed of three subunits, alpha,

beta, and gamma in the stoichiometric ratio of 2:1:1

respectively (Greene et al., 1969). The alpha subunit has

no known biological function (Isackson and Bradshaw, 1984).

The beta subunit is responsible for the outgrowth of

sympathetic and sensory ganglia from chicken embryo

(Levi-Montalcini, 1965). The gamma subunit is a serine

class enzyme (Orenstein et al., 1978) which is responsible

for complement activation (Boyle and Young, 1982). This

complement activating ability of gamma NGF was shown to be







9

inhibited by Cl-Inh (Boyle and Young, 1982). To date, gamma

NGF has been reported to mediate a number oF activities

associated with inflammation, i.e., enhanced wound healing

(Li et al., 1980), and both in vivo and in vitro

leukocyte chemotaxis (Boyle et al., 1985; Gee et al., 1983).

However, gamma NGF is not known to be a member of any

physiological cascade system. Similarly, Cl-Inh is not

known to inhibit proteases not associated with cascade

systems. In order to determine if Cl-Inh was acting as a

serpin (or nexin), the nature of the interaction of gamma

NGF and Cl-Inh seen in the complement assay needed to be

examined. In this study, purified Cl-Inh was shown to

inhibit the enzymatic activity of purified gamma NGF in a

time-dependent manner with 1:1 stoichiometry. The

interaction of gamma NGF with Cl-Inh resulted in the

formation of a covalent bonded enzyme-inhibitor complex.

The properties of Cl-Inh reported in this study expand

our understanding of its physiological role as a serum

protease inhibitor. The results presented demonstrate that

Cl-Inh displays the general characteristics used to define a

protease nexin, in addition to regulating Cl activation and

the enzyme activity of fully activated Cl and nerve growth

factor.















CHAPTER II
SPONTANEOUS ACTIVATION OF PARTIALLY PURIFIED SERUM Cl


Introduction

Zymogen Cl has been shown to spontaneously activate in

the absence of either soluble or cell bound immune complexes

(Borsos et al., 1964; Cooper, 1983; Lepow et al., 1965;

Ziccardi, 1982a). Using a synthetic substrate to measure

the esterolytic activity of Cl, Lepow et al. (1965)

demonstrated that serum Cl isolated by euglobulin

fractionation spontaneously activated in a classical

autocatalytic manner. That is, zymogen Cl served as a

substrate for aCl and the enzymatic activity of the aCl led

to the conversion of zymogen Cl to aCl. By contrast, in a

recent series of experiments, Ziccardi (1982a) showed that

the activation of zymogen Cl was a concentration independent

process that was not influenced by the presence of aCl. In

those experiments zymogen Cl was reconstituted from its

isolated zymogen subunits (Cooper and Ziccardi, 1977) and

activation was measured by changes in the physicochemical

properties of Clr and Cls (Arlaud et al., 1980; Sim, 1981;

Sim and Porter, 1976; Valet and Cooper, 1974). This assay

requires a large input of zymogen Cl and is not as sensitive

as the Cl hemolytic titration (Borsos and Rapp, 1967) or

synthetic substrate enzymatic cleavage assays (Lepow et al.,

10







11

1965). The reason for the discrepancy between Lepow's and

Ziccardi's findings has not been established but may be

related to the source of zymogen Cl used and/or the method

chosen for measuring Cl activation.

In the studies presented here, the Cl hemolytic assay

was was shown to be an effective method to discriminate

zymogen from activated Cl and was used to monitor the

activation of serum Cl. Partially purified serum Cl was

shown to spontaneously activate in a concentration-dependent

manner and that the addition of adCl to the zymogen Cl did

not increase the extent of spontaneous activation. These

studies suggest a model of Cl activation that is dependent

on an intermolecular interaction between zymogen Cl

molecules which is independent of the enzymatic activity of

either the aClr or aCls subunits.



Materials and Methods

Reagents

Sheep erythrocytes (E) were collected and washed as

described in Borsos and Rapp (1967). Rabbit antiserum rich

in IgM anti-E was produced and depleted of IgG by passage

over a Protein A-Sepharose (Pharmacia, Upsala, Sweden)

affinity column. Ascites fluids containing mouse monoclonal

IgM antibodies specific for E were derived from a hybridoma

(TIB 110) purchased from ATCC, Washington, DC (contributed

by R. Raschke). Human serum was prepared from outdated

plasma (Civitan Regional Blood Center, Gainesville, FL).







12

Guinea pig complement was obtained from Pel-Freez

Biologicals, Rogers, AR. Isolated complement components

were purchased from Cordis Biologicals, Miami, FL. Isotonic

veronal buffered saline containing metals (VBS) and

sucrose-veronal buffer (SVB) (pH 7.35, with 0.1% gelatin

(Difco), 0.001 M Mg ++, and 0.00015 M Ca ++), and

veronal-buffered saline containing 0.01 M ethylenediamine

tetraacetic acid (VBS-EDTA) were prepared as described in

Borsos and Rapp (1967). A reagent which was devoid of whole

complement activity but provided functional complement

components C3-C9 was obtained by diluting serum in VBS-EDTA

(CEDTA) (Borsos and Rapp, 1967). Indicator cells for the Cl

hemolytic assay were prepared by the method of Borsos and

Rapp (1967) in which E were sensitized with the antibody

solution, coated with aCl, incubated in CEDTA, and washed.

This procedure yields antibody sensitized E (EA) with the

activated fourth component of complement bound to their

surfaces (EAC4b)(contributed by the added serum), without

contaminating Cl and C2 (removed by the added VBS-EDTA) or

any later acting complement components (Borsos and Rapp,

1967).



Isolation of Serum Fraction Rich in Zymogen Cl

Fractions rich in zymogen Cl were isolated from human

serum by ion-exchange chromatography. Human serum diluted

1:2 in distilled water and applied to a column of DE-52

cellulose equilibrated at pH 7.3 with 0.01 M








13

N-2-hydroxlethyl piperazine-N'-2-ethanesulfonic acid (HEPES)

and 0.075 M NaCi was eluted with equlibration buffer at

4C. This procedure depletes the serum of Cl-Inh (see

Chapter III) and yields a preparation of zymogen Cl that

will spontaneously self-activate (see Figure 8). Throughout

the procedures, the isolation and activation state of the Cl

was monitored using the hemolytic assay described below.



Measurement of the Hemolytic Activity of aCl

The assay for aCl was carried out by incubation of

dilutions of a sample containing an unknown amount of aCl

with 1.5x107 EAC4b in a total volume of .3 ml of SVB

(p=0.065) (Borsos and Rapp, 1967) for 20 minutes at 37C.

The hemolytic reaction was completed by the sequential

addition of excess C2 for 5 minutes at 37C followed by the

addition of excess late acting components (CEDTA) for 1 hour

at 37C. Under these conditions, the reaction is at

endpoint (Borsos and Rapp, 1967). The number of effective

aCl molecules was determined by measuring the extent of

lysis of the EAC4b indicator cells by spectrophotometrically

monitoring the cell free supernatants at 412 nm for the

release of hemoglobin.



Analysis of Cl Activation Status from Hemolytic Activity

Data

The functional activity of serum Cl and fully activated

Cl has been extensively studied and characterized (for







14

review see Loos (1982)). It has been shown from dose

response and kinetic experiments that every aCl molecule is

capable of generating a lytic site (Borsos and Rapp, 1963;

Colten et al., 1967) and that one such site is necessary and

sufficient for cell lysis (Borsos and Mayer, 1962). Tke

hemolytic assay (described above) provided all the other

components necessary for completing the lysis of all the

cells in the reaction mixture, i.e. pseudo-first order for

aCl activity, and the extent of hemolysis was measured at

end point. The physical conditions, eg., pH, ionic

strength, and temperature, of each reaction step were

carefully controlled to ensure the reproducibility of the

reaction. In meeting these criteria the distribution of aCl

molecules among the cells would conform to the predictions

of the Poisson distribution (Rapp, 1955) and consequently it

was possible to relate the percentage of cells lysed (y) to

the absolute number of hemolytically effective aCl molecules

added to the assay by the relationship

Z = In ( 1 y ).

In this equation Z represents the average number of lytic

sites per cell (Borsos and Mayer, 1962) and the total number

of hemolytically effective aCl molecules in the system was

determined by multiplying Z times the number of cells in the

assay. Further considerations of the activity of aCl and

serum Cl in the hemolytic assay are dealt with more fully in

the results and discussion sections below.







15

Results

Activity of Serum and Activated Cl in the Hemolytic

Assay

In the Cl hemolytic assay, aCl yielded one-hit

characteristics in which there was a 1:1 relationship

between the relative amount of aCl added to the assay and

the number of effective Cl induced lytic sites produced (see

Figure 3). This relationship was observed irrespective of

whether aCl was diluted or the number of indicator cells was

changed (see Figure 4 and Table 1). These findings fully

confirm the previous work of others (Borsos et al., 1980;

Borsos et al., 1964; Loos, 1982).

Serum Cl titrated under identical conditions did not

follow a similar pattern. As shown in Figure 3, the

quantity of aCl activity detected fell off more quickly than

predicted as serum was diluted. This was not due to the

heterogeneous nature of the binding specificities of the

sensitizing antibodies, because indicator cells prepared

with monoclonal antibodies yielded similar results (see

Figure 5). Additionally, the excessive loss of Cl activity

upon diluting serum Cl was not due solely to dissociation

since, as is shown in Figure 4 and Table 1, the total number

of effective Cl molecules detected at any dilution was

dependent upon the indicator cell number used in the assay.

In agreement with earlier studies (Loos et al., 1972), mild

treatment of serum Cl coated EAC4b indicator cells with

trypsin resulted in the appearance of an increased titer of






















2 .0 i- i i 1 i i ii i iii- i- ii


1.0-
0.8-
Q8: / E -

0.6-

Q4-


0.2-


-'.10-
c .08-
.06

.04-


.02


I I I 1 I II ... I I I 11! 111 1 I It I I
I 2 5 10 20 50 100 200 500 1000

RELATIVE CI CONCENTRATION


Figure 3. THE HEMOLYTIC TITRATION OF SERUM Cl AND ACTIVATED
Cl. Various dilutions of either serum or activated Cl were
incubated with EAC4b for 20 minutes at 37C. The extent of
hemolysis was determined at end point following addition of
excess C2 and CEDTA. ( serum and a activated Cl)








17















2.0-
A 8
1.0 E
0.8
0.6-



''0.2-
o3 / /


0.2

VS
NO.I 1
L.08-
.-.0E

1.04


I.02

A
I 2 4 6 1 2 4 6 8
RELATIVE Cl CONCENTRATION

Figure 4. MEASUREMENT OF SERUM AND ACTIVATED Cl ACTIVITY AS
A FUNCTION OF INDICATOR CELL NUMBER. Various dilutions of
either activated Cl (panel A) or serum (panel B) were
measured for Cl-activity using various concentrations of
EAC4b indicator cells. ( 24x10 6.0xlO and
0 1.5x10 )

























Table 1. MEASUREMENT OF SERUM AND ACTIVATED Cl ACTIVITY AS
A FUNCTION OF INDICATOR CELL NUMBER. A fixed dilution of
activated Cl or serum Cl was assayed using various
concentrations of indicator cells (EAC4b). These results
were calculated from the data presented in Figure 4.


Number of EAC4b
Target Cells


Number of Hemolytic Effective
Cl Molecules/ml Measured
Activated Cl Serum Cl


1.5 x 107

6.0 x 107

24.0 x 107


3.00 x 107

2.88 x 107

2.88 x 107


17.0 x 107

13.8 x 107

7.2 x 107







19










I I I I I I | !

3.0-


2.0




as)
I 0.8: //'

N 0.6
Li






0.2 ,




2 5 10 20 50 100
RELATIVE C I CONCENTRATION

Figure 5. HEMOLYTIC ACTIVITY OF SERUM ON POLY- AND
MONOCLONAL ANTIBODY COATED INDICATOR CELLS.- Various
dilutions of serum were assayed for Cl activity on EAC4b
indicator cells that had been prepared with either
polyclonal rabbit anti-E ( A ) or monoclonal mouse anti-E
( ). The activity of activated Cl on monoclonal
sensitized indicator cells ( ) was included as a
reference.








20

Cl that had characteristics of aCl (see Figure 6). These

findings confirmed those of Loos and colleagues (Borsos et

al., 1980; Loos, 1982; Loos et al., 1972) and demonstrated

that the Cl hemolytic assay was highly sensitive and was

able to distinguish native and activated Cl in the presence

of many other serum proteins. In other experiments it was

confirmed that aCl bound to indicator cells at 4C and

shifted to 37C yielded activity without a lag, while Cl

activity in serum that had been bound to EAC4b cells at 4C

and moved to 37C demonstrated a lag period that was

consistent with the time required for Cl to undergo an

activation process (see Figure 7). These functional

properties of native and activated Cl have been used in the

remainder of the study to follow the activation status of

various preparations of serum Cl.



Spontaneous Activation of Partially Purified Serum Cl

Human serum was diluted 1:2 with distilled water and

applied to a column of DE-52 previously equilibrated with

.075 M NaCI and 10 mM HEPES pH 7.4 at 4C. The

pass-through material contained approximately 80% of the

serum proteins, as monitored by OD280 and include the

majority of the measurable Cl activity. The amount of Cl

activity in a single dilution of the partially purified

serum Cl was monitored in the Cl hemolytic assay, along with

appropriate dilutions of unfractionated serum and aCl. When

incubated for various times at 37C, the relative amount of










!iO.O ----- i T i- i ,i1 i ---- -- l- I I I I I I
8.0-
6.0
/'
4.0 / -


0/ /

r-/ -/ / W
Q6 ,,/y J .~
1.0 / /
i////
0.8/


> 0.4- / /
N / /

>'Q2 -
/// / /
1~, Q2 ///
I .I 0
C//
.08 //
.06 -/

.04


.02-

1 I t t t II I I I1 1 I I 1 1 1 1 1 1 I I I... ...I.....I...I
2 5 20 50 100 200 500 1000

RELATIVE CI CONCENTRATION

Figure 6. EFFECT OF TRYPSIN TREATMENT ON THE IIENIOLYTIC
ACTIVITY OF CELL BOUND SERUM Cl AND ACTIVATED Cl. Various
dilutions of serum and activated Cl were incubated with
EAC4b for 20 minutes at 37C in SVB buffer. The cells were
washed and treated with trypsin (1 pg/ml) (closed symbols)
or buffer (open symbols) for 5 minutes at 27C. The cells
were again washed, resuspended in SVB, and assayed for cell
bound Cl activity as described in the Methods. ( and 0 -
serum treated cells, a and 0 activated Cl treated cells)












































































I^ru CM-
5 [nIVA 2:( CA ; ul -


1--- _

,-4 a)O
r-q Em 0 $
u T3 14-1 41 3
Crd 0 4Ji u
Z ro cc0 3(a -11 U
4G vU u 0) 0
4J O-4J U
a > i U n Do
0 -, r -4 r- 0 a
> 0CO -
a 4J W W 4-1 04


4) 0) d 41 -
&T4 4J -4 X %i 4-) xr
0 4 Q) 0 4-)

HHo *-4 rOO
0 Z .0 c I (v -4


H H 3 -4 3 0 01
> -E-- d W C : rA >i
H Xc -4 U) r-4 r-4
E-4 r 4.i U) M4 C 0
w4 c m7 -4
OC> (0 r. C l 0
0 a!) o (1) (a
U r0 4-) +- 4- H
H u) a) 0 c
rj U) w 0 c =%
X :?3 Z a U-,q 0 0
0 U -U -4 4d4 4 It





~OLo u~4
r-4)
U ) M ) 4 0) 4 0) C




a <' a > 4-J a) E
0 '0 a0 ^ )







OiH S WUOEiX (
4-J T$ -.4 C U r-4
Z 0-441 W14J 4J -4 -T
0 E-iU o -. r C)
HMU T 3'o 3S 0
E-4 Q c$T c;

a M to $4 (u
Ur- (U4 o a)i' a

u Ma (d a) 0) M -4^
> 4-1 5n
W~( C (L) it 34Ji a)
E-f (0 41>< 41)
QU $4 > roi*< H'
M lE-4 U 0-< Q (n
*-l U c! Ql) r- w C
(n4 < --' *: SMu U .DS







23

Cl activity in a single dilution of partially purified serum

Cl increased, while the relative Cl activity found in serum

or aCl dilutions remained constant (see Figure 8). This

activation of partially purified serum Cl upon incubation at

37C could be seen more dramatically when the various Cl

solutions were titered in the Cl hemolytic assay. In Figure

9 it can be seen that the partially purified serum Cl

expressed an increased titer of Cl activity in the hemolytic

assay and demonstrated the characteristics of aCl.

Unfractionated serum incubated under identical conditions

did not self-activate (data not shown). Similar results

were obtained when Cl was partially purified from serum by

precipitation with 40% saturated ammonium sulphate or by

ultracentrifugation (100,000 x G) (data not shown). These

results suggest that Cl can be partially purified from serum

and remain zymogen; however, once separated from certain

serum proteins, this zymogen Cl undergoes self-activation.



Concentration Dependence of Serum Cl Activation

In the next series of experiments, the effect of the Cl

concentration on the process of spontaneous activation of

the DE-52 separated material was tested. Serum Cl was

isolated as described above, and divided into three pools,

and diluted 1:4, 1:16, and 1:64 respectively. Each pool was

preincubated at 37C for various times before testing at

equal final dilutions for Cl activity. The results

presented in Figure 10 indicate that the rate of spontaneous


















r-1 m oerum
W 0.6- a-Activated CI
O-J-

> 0.5
N


0.3-
UA



0.1
I t I I
2 5 10 20 60

TIME (min)


Figure 8. HEMOLYTIC ACTIVITY OF SERUM, ACTIVATED Cl, AND
PARTIALLY PURIFIED SERUM Cl AFTER INCUBATION AT 37C. A
single dilution (1:4 in VBS) of activated Cl ( U ), serum
( A ), and partially purified serum Cl ( 0 ) were incubated
37C. At appropriate times, aliquots were taken, diluted,
and assayed for Cl activity in the hemolytic assay. A
single dilution of each source is represented at each time
point.





















I I I

/


S0.6-

N 0.4-
L-I


I 0--0'
S0.2- 0-2'
7" a-5'
Ml- I0'

0.1- A-20'
0"10 o-60'



I 2 4

RELATIVE CI CONCENTRATION
Figure 9. HEMOLYTIC ACTIVITY OF PARTIALLY PURIFIED SERUM Cl
AFTER INCUBATION AT 37C. A single dilution (1:4 in VBS)
of partially purified serum Cl was incubated at 37C. At
various times, aliquots were removed and diluted. The
samples were then titered in the Cl hemolytic assay.
( 0 0, 0 2, A 5, a 10, A- 20, O 60 minutes)







































































0 cq (R lq
0 0n Z]
Odo o
[anlIDA ZJ


o C aJ

OD (1 4 41 t|-
4~4~J

0o .-u x -,.
<[C T 0.. C- o
W i>,~ rO 34J rl



,-4 r- c >4-
ZZO O--A ol 0 u >a!54
z Qs ( U) -o C
1 k *<(1 0 U) 1s 4. M 4J 0-
\" 0 0 --r c 0 (0 E










O. z oz I
U4 0 t 4J <
0 -4 u r-4 I



--04 : 0 4 -) 0
-- < *- - "' u u' s


IL = W -C .-W 0 0 -
r -4,,-1 0 ,-r 1 t


0-.4 c m 0 p 0
Z CA 4 044 4jt~)
Si-~ < oji* c o











0 -A 4- (P.
I ....... I, i I i H O l n i
H C 4-) I4

























d oo Qq
( ,C -)) u -o
U 4 E 0 a) MUC> i
4J1 -^ O 4 .)

L 4^-" U i- U)r (f) ^i0 c 0
-s. '^ r > 2 t 6i QC 0t'
\. ~~~E -4, -4 4J -4 '+ u
E-4 r-4 4 41 4. ^
<\ u^ u (1i- )^ : ff (d) -
1- 44.. a-, Uo o ror-I (a C S
CD, MJ FX 0 r01 M W w
W 0^ w (1 r- CtOi

: 0 i -l W 40 Cr-4 0 0
*- H l- r- 4 **c 44- 4 -4
E-4 a;rlr() 4J -) W r-l
-<- x q w :j C1 o >4
3 Ha -4 c 4 t E- 4-) a) e4 (0 5
z o 04 a) P^ >



C-4 co (Du- It
-'^0 0 0. 0- OrIX: 1- ^--
u< ) ) J 4








27
activation of partially purified serum Cl is influenced by

the concentration of the Cl in the system, i.e., the

activation was concentration dependent since after 20

minutes at 370C the 1:4 dilution was activated (slope = 1),

while the 1:64 dilution was not (slope>l). Upon further

incubation, however, the 1:64 dilution was shown to

self-activate. As all the solutions were assayed at the

same final dilutions, the rate of activation was clearly

dependent on the concentration of the partially purified

serum Cl in the preincubation solution. These results

suggest that some interaction between Cl molecules is

required for activation.



Effects of Activated Cl on Serum Cl Activation

In earlier experiments, using euglobulin precipitates,

Lepow et al. (1965) suggested Cl would activate by an

autocatalytic reaction, i.e., aCl acted on zymogen Cl to

generate more aCl activity. To test this possibility, a low

level of aCl was added to a dilution of partially purified

serum Cl and incubated at 37C. As shown in Figure 11,

addition of aCl to this mixture did not result in increased

level of activation of the partially purified serum Cl. The

level of zymogen Cl chosen for this experiment was below the

concentration that would self-activate during the time

course of the experiment (see Figure 10). Rather, the

resultant Cl activity of the mixture represented the sum of

the respective Cl activities of the zymogen and added aCl.

















I I I I


4.0- /
/
/


2.0 /
/
/
ao /
~/
/
oI- 8 /
>/
N 1.0


I/
0.6 /

I/
0.4/




02 /





0.4 0.6 0.8 I 2 4 6 8 10

RELATIVE Cl CONCENTRATION

Figure 11. EFFECT OF ENZYMATICALLY ACTIVE Cl ON THE
SPONTANEOUS ACTIVATION OF PARTIALLY PURIFIED SERUM Cl.
Activated Cl or buffer was added to partially purified serum
Cl (diluted 1:400 in VBS), incubated 20 minutes at 37C,
and assayed for Cl activity by the hemolytic assay.
( partially purified serum Cl, - activated Cl,
&- partially purified serum Cl + activated Cl,
- - additive activity of partially purified serum Cl and
activated Cl separately)







29

Taken together, the results in Figures 10 and 11 indicate

that Cl activation is neither a concentration independent

autoactivation reaction nor a classical autocatalytic

reaction. This suggests that Cl activation requires the

interaction of more than one Cl molecule but the interaction

is not one that is characteristic of an enzyme-substrate

interaction. These findings are consistent with the

observation that Cl can be demonstrated in self-activatable

and non-self-activating forms (Borsos et al., 1980; Loos et

al., 1972).



Discussion

When assayed under pseudo-first order conditions in

hemolytic assays, the fully activated form of Cl exhibits

all the characteristics of a single hit reaction (Borsos and

Mayer, 1962). The behavior of activated Cl in the hemolytic

assay is summarized below:

1) Activated Cl is 100% efficient in the generation of

lytic sites on the sensitized red cell membrane (Colten et

al., 1967).

2) There is a direct proportionality between the

quantity of Cl added and the number of lytic sites generated

(Colten et al., 1967).

3) The number of effective Cl molecules measured in the

hemolytic assay is independent of the number of indicator

cells used (Loos, 1982).








30

These results indicate that Cl conforms to all the

predictions of the one hit theory of immune hemolysis

originally proposed by Borsos and Mayer (1962).

In contrast, when Cl in serum is measured in the

hemolytic assay under conditions that are pseudo-first order

for Cl its behavior does not conform to the predictions of

the one hit theory. The results observed with serum

(zymogen) Cl are summarized below:

1) The number of lytic sites generated is not directly

proportional to the quantity of serum Cl added (Heinz et

al., 1984; Hoffman and Stevland, 1971; Loos, 1982; Ratnoff

and Lepow, 1957).

2) Using the same dilution of serum Cl, the measured Cl

activity is dependent on the number of indicator cells used

in the assay (Loos, 1982; Loos et al., 1972).

3) Two populations of serum Cl can be detected on

indicator cells incubated with serum. The first, a

productive self-activatable Cl molecule, and the second, a

non-self-activatable molecule. The latter form of Cl may be

converted to active Cl by mild treatment with trypsin or

other proteases (Loos, 1982; Loos et al., 1972).

The differences between aCl activity and zymogen Cl

activity were not due to the heterogeneous polyclonall)

nature of the antiserum normally used to prepare the EAC4b

indicator cells used in the Cl hemolytic assay. The results

presented here demonstrate that the behavior of both aCl and

serum Cl was not changed when monoclonal sensitizing







31

antibodies rather than the polyclonal were used to prepare

the indicator cells.

The differences in aCl and zymogen Cl activity were

used to monitor the activation of Cl either in serum or

following partial purification. It was shown that serum Cl

could be partially purified from other serum proteins

without being fully activated in the process. This

partially purified serum Cl was seen to undergo a

spontaneous self-activation process that was

time-and-concentration-dependent. This activation required

some form of Cl to Cl interaction but activation was not

promoted in a classic autocatalytic fashion. These findings

are distinct from either the results of Ziccardi (1982a) or

of Lepow et al. (1965). The disparity between the findings

reported here and those of Ziccardi may be due to the

characteristics of the assays used to measure Cl activation.

The change in activation rate seen in these studies was

most noticeable at concentrations of Cl below the detection

limits of the physicochemical assay used by Ziccardi (Cooper

and Ziccardi, 1977; Ziccardi, 1982a; Ziccardi and Cooper,

1976). Moreover, due to the one-hit nature of the aCl

molecule, the hemolytic assay has the potential of being

more sensitive for aCl than zymogen Cl. This has been shown

in a similar hemolytic assay system in which IgG sensitizing

antibodies are compared to IgM sensitizing antibodies for

the ability to induce complement mediated lysis. The

fixation of Cl to IgG is a two-hit phenomenon in which one







32

Cl molecule must bind two IgG molecules on the surface of

the target cell to induce activation (slope=2 in the

hemolytic assay). The activation of Cl on IgM sensitizing

antibodies is a one-hit phenomenon (slope=l) and as such has

been shown to be at least 100 times more efficient than IgG

in activating Cl. Thus, a contamination of 1% specific IgM

in an IgG sensitizing antibody reagent would completely mask

the contribution of the IgG antibodies in the assay

(slope=l). Similarly, adC is more effective than zymogen Cl

in the Cl hemolytic assay.

In agreement with Ziccardi's findings (Ziccardi,

1982a), however, in this study no evidence for an

autocatalytic activation reaction, in which the the rate of

activation would increase with the addition of aCl, was

observed. It is possible that Hageman factor, which has

since been shown to spontaneously activate in a classical

autocatalytic manner (Silverberg et al., 1980) which can

then activate serum Cl (Ghebrehiwet et al., 1983), could be

responsible for the autocatayltic generation of enzyme

activity in the euglooulin preparation used in Lepow's

studies. Furthermore, the synthetic substrate,

N-acetyl-l1-tyrosine ethyl ester (ATEE), used in these

studies is not particularly sensitive in detection activity

of Cl (McRae et al., 1982), therefore the presence of any

contaminating autocatalytically activated enzyme that is

more active (i.e. Hageman factor) would mask the low level

of serum Cl activity.







33

The differences reported in the various studies will

only be resolved when methods become available to compare

the Cl enzyme activity detected in the hemolytic assay,

synthetic substrate assay, and physicochemical assay can all

be carried out using the same preparation and the same

concentration of zymogen Cl under the same experimental

conditions.

The apparent lack of spontaneous activation seen with

Cl in unfractionated serum will be addressed in the

following section (Chapter III).
















CHAPTER III
REGULATION OF Cl ACTIVATION


Introduction

The first component of complement (Cl) is present in

normal serum in a zymogen form (Lepow et al., 1965). Upon

binding to the Fc region of complement fixing antibodies

bound to antigen, serum Cl undergoes a series of

intramolecular events producing activated Cl (aCl) (Borsos

et al., 1964; Heinz et al., 1984; Loos et al., 1972; Loos

and Hill, 1977; Reid, 1983). The enzymatic activity of the

aCl molecule is then capable of cleaving C4 and C2 (Patrick

et al., 1970), thereby initiating the classical complement

pathway. Other agents, eg. bacterial products (Clas and

Loos, 1981; Phillips et al., 1972) and various

compounds(Cyong et al., 1982), are also capable of

activating serum Cl. The aCl molecule is regulated in serum

by Cl-Inh (Lepow et al., 1965) which binds to the activated

subunits (aClr and aCls) inhibiting their esterolytic

activity (Harpel and Cooper, 1975), with resultant

dissociation of the aCl molecule (Harpel and Cooper, 1975;

Minta and Aziz, 1981). In order to study the activation

process of Cl, a number of researchers have attempted to

isolate Cl in its zymogen form (Borsos et al., 1964; Lepow

et al., 1965). Typically, these attempts to isolate whole

34







35

serum Cl have yielded exclusively aCl, which led to the

conclusion that upon isolation serum Cl undergoes a

spontaneous activation process (Lepow et al., 1965;

Ziccardi, 1982a). The reason that the activation of

isolated serum Cl does not occur in unfractionated serum is

that 1) the isolation process caused the serum Cl molecules

to become activated, or that 2) the isolation process

separated the serum Cl away from some form of regulation

normally found in serum. In initial experiments, sham

isolation procedures, in which serum was exposed to gel

chromatography, euglobulin fraction, etc., but was left

unfractionated, the Cl failed to spontaneously activate. An

example of this observation was that ultracentrifugation of

serum resulted in a pellet rich in Cl activity, which when

resuspended in the supernatant displayed the properties of

zymogen Cl. However, resuspending the pellet in buffer

yielded Cl activity that had the characteristics of the

fully activated molecule (data not shown). The isolation

process was therefore not the sole cause for the observed

spontaneous activation.

The autoactivation of serum Cl may have an in vivo

corollary in the disease hereditary angioedema (HAE), in

which patients are shown to have low levels of functional

Cl-Inh (Gadek et al., 1980). These individuals experience

episodes of spontaneous activation of their serum Cl

(Gelfand et al., 1979), with serious physiological

consequences. The episodic nature of this disease and the







36

zymogen nature of the Cl found in asymptomatic HAE patients

(Gelfand et al., 1979), raise important questions about the

regulation of serum Cl activity in serum. A naturally

occurring factor (possibly Cl-Inh) in serum that regulates

the activation of native serum Cl would therefore be an

attractive hypothesis.

Other workers (Ziccardi, 1982b) using serum Cl reformed

from isolated Cl subunits have shown that Cl-Inh inhibited

the spontaneous activation of serum Cl to aCl. Those

studies must be viewed with caution because the serum Cl
125
molecules are reformed using modified ( I labeled)

subunits and the reformed serum Cl molecules differed

significantly from serum Cl in a number of physical

characteristics, specifically, sedimentation coeffecients

(16S and 12S for serum Cl and refomed zymogen Cl

respectively)(Colten et al., 1967; Cooper and Ziccardi,

1977), Cl-Inh kinetics (Loos, 1982; Ziccardi and Cooper,

1976), and spontaneous activation kinetics (see Chapter II;

Ziccardi, 1982a). Furthermore, the assays used in those

studies monitored the average physical changes-of

populations of Cl subunits during the course of spontaneous

activation. This may or may not directly reflect the

functional changes associated with the activation of serum

Cl to aCl.

In the previous section it was shown, using the Cl

hemolytic titration assay, that crude preparations of

unmodified serum Cl from serum were able to spontaneously








37

convert to aCl. In the experiments described in this

section, various serum fractions, including purified Cl-Inh,

were shown to inhibit this spontaneous autoactivation. This

activity of Cl-Inh to inhibit serum Cl autoactivation was

distinct from its irreversible covalent interaction with

fully activated Cl.



Materials and Methods

Reagents

The reagents and materials used in these studies were

identical to those reported in the previous chapter.



Cl Assays

Titration of Cl hemolytic activity was performed by the

methods described in the previous chapter.



Isolation and Measurement of Cl-Inh

Human Cl-Inh was purified essentially by the method of

Harrison (1984), using DE-52 in place of DEAE ion-exchange

chromatography. Specifically, outdated plasma was converted

to serum by the addition of excess Ca ++ and the

supernatant of a 5% polyethylene glycol precipitation was

passed over a lysine-sepharose affinity column. The

pass-tnrougn traction was t.nen suojectea to ion-exchange

cnromatograpny on DE-bZ equiilbrated in 0.1 M KCl and 10 mM

phosphate pH 7.0 with 0.01 M benzamidine. Bound proteins

were eluted with a linear gradient of 0.1 to .25 M KC1 in








38
the same phosphate buffer. Functional levels of Cl-Inh in

various fractions were monitored by the method of Gigli,

et.al. (1968). A radioimmunoassay was developed to measure

the antigenic levels of Cl-Inh in the various fractions. In

this assay aliquots of solutions containing unknown amounts

of Cl-Inh were added to Immuno-beads (Bio-Rad) coupled with

partially purified Cl-Inh followed by the addition of

commercially prepared Cl-Inh specific rabbit antiserum

(Pel-Freze). After incubation at 37C for 45 minutes and

extensive washing of the beads in VBS-EDTA, the amount of

rabbit antibodies remaining bound to the beads was

determined by the amount of 125I-Protein A that would

specifically bind to the beads in a 45-minute incubation at

37C. The relative concentration of Cl-Inh in the unknown

was determined by relating the inhibition of the binding of

the rabbit antibodies to the beads caused by the fluid phase

Cl-Inh compared to the standard curve of inhibition

generated by the addition of standard amounts of partially

purified Cl-Inh (see Figure 12). This assay could detect

concentrations of Cl-Inh in solutions of less than 1 pg/ml

(data not shown). Fractions showing Cl-Inh activity were

pooled and applied to a G-150 molecular sieving column.

Fractions containing Cl-Inh activity were pooled and further

purified by hydroxylapatite column chromatography.

Fractions containing Cl-Inh activity were pooled. This

material appeared as a single homogeneous band in SDS-PAGE

(Laemmeli, 1970) (Figure 13).









39









(0
N
p )
^0 lU
4-J 0
t, 0U) Efl
\3 a ro (a
,z 3:
\ ." 2 --i
SI4 , --
\ ~~I \i H 0 't-

0 (d 0 .0
\ a) () V4 C
\-4 U) ( 4 .a H
\ 0 C "*- r-
\ (J2'-4 dP
.4J rO 0(D4




\ ~~~. ii 1i 4JO *0 -
o 0 irl*4- tP
\ ~E-4i C:C-
H >0- l) 0
00 4-,A UC%

< HC14O C-










-, U,- O .>
H H 4 m I )

4J -U E- M 4J-
\ q a) (0
Z (d I 4-1 a) 6
Sm 0 0 4J -4
\(NI 4J _SIO C >li NX
\ O.OM -4 0 0
WLU Hd--mn 0 Q0
W U4'~~
r ~ ~ U.U -- **ai'
0 r-4 >i4-4 4J-
04-0 0*-1
\C4 0 0 3 aU)
\r-4-4 C) Ua) lrto
U)44 0 M
\C.4J U4 W
\ cnjce i c; a) U)
e -4 0 X- 4
r14 -A U 5 4j 0







o 0 0 0
0C (0 It CM
ONIQNIO Vd-lczi AO N011191HNI %









1 2 3 4


I-I


1


~6.


4'


Figure 13. SDS-PAGE OF THE VARIOUS STEPS IN THE
PURIFICATION OF Cl-INH. Pictured is the Coomassie stain of
a SDS-PAGE analysis of aliquots from the various stages of
the purification of Cl-Inh from human serum (lanes 1 and 8 -
molecular weight standards, lane 2 serum, lane 3 PEG
supernate, lane 4 lysine-sepharose flow through fraction,
lane 5 DE-52 eluate pool, lane 6 G150 eluate pool, lane
7 hydroxylapatite eluate pool)


5 6


7 8


Jh












Human Serum Fractionation

Partially purified serum Cl was prepared by

ion-exchange chromatography described in the previous

chapter. Proteins bound to the ion-exchange column were

eluted by washing the column with equlibration buffer

containing 200 mM NaCI. Other sources of Cl depleted human

serum fractions were produced by eugologulin precipitation

supernatantt of serum dialysed against water) and

ultracentrifugation supernatantt of serum spun for 4 hours

at 100,000 x G).



Results

Inhibition of Spontaneous Activation of Zymogen Cl by

Serum Fractions

To address the role of serum factors in maintaining the

zymogen nature of unfractionated serum Cl, Cl depleted serum

fractions were assayed for their ability to inhibit the

spontaneous activation of partially purified serum Cl (DE-52

pass-thorugh). Human serum was fractionated to remove Cl

activity by euglobulin precipitation (eublobulin

supernatant), ultracentrifugation supernatantt of 100,000 x

G), or ion-exchange chromatography (DE-52 eluate). The

various serum fractions were added to partially purified

serum Cl. The solutions were incubated one hour at 37C

and the ability of each fraction to inhibit the conversion

of serum Cl activity (as shown by a Cl titration slope of







42
1.3) to aCl activity (as shown by a Cl titration slope of

1.0) in subsequent Cl hemolytic titration assays was

measured. As can be seen in Figure 14, at the highest

concentrations tested all three Cl depleted serum fractions

(euglobulin supernatant, ultracentrifuged supernatant, and

DE-52 eluate) were able to inhibit the spontaneous

conversion of serum Cl to aCl. Inhibition of the

spontaneous activation was shown to be dependent on the

concentration of the serum fraction used(l:100 dilutions of

the various serum fractions yielded adCl activity).

Spontaneous serum Cl activation was therefore inhibited by

some factors) present in the various serum fractions.



Inhibition of Spontaneous Activation of Zymogen Cl by

Cl-Inh

Human Cl-Inh was purified to a single band in SDS-PAGE

(see Figure 13). Addition of Cl-Inh to partially purified

zymogen Cl (DE-52 pass-through) inhibited the spontaneous

activation of Cl (Figure 15A). This inhibition of the

spontaneous activation of serum Cl by Cl-Inh was

concentration dependent. When the relative amounts of

Cl-Inh in the various dilutions of the various serum

fractions used in Figure 14 were considered, the

concentration of Cl-Inh needed to inhibit by 50% the

activation of the serum Cl was approximately equal to the

Cl-Inh concentration in the crude serum fraction from DE-52

eluate (Figure 14).











































































CenIDA ZJ ( A I ) Ul-


,- 3 (0
rFZ) C
0 C: U rt. 0-
00 a) (0 3 4-.)
z .-14 r- r_4 J__ 11 >1 a 0
0 4J M U4-) E 0 A- t-i 4),4-
H U 3 0 r-4 0 E -_4

I-i (a .1.1 u-i M I .i nj .
H 0 4- 4 -4--
0 mt 0 0 C4- *-i C p
H 4J dp | 1 -i .I a
E-4 ez :Jo C CD C: W (0 4
0 0 6 n -
S-- 0 C 3 3r-00
Z) C Sd -4J 4-) (d 0
0 CJ) 0 ) 4 44 w .-i
0 C 044 0 4J ..
Z 0 O E- )i 0 I-- C'
< -4U 0 ., U) U4 -)
E-4 W 4J a)- )C-UM U ro0
Z M (d *r-4 (04J U (d CO
0 > a >4 4J (0 MB $4 r-4 (d
0 '-4 cO r W ,0 0-.IJU -
CI ^Ur-4 M Cr-4 r-4
M- CT (1 3: 4 Q) rO 0 r-4
4. 4J -4) Q ) a) 0 00
w C .-I ) 04 r4 o 4) w w
cr. E-4 Q) C: 4J C! :JU nj 4 ) -4u
5 ( >i 0 U) (d l4 ) c
Z Z-^4 -q-H z 00
0 O 4J 0 4J C r-4 44 r -4 U
U- (u E u -4u Z U "0 -,-
0 Co 0' no () -4r-4 -Ir-
z ZXCCr-C)^3 e+J5-4 4
^: 0 a) (d 44lX 0 -': C -44J 4
o H rd 0 i4 (d W (d W
0 E-4 ) r- ) U r-4 I Q) 4) 3 cl
U 4j U >4 3 t7> WO M 04 rU
- c( $4( C QU) t
o 5 C6O a) (D) ro w >i>U)
F4 CC0 tO U) -4 ) (U) Ur-4 o 4J
04 d 1 0) -If0-4 U4 W
W a)U U) En C44-i (d () C
ZD U) wn r. C < rd; -q^ co -q CC 0
a- )0 04 O 4J --
c H r C r -4 ^ :3 1 WI QU) 4J
0) 4nM C J W (L) 04 (al -l
< .-1 (0Cr r_4 U 04U)'t
-j U') 4-4(1) > (ta U >,-
llf M -4t (d 04 r-4- 0 44
0-4 1 3-4 4) 4J W) C
j C>i 4-J -4 (O0 4C0
4 J s4- Qj 0) -, 0)
>4 > -0 WU M 'r-- 4 3U4J
rta W-4 -4 0 04C CO M





0 W tZ U4J rj4Ji a).. 0 d U $t -C
U -) U) X ---4 () S W- 4U '
E-4 0 4J1 Mf 4J -4 '+44 XC -4 *
UM "i 3o e$4- ul- a) g
M 44) 04 U(U -3r)UW r





-H~ a)x4 04 -q it 04 cr 0 0)
,-I rO M~-- 3J 4 3 4 C 0 1
D4 HU w > >i' al I" V >
Ct i **l O i- a) 4.J 4J 0) ij^q )

-4) a) n3 U () 0<- M'- :1i -4 C-i (i

&






44









A B
I I I I1


1.0 0-
0.80
0.60
S0.40





0.10
S0.08 -
0.06
0.04


I 2 4 8 1 2 4 8
RELATIVE C I CONCENTRATION
Figure 15. EFFECTS OF PURIFIED Cl-INH ON THE SPONTANEOUS
ACTIVATION OF PARTIALLY PURIFIED SERUM Cl AND ON ACTIVATED
PARTIALLY PURIFIED SERUM Cl. Purified Cl-Inh (at various
dilutions) or buffer was added to partially purified serum
Cl which had been preincubated for 20 minutes at either 4C
(panel A) or 37C (panel B). Following a 20 minute
incubation at 37C the resultant Cl activity was monitored
in the hemolytic assay (dilutions of Cl-Inh used: 1:5
( o ), 1:25 ( ), 1:125 ( 0 ), 1:625 ( A ), buffer ( &,)).)
For these studies, the undiluted preparation of Cl-I'nh
contained an equivalent concentration of inhibitor normally
present in human serum (0.1 mg/ml)








45

The action of Cl-Inh in preventing the conversion of

partially purified serum Cl activity to aCl activity in the

preincubation step was not due to its action on aCl. As

mentioned above, addition of Cl-Inh to partially purified

serum Cl solutions, followed by incubation for 20 minutes at

37C, yielded Cl activity similar to that found in serum

(Figure 15A). However, in parallel studies if the serum Cl

was preincubated in the absence of Cl-Inh for 20 minutes at

37C, then further incubation in the presence of Cl-Inh did

not result in the appearance of serum Cl activity (Figure

15B). Rather, under these conditions, Cl-Inh reduced the

aCl activity in a concentration dependent fashion. Similar

results were seen when Cl-Inh was added to purified aCl

(data not shown). The only difference in the experiment

that generated the results of Figure 15A and 15B was that

the serum Cl of 15A was preincubated on ice for 20 minutes

while the serum Cl of 15B was preincubated for 20 minutes at

37 prior to the addition of the Cl-Inh. These results

showed that the action of Cl-Inh in inhibiting the

conversion of partially purified serum Cl to aCl activity

was not due to the interaction of Cl-Inh with aCl alone or

an artefact of the assay system. Furthermore, the amount of

Cl-Inh needed to inhibit the activation of the partially

purified serum Cl was less than that needed to inhibit the

Cl activity after the spontaneous activation (see Figure 15

A and B).







46

Discussion

These studies extend the observations made in the

previous chapter in which crude isolates of Cl from serum,

upon incubation at 37C, spontaneously converted to

activated Cl, as measured functionally in the hemolytic

assay. This spontaneous autoactivation was not observed

with unfractionated serum Cl, i.e. human serum. There are

two potential explanations for why spontaneous Cl activation

does not occur in serum: 1) the isolation process somehow

altered the serum Cl molecule thus allowing it to self

activate, or 2) in the process of isolation, some regulatory

molecule was separated from the serum Cl, thus allowing the

serum Cl molecule to self-activate, a process normally

regulated in serum. The studies reported here support the

second hypothesis that there are regulatory factors) found

in serum that inhibit the spontaneous activation of Cl.

Partially purified serum Cl that self-activated when

incubated in buffer failed to activate when admixed with Cl

depleted crude serum fractions.

Due to the relationship of functional Cl-Inh levels and

Cl activation seen in HAE patients and the results from

researchers using reassociated Cl molecules, Cl-Inh was

examined as the most likely regulator of serum Cl

self-activation. It was found that when Cl-Inh was

incubated with partially purified serum Cl, activation could

be prevented. This regulatory activity of Cl-Inh was







47
independent of its irreversible action on aCl. The action

of Cl-Inh is represented schematically in Figure 16.

The regulation of the conversion of zymogen Cl to aCl

by Cl-Inh was concentration dependent. The concentration of

Cl-Inh needed to maintain serum Cl from spontaneous

activation was less that that required to inactivate the

total quantity of activated Cl that could be generated in

the system. That is, if activation was allowed to occur

prior to adding Cl-Inh, then the minimal level previously

shown to inhibit spontaneous activation was insufficient to

inactivate all the aCl. This leads to the hypothesis that

Cl-Inh (Figure 16, Step I) is more effective in regulating

the generation of Cl activity in serum than it is as an

inactivator of fully activated Cl (Figure 16, Step II).

These studies have led to similar conclusions to those of

Ziccardi (1982a).

The ability of a low level of Cl-Inh to maintain serum

Cl in its zymogen form may account for the episodic nature

of the edema in HAE patients. Those patients usually have

10%-15% of the normal level of Cl-Inh. From these studies it

would be predicted that these patients have levels of Cl-Inh

sufficient to prevent continual spontaneous activation of

their serum Cl. However, if an enzyme regulated by Cl-Inh

(Figure 1) was activated in these patients there would be

insufficient Cl-Inh to effectively regulate such an enzyme.





















CI-Inh

(I)

zymogen Cl ---> activated Cl


reversible


Cl -Inh


(II)


inactivated Cl


irreversible


Figure 16. SCHEMATIC OF THE REGULATORY EFFECTS OF Cl-INH ON
Cl ACTIVITY















CHAPTER IV
INACTIVATION OF THE PROTEOLYTIC ACTIVITY OF NGF BY Cl-INH


Introduction

In 1959 Levy and Lepow demonstrated an inhibitor of aCl

in human serum, which they named the Cl esterase inhibitor

(Cl-Inh). The Cl-Inh molecule has since been shown to

interact and inhibit a number of other serum proteases

involved in homeostatic cascade reactions (Travis and

Salvesen, 1983). Other serum serine esterases, e.g.

trypsin, chymotrypsin, etc., are unaffected by Cl-Inh

(Harpel and Cooper, 1975; Travis and Salvesen, 1983). The

nature of the interaction and inhibition involves the

formation of a covalent Cl-Inh-enzyme complex (Travis and

Salvesen, 1983) near the active site on the enzyme (Travis

and Salvesen, 1983). The rate of inactivation of aCl by

Cl-Inh (Sim et al., 1979; Travis and Salvesen, 1983)

compared to the enzymatic activity of aCl in cleaving C4 and

C2 (Patrick et al., 1970) has led to speculation that the

primary function of Cl-Inh is not in regulating aCl activity

(Gigli et al., 1970; Travis and Salvesen, 1983). Other

serum proteases, eg. kallikrein, are more effectively

regulated by Cl-Inh (Gigli et al., 1970). Evidence reported

elsewhere (Ziccardi, 1982b, see Chapter III) suggests that

Cl-Inh has the ability to regulate zymogen Cl activation

49







50

independent of its esterase inhibiting ability. Clearly,

Cl-Inh is a multi-functional molecule and a full

understanding of its regulatory capabilities requires

further investigation.

Recently, nerve growth factor (NGF) was shown to

substitute for aCl in initiating the classical complement

pathway (Boyle and Young, 1982). This ability to cleave and

activate C4 and C2 was dependent on the serine esterase

activity of the gamma subunit of NGF (gamma NGF) (Boyle and

Young, 1982). This activity was inhibited by preincubation

of activated gamma NGF with Cl-Inh (Boyle and Young, 1982).

Nerve growth factor is naturally found as a

multimolecular complex of three polypeptide subunits (alpha,

beta, and gamma chains, in a 2:1:1 molar ratio) (Bradshaw

and Young, 1976; Greene et al., 1969). It was first

described by Levi-Montalcini and Booker (1960) as a soluble

factor that stimulated neurite outgrowth. This nerve growth

inducing activity is mediated by the beta subunit of the

molecule (Levi-Montalcini, 1965). No known biological

activity has been reported for the alpha subunit (Isackson

and Bradshaw, 1984). The gamma subunit in addition to its

role in complement activation has been shown to act as a

plasminogen activator (Orenstein et al., 1978) and topical

administration of enzymatically active NGF has been shown to

enhance the rate of wound healing in mice (Li et al., 1980).

Other researchers have reported that the gamma subunit of

NGF stimulated polymorphonuclear phagocyte chemotaxis in







51

vitro (Gee et al., 1983) and in vivo (Boyle et al.,

1985). From these studies it would appear that gamma NGF

may have important physiological roles in the inflammation

response and wound repair.

A series of cell secreted protease inhibitors (nexins)

have recently been described that bind and inactivate

specific serine esterases (Baker et al., 1980; Knauer et

al., 1983). In binding to the enzyme, the nexin inactivates

the enzyme and leads to efficient clearance via ingestion by

cells expressing receptors for the enzyme-nexin complexes

(Knauer et al., 1983). Two proteins that function as nexins

in completing with activated gamma NGF have been reported

and partially characterized (Knauer et al., 1982).

Experiments presented in this chapter examine the

interaction of Cl-Inh with gamma NGF. It was found that

purified Cl-Inh was able to inhibit the esterolytic activity

and form covalent complexes with gamma NGF in a time

dependent fashion, with a 1:1 stoichiometry. These results

are similar to those reported for the interactions between

Cl-Inh and other enzymes.



Materials and Methods

Reagents

Cl-Inh was purified to homogeneity by the method of

Harrison (1984) as reported in Chapter III. Purified gamma

NGF from mouse submaxillary salivary glands was provided by

Dr. M. Young.









Enzymatic Assay

Various amounts of purified Cl-Inh were incubated with

2.6 pg of gamma NGF and the enzymatic activity of the gamnma

NGF was measured by monitoring the cleavage of the synthetic

substrate, N-alpha-benzoyl-1-arginine-p-nitroanilide

(1-BAPNA). The gamma NGF containing solution was added to a

1 mM solution of 1-BAPNA in 10 mM Tris(hydroxymethyl)

aminomethane (Tris) buffer, pH 7.4, and the cleavage of the

substrate was determined by an increase in the OD410 of

the solution, using a Varian spectrophotometer, with the

cuvette jacketed at 24C.



Physiochemical Assay
125
Gamma NGF was labeled with 125I by the

lactoperoxidase method. Radiolabeled gamma NGF and Cl-Inh

were incubated together under various conditions and the

solutions were subjected to SDS-PAGE (Weber and Osborn,

1969). The slab gels were dried and the radiolabeled gamma

NGF was visualized by autoradiography. The relative amounts

of radioactivity migrating at different rates in the gels

were measured by scanning the autoradiographs on a Joyce

Lobel densitometer. Molecular weights were determined from

Coomasse R-250 stained reference wells containing molecular

weight standards (Sigma Chemicals).







53
Results

Inhibition of Gamma NGF Enzymatic Activity by Cl-Inh

Preincubation of active gamma NGF with a ninefold molar

excess Cl-Inh for one hour at 37C effectively inhibited

the ability of the gamma NGF to subsequently cleave the

synthetic substrate 1-BAPNA (see Figure 17). The inhibition

of gamma NGF by Cl-Inh was complete within one hour of

incubation at 37C. The stoichiometry of the gamma

NGF-Cl-Inh interaction was consequently determined following

a one hour preincubation at 37C. The results of adding

various amounts of Cl-Inh to a constant amount gamma NGF,

plotted in Figure 18, showed an approximately 1:1

stoichiometry for the inhibition of gamma NGF by Cl-Inh. In

the next group of experiments the rate of inactivation of

gamma NGF by Cl-Inh was examined. In preliminary

experiments it was shown that addition of 1 mM 1-BAPNA to

gamma NGF prior to addition of Cl-Inh to the mixture

prevented the interaction of Cl-Inh with gamma NGF (data not

shown). This blockage of Cl-Inh activity was due most

probably to competitive inhibition for the active site of

the gamma NGF. Therefore, the time course experiment was

not affected by the interaction of Cl-Inh and gamma NGF

during the enzyme assay. Using a 1:2 molar ratio of Cl-Inh

to gamma NGF, the rate of inhibition of the gamma NGF enzyme

activity was measured at 24C (see Figure 19). These

results are similar to those reported for the interaction of

Cl-Inh with aCls (Harpel and Cooper, 1975), and suggest the




























.03-



.02




0 .01 ,


0






4 5
TIME


Figure 17. ENZYMATIC ACTIVITY OF NGF AFTER PREINCUBATION
WITH Cl-INH. Cleavage of the synthetic substrate 1-BAPNA,
as measured by the generation of OD4. was monitored
for the enzymatic activity of gamma NF after incubation for
one hour at 37C in buffer (top tracing) or after the
addition of 9 molar excess of purified Cl-Inh (lower
tracing).























100'




80




60-


0 ^
<40-


0 N
z \
N


\
\
\
.2 .4 .6 .8 1.0 1.2
CCI-INH]
L NGFJ


Figure 18. CONCENTRATION DEPENDENCE OF THE INHIBITION OF
THE ENZYMATIC ACTIVITY OF NGF BY Cl-INH. Various molar
equivalents of Cl-Inh were added to a constant amount of
gamma NGF, incubated one hour.at 37C, and then assayed for
...- remaining enzymatic activity in Cleaving 1-BAPNA. The graph
depicts the amount of gamma NGF activity remaining after
incubation with various molar amounts of Cl-Inh .




























I-


CD
40

Z
LL
z
3O
030- 1b
Z
0


20-




10
z i









.5 1 2 4 8 16

TIME AT 24 C

Figure 19. TIME COURSE OF THE INHIBITION OF THE ENZYMATIC
ACTIVITY OF NGF BY Cl-INH. Gamma NGF was added to 1/2 molar
equivalent of Cl-Inh and incubated at 24C for various
times. The enzymatic activity of the gamma NGF was then
analyzed by monitoring the cleavage of 1-BAPNA
spectrophotometrically. The graph depicts the amount of
gamma NGF activity remaining with time (minutes).







57

interaction of gamma NGF and Cl-Inh requires an enzymatic

reaction.



Complex Formation between Gamma NGF and Cl-Inh

The apparent enzymatic nature of the interaction of

gamma NGF and Cl-Inh and the 1:1 stoichiometry described

above led to the hypothesis that the enzymatic activity of

gamma NGF was inhibited by the formation of a stable complex

with Cl-Inh. Experiments were designed to monitor the

formation of a covalent complex formed between gamma NGF

(26,000 molecular weight) and Cl-Inh (97,000 molecular

weight) using SDS-PAGE analysis. In these experiments,
1251-gamma NGF was incubated with Cl-Inh for one hour at

37C and the mixture was electrophoresed in SDS-PAGE using

the Weber and Osborn buffer system (phosphate buffer pH

7.2). Incubation of 125I-gamma NGF with an 80-fold molar

excess Cl-Inh for one hour at 37C yielded greater than 97%

of the 125I migrating in a high molecular weight complex

of approximately 125,000 Daltons (Figures 20 and 21). In

order to investigate the nature of the bond formed in the

high molecular weight complex, the effects of hydroxylamine

and high pH on the stability of the complex were examined.

Preformed high molecular complexes were formed by the
125
incubation of 125I-gamma NGF with 1/2 molar equilivant of

Cl-Inh. An aliquot was removed and further incubated for 20

minutes at 37C in .1 M hydroxylamine pH 10.5 and

electrophoresed. The results (see Figure 22) show that the


















I 2 3


116,000-
97,000-



66,000-


45,000-

29,000-


yFure 20. AUTORADIOGRAPH OF SOS-PAGE ANALYSIS OF
I-GAMMA NGF INTERACTION WITH Cl-INH. Radiolabeled
gamma NGF was incubted with various amounts of Cl-Inh (lane
1 none, lane 2 1/2 molar equivalent of cl-Inh, lane 3 -
80 molar excess Cl-Inh) and incubated one hour at 370C,
added to 2% SOS, and electrophoresed.



















ci I I I
Ci

Uj

-- I --
-J













Rf


125-
Figure 21. DENSITOMETRY OF SDS-PAGE OF -I-NGF INCUBATED
WITH VARIOUS CONCENTRATIONS OF Cl-INH. The autoradiograph
from Figure 20 was scanned with a densitometer. The
relative areas under the curves had been shown in previous
experiments to obey Beer's law, and thus were analysed to
estimate the relative amounts of radioactivity in each peak
(panel A No Cl-Inh added, panel B- 1/2 molar equivalent of
Cl-Inh added, and panel C 80 molar excess Cl-Inh added).
Eectrophoresis proceeded from right to left.



















I 2



1169000"6
97,000-




66,000 -



45,000 -

fe -,. 9.000-L l
29n,O0-w S l

]s .





Figure 22. AUTORADIOGRAPH OF SDS-PAGE ANALYSIS OF
HYDROXYLAMINE TREATED HIGH MOLECULAR WEIGHT COMPLEXES.
Radiolabeled gamma NGF was incubated with 1/2 molar
equivilant of Cl-Inh and incubated one hour at 37C (lane
1). An aliquot was removed and was made .1 M hydroxylamine
(pH 10.5) and incubated an additional 20 minutes at 37C
(lane 2). The solutions were then subjected to SDS-PAGE
separation and autoradiographed.







61
high molecular weight complex was labile to treatment with

hydroxylamine at pH 10.5, which is characteristic of an acyl
125
linkage between the Cl-Inh and 125I-NGF in the high

molecular weight complex.



Discussion

The proteolytic activity of gamma NGF, as measured by

the cleavage of the synthetic substrate 1-BAPNA, has been

shown to be inhibited by Cl-Inh. This inhibition was time

dependent with a stoichiometry of approximately 1:1. The

interaction between gamma NGF and Cl-Inh was shown to result

in the formation of a covalent bonded complex, as judged by

SDS-PAGE. These results would be consistent with a model of

interaction in which the inhibition of the enzymatic

activity of gamma NGF was inhibited by the association of

Cl-Inh near the active site of the molecule, followed by the

cleavage of Cl-Inh and covalent coupling to gamma NGF.

There are a number of predictions associated with such a

model. First, the cleavage of Cl-Inh by gamma NGF should

generate a new amino terminal in the Cl-Inh. Second, a new

Cl-Inh fragment should be produced by the completing of

gamma NGF and Cl-Inh. Finally, the formation of the

covalent bond should be localized to a specific polypeptide

chain(s) in the gamma NGF.

The bond formed in the interaction of gamma NGF and

Cl-Inh was shown to be susceptible to the action of

hydroxylamine at high pH. These results are similar to







62

those seen with the acyl linkage between gamma NGF and the

protease nexins (Baker et al., 1980) or the characteristic

serpin interaction with their appropriate serine esterase.

These properties suggest that Cl-Inh may have many of the

characteristics of a protease nexin. Further studies are

required to determine if any cell type has specific

receptors for Cl-Inh-gamma NGF complexes that would indicate

that Cl-Inh closely mimicked the biological activity of a

cell-secreted nexin.















CHAPTER V
SUMMARY AND CONCLUSIONS


This investigation has examined protein interactions

important in the activation, regulation, and inactivation of

serum proteases. In addressing these topics, three basic

questions were posed and answered. First, what is the

nature of the spontaneous activation reaction of partially

purified serum Cl? Second, what is the role of Cl-Inh in

regulating the activation of serum Cl? Finally, wnat is tne

nature or tne interaction or c-inn witn tne activated gamma

subunit or NGF?

The process or serum Ci activation was shown to require

intermolecular interactions. This process was not

autocatalytic in that the addition of enzymatically active

Cl did not alter the extent of serum Cl activation. The

studies presented here confirm the work of others (Borsos et

al., 1980; Borsos et al., 1964; Heinz et al., 1984; Loos et

al., 1972) and clearly show that the level of activatable

zymogen Cl is not simply limited by the association of the

Clq subunits with Clr2s2 complexes, as has been

proposed by Ziccardi (1982a). The concentration dependence

of the Cl activation reaction has not been adequately

explained. Borsos et al. (Borsos et al., 1980; Loos et al.,

1972) suggest that heterogeneity in the Clq subunits could

63







64

account for the observed behavior. The specific subunit

interactions needed for activation of macromolecular serum

Cl were not established in this study. This is an important

question that merits further investigation.

In this investigation, the activity of Cl-Inh was shown

to be broader than previously appreciated. The serine

esterase inhibitor activity of Cl-Inh has been well

documented (Harpel and Cooper, 1975). However, the rate of

inactivation of aCl by Cl-Inh is slow (Sim and Porter, 1976)

compared to the activity of aCl on C4 and C2 (Patrick et

al., 1970), and it is thought that the true physiological

role of the serpin activity of Cl-Inh is in inactivating the

enzymatic activity of kallikrein (Gigli et al., 1970). The

results reported in this study present a new role of Cl-Inh

in regulating serum Cl activity which entails the regulation

of the activation of zymogen Cl. This regulation of serum

Cl activation was distinct from the activity of Cl-Inh in

inactivating aCl. In regulating both the activation and

resultant activity of serum Cl, Cl-Inh is able to

effectively control the initiation of the classical

complement pathway. Furthermore, it has been seen by others

(Harpel and Cooper, 1975; Minta and Aziz, 1981) that the

interaction of aCl with Cl-Inh not only inactivates the aClr

and adCls subunits, but also dissociates the completed

subunits from the Clq subunit. Though enhanced clearance of

complexes of Cl-Inh with aClr or aCls has not been observed,

the dissociation of the aCl macromolecule would enhance the







65

binding of Clq to Clq receptors on phagocytes. Thus, Cl-Inh

may be considered a nexin of sorts, in that by binding to

the adCl macromolecule and dissociating it, the Clq is more

readily cleared by binding to specific cell receptors.

The interaction of Cl-Inh with gamma NGF led to the

inactivation of the enzymatic activity of the gamma NGF.

Associated with the inactivation was the formation of a

gamma NGF-Cl-Inh complex that was resistant to SDS and

susceptible to hydroxylamine treatment. This interaction is

similar to that seen with Cl-Inh and aCls (Harpel and

Cooper, 1975). In binding the aCls, Cl-Inh becomes cleaved

and new carboxy and amino terminals are formed (Salvesen et

al., 1985). By sequencing these new fragments a partial

amino acid sequence of the enzymatic target site on the

Cl-Inh has been reported (Carrell, 1984; Salvesen et al.,

1985). Other than an arginine residue where the serine

esterase cleaves, this target site sequence is unrelated to

any other enzymatic target site reported for other serpins

(Carrell, 1984). The similarities between the target sites

of the various enzymes inhibited by Cl-Inh could provide

information regarding the specificities of both the

proteases and the Cl-Inh inactivation activity.

Additionally, knowledge of the fine structure of the

enzymatic target site of some serpins has led to new

research in producing more effective protease inhibitors by

site specific mutagenesis and genetic recombination

(Rosenberg et al., 1984). The production of a molecule that








66

could more effectively control the activation of the

classical complement pathway would have great benefits in

the treatment of HAE and other disorders resulting in

inappropriate Cl activation.

The interaction of Cl-Inh with the gamma subunit of

nerve growth factor is characteristic of that of most serine

protease inhibitors (Travis and Salvesen, 1983). There is a

1:1 stoichiometric complex formed between Cl-Inh and NGF and

this complex is stabilized by the formation of a covalent

bond between the two proteins. Recently, a new group of

cell secreted protease inhibitors known as protease nexins

(PN) have been described (Baker et al., 1980). Three

distinct protease nexins (PNl, PN2, and PN3) have been

identified (Knauer et al., 1983) and their physiological

function appears to be inhibition of the enzyme activity of

selected proteases by complex formation followed by

enhancement of the specific clearance of PN-protease

complexes by a variety of cells (Knauer et al., 1983). One

of these nexins, PN3, reacts in a highly selective manner

with the gamma subunit of NGF (Knauer et al., 1982). The

Cl-Inh-NGF and the PN3-NGF complexes share a number of

similarities and future studies should be directed towards

determining whether Cl-inactivator-NGF complexes have other

biological properties of PN3-NGF complexes.

In the comparison of the properties of nerve growth

factor and Cl a number of similarities have emerged. First,

both can activate the classical complement pathway (Boyle







67

and Young, 1982; Lepow et al., 1965) and their enzyme

activity can be regulated by the Cl inhibitor protein (Boyle

and Young, 1982; Lepow et al., 1965). Second, both are

produced as zymogens (Lepow et al., 1965; Young and Koroly,

1980). Third, both zymogen NGF and Cl undergo

self-activation reactions (Lepow et al., 1965; Young and

Koroly, 1980; Ziccardi, 1982a). In the case of NGF this is

a classical autocatalytic reaction in which the removal of

the Zn++ associated with the zymogen (Young and Koroly,

1980) allows the NGF to spontaneously activate in a biphasic

reaction (Young and Koroly, 1980). Initially, there is a

low level of intramolecular spontaneous activation (Young

and Koroiy, iuu), roiowea oy a rapid ciassica

autocataiytic reaction, in wnicn zymogen IS NGF serves as a

substrate ot activated 7S NGF to produce more activated NGF

(Young and Koroly, 1980). The autocatalysis ot the molecule

can be inhibited by inhibiting the esterolytic activity of

the gamma subunit (Young and Koroly, 1980). One would

predict that addition of Cl-Inh to zymogen 7S NGF would

inhibit the autocatalytic activation (enzymatic activation).

However, because of the results reported for Cl in Chapter

III, the possibility would also exist that Cl-Inh may

inhibit the first phase of the activation (spontaneous

activation) of 7S NGF. The regulation of activation of

other zymogen serum proteases by Cl-Inh warrants further

investigation.







68

The many parallels and similarities between the

structure, function, and regulation of NGF and Cl suggest

that nerve growth factor may play an important physiological

role totally distinct from its trophic effect on the sensory

and sympathetic nervous system. The possibility that NGF

may be part of a heretofore unrecognized system involved in

homeostatic reactions merits closer scrutiny. To date

Cl-Inh has only been recognized to regulate enzymes within

cascade systems, see Figure 1, and this raises the

possibility that NGF is part of some unknown cascade system.

A number of recent observations indicating that NGF is

chemotactic for polymorphonuclear leukocytes in vivo

(Boyle and Young, 1982) and in vitro (Gee et al., 1983)

coupled with the reported ability of the growth factor to

accelerate wound healing (Li et al., 1980) suggest NGF may

be a potential mediator of a variety of normal responses

subsequent to tissue damage.















BIBLIOGRAPHY


Arlaud, G.J., C.L. Villiers, S. Chesne, and M.G. Colomb.
1980. Purified proenzyme Cir: Some characteristics of its
activation and subsequent proteolytic cleavage. Biochem.
et Biophys. Acta 616:116-129.

Baker, J.B., D.A. Low, R.L. Simmer, and D.D. Cunningham.
1980. Protease-nexin: A cellular component that links
thrombin and plasminogen activator and mediates their
binding to cells. Cell 21:37-45.

Borsos, T., M. Loos, R.M. Chapuis, R. Medicus, and J.
Isliker. 1980. A novel way of relating the structure of
Clq to the hemolytic activity of the first component of
complement. Molec. Immunol. 17:1415-1421.

Borsos, T., and M.M. Mayer. 1962. Mechanism of action of
guinea pig complement. In Mechanism of Cell and
Tissue Damage Produced by Immune Reactions IInd
International Symposium on Immunopathology. Benno
Schwabe Co., Basel (Switzerland). pp. 13-22.

Borsos, T., and H.J. Rapp. 1963. Chromatographic separation
of the first component of complement and its assay in a
molecular basis. J. Immunol. 91:851-858.

Borsos, T., and H.J. Rapp. 1967. Immune hemolysis: A
simplified method for the preparation of EAC'4 with
guinea pig or human complement. J. Immunol. 99:263-268.

Borsos, T., H.J. Rapp, and U.L. Walz. 1964. Action of the
first component of complement: Activation of C'l to C'la
in the hemolytic system. J. Immunol. 92:108-112.

Boyle, M.D.P., M.J.P. Lawman, A.P. Gee, and M. Young. 1985.
Nerve growth factor: A chemotactic factor for
polymorphonuclear leukocytes in vivo. J. Immunol.
134:564-568.

Boyle, M.D.P., and M. Young. 1982. Nerve growth factor:
Activation of the Classical complement pathway by
specific substitution for component Cl. Proc. Natl. Acad.
Sci. 79:2519-2522.









Bradshaw, R.A., and M. Young. 1976. Nerve growth factor:
Recent developments and perspectives. Biochem. Pharmacol.
25:1445-1449.

Carrell, R.W.. 1984. Therapy by instant evolution. Nature
312:14.

Clas, F., and M. Loos. 1981. Antibody-independent binding of
the first component of complement (Cl) and its
subcomponent Clq to the S and R forms of Salmonella
minnesota. Infect, and Immunity 31:1138-1144.

Colten, H.R., T. Borsos, and H.J. Rapp. 1967. Efficiency of
the first component of complement (C'l) in the hemolytic
reaction. Science 158:1590-1592.

Cooper, N.R. 1983. Activation and regulation of the first
complement component. Fed. Proc. 42:134-138.

Cooper, N.R., and R.J. Ziccardi. 1977. Reconstitution of Cl
in native, proenzyme form and its use in a quantitative
Cl activation test. J. Immunol. 119:1664-1667.

Cyong, C.-C., S.S. Witkin, B. Rieger, L. Barbarese, R.A.
Good, and N.K. Day. 1982. Antibody-independent complement
activation by myelin via the classical complement
pathway. J. Exp. Med. 155:587-598.

Donaldson, V.H. 1970. Blood coagulation and related plasma
enzymes in inflammation. Ser. Haemat. 3:39-95.

Donaldson, V.H. and R.A. Harrison. 1982. Complexes between
Cl-inhibitor, kallikrein, high molecular weight
kininogen, plasma thromboplastin antecedent, and plasmin
in normal plasma and hereditary angioneurotic edema
plasmas containing dysmorphic Cl-inhibitors: Role of cold
activation. Blood 60:121-129.

Gadek, J.E., S.W. Hosea, J.A. Gelfand, M. Santaella, M.
Wickerhauser, D.C. Triantapyhllopoulos, and M.M. Frank.
1980. Replacement therapy in heredity angioedema:
Successful treatment of acute episodes of angioedema with
partially purified Cl inhibitor. New England J. Med.
302:542-546.

Gee, A.P., M.D.P. Boyle, K. Munger, M.J.P Lawman, and M.
Young. 1983. Nerve growth factor: Stimulation of
polymorphonuclear leukocytes in vitro. Proc. Natl.
Acad. Sci. USA 80:7215-7218.

Gelfand, J.A., G.R. Boss, C.L. Conley, R. Reinhart, and M.M.
Frank. 1979. Aquired Cl esterase inhibitor deficiency and
angioedema: A review. Medicine 58:321-328.









Ghebrehiwet, B., B.P. Randazzo, J.T. Dunn, M. Silverberg,
and A.P. Kaplan. 1983. Mechanisms of activation of the
classical pathway of complement by Hageman factor
fragment. J. Clin. Invest. 71:1450-1456.

Gigli, I., J.W. Mason, R.W. Coleman, and K.F. Austen. 1970.
Interaction of plasma kallikrein with the Cl inhibitor.
J. Immonol. 104:574-581.

Gigli,I., S. Ruddy, and K.L. Austen. 1968. The
stoichiometric measurement of the serum inhibitor of the
first component of complement by the inhibition of immune
hemolysis. J. Immunol. 100:1154-1164.

Greene, L.A., E.M. Shooter, and S. Varon. 1969. Subunit
interaction and enzymatic activity of mouse 7S nerve
growth factor. Biochemistry 8:3735-3741. ,

Harpel, P.C., and N.R. Cooper. 1975. Studies on human plasma
Cl inactivator-enzyme interactions. I. Mechanisms of
interaction with aCls, plasmin, and trypsin. J. Clin.
Invest. 55:593-603.

Harrison, R.A. 1984. Human Cl inhibitor: Improved isolation
and preliminary structural characterization. Biochemistry
22:5001-5007.

Heinz, H.-P., R. Burger, M.D. Golan, and M. Loos. 1984.
Activation of the first component of complement, Cl, by a
monoclonal antibody recognizing the C chain of Clq. J.
Immunol. 132:804-808.

Hoffman, L.G., and N.P. Stevland. 1971. Difference in dose
response between native and activated first component of
guinea pig complement (Cl): Reaction with aCl inactivator
offers no explanation. Immunochemistry 8:499-508.

Isackson, P.J., and R.A. Bradshaw. 1984. The alpha- and
beta-subunit of mouse 7S nerve growth factor is an
inactive serine protease. J. Biol. Chem. 259:5380-5383.

Knauer,D.J., K.M. Scaparro, and D.D. Cunningham. 1982. The
gamma subunit of 7S nerve growth factor binds to cells
via complexes formed with two cell-secreted nexins. J.
Biol. Chem. 257:15098-15104.

Knauer, D.J., J.A. Thompson, and D.D. Cunningham. 1983.
Protease nexins: Cell-secreted proteins that mediate the
binding, internalization, and degradation of regulatory
serine proteases. J. Cell Physiol. 117:385-396.

Laemmli, U.K. 1970. Cleavage of structural proteins during
the assembly of the head of bacteriophage T4. Nature
227:680-685.









Lepow, I.H., G.B. Naff, and J. Pensky. 1965. Mechanisms of
activation of C'l and inhibition of C'l esterase. In
CIBA Foundation Symposium: Complement, ed.
Wolstenholme, G.E.W., and Julie Knight. Little Brown and
Co., Boston, MA. pp. 74-98.

Levi-Montalcini,R. 1965. The nerve growth factor: Its mode
of action on sensory and sympathetic nerve cells. Harvey
Lect. 60:217-259.

Levi-Montalcini, R., and B. Booker. 1960. Excessive growth
of sympathetic ganglia evoked by a protein isolated from
mouse salivary gland. Proc. Natl. Acad. Sci. USA
46:373-384.

Levy, L.R. and I.H. Lepow. 1959. Assay and properties of
serum inhibitor of C'l esterase. Proc. Soc. Exp. Biol.
Med. 101:608-611.

Li, A.K.C., M.J. Koroly, M.E. Schattenkerk, R.A. Malt, and
M. Young. 1980. Nerve growth factor: Acceleration of the
rate of wound healing in mice. Proc. Natl. Acad. Sci. USA
77:4379-4381.

Loos, M. 1982. The classical complement pathway: Mechanism
of activation of the first component by antigen-antibody
complexes. Prog. in Allergy 30:135-192.

Loos, M., T. Borsos, and H.J. Rapp. 1972. Activation of the
first component of complement: Evidence for an internal
activation step. J. Immunol. 108:683-688.

Loos, M.,. and H.-U. Hill. 1977. Different behaviour of
native and purified Clq, a subcomponent of the first
component of complement (Cl). Monogr. in Allergy
12:46-51.

McRae, B.J., T.-Y. Lin, and J.C. Powers. 1981. Mapping the
substrate binding site of human aClr and aCls with
peptide thioesters. J. Biol. Chem. 256:12362-12366.

Minta, J.O., and E. Aziz. 1981. Analysis of the reactive
site peptide bond in aCl-inhibitor by chemical
modification of tyrosyl, lysyl, and arginyl residues: The
essential role of lysyl residues in the functional
activity of aCl-Inh. J. Immunol. 126:250-255.

Orenstein, N.S., H.F. Dvorak, M.H. Blanchard, and M. Young.
1978. Nerve growth factor: A protease that can activate
plasminogen. Proc. Natl. Acad. Sci. USA 75:5497-5500.










Patrick, R.A., S.B. Taubman, and I.H. Lepow. 1970. Cleavage
of the fourth component of human complement (C4) by
activated Cis. Immunochemistry 7:217-225.

Pensky, J., L.R. Levy, and I.H. Lepow. 1961. Partial
purification of serum inhibitor of C'l-esterase. J. Biol.
Chem. 236:1674-1679.

Phillips, J.K., R. Snyderman, and S.E. Mergenhagen. 1972.
Activation of complement by endotoxin: A role for gamma2
globulin, Cl, C4 and C2 in the consumption of terminal
complement components by endotoxin-coated erythrocytes.
J. Immunol. 109:334-341.

Rapp, H.J. 1955. A mathematical theory and analysis of
immune hemolysis. Dissertation, Johns Hopkins University,
School of Hygiene and Public Health.

Ratnoff, O.D., and I.H. Lepow. 1957. Some properties of an
esterase derived from preparations of the first component
of complement. J. Exper. Med. 106:327-342.

Reid, K.B.M. 1983. Proteins involved in the activation and
control of the two pathways of human complement. Proc.
Royal Acad. Sci. 11:1-12.

Rosenberg, S., P.J. Barr, R.C. Najarin, and R.A. Hallewell.
1984. Synthesis in yeast of a functional
oxidation-resistant mutant of human alpha1-antitrypsin.
Nature 312:77-80.

Salvesen, G.S., J.J. Catanese, L.F. Kress, and J. Travis.
1985. Primary structure of the reactive site of human
Cl-inhibitor. J. Biol. Chem. 260:2432-2436.

Schapira, M., L.D. Silver, C.F. Scott, A.H. Schaier, L.J.
Prograis, J.G. Curd, and R.W. Coleman. 1983.
Prekallikrein activation and high-molecular-weight
kinigen consumption in hereditary angioedema. New England
J. Med. 308:1050-1053.

Server, A.C., and E.M. Shooter. 1977. Nerve growth factor.
Adv. Prot. Chem. 31:339-409.

Sliverberg,M., J.T. Dunn, L. Garen, and A.P. Kaplan. 1980.
Autoactivation of human Hageman factor: Demonstration
utilizing a synthetic substrate. J. Biol. Chem.
255:7281-7286.

Sim, R.B. 1981. The human complement system serine proteases
aClr and aCls and their proenzymes. Meth. Enzymol.
80:26-42.









Sim, R.B., G.J. Arlaud, and M.G. Colomb. 1979. Kinetics of
reaction of human Cl-inhibitor with the human complement
system proteases Clr and Cls. Biochem. et Biophys. Acta
612:433-449.

Sim, R.B., and R.R. Porter. 1976. Isolation and comparison
of the proenzyme and activated forms of the human serum
components Cir and Cis. Biochem. Soc. Trans. 4:127-129.

Travis, J., and G.S. Salvesen. 1983. Human plasma protease
inhibitors. Ann. Rev. Biochem. 52:655-709.

Valet, G., and N.R. Cooper. 1974. Isolation and
characterization of the proenzyme form of the Clr subunit
of the first complement component. J. Immunol.
112:1667-1673.

Varon, S., J. Nomura, and E.M. Shooter. 1968. Reversible
dissociation of the mouse nerve growth factor protein
into different subunits. Biochemistry 7:1297-1303.

Weber K., and M. Osborn. 1969. The reliability of molecular
weight determinations by dodecyl sulfate-polyacrylamide
gel electrophoresis. J. Biol. Chem. 244:4406-4412.58.

Young, M., and M.J. Koroly. 1980. The nerve growth factor
zymogen. Stoichiometry of the active site serine and the
role of Zn(II) in controlling autocatalytic
self-activation. Biochemistry 19:5316-5322.

Ziccardi, R.J. 1982a. A new role for aCl-inhibitor in
homeostasis: Control of activation of the first component
of human complement. J. Immunol. 128:2505-2508.

Ziccardi, R.J. 1982b. Spontaneous activation of the first
component of human complement (Cl) by an intramolecular
autocatalytic mechanism. J. Immunol. 128:2500-2504.

Ziccardi, R.J., and N.R. Cooper. 1976. Physicochemical and
functional characterization of the Clr subunit of the
first complement component. J. Immunol. 116:496-503.















BIOGRAPHICAL SKETCH


Ervin L.W. Faulmann, Jr., was born on February 10,

1949, to Ervin and Leithel Faulmann. He was raised and

educated in southern California and graduated from Alhambra

High School in 1967. He attended California Polytechnic

State University at San Luis Obispo. After a brief period

of working on ranches in Montana, he returned to school and

received a Bachelor of Science degree in microbiology from

Montana State University in 1976. He entered the Department

of Immunology and Medical Microbiology at the University of

Florida College of Medicine for post graduate studies.

During the course of those studies he worked at the

University Medical Center in Jackson, Mississippi, under Dr.

L. William Clem and completed the work on the effects of

temperature on the mitogen responses of catfish leukocytes,

for which he received a Master of Science degree from the

University of Florida. He returned to the Department of

Immunology and Medical Microbiology and and continued his

studies under Dr. Michael D.P. Boyle. On October 24, 1982,

he married Phoebe Castle Hellems. Following the completion

of his dissertation, he intends to continue research in

immunology in hopes of improving our understanding of

molecular interactions associated with disease processes.

75









I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosophy.



Mi'chael D.P. Boy'9= Chairman
Associate Prof ssor of
Immunology and Medical
Microbiology


I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosophy.



U \/
Michael Youn i
Professor of Bioc istry
and Molecular Biology


I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosophy.



Adrian f. P ...
Assistant Professor of
Immunology and Medical
Microbiology

I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doc or of Ph osophy.



SPar er 'IA.' 9ma~ I
Professor o Immunology and
Medical Microbiology










I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosophy.




Edward M. Hoffman/n/
Professor of Mic biology
and Cell- Science


I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosophy.




aie-sB. Flanegan(/
Associate Professo)-/
Immunology and Medical
Microbiology



This dissertation was submitted to the Graduate Faculty of
the College of Medicine and to the Graduate School and was
accepted as partial fulfillment of the requirements for the
degree of Doctor of Philosophy.


May, 1985

eaon, College of Medicine t2-




Dean, Graduate School




Full Text
-In (l-y) Cz ValueH
19
Figure 5. HEMOLYTIC ACTIVITY OF SERUM ON POLY- AND
MONOCLONAL ANTIBODY COATED INDICATOR CELLS. Various
dilutions of serum were assayed for Cl activity on EAC4b
indicator cells that had been prepared with either
polyclonal rabbit anti-E (A ) or monoclonal mouse anti-E
( 9 ) The activity of activated Cl on monoclonal
sensitized indicator cells ( 0 ) was included as a
reference.


42
1.3) to aCl activity (as shown by a Cl titration slope of
1.0) in subsequent Cl hemolytic titration assays was
measured. As can be seen in Figure 14, at the highest
concentrations tested all three Cl depleted serum fractions
(euglobulin supernatant, ultracentrifuged supernatant, and
DE-52 eluate) were able to inhibit the spontaneous
conversion of serum Cl to aCl. Inhibition of the
spontaneous activation was shown to be dependent on the
concentration of the serum fraction used(l:100 dilutions of
the various serum fractions yielded aCl activity).
Spontaneous serum Cl activation was therefore inhibited by
some factor(s) present in the various serum fractions.
Inhibition of Spontaneous Activation of Zymogen Cl by
Cl-Inh
Human Cl-Inh was purified to a single band in SDS-PAGE
(see Figure 13). Addition of Cl-Inh to partially purified
zymogen Cl (DE-52 pass-through) inhibited the spontaneous
activation of Cl (Figure 15A). This inhibition of the
spontaneous activation of serum Cl by Cl-Inh was
concentration dependent. When the relative amounts of
Cl-Inh in the various dilutions of the various serum
fractions used in Figure 14 were considered, the
concentration of Cl-Inh needed to inhibit by 50% the
activation of the serum Cl was approximately equal to the
Cl-Inh concentration in the crude serum fraction from DE-52
eluate (Figure 14).


47
independent of its irreversible action on aCl. The action
of Cl-Inh is represented schematically in Figure 16.
The regulation of the conversion of zymogen Cl to aCl
by Cl-Inh was concentration dependent. The concentration of
Cl-In'n needed to maintain serum Cl from spontaneous
activation was less that that required to inactivate the
total quantity of activated Cl that could be generated in
the system. That is, if activation was allowed to occur
prior to adding Cl-Inh, then the minimal level previously
shown to inhibit spontaneous activation was insufficient to
inactivate all the aCl. This leads to the hypothesis that
Cl-Inh (Figure 16, Step I) is more effective in regulating
the generation of Cl activity in serum than it is as an
inactivator of fully activated Cl (Figure 16, Step II).
These studies have led to similar conclusions to those of
Ziccardi (1982a).
The ability of a low level of Cl-Inh to maintain serum
Cl in its zymogen form may account for the episodic nature
of the edema in HAE patients. Those patients usually have
10%-15% of the normal level of Cl-Inh. From these studies it
would be predicted that these patients have levels of Cl-Inh
sufficient to prevent continual spontaneous activation of
their serum Cl. However, if an enzyme regulated by Cl-Inh
(Figure 1) was activated in these patients there would be
insufficient Cl-Inh to effectively regulate such an enzyme.


30
These results indicate that Cl conforms to all the
predictions of the one hit theory of immune hemolysis
originally proposed by Borsos and Mayer (1962).
In contrast, when Cl in serum is measured in the
hemolytic assay under conditions that are pseudo-first order
for Cl its behavior does not conform to the predictions of
the one hit theory. The results observed with serum
(zymogen) Cl are summarized below:
1) The number of lytic sites generated is not directly
proportional to the quantity of serum Cl added (Heinz et
al., 1984; Hoffman and Stevland, 1971; Loos, 1982; Ratnoff
and Lepow, 1957).
2) Using the same dilution of serum Cl, the measured Cl
activity is dependent on the number of indicator cells used
in the assay (Loos, 1982; Loos et al., 1972).
3) Two populations of serum Cl can be detected on
indicator cells incubated with serum. The first, a
productive self-activatable Cl molecule, and the second, a
non-self-activatable molecule. The latter form of Cl may be
converted to active Cl by mild treatment with trypsin or
other proteases (Loos, 1982; Loos et al., 1972).
The differences between aCl activity and zymogen Cl
activity were not due to the heterogeneous (polyclonal)
nature of the antiserum normally used to prepare the EAC4b
indicator cells used in the Cl hemolytic assay. The results
presented here demonstrate that the behavior of both aCl and
serum Cl was not changed when monoclonal sensitizing


54
Figure 17. ENZYMATIC ACTIVITY OF NGF AFTER PREINCUBATION
WITH Cl-INH. Cleavage of the synthetic substrate 1-BAPNA,
as measured by the generation of OD^,was monitored
for the enzymatic activity of gamma NGF after incubation for
one hour at 37C in buffer (top tracing) or after the
addition of 9 molaf excess of purified Cl-Inh (lower
tracing).


52
Enzymatic Assay
Various amounts of purified Cl-Inh were incubated with
2.6 pg of gamma NGF and the enzymatic activity of the gamma
NGF was measured by monitoring the cleavage of the synthetic
substrate, N-alpha-benzoyl-l-arginine-p-nitroanilide
(1-BAPNA). The gamma NGF containing solution was added to a
1 mM solution of 1-BAPNA in 10 mM Tris(hydroxymethyl)
aminomethane (Tris) buffer, pH 7.4, and the cleavage of the
substrate was determined by an increase in the 041g of
the solution, using a Varian spectrophotometer, with the
cuvette jacketed at 24C.
Physiochemical Assay
125
Gamma NGF was labeled with I by the
lactoperoxidase method. Radiolabeled gamma NGF and Cl-Inh
were incubated together under various conditions and the
solutions were subjected to SDS-PAGE (Weber and Osborn,
1969). The slab gels were dried and the radiolabeled gamma
NGF was visualized by autoradiography. The relative amounts
of radioactivity migrating at different rates in the gels
were measured by scanning the autoradiographs on a Joyce
Lobel densitometer. Molecular weights were determined from
Coomasse R-250 stained reference wells containing molecular
weight standards (Sigma Chemicals).


62
those seen with the acyl linkage between gamma NGF and the
protease nexins (Baker et al., 1980) or the characteristic
serpin interaction with their appropriate serine esterase.
These properties suggest that Cl-Inh may have many of the
characteristics of a protease nexin. Further studies are
required to determine if any cell type has specific
receptors for Cl-Inh-gamma NGF complexes that would indicate
that Cl-Inh closely mimicked the biological activity of a
cell-secreted nexin.


55
CCI-INffl
C NGFJ
Figure 18. CONCENTRATION DEPENDENCE OF THE INHIBITION OF
THE ENZYMATIC ACTIVITY OF NGF BY Cl-INH. Various molar
equivalents of Cl-Inh were added to a constant amount of
gamma NGF, incubated one hour.at 37C, and then assayed for
remaining enzymatic activity in Cleaving 1-BAPNA. The graph
depicts the amount of gamma NGF activity remaining after
incubation with various molar amounts of Cl-Inh .


-In (l-y) Cz Valued
16
y-i.
Figure 3. THE HEMOLYTIC TITRATION OF SERUM Cl AND ACTIVATED
Cl. Various dilutions of either serum or activated Cl were
incubated with EAC4b for 20 minutes at 37C. The extent of
hemolysis was determined at end point following addition of
excess C2 and CEDTA. ( serum and ffi activated Cl)


CHAPTER II
SPONTANEOUS ACTIVATION OF PARTIALLY PURIFIED SERUM Cl
Introduction
Zymogen Cl has been shown to spontaneously activate in
the absence of either soluble or cell bound immune complexes
(Borsos et al., 1964; Cooper, 1983; Lepow et al., 1965;
Ziccardi, 1982a). Using a synthetic substrate to measure
the esterolytic activity of Cl, Lepow et al. (1965)
demonstrated that serum Cl isolated by euglobulin
fractionation spontaneously activated in a classical
autocatalytic manner. That is, zymogen Cl served as a
substrate for aCl and the enzymatic activity of the aCl led
to the conversion of zymogen Cl to aCl. By contrast, in a
recent series of experiments, Ziccardi (1982a) showed that
the activation of zymogen Cl was a concentration independent
process that was not influenced by the presence of aCl. In
those experiments zymogen Cl was reconstituted from its
isolated zymogen subunits (Cooper and Ziccardi, 1977) and
activation was measured by changes in the physicochemical
properties of Clr and Cls (Arlaud et al., 1980; Sim, 1981;
Sim and Porter, 1976; Valet and Cooper, 1974). This assay
requires a large input of zymogen Cl and is not as sensitive
as the Cl hemolytic titration (Borsos and Rapp, 1967) or
synthetic substrate enzymatic cleavage assays (Lepow et al.,
10


5
to regulate the enzyme activities of the first component of
the classical pathway of complement (Cl) (Levy and Lepow,
1959). Serum Cl exists as a zymogen that can be activated
to produce a proteolytically active form (aCl) which cleaves
C4 and C2 (Patrick et al., 1970). The Cl-Inh has been shown
to form a covalent bond with the activated subunits of aCl,
activated Clr (aClr) and Cls (aCls), leading to the
inactivation of the enzyme and the dissociation of Cl into
its subunits (Harpel and Cooper, 1975; Minta and Aziz,
1981).
Hereditary angioedema (HAE) is a genetic deficiency of
functional Cl-Inh, which is associated in patients with
episodic bouts of swelling. Donaldson has attributed the
edema to uncontrolled production of a kinin produced by the
cleavage of C2 (i.e., C2b) (Donaldson, 1970). Others have
suggested the swelling results from Hageman factor
activation (Schapira et al., 1983). As a serpin, Cl-Inh is
thought to operate by being first cleaved by the specific
protease and then forming a covalent bond with an
appropriate amino acid within the active site of the serine
esterase. The possibility that a variety of functionally
distinct Cl-Inh may exist is suggested by the observation
that Cl-Inh from sera of certain HAE patients fail to block
aCl activity, but continue to inhibit kallikrein and plasmin
effectively (Donaldson, 1982; Travis and Salvasen, 1983).
Similar variations of the range of activity for other
serpins have been observed (Travis and Salvasen, 1983). The


70
Bradshaw, R.A., and M. Young. 1976. Nerve growth factor:
Recent developments and perspectives. Biochem. Pharmacol.
25:1445-1449.
Carrell, R.W.. 1984. Therapy by instant evolution. Nature
312:14.
Cas, F., and M. Loos. 1981. Antibody-independent binding of
the first component of complement (Cl) and its
subcomponent Clq to the S and R forms of Salmonella
minnesota. Infect, and Immunity 31:1138-1144.
Colten, H.R., T. Borsos, and H.J. Rapp. 1967. Efficiency of
the first component of complement (C'l) in the hemolytic
reaction. Science 158:1590-1592.
Cooper, N.R. 1983. Activation and regulation of the first
complement component. Fed. Proc. 42:134-138.
Cooper, N.R., and R.J. Ziccardi. 1977. Reconstitution of Cl
in native, proenzyme form and its use in a quantitative
Cl activation test. J. Immunol. 119:1664-1667.
Cyong, C.-C., S.S. Witkin, B. Rieger, L. Barbarese, R.A.
Good, and N.K. Day. 1982. Antibody-independent complement
activation by myelin via the classical complement
pathway. J. Exp. Med. 155:587-598.
Donaldson, V.H. 1970. Blood coagulation and related plasma
enzymes in inflammation. Ser. Haemat. 3:39-95.
Donaldson, V.H. and R.A. Harrison. 1982. Complexes between
Cl-inhibitor, kallikrein, high molecular weight
kininogen, plasma thromboplastin antecedent, and plasmin
in normal plasma and hereditary angioneurotic edema
plasmas containing dysmorphic Cl-inhibitors: Role of cold
activation. Blood 60:121-129.
Gadek, J.E., S.W. Hosea, J.A. Gelfand, M. Santaella, M.
Wickerhauser, D.C. Triantapyhllopoulos, and M.M. Frank.
1980. Replacement therapy in heredity angioedema:
Successful treatment of acute episodes of angioedema with
partially purified Cl inhibitor. New England J. Med.
302:542-546.
Gee, A.P., M.D.P. Boyle, K. Munger, M.J.P Lawman, and M.
Young. 1983. Nerve growth factor: Stimulation of
polymorphonuclear leukocytes in vitro. Proc. Natl.
Acad. Sci. USA 80:7215-7218.
Gelfand, J.A., G.R. Boss, C.L. Conley, R. Reinhart, and M.M.
Frank. 1979. Aquired Cl esterase inhibitor deficiency and
angioedema: A review. Medicine 58:321-328.


32
Cl molecule must bind two IgG molecules on the surface of
the target cell to induce activation (slope=2 in the
hemolytic assay). The activation of Cl on IgM sensitizing
antibodies is a one-hit phenomenon (slope=l) and as such has
been shown to be at least 100 times more efficient than IgG
in activating Cl. Thus, a contamination of 1% specific IgM
in an IgG sensitizing antibody reagent would completely mask
the contribution of the IgG antibodies in the assay
(slope=l). Similarly, aCl is more effective than zymogen Cl
in the Cl hemolytic assay.
In agreement with Ziccardi's findings (Ziccardi,
1982a), however, in this study no evidence for an
autocatalytic activation reaction, in which the the rate of
activation would increase with the addition of aCl, was
observed. It is possible that Hageman factor, which has
since been shown to spontaneously activate in a classical
autocatalytic manner (Silverberg et al., 1980) which can
then activate serum Cl (Ghebrehiwet et al., 1983), could be
responsible for the autocatayltic generation of enzyme
activity in the euglobulin preparation used in Lepow's
studies. Furthermore, the synthetic substrate,
N-acetyl-l-tyrosine ethyl ester (ATEE), used in these
studies is not particularly sensitive in detection activity
of Cl (McRae et al., 1982), therefore the presence of any
contaminating autocatalytically activated enzyme that is
more active (i.e. Hageman factor) would mask the low level
of serum Cl activity.


ACKNOWLEDGEMENTS
I wish to express my thanks to Dr. Michael D.P. Boyle
for his guidance, support, and understanding during my
training under him. It has been a privilege to work with
Mike, with his wit, enthusiasism, intelligence and constant
pursuit of "What's new and exciting" he enhances the world
around him.
I would also like to express my appreciation to the
graduate students (both past and present), technicians, and
faculty with whom I have worked the last three years. Their
able assistance, personal sharing, and willingness to put up
with my "helping behaviors" are deeply appreciated.
My parents, Leithel and Ervin Faulmann continue to be a
supporting factor in my life. Though time and distance do
not allow direct contact, I know that their hopes and
prayers are for what I feel is best for me. I find that
virtually all of my personal strengths stem directly from
the foundations that they provided me with many years ago.
I thank them.
Finally, I would like to thank my companion in life,
Ms. Phoebe Castle Faulmann. Her support of my endeavors was
essential for all my work in pursuing my goals. The
strength of our relationship remains not in looking at each
other, but in looking in the same direction. To her, my
wife, and my friend, I would like to dedicate this and all
endeavors.
IX


27
activation of partially purified serum Cl is influenced by
the concentration of the Cl in the system, i.e., the
activation was concentration dependent since after 20
minutes at 37C the 1:4 dilution was activated (slope = 1),
while the 1:64 dilution was not (slope>l). Upon further
incubation, however, the 1:64 dilution was shown to
self-activate. As all the solutions were assayed at the
same final dilutions, the rate of activation was clearly
dependent on the concentration of the partially purified
serum Cl in the preincubation solution. These results
suggest that some interaction between Cl molecules is
required for activation.
Effects of Activated Cl on Sejrum Cl Activation
In earlier experiments, using euglobulin precipitates,
Lepow et al. (1965) suggested Cl would activate by an
autocatalytic reaction, i.e., aCl acted on zymogen Cl to
generate more aCl activity. To test this possibility, a low
level of aCl was added to a dilution of partially purified
serum Cl and incubated at 37C. As shown in Figure 11,
addition of aCl to this mixture did not result in increased
level of activation of the partially purified serum Cl. The
level of zymogen Cl chosen for this experiment was below the
concentration that would self-activate during the time
course of the experiment (see Figure 10). Rather, the
resultant Cl activity of the mixture represented the sum of
the respective Cl activities of the zymogen and added aCl.


REGULATION OF SERUM PROTEASE ACTIVITY:
ACTION OF Cl-INHIBITOR ON THE ACTIVITIES
OF SERUM Cl AND NERVE GROWTH FACTOR
By
ERVIN L.W. FAULMANN, JR.
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
1985


45
The action of Cl-Inh in preventing the conversion of
partially purified serum Cl activity to aCl activity in the
preincubation step was not due to its action on aCl. As
mentioned above, addition of Cl-Inh to partially purified
serum Cl solutions, followed by incubation for 20 minutes at
37C, yielded Cl activity similar to that found in serum
(Figure 15A). However, in parallel studies if the serum Cl
was preincubated in the absence of Cl-Inh for 20 minutes at
37C, then further incubation in the presence of Cl-Inh did
not result in the appearance of serum Cl activity (Figure
15B). Rather, under these conditions, Cl-Inh reduced the
aCl activity in a concentration dependent fashion. Similar
results were seen when Cl-Inh was added to purified aCl
(data not shown). The only difference in the experiment
that generated the results of Figure 15A and 15B was that
the serum Cl of 15A was preincubated on ice for 20 minutes
while the serum Cl of 15B was preincubated for 20 minutes at
37 prior to the addition of the Cl-Inh. These results
showed that the action of Cl-Inh in inhibiting the
conversion of partially purified serum Cl to aCl activity
was not due to the interaction of Cl-Inh with aCl alone or
an artefact of the assay system. Furthermore, the amount of
Cl-Inh needed to inhibit the activation of the partially
purified serum Cl was less than that needed to inhibit the
Cl activity after the spontaneous activation (see Figure 15
A and B).


CHAPTER V
SUMMARY AND CONCLUSIONS
This investigation has examined protein interactions
important in the activation, regulation, and inactivation of
serum proteases. In addressing these topics, three basic
questions were posed and answered. First, what is the
nature of the spontaneous activation reaction of partially
r'*
purified serum Cl? Second, what is the role of Cl-Inh in
regulating the activation of serum Cl? Finally, wnat is tne
nature or tne interaction or ei-lnn witn tne activated gamma
subunit or NGF?
The process or serum Cl activation was shown to require
intermolecular interactions. This process was not
autocatalytic in that the addition of enzymatically active
Cl did not alter the extent of serum Cl activation. The
studies presented here confirm the work of others (Borsos et
al., 1980; Borsos et al., 1964; Heinz et al. 1984; Loos et
al., 1972) and clearly show that the level of activatable
zymogen Cl is not simply limited by the association of the
Clq subunits with Clr2s2 complexes, as has been
proposed by Ziccardi (1982a). The concentration dependence
of the Cl activation reaction has not been adequately
explained. Borsos et al. (Borsos et al., 1980; Loos et al.,
1972) suggest that heterogeneity in the Clq subunits could
63


20
Cl that had characteristics of aCl (see Figure 6). These
findings confirmed those of Loos and colleagues (Borsos et
al., 1980; Loos, 1982; Loos et al., 1972) and demonstrated
that the Cl hemolytic assay was highly sensitive and was
able to distinguish native and activated Cl in the presence
of many other serum proteins. In other experiments it was
confirmed that aCl bound to indicator cells at 4C and
shifted to 37C yielded activity without a lag, while Cl
activity in serum that had been bound to EAC4b cells at 4C
and moved to 37C demonstrated a lag period that was
consistant with the time required for Cl to undergo an
activation process (see Figure 7). These functional
properties of native and activated Cl have been used in the
remainder of the study to follow the activation status of
various preparations of serum Cl.
Spontaneous Activation of Partially Purified Serum Cl
Human serum was diluted 1:2 with distilled water and
applied to a column of DE-52 previously equilibrated with
.075 M NaCl and 10 mM HEPES pH 7.4 at 4C. The
pass-through material contained approximately 80% of the
serum proteins, as monitored by OD2gQ and include the
majority of the measurable Cl activity. The amount of Cl
activity in a single dilution of the partially purified
serum Cl was monitored in the Cl hemolytic assay, along with
appropriate dilutions of unfractionated serum and aCl. When
incubated for various times at 37C, the relative amount of


48
Cl-Inh
(I)
Cl-Inh
(II)
V
zymogen Cl X> activated Cl
inactivated Cl
reversible
irreversible
Figure 16. SCHEMATIC OF THE REGULATORY EFFECTS OF Cl-INH ON
Cl ACTIVITY


72
Lepow, I.H., G.B. Naff, and J. Pensky. 1965. Mechanisms of
activation of C'l and inhibition of C'l esterase. In
CIBA Foundation Symposium; Complement, ed.
Wolstenholme, G.E.W., and Julie Knight. Little Brown and
Co., Boston, MA. pp. 74-98.
Levi-Montalcini,R. 1965. The nerve growth factor: Its mode
of action on sensory and sympathetic nerve cells. Harvey
Lect. 60:217-259.
Levi-Montalcini, R., and B. Booker. 1960. Excessive growth
of sympathetic ganglia evoked by a protein isolated from
mouse salivary gland. Proc. Natl. Acad. Sci. USA
46:373-384.
Levy, L.R. and I.H. Lepow. 1959. Assay and properties of
serum inhibitor of C'l esterase. Proc. Soc. Exp. Biol.
Med. 101:608-611.
Li, A.K.C., M.J. Koroly, M.E. Schattenkerk, R.A. Malt, and
M. Young. 1980. Nerve growth factor: Acceleration of the
rate of wound healing in mice. Proc. Natl. Acad. Sci. USA
77:4379-4381.
Loos, M. 1982. The classical complement pathway: Mechanism
of activation of the first component by antigen-antibody
complexes. Prog, in Allergy 30:135-192.
Loos, M., T. Borsos, and H.J. Rapp. 1972. Activation of the
first component of complement: Evidence for an internal
activation step. J. Immunol. 108:683-688.
Loos, M., and H.-U. Hill. 1977. Different behaviour of
native and purified Clq, a subcomponent of the first
component of complement (Cl). Monogr. in Allergy
12:46-51.
McRae, B.J., T.-Y. Lin, and J.C. Powers. 1981. Mapping the
substrate binding site of human aClr and aCls with
peptide thioesters. J. Biol. Chem. 256:12362-12366.
Minta, J.O., and E. Aziz. 1981. Analysis of the reactive
site peptide bond in aCl-inhibitor by chemical
modification of tyrosyl, lysyl, and arginyl residues: The
essential role of lysyl residues in the functional
activity of aCl-Inh. J. Immunol. 126:250-255.
Orenstein, N.S., H.F. Dvorak, M.H. Blanchard, and M. Young.
1978. Nerve growth factor: A protease that can activate
plasminogen. Proc. Natl. Acad. Sci. USA 75:5497-5500.


4
Complement
System
C4 C2
a Cl #->NJ
V
classical
pathway
Clotting
System
factor IX
PTA (factor XI) nJ/
activation
Kinin
Generating
System
kininogen
kallikrein ^ ^ \1/
kinin
Fibrinolytic
System
fibrin split products
plasmln
Cl activation
Figure 1. INHIBITORY FUNCTIONS OF Cl-INH.


I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosophy.
Michael D.P. Boy
Associate Professor of
Immunology and Medical
Microbiology
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosophy.
^ m Michael Young^
Professor of Biocm?mistry
and Molecular Biology
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosophy.
/ .
/
/ /D ^
Adrian" P. tiee~ Jj
Assistant Professor of
Immunology and Medical
Microbiology
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Ph^i^osophy.
Medical Microbiology


12
Guinea pig complement was obtained from Pel-Freez
Biologicals, Rogers, AR. Isolated complement components
were purchased from Cordis Biologicals, Miami, FL. Isotonic
veronal buffered saline containing metals (VBS) and
sucrose-veronal buffer (SVB) (pH 7.35, with 0.1% gelatin
(Difco), 0.001 M Mg++, and 0.00015 M Ca++), and
veronal-buffered saline containing 0.01 M ethylenediamine
tetraacetic acid (VBS-EDTA) were prepared as described in
Borsos and Rapp (1967). A reagent which was devoid of whole
complement activity but provided functional complement
components C3-C9 was obtained by diluting serum in VBS-EDTA
(CEDTA) (Borsos and Rapp, 1967). Indicator cells for the Cl
hemolytic assay were prepared by the method of Borsos and
Rapp (1967) in which E were sensitized with the antibody
solution, coated with aCl, incubated in CEDTA, and washed.
This procedure yields antibody sensitized E (EA) with the
activated fourth component of complement bound to their
surfaces (EAC4b)(contributed by the added serum), without
contaminating Cl and C2 (removed by the added VBS-EDTA) or
any later acting complement components (Borsos and Rapp,
1967 ) .
Isolation of Serum Fraction Rich in Zymogen Cl
Fractions rich in zymogen Cl were isolated from human
serum by ion-exchange chromatography. Human serum diluted
1:2 in distilled water and applied to a column of DE-52
cellulose equilibrated at pH 7.3 with 0.01 M


58
116,000-
97,000-
66,000-
45,000-
29,000-
I 2 3
Figure 20. AUTORADIOGRAPH OF SDS-PAGE ANALYSIS OF
i23I-GAMMA NGF INTERACTION WITH Cl-INH. Radiolabeled
gamma NGF was incubted with various amounts of Cl-Inh (lane
1 none, lane 2 1/2 molar equivalent of cl-Inh, lane 3 -
80 molar excess Cl-Inh) and incubated one hour at 37C,
added to 2% SDS, and electrophoresed.


17
C O NCENTR ATION
Figure 4. MEASUREMENT OF SERUM AND ACTIVATED Cl ACTIVITY AS
A FUNCTION OF INDICATOR CELL NUMBER. Various dilutions of
either activated Cl (panel A) or serum (panel B) were
measured for Cl activity using various concentrations of
EAC4b indicator cells. ( a 24x10, 3 6.0x10, and
1.5x10 )


46
Discussion
These studies extend the observations made in the
previous chapter in which crude isolates of Cl from serum,
upon incubation at 37C, spontaneously converted to
activated Cl, as measured functionally in the hemolytic
assay. This spontaneous autoactivation was not observed
with unfractionated serum Cl, i.e. human serum. There are
two potential explanations for why spontaneous Cl activation
does not occur in serum: 1) the isolation process somehow
altered the serum Cl molecule thus allowing it to self
activate, or 2) in the process of isolation, some regulatory
molecule was separated from the serum Cl, thus allowing the
serum Cl molecule to self-activate, a process normally
regulated in serum. The studies reported here support the
second hypothesis that there are regulatory factor(s) found
in serum that inhibit the spontaneous activation of Cl.
Partially purified serum Cl that self-activated when
incubated in buffer failed to activate when admixed with Cl
depleted crude serum fractions.
Due to the relationship of functional Cl-Inh levels and
Cl activation seen in HAE patients and the results from
researchers using reassociated Cl molecules, Cl-Inh was
examined as the most likely regulator of serum Cl
self-activation. It was found that when Cl-Inh was
incubated with partially purified serum Cl, activation could
be prevented. This regulatory activity of Cl-Inh was


RELATIVE Cl CONCENTRATION
Figure 14. EFFECT OF VARIOUS SERUM FRACTIONS ON THE SPONTANEOUS ACTIVATION OF
PARTIALLY PURIFIED SERUM Cl. In separate experiments, various serum fractions
were added to partially purified serum Cl and then incubated one hour at 37C.
The resulting Cl activity was then assayed hemolytically. The amount of Cl
activity contributed by the various serum fractions was less than 20% of the Cl
activity of the mixture. (Panel A euglobulin supernatant of human serum added
to partially purified serum Cl, panel B .2 /M NaCl eluate of the DE-52 column
used to produce the partially purified serum Cl added back to the partially
purified serum Cl, and panel C supernatant of the ultracentrifugation of
diluted human serum added to partially purified serum Cl. In all experiments
the various serum fractions were assayed at 1:10 ( O ) and 1:100 ( ) dilutions
and are graphed with the unfractionated serum control ( )).


-ln(l-y) CZ value!
44
A B
RELATIVE Cl CONCENTRATION
Figure 15. EFFECTS OF PURIFIED Cl-INH ON THE SPONTANEOUS
ACTIVATION OF PARTIALLY PURIFIED SERUM Cl AND ON ACTIVATED
PARTIALLY PURIFIED SERUM Cl. Purified Cl-Inh (at various
dilutions) or buffer was added to partially purified serum
Cl which had been preincubated for 20 minutes at either 4C
(panel A) or 37C (panel B). Following a 20 minute
incubation at 37C the resultant Cl activity was monitored
in the hemolytic assay (dilutions of Cl-Inh used: 1:5
( ), 1:25 ( ), 1:125 ( O ), 1:625 ( A ), buffer ( A*v) ) )
For these studies, the. undiluted preparation of Cl-lTih
contained an equivalent concentration of inhibitor normally
present in human serum (0.1 mg/ml)


24
Figure 8. HEMOLYTIC ACTIVITY OF SERUM, ACTIVATED Cl, AND
PARTIALLY PURIFIED SERUM Cl AFTER INCUBATION AT 37C. A
single dilution (1:4 in VBS) of activated Cl ( ), serum
( A ), and partially purified serum Cl ( 9 ) were incubated
37C. At appropriate times, aliquots were taken, diluted,
and assayed for Cl activity in the hemolytic assay. A
single dilution of each source is represented at each time
point.


CHAPTER III
REGULATION OF Cl ACTIVATION
Introduction
The first component of complement (Cl) is present in
normal serum in a zymogen form (Lepow et al., 1965). Upon
binding to the Fc region of complement fixing antibodies
bound to antigen, serum Cl undergoes a series of
r' ,
intramolecular events producing activated Cl (aCl) (Borsos
et al., 1964; Heinz et al., 1984; Loos et al., 1972; Loos
and Hill, 1977; Reid, 1983). The enzymatic activity of the
aCl molecule is then capable of cleaving C4 and C2 (Patrick
et al., 1970), thereby initiating the classical complement
pathway. Other agents, eg. bacterial products (Clas and
Loos, 1981; Phillips et al., 1972) and various
compounds(Cyong et al., 1982), are also capable of
activating serum Cl. The aCl molecule is regulated in serum
by Cl-Inh (Lepow et al., 1965) which binds to the activated
subunits (aClr and aCls) inhibiting their esterolytic
activity (Harpel and Cooper, 1975), with resultant
dissociation of the aCl molecule (Harpel and Cooper, 1975;
Minta and Aziz, 1981). In order to study the activation
process of Cl, a number of researchers have attempted to
isolate Cl in its zymogen form (Borsos et al., 1964; Lepow
et al., 1965). Typically, these attempts to isolate whole
34


TABLE OF CONTENTS
Page
ACKNOWLEDGEMENTS ii
ABSTRACT iv
CHAPTERS
IINTRODUCTION 1
IISPONTANEOUS ACTIVATION OF PARTIALLY PURIFIED
SERUM Cl 10
Introduction 10
Materials and Methods 11
Results 15
Discussion 29
IIIREGULATION OF Cl ACTIVATION 34
Introduction 34
Materials and Methods 37
Results 41
Discussion 46
IVINACTIVATION OF THE PROTEOLYTIC ACTIVITY OF
NGF BY Cl-INH 49
Introduction 49
Materials and Methods 51
Results 53
Discussion 61
VSUMMARY AND CONCLUSIONS ..... 63
BIBLIOGRAPHY 69
BIOGRAPHICAL SKETCH 75
iii


57
interaction of gamma NGF and Cl-Inh requires an enzymatic
reaction.
Complex Formation between Gamma NGF and Cl-Inh
The apparent enzymatic nature of the interaction of
gamma NGF and Cl-Inh and the 1:1 stoichiometry described
above led to the hypothesis that the enzymatic activity of
gamma NGF was inhibited by the formation of a stable complex
with Cl-Inh. Experiments were designed to monitor the
formation of a covalent complex formed between gamma NGF
(26,000 molecular weight) and Cl-Inh (97,000 molecular
weight) using SDS-PAGE analysis. In these experiments,
125
I-gamma NGF was incubated with Cl-Inh for one hour at
37C and the mixture was electrophoresed in SDS-PAGE using
the Weber and Osborn buffer system (phosphate buffer pH
12 5
7.2). Incubation of I-gamma NGF with an 80-fold molar
excess Cl-Inh for one hour at 37C yielded greater than 97%
125
of the I migrating in a high molecular weight complex
of approximately 125,000 Daltons (Figures 20 and 21). In
order to investigate the nature of the bond formed in the
high molecular weight complex, the effects of hydroxylamine
and high pH on the stability of the complex were examined.
Preformed high molecular complexes were formed by the
125
incubation of I-gamma NGF with 1/2 molar equilivant of
Cl-Inh. An aliquot was removed and further incubated for 20
minutes at 37C in .1 M hydroxylamine pH 10.5 and
electrophoresed. The results (see Figure 22) show that the


74
Sim, R.B., G.J. Arlaud, and M.G. Colomb. 1979. Kinetics of
reaction of human Cl-inhibitor with the human complement
system proteases Clr and Cls. Biochem. et Biophys. Acta
612:433-449.
Sim, R.B., and R.R. Porter. 1976. Isolation and comparison
of the proenzyme and activated forms of the human serum
components Clr and Cls. Biochem. Soc. Trans. 4:127-129.
Travis, J., and G.S. Salvesen. 1983. Human plasma protease
inhibitors. Ann. Rev. Biochem. 52:655-709.
Valet, G., and N.R. Cooper. 1974. Isolation and
characterization of the proenzyme form of the Clr subunit
of the first complement component. J. Immunol.
112:1667-1673.
Varn, S., J. Nomura, and E.M. Shooter. 1968. Reversible
dissociation of the mouse nerve growth factor protein
into different subunits. Biochemistry 7:1297-1303.
Weber K., and M. Osborn. 1969. The reliability of molecular
weight determinations by dodecyl sulfate-polyacrylamide
gel electrophoresis. J. Biol. Chem. 244:4406-4412.58.
Young, M., and M.J. Koroly. 1980. The nerve growth factor
zymogen. Stoichiometry of the active site serine and the
role of Zn(II) in controlling autocatalytic
self-activation. Biochemistry 19:5316-5322.
Ziccardi, R.J. 1982a. A new role for aCl-inhibitor in
homeostasis: Control of activation of the first component
of human complement. J. Immunol. 128:2505-2508.
Ziccardi, R.J. 1982b. Spontaneous activation of the first
component of human complement (Cl) by an intramolecular
autocatalytic mechanism. J. Immunol. 128:2500-2504.
Ziccardi, R.J., and N.R. Cooper. 1976. Physicochemical and
functional characterization of the Clr subunit of the
first complement component. J. Immunol. 116:496-503.


35
serum Cl have yielded exclusively aCl, which led to the
conclusion that upon isolation serum Cl undergoes a
spontaneous activation process (Lepow et al., 1965;
Ziccardi, 1982a). The reason that the activation of
isolated serum Cl does not occur in unfractionated serum is
that 1) the isolation process caused the serum Cl molecules
to become activated, or that 2) the isolation process
separated the serum Cl away from some form of regulation
normally found in serum. In initial experiments, sham
isolation procedures, in which serum was exposed to gel
f'
chromatography, euglobulin fraction, etc., but was left
unfractionated, the Cl failed to spontaneously activate. An
example of this observation was that ultracentrifugation of
serum resulted in a pellet rich in Cl activity, which when
resuspended in the supernatant displayed the properties of
zymogen Cl. However, resuspending the pellet in buffer
yielded Cl activity that had the characteristics of the
fully activated molecule (data not shown). The isolation
process was therefore not the sole cause for the observed
spontaneous activation.
The autoactivation of serum Cl may have an in vivo
corollary in the disease hereditary angioedema (HAE), in
which patients are shown to have low levels of functional
Cl-Inh (Gadek et al., 1980). These individuals experience
episodes of spontaneous activation of their serum Cl
(Gelfand et al., 1979), with serious physiological
consequences. The episodic nature of this disease and the


71
Ghebrehiwet, B., B.P. Randazzo, J.T. Dunn, M. Silverberg,
and A.P. Kaplan. 1983. Mechanisms of activation of the
classical pathway of complement by Hageman factor
fragment. J. Clin. Invest. 71:1450-1456.
Gigli, I., J.W. Mason, R.W. Coleman, and K.F. Austen. 1970.
Interaction of plasma kallikrein with the Cl inhibitor.
J. Immonol. 104:574-581.
Gigli,I., S. Ruddy, and K.L. Austen. 1968. The
stoichiometric measurement of the serum inhibitor of the
first component of complement by the inhibition of immune
hemolysis. J. Immunol. 100:1154-1164.
Greene, L.A., E.M. Shooter, and S. Varn. 1969. Subunit
interaction and enzymatic activity of mouse 7S nerve
growth factor. Biochemistry 8:3735-3741.
Harpel, P.C., and N.R. Cooper. 1975. Studies on human plasma
Cl inactivator-enzyme interactions. I. Mechanisms of
interaction with aCls, plasmin, and trypsin. J. Clin.
Invest. 55:593-603.
Harrison, R.A. 1984. Human Cl inhibitor: Improved isolation
and preliminary structural characterization. Biochemistry
22:5001-5007.
Heinz, H.-P., R. Burger, M.D. Golan, and M. Loos. 1984.
Activation of the first component of complement, Cl, by a
monoclonal antibody recognizing the C chain of Clq. J.
Immunol. 132:804-808.
Hoffman, L.G., and N.P. Stevland. 1971. Difference in dose
response between native and activated first component of
guinea pig complement (Cl): Reaction with aCl inactivator
offers no explaination. Immunochemistry 8:499-508.
Isackson, P.J., and R.A. Bradshaw. 1984. The alpha- and
beta-subunit of mouse 7S nerve growth factor is an
inactive serine protease. J. Biol. Chem. 259:5380-5383.
Knauer,D.J., K.M. Scaparro, and D.D. Cunningham. 1982. The
gamma subunit of 7S nerve growth factor binds to cells
via complexes formed with two cell-secreted nexins. J.
Biol. Chem. 257:15098-15104.
Knauer, D.J., J.A. Thompson, and D.D. Cunningham. 1983.
Protease nexins: Cell-secreted proteins that mediate the
binding, internalization, and degradation of regulatory
serine proteases. J. Cell Physiol. 117:385-396.
Laemmli, U.K. 1970. Cleavage of structural proteins during
the assembly of the head of bacteriophage T4. Nature
227:680-685.


REGULATION OF SERUM PROTEASE ACTIVITY:
ACTION OF Cl-INHIBITOR ON THE ACTIVITIES
OF SERUM Cl AND NERVE GROWTH FACTOR
By
ERVIN L.W. FAULMANN, JR.
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
1985

ACKNOWLEDGEMENTS
I wish to express my thanks to Dr. Michael D.P. Boyle
for his guidance, support, and understanding during my
training under him. It has been a privilege to work with
Mike, with his wit, enthusiasism, intelligence and constant
pursuit of "What's new and exciting" he enhances the world
around him.
I would also like to express my appreciation to the
graduate students (both past and present), technicians, and
faculty with whom I have worked the last three years. Their
able assistance, personal sharing, and willingness to put up
with my "helping behaviors" are deeply appreciated.
My parents, Leithel and Ervin Faulmann continue to be a
supporting factor in my life. Though time and distance do
not allow direct contact, I know that their hopes and
prayers are for what I feel is best for me. I find that
virtually all of my personal strengths stem directly from
the foundations that they provided me with many years ago.
I thank them.
Finally, I would like to thank my companion in life,
Ms. Phoebe Castle Faulmann. Her support of my endeavors was
essential for all my work in pursuing my goals. The
strength of our relationship remains not in looking at each
other, but in looking in the same direction. To her, my
wife, and my friend, I would like to dedicate this and all
endeavors.
IX

TABLE OF CONTENTS
Page
ACKNOWLEDGEMENTS ii
ABSTRACT iv
CHAPTERS
IINTRODUCTION 1
IISPONTANEOUS ACTIVATION OF PARTIALLY PURIFIED
SERUM Cl 10
Introduction 10
Materials and Methods 11
Results 15
Discussion 29
IIIREGULATION OF Cl ACTIVATION 34
Introduction 34
Materials and Methods 37
Results 41
Discussion 46
IVINACTIVATION OF THE PROTEOLYTIC ACTIVITY OF
NGF BY Cl-INH 49
Introduction 49
Materials and Methods 51
Results 53
Discussion 61
VSUMMARY AND CONCLUSIONS ..... 63
BIBLIOGRAPHY 69
BIOGRAPHICAL SKETCH 75
iii

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
REGULATION OF SERUM PROTEASE ACTIVITY:
ACTION OF Cl-INHIBITOR ON THE ACTIVITIES
OF SERUM Cl AND NERVE GROWTH FACTOR
By
ERVIN L.W. FAULMANN, JR.
MAY 1985
Chairman: Michael D.P. Boyle
Major Department: Immunology and Medical Microbiology
The interactions between serum proteases (serpins) and
serum protease inhibitors are the basis of the regulation of
many homeostatic processes. Evidence from a number of
researchers suggests that protease inhibitors may have
physiological roles in addition to inactivating proteolytic
enzymes.
The first component of the classical complement pathway
(Cl) exists in serum in a precurser zymogen form. During
purification Cl undergoes spontaneous activation yielding
activated Cl (aCl); however the mechanism of this
autoactivation process is unclear.
In this study, by means of the Cl hemolytic assay,
partially purified serum Cl was shown to spontaneously
activate in a time-and-concentration-dependent manner.
Addition of aCl did not alter the level of Cl activation.
iv

These results suggest a model of: serum Cl activation that
requires intermolecular interactions between the Cl
molecules independent of the enzymatic activity of aCl
produced. However, addition of purified Cl esterase
inhibitor protein (Cl-Inh), a serpin that inhibits the
enzymatic activity of aCl and a variety of cascade
proteases, inhibited the spontaneous activation of serum Cl
in a concentration-dependent manner. The effect of Cl-Inh
regulating Cl activation was independent of its activity in
inactivating aCl. These results document a new role of
r
Cl-Inh in which it reversibly inhibits the activation of
zymogen Cl as compared to its irreversible inhibition of
aCl.
The role of Cl-Inh in inhibiting another classical
complement pathway activator, the gamma subunit of nerve
growth factor (gamma NGF), was also examined. The enzymatic
activity of gamma NGF was inhibited by Cl-Inh in a
time-dependent manner with 1:1 stoichiometry. The
interaction of Cl-Inh with gamma NGF produced a sodium
dodecyl sulfate (SDS) stable complex that was suceptible to
cleavage by hydroxylamine, thus suggesting a covalent bond
between the protease and protease inhibitor. This reaction
between gamma NGF and Cl-Inh is similar to that between
gamma NGF and a newly described class of cell secreted
protease inhibitors, called nexins, and provides new insight
as to the possible physiological roles of both Cl-Inh and
gamma NGF.
v

CHAPTER I
INTRODUCTION
The activation and control of specific serum proteases
is the basis of many homeostatic processes. These processes
include the clotting system, the fibrinolytic system, the
kinin generating system, and the complement system, all of
which have important roles in inflammation and in the normal
response to tissue damage. Each of these processes entails
the serial interaction of specific serum proteins, in which
the activation of one component can initiate a chain
reaction (cascade reaction) culminating in the generation of
a specific effector molecule, e.g., the activation of
Hageman factor initiates a series of reactions that can
result in the formation of a fibrin clot. Many of the
interactions of these cascades are mediated through the
activation of a proteolytic enzyme from an inactive
precursor (zymogen) molecule. At various points along these
cascade reactions there are specific amplification steps in
which one activated molecule can stimulate the activation of
a large number of subsequent components in the cascade.
During the course of these cascade reactions biologically
active split products are produced, e.g., the anaphylatoxins
of the complement system. It is clear that if such
pharmacologically active split products are produced
1

2
unregulated there would be serious physiological
consequences. Serum protease inhibitors provide an
important level of regulation by inactivating specific
proteolytic enzymes associated with key amplifying roles in
these cascade systems.
Recent evidence suggests that protease inhibitors may
have physiological roles in addition to the inactivation of
proteolytic enzymes. In 1980, Baker et al. described the
secretion of a protein from normal human fibroblasts that
formed a covalent acyl linkage with urokinase and thrombin.
The resulting proteolytically inactive complexes were
internalized and degraded (Knauer et al., 1983). These
regulating anti-proteases were found to be secreted by a
wide range of normal and transformed cell lines and were
named nexins. Detailed studies using a variety of sources
of nexins and a number of serine class proteases have
indicated the existance of three distinct protease nexins
(PNs). These regulating proteins designated PNl, PN2, and
PN3 share many of the functional characteristics of the
serpins, but appear to be both physicochemically and
antigenically distinct proteins (Knauer et al., 1983). The
three groups, PNl, PN2, and PN3, differ in the enzymes which
they inhibit. The first, PNl, is a protein of molecular
weight of 38,000 Daltons that binds thrombin and urokinase.
Complexes with either of these enzymes bind to the surface
receptors on a variety of cell types and are rapidly
internalized and degraded. In the presence of heparin, cell

3
binding is enhanced (Knauer et al., 1983). The second, PN2,
is a protein of 93,000 Daltons with selective reactivity for
the epidermal growth factor (EGF)-binding protein (Knauer et
al., 1983). These complexes are also cleared by a variety
of cell types but their uptake is not augmented by addition
of heparin. The last, PN3, is a 31,000 Dalton protein with
selective reactivity for the gamma subunit of nerve growth
factor (gamma NGF) and the resulting gamma NGF-PN3 complexes
are rapidly cleared by a variety of cells. As in
PN2-containing complexes, clearance is not augmented by
heparin. The binding specificities of the PNs are not
absolute and some reactivity of gamma NGF and PNl and PN2
has also been reported (Knauer et al., 1983). The
physiological function of the nexins is not clear. The
current findings suggest that they act to regulate the
biological activities of certain serine proteases by forming
complexes that are rapidly cleared and degraded. At
present, no stimulatory biological activity has been
reported for any PN-protease complex.
Of particular interest to this study is the Cl esterase
inhibitor protein (Cl-Inh) that regulates a variety of
cascade systems associated with the host response to tissue
damage and inflammation (see Figure 1). Human Cl-Inh is a
single-chain glycoprotein with a molecular weight of 116,000
Daltons (Harrison, 1984) (97,000 Daltons in SDS-PAGE (Harpel
and Cooper, 1975)), containing about 35% carbohydrate
(Harrison, 1984). It was first recognized for its ability

4
Complement
System
C4 C2
a Cl #->NJ
V
classical
pathway
Clotting
System
factor IX
PTA (factor XI) nJ/
activation
Kinin
Generating
System
kininogen
kallikrein ^ ^ \1/
kinin
Fibrinolytic
System
fibrin split products
plasmln
Cl activation
Figure 1. INHIBITORY FUNCTIONS OF Cl-INH.

5
to regulate the enzyme activities of the first component of
the classical pathway of complement (Cl) (Levy and Lepow,
1959). Serum Cl exists as a zymogen that can be activated
to produce a proteolytically active form (aCl) which cleaves
C4 and C2 (Patrick et al., 1970). The Cl-Inh has been shown
to form a covalent bond with the activated subunits of aCl,
activated Clr (aClr) and Cls (aCls), leading to the
inactivation of the enzyme and the dissociation of Cl into
its subunits (Harpel and Cooper, 1975; Minta and Aziz,
1981).
Hereditary angioedema (HAE) is a genetic deficiency of
functional Cl-Inh, which is associated in patients with
episodic bouts of swelling. Donaldson has attributed the
edema to uncontrolled production of a kinin produced by the
cleavage of C2 (i.e., C2b) (Donaldson, 1970). Others have
suggested the swelling results from Hageman factor
activation (Schapira et al., 1983). As a serpin, Cl-Inh is
thought to operate by being first cleaved by the specific
protease and then forming a covalent bond with an
appropriate amino acid within the active site of the serine
esterase. The possibility that a variety of functionally
distinct Cl-Inh may exist is suggested by the observation
that Cl-Inh from sera of certain HAE patients fail to block
aCl activity, but continue to inhibit kallikrein and plasmin
effectively (Donaldson, 1982; Travis and Salvasen, 1983).
Similar variations of the range of activity for other
serpins have been observed (Travis and Salvasen, 1983). The

6
altered activity has been shown to be associated with
mutations in the amino acid sequence of the target site on
the serpin recognized by the specific enzyme (Carrell,
1984). During a disease episode, the serum Cl of HAE
patients is fully activated and there is no measurable C4
and C2 (Gelfand et al., 1979). Asymptomatic HAE patients,
however, exibit low but measurable levels of C4 and C2
(Gelfand et al., 1979), hence their Cl is not fully
activated. The low levels of functional Cl-Inh and the
anomalous episodic activation of the serum Cl of HAE
patients appear to be causally related. Clinical
administration of Cl-Inh reverses the disease process and
the patient's C4 and C2 levels are restored (Gadek et al.,
1980 ) .
The main focus of this study was to examine the
activation of serum Cl and determine the role of Cl-Inh in
regulating the activation process. Previous studies have
reported the autocatalytic activation of partially purified
serum Cl, in which aCl reacts enzymatically with zymogen Cl
resulting in the conversion of zymogen Cl to aCl (see Figure
2) (Lepow et al., 1965). In this case, Cl-Inh would
regulate the activation of Cl by inhibiting the enzymatic
activity of aCl. A recent report has shown that purified
zymogen Cl spontaneously activates by a intramolecular event
(concentration independent) that was not autocatalytic
(Ziccardi, 1982a). Furthermore, Cl-Inh was shown to inhibit
this spontaneous activation, though the mechanism of this

7
Figure 2.
ACTIVATION
Z d Slow u
ZCI
%
fast
Cl INA
* aCI
SCHEMATIC OF THE CLASSICAL AUTOCATALYTIC
MODEL.

8
inhibition remains unclear (Ziccardi, 1982b). This study
was initiated to resolve the apparent controversy regarding
the spontaneous activation of zymogen Cl and the role Cl-Inh
may play in it. Partially purified serum Cl was shown to
undergo spontaneous activation in a
time-and-concentration-dependent manner. This spontaneous
activation was independent of the enzymatic activity of
exogenous aCl. Serum fractions depleted of Cl and purified
Cl-Inh were shown to inhibit the spontaneous activation of
serum Cl. These studies suggest a model of Cl activation
that is dependent on an intermolecular interaction between
zymogen Cl molecules that is independent of the enzymatic
activity of either the Clr or Cls subunits.
Recently, it has been reported that the high molecular
weight form of NGF can substitute for activated Cl in
activating the classical complement pathway cleaving C4 and
C2 (Boyle and Young, 1982). High molecular weight NGF is a
multisubunit protein composed of three subunits, alpha,
beta, and gamma in the stoichiometric ratio of 2:1:1
respectively (Greene et al., 1969). The alpha subunit has
no known biological function (Isackson and Bradshaw, 1984).
The beta subunit is responsible for the outgrowth of
sympathetic and sensory ganglia from chicken embryo
(Levi-Montalcini, 1965). The gamma subunit is a serine
class enzyme (Orenstein et al., 1978) which is responsible
for complement activation (Boyle and Young, 1982). This
complement activating ability of gamma NGF was shown to be

9
inhibited by Cl-Inh (Boyle and Young, 1982). To date, gamma
NGF has been reported to mediate a number of activities
associated with inflammation, i.e., enhanced wound healing
(Li et al., 1980), and both i_n vivo and in vitro
leukocyte chemotaxis (Boyle et al., 1985; Gee et al., 1983).
However, gamma NGF is not known to be a member of any
physiological cascade system. Similarly, Cl-Inh is not
known to inhibit proteases not associated with cascade
systems. In order to determine if Cl-Inh was acting as a
serpin (or nexin), the nature of the interaction of gamma
NGF and Cl-Inh seen in the complement assay needed to be
examined. In this study, purified Cl-Inh was shown to
inhibit the enzymatic activity of purified gamma NGF in a
time-dependent manner with 1:1 stoichiometry. The
interaction of gamma NGF with Cl-Inh resulted in the
formation of a covalent bonded enzyme-inhibitor complex.
The properties of Cl-Inh reported in this study expand
our understanding of its physiological role as a serum
protease inhibitor. The results presented demonstrate that
Cl-Inh displays the general characteristics used to define a
protease nexin, in addition to regulating Cl activation and
the enzyme activity of fully activated Cl and nerve growth
factor.

CHAPTER II
SPONTANEOUS ACTIVATION OF PARTIALLY PURIFIED SERUM Cl
Introduction
Zymogen Cl has been shown to spontaneously activate in
the absence of either soluble or cell bound immune complexes
(Borsos et al., 1964; Cooper, 1983; Lepow et al., 1965;
Ziccardi, 1982a). Using a synthetic substrate to measure
the esterolytic activity of Cl, Lepow et al. (1965)
demonstrated that serum Cl isolated by euglobulin
fractionation spontaneously activated in a classical
autocatalytic manner. That is, zymogen Cl served as a
substrate for aCl and the enzymatic activity of the aCl led
to the conversion of zymogen Cl to aCl. By contrast, in a
recent series of experiments, Ziccardi (1982a) showed that
the activation of zymogen Cl was a concentration independent
process that was not influenced by the presence of aCl. In
those experiments zymogen Cl was reconstituted from its
isolated zymogen subunits (Cooper and Ziccardi, 1977) and
activation was measured by changes in the physicochemical
properties of Clr and Cls (Arlaud et al., 1980; Sim, 1981;
Sim and Porter, 1976; Valet and Cooper, 1974). This assay
requires a large input of zymogen Cl and is not as sensitive
as the Cl hemolytic titration (Borsos and Rapp, 1967) or
synthetic substrate enzymatic cleavage assays (Lepow et al.,
10

11
1965). The reason for the discrepancy between Lepow's and
Ziccardi's findings has not been established but may be
related to the source of zymogen Cl used and/or the method
chosen for measuring Cl activation.
In the studies presented here, the Cl hemolytic assay
was was shown to be an effective method to discriminate
zymogen from activated Cl and was used to monitor the
activation of serum Cl. Partially purified serum Cl was
shown to spontaneously activate in a concentration-dependent
manner and that the addition of aCl to the zymogen Cl did
not increase the extent of spontaneous activation. These
studies suggest a model of Cl activation that is dependent
on an intermolecular interaction between zymogen Cl
molecules which is independent of the enzymatic activity of
either the aClr or aCls subunits.
Materials and Methods
Reagents
Sheep erythrocytes (E) were collected and washed as
described in Borsos and Rapp (1967). Rabbit antiserum rich
in IgM ant.i-E was produced and depleted of IgG by passage
over a Protein A-Sepharose (Pharmacia, Upsala, Sweden)
affinity column. Ascites fluids containing mouse monoclonal
IgM antibodies specific for E were derived from a hybridoma
(TIB 110) purchased from ATCC, Washington, DC (contributed
by R. Raschlce) Human serum was prepared from outdated
plasma (Civitan Regional Blood Center, Gainesville, FL).

12
Guinea pig complement was obtained from Pel-Freez
Biologicals, Rogers, AR. Isolated complement components
were purchased from Cordis Biologicals, Miami, FL. Isotonic
veronal buffered saline containing metals (VBS) and
sucrose-veronal buffer (SVB) (pH 7.35, with 0.1% gelatin
(Difco), 0.001 M Mg++, and 0.00015 M Ca++), and
veronal-buffered saline containing 0.01 M ethylenediamine
tetraacetic acid (VBS-EDTA) were prepared as described in
Borsos and Rapp (1967). A reagent which was devoid of whole
complement activity but provided functional complement
components C3-C9 was obtained by diluting serum in VBS-EDTA
(CEDTA) (Borsos and Rapp, 1967). Indicator cells for the Cl
hemolytic assay were prepared by the method of Borsos and
Rapp (1967) in which E were sensitized with the antibody
solution, coated with aCl, incubated in CEDTA, and washed.
This procedure yields antibody sensitized E (EA) with the
activated fourth component of complement bound to their
surfaces (EAC4b)(contributed by the added serum), without
contaminating Cl and C2 (removed by the added VBS-EDTA) or
any later acting complement components (Borsos and Rapp,
1967 ) .
Isolation of Serum Fraction Rich in Zymogen Cl
Fractions rich in zymogen Cl were isolated from human
serum by ion-exchange chromatography. Human serum diluted
1:2 in distilled water and applied to a column of DE-52
cellulose equilibrated at pH 7.3 with 0.01 M

13
N-2-hydroxlethyl piperazine-N'-2-ethanesulfonic acid (HEPES)
and 0.075 M NaCl was eluted with equlibration buffer at
4C. This procedure depletes the serum of Cl-Inh (see
Chapter III) and yields a preparation of zymogen Cl that
will spontaneously self-activate (see Figure 8). Throughout
the procedures, the isolation and activation state of the Cl
was monitored using the hemolytic assay described below.
Measurement of the Hemolytic Activity of aCl
The assay for aCl was carried out by incubation of
dilutions of a sample containing an unknown amount of aCl
with 1.5xl07 EAC4b in a total volume of .3 ml of SVB
(p=0.065) (Borsos and Rapp, 1967) for 20 minutes at 37C.
The hemolytic reaction was completed by the sequential
addition of excess C2 for 5 minutes at 37C followed by the
addition of excess late acting components (CEDTA) for 1 hour
at 37C. Under these conditions, the reaction is at
endpoint (Borsos and Rapp, 1967). The number of effective
aCl molecules was determined by measuring the extent of
lysis of the EAC4b indicator cells by spectrophotometrically
monitoring the cell free supernatants at 412 nm for the
release of hemoglobin.
Analysis of Cl Activation Status from Hemolytic Activity
Data
The functional activity of serum Cl and fully activated
Cl has been extensively studied and characterized (for

14
review see Loos (1982)). It has been shown from dose
response and kinetic experiments that every aCl molecule is
capable of generating a lytic site (Borsos and Rapp, 1963;
Colten et al., 1967) and that one such site is necessary and
sufficient for cell lysis (Borsos and Mayer, 1962). The
hemolytic assay (described above) provided all the other
components necessary for completing the lysis of all the
cells in the reaction mixture, i.e. pseudo-first order for
aCl activity, and the extent of hemolysis was measured at
end point. The physical conditions, eg., pH, ionic
v -
strength, and temperature, of each reaction step were
carefully controlled to ensure the reproducibility of the
reaction. In meeting these criteria the distribution of aCl
molecules among the cells would conform to the predictions
of the Poisson distribution (Rapp, 1955) and consequently it
was possible to relate the percentage of cells lysed (y) to
the absolute number of hemolytically effective aCl molecules
added to the assay by the relationship
Z = In ( 1 y ) .
In this equation Z represents the average number of lytic
sites per cell (Borsos and Mayer, 1962) and the total number
of hemolytically effective aCl molecules in the system was
determined by multiplying Z times the number of cells in the
assay. Further considerations of the activity of aCl and
serum Cl in the hemolytic assay are dealt with more fully in
the results and discussion sections below.

15
Results
Activity of Serum and Activated Cl in the Hemolytic
Assay
In the Cl hemolytic assay, aCl yielded one-hit
characteristics in which there was a 1:1 relationship
between the relative amount of aCl added to the assay and
the number of effective Cl induced lytic sites produced (see
Figure 3). This relationship was observed irrespective of
whether aCl was diluted or the number of indicator cells was
changed (see Figure 4 and Table 1). These findings fully
confirm the previous work of others (Borsos et al., 1980;
Borsos et al., 1964; Loos, 1982).
Serum Cl titrated under identical conditions did not
follow a similar pattern. As shown in Figure 3, the
quantity of aCl activity detected fell off more quickly than
predicted as serum was diluted. This was not due to the
heterogeneous nature of the binding specificities of the
sensitizing antibodies, because indicator cells prepared
with monoclonal antibodies yielded similar results (see
Figure 5). Additionally, the excessive loss of Cl activity
upon diluting serum Cl was not due solely to dissociation
since, as is shown in Figure 4 and Table 1, the total number
of effective Cl molecules detected at any dilution was
dependent upon the indicator cell number used in the assay.
In agreement with earlier studies (Loos et al., 1972), mild
treatment of serum Cl coated EAC4b indicator cells with
trypsin resulted in the appearance of an increased titer of

-In (l-y) Cz Valued
16
y-i.
Figure 3. THE HEMOLYTIC TITRATION OF SERUM Cl AND ACTIVATED
Cl. Various dilutions of either serum or activated Cl were
incubated with EAC4b for 20 minutes at 37C. The extent of
hemolysis was determined at end point following addition of
excess C2 and CEDTA. ( serum and ffi activated Cl)

17
C O NCENTR ATION
Figure 4. MEASUREMENT OF SERUM AND ACTIVATED Cl ACTIVITY AS
A FUNCTION OF INDICATOR CELL NUMBER. Various dilutions of
either activated Cl (panel A) or serum (panel B) were
measured for Cl activity using various concentrations of
EAC4b indicator cells. ( a 24x10, 3 6.0x10, and
1.5x10 )

18
Table 1. MEASUREMENT OF SERUM AND ACTIVATED Cl ACTIVITY AS
A FUNCTION OF INDICATOR CELL NUMBER. A fixed dilution of
activated Cl or serum Cl was assayed using various
concentrations of indicator cells (EAC4b). These results
were calculated from the data presented in Figure 4.
Number of EAC4b
Target Cells
1.5 x 107
6.0 x 107
24.0 x 107
Number of Hemolytic Effective
Cl Molecules/ml Measured
Activated Cl Serum Cl
3.00 x 107
2.88 x 107
2.88 x 107
17.0 x 107
13.8 x 107
7.2 x 107

-In (l-y) Cz ValueH
19
Figure 5. HEMOLYTIC ACTIVITY OF SERUM ON POLY- AND
MONOCLONAL ANTIBODY COATED INDICATOR CELLS. Various
dilutions of serum were assayed for Cl activity on EAC4b
indicator cells that had been prepared with either
polyclonal rabbit anti-E (A ) or monoclonal mouse anti-E
( 9 ) The activity of activated Cl on monoclonal
sensitized indicator cells ( 0 ) was included as a
reference.

20
Cl that had characteristics of aCl (see Figure 6). These
findings confirmed those of Loos and colleagues (Borsos et
al., 1980; Loos, 1982; Loos et al., 1972) and demonstrated
that the Cl hemolytic assay was highly sensitive and was
able to distinguish native and activated Cl in the presence
of many other serum proteins. In other experiments it was
confirmed that aCl bound to indicator cells at 4C and
shifted to 37C yielded activity without a lag, while Cl
activity in serum that had been bound to EAC4b cells at 4C
and moved to 37C demonstrated a lag period that was
consistant with the time required for Cl to undergo an
activation process (see Figure 7). These functional
properties of native and activated Cl have been used in the
remainder of the study to follow the activation status of
various preparations of serum Cl.
Spontaneous Activation of Partially Purified Serum Cl
Human serum was diluted 1:2 with distilled water and
applied to a column of DE-52 previously equilibrated with
.075 M NaCl and 10 mM HEPES pH 7.4 at 4C. The
pass-through material contained approximately 80% of the
serum proteins, as monitored by OD2gQ and include the
majority of the measurable Cl activity. The amount of Cl
activity in a single dilution of the partially purified
serum Cl was monitored in the Cl hemolytic assay, along with
appropriate dilutions of unfractionated serum and aCl. When
incubated for various times at 37C, the relative amount of

21
Figure 6. EFFECT OF TRYPSIN TREATMENT ON THE HEMOLYTIC
ACTIVITY OF CELL BOUND SERUM Cl AND ACTIVATED Cl. Various
dilutions of serum and activated Cl were incubated with
EAC4b for 20 minutes at 37C in SVB buffer. The cells were
washed and treated with trypsin (1 pg/ml) (closed symbols)
or buffer (open symbols) for 5 minutes at 27C. The cells
were again washed, resuspended in SVB, and assayed for cell
bound Cl activity as described in the Methods. ( $ and o -
serum treated cells, m and activated Cl treated cells)

y ) Cz Valu]
were removed and added to buffer containing excess C2 at
37C. After an additional 5 minute incubation at 37C,
CEDTA was added and the reaction was allowed to proceed for
60 minutes, after which the amount of lysis that had occured
was determined by 0D4^2*

23
Cl activity in a single dilution of partially purified serum
Cl increased, while the relative Cl activity found in serum
or aCl dilutions remained constant (see Figure 8). This
activation of partially purified serum Cl upon incubation at
37C could be seen more dramatically when the various Cl
solutions were titered in the Cl hemolytic assay. In Figure
9 it can be seen that the partially purified serum Cl
expressed an increased titer of Cl activity in the hemolytic
assay and demonstrated the characteristics of aCl.
Unfractionated serum incubated under identical conditions
did not self-activate (data not shown). Similar results
were obtained when Cl was partially purified from serum by
precipitation with 40% saturated ammonium sulphate or by
ultracentrifugation (100,000 x G) (data not shown). These
results suggest that Cl can be partially purified from serum
and remain zymogen; however, once separated from certain
serum proteins, this zymogen Cl undergoes self-activation.
Concentration Dependence of Serum Cl Activation
In the next series of experiments, the effect of the Cl
concentration on the process of spontaneous activation of
the DE-52 separated material was tested. Serum Cl was
isolated as described above, and divided into three pools,
and diluted 1:4, 1:16, and 1:64 respectively. Each pool was
preincubated at 37C for various times before testing at
equal final dilutions for Cl activity. The results
presented in Figure 10 indicate that the rate of spontaneous

24
Figure 8. HEMOLYTIC ACTIVITY OF SERUM, ACTIVATED Cl, AND
PARTIALLY PURIFIED SERUM Cl AFTER INCUBATION AT 37C. A
single dilution (1:4 in VBS) of activated Cl ( ), serum
( A ), and partially purified serum Cl ( 9 ) were incubated
37C. At appropriate times, aliquots were taken, diluted,
and assayed for Cl activity in the hemolytic assay. A
single dilution of each source is represented at each time
point.

25
RELATIVE Cl CONCENTRATION
Figure 9. HEMOLYTIC ACTIVITY OF PARTIALLY PURIFIED SERUM Cl
AFTER INCUBATION AT 37C. A single dilution (1:4 in VBS)
of partially purified serum Cl was incubated at 37C. At
various times, aliquots were removed and diluted. The
samples were then titered in the Cl hemolytic assay.
( 0, 2, A 5, a 10, A 20, o 60 minutes)

RELATIVE Cl CONCENTRATION
Figure 10. EFFECT OF DILUTION ON THE RATE OF SPONTANEOUS
ACTIVATION OF PARTIALLY PURIFIED SERUM Cl. Partially
purified serum Cl was split into three aliquots and diluted
either 1:4 (panel A), 1:16 (panel B), or 1:64 (panel C).
These three solutions were then incubated at 37C and at
various times, aliquots were removed and diluted further
(1:512 final dilution) and stored on ice. At the end of the
incubation all aliquots were assayed for Cl activity in the
hemolytic assay. ( A 0, 20, 60 minutes)

27
activation of partially purified serum Cl is influenced by
the concentration of the Cl in the system, i.e., the
activation was concentration dependent since after 20
minutes at 37C the 1:4 dilution was activated (slope = 1),
while the 1:64 dilution was not (slope>l). Upon further
incubation, however, the 1:64 dilution was shown to
self-activate. As all the solutions were assayed at the
same final dilutions, the rate of activation was clearly
dependent on the concentration of the partially purified
serum Cl in the preincubation solution. These results
suggest that some interaction between Cl molecules is
required for activation.
Effects of Activated Cl on Sejrum Cl Activation
In earlier experiments, using euglobulin precipitates,
Lepow et al. (1965) suggested Cl would activate by an
autocatalytic reaction, i.e., aCl acted on zymogen Cl to
generate more aCl activity. To test this possibility, a low
level of aCl was added to a dilution of partially purified
serum Cl and incubated at 37C. As shown in Figure 11,
addition of aCl to this mixture did not result in increased
level of activation of the partially purified serum Cl. The
level of zymogen Cl chosen for this experiment was below the
concentration that would self-activate during the time
course of the experiment (see Figure 10). Rather, the
resultant Cl activity of the mixture represented the sum of
the respective Cl activities of the zymogen and added aCl.

28
Figure 11. EFFECT OF ENZYMATICALLY ACTIVE Cl ON THE
SPONTANEOUS ACTIVATION OF PARTIALLY PURIFIED SERUM Cl.
Activated Cl or buffer was added to partially purified serum
Cl (diluted 1:400 in VBS), incubated 20 minutes at 37C,
and assayed for Cl activity by the hemolytic assay.
( 9 partially purified serum Cl, activated Cl,
A- partially purified serum Cl + activated Cl,
- - additive activity of partially purified serum Cl and
activated Cl separately)

29
Taken together, the results in Figures 10 and 11 indicate
that Cl activation is neither a concentration independent
autoactivation reaction nor a classical autocatalytic
reaction. This suggests that Cl activation requires the
interaction of more than one Cl molecule but the interaction
is not one that is characteristic of an enzyme-substrate
interaction. These findings are consistent with the
observation that Cl can be demonstrated in self-activatable
and non-self-activating forms (Borsos et al., 1980; Loos et
al., 1972).
r1
Discussion
When assayed under pseudo-first order conditions in
hemolytic assays, the fully activated form of Cl exhibits
all the characteristics of a single hit reaction (Borsos and
Mayer, 1962). The behavior of activated Cl in the hemolytic
assay is summarized below:
1) Activated Cl is 100% efficient in the generation of
lytic sites on the sensitized red cell membrane (Colten et
al., 1967).
2) There is a direct proportionality between the
quantity of Cl added and the number of lytic sites generated
(Colten et al., 1967).
3) The number of effective Cl molecules measured in the
hemolytic assay is independent of the number of indicator
cells used (Loos, 1982).

30
These results indicate that Cl conforms to all the
predictions of the one hit theory of immune hemolysis
originally proposed by Borsos and Mayer (1962).
In contrast, when Cl in serum is measured in the
hemolytic assay under conditions that are pseudo-first order
for Cl its behavior does not conform to the predictions of
the one hit theory. The results observed with serum
(zymogen) Cl are summarized below:
1) The number of lytic sites generated is not directly
proportional to the quantity of serum Cl added (Heinz et
al., 1984; Hoffman and Stevland, 1971; Loos, 1982; Ratnoff
and Lepow, 1957).
2) Using the same dilution of serum Cl, the measured Cl
activity is dependent on the number of indicator cells used
in the assay (Loos, 1982; Loos et al., 1972).
3) Two populations of serum Cl can be detected on
indicator cells incubated with serum. The first, a
productive self-activatable Cl molecule, and the second, a
non-self-activatable molecule. The latter form of Cl may be
converted to active Cl by mild treatment with trypsin or
other proteases (Loos, 1982; Loos et al., 1972).
The differences between aCl activity and zymogen Cl
activity were not due to the heterogeneous (polyclonal)
nature of the antiserum normally used to prepare the EAC4b
indicator cells used in the Cl hemolytic assay. The results
presented here demonstrate that the behavior of both aCl and
serum Cl was not changed when monoclonal sensitizing

31
antibodies rather than the polyclonal were used to prepare
the indicator cells.
The differences in aCl and zymogen Cl activity were
used to monitor the activation of Cl either in serum or
following partial purification. It was shown that serum Cl
could be partially purified from other serum proteins
without being fully activated in the process. This
partially purified serum Cl was seen to undergo a
spontaneous self-activation process that was
time-and-concentration-dependent. This activation required
some form of Cl to Cl interaction but activation was not
promoted in a classic autocatalytic fashion. These findings
are distinct from either the results of Ziccardi (1982a) or
of Lepow et al. (1965). The disparity between the findings
reported here and those of Ziccardi may be due to the
characteristics of the assays used to measure Cl activation.
The change in activation rate seen in these studies was
most noticeable at concentrations of Cl below the detection
limits of the physicochemical assay used by Ziccardi (Cooper
and Ziccardi, 1977; Ziccardi, 1982a; Ziccardi and Cooper,
1976). Moreover, due to the one-hit nature of the aCl
molecule, the hemolytic assay has the potential of being
more sensitive for aCl than zymogen Cl. This has been shown
in a similar hemolytic assay system in which IgG sensitizing
antibodies are compared to IgM sensitizing antibodies for
the ability to induce complement mediated lysis. The
fixation of Cl to IgG is a two-hit phenomenon in which one

32
Cl molecule must bind two IgG molecules on the surface of
the target cell to induce activation (slope=2 in the
hemolytic assay). The activation of Cl on IgM sensitizing
antibodies is a one-hit phenomenon (slope=l) and as such has
been shown to be at least 100 times more efficient than IgG
in activating Cl. Thus, a contamination of 1% specific IgM
in an IgG sensitizing antibody reagent would completely mask
the contribution of the IgG antibodies in the assay
(slope=l). Similarly, aCl is more effective than zymogen Cl
in the Cl hemolytic assay.
In agreement with Ziccardi's findings (Ziccardi,
1982a), however, in this study no evidence for an
autocatalytic activation reaction, in which the the rate of
activation would increase with the addition of aCl, was
observed. It is possible that Hageman factor, which has
since been shown to spontaneously activate in a classical
autocatalytic manner (Silverberg et al., 1980) which can
then activate serum Cl (Ghebrehiwet et al., 1983), could be
responsible for the autocatayltic generation of enzyme
activity in the euglobulin preparation used in Lepow's
studies. Furthermore, the synthetic substrate,
N-acetyl-l-tyrosine ethyl ester (ATEE), used in these
studies is not particularly sensitive in detection activity
of Cl (McRae et al., 1982), therefore the presence of any
contaminating autocatalytically activated enzyme that is
more active (i.e. Hageman factor) would mask the low level
of serum Cl activity.

33
The differences reported in the various studies will
only be resolved when methods become available to compare
the Cl enzyme activity detected in the hemolytic assay,
synthetic substrate assay, and physicochemical assay can all
be carried out using the same preparation and the same
concentration of zymogen Cl under the same experimental
conditions.
The apparent lack of spontaneous activation seen with
Cl in unfractionated serum will be addressed in the
following section (Chapter III).

CHAPTER III
REGULATION OF Cl ACTIVATION
Introduction
The first component of complement (Cl) is present in
normal serum in a zymogen form (Lepow et al., 1965). Upon
binding to the Fc region of complement fixing antibodies
bound to antigen, serum Cl undergoes a series of
r' ,
intramolecular events producing activated Cl (aCl) (Borsos
et al., 1964; Heinz et al., 1984; Loos et al., 1972; Loos
and Hill, 1977; Reid, 1983). The enzymatic activity of the
aCl molecule is then capable of cleaving C4 and C2 (Patrick
et al., 1970), thereby initiating the classical complement
pathway. Other agents, eg. bacterial products (Clas and
Loos, 1981; Phillips et al., 1972) and various
compounds(Cyong et al., 1982), are also capable of
activating serum Cl. The aCl molecule is regulated in serum
by Cl-Inh (Lepow et al., 1965) which binds to the activated
subunits (aClr and aCls) inhibiting their esterolytic
activity (Harpel and Cooper, 1975), with resultant
dissociation of the aCl molecule (Harpel and Cooper, 1975;
Minta and Aziz, 1981). In order to study the activation
process of Cl, a number of researchers have attempted to
isolate Cl in its zymogen form (Borsos et al., 1964; Lepow
et al., 1965). Typically, these attempts to isolate whole
34

35
serum Cl have yielded exclusively aCl, which led to the
conclusion that upon isolation serum Cl undergoes a
spontaneous activation process (Lepow et al., 1965;
Ziccardi, 1982a). The reason that the activation of
isolated serum Cl does not occur in unfractionated serum is
that 1) the isolation process caused the serum Cl molecules
to become activated, or that 2) the isolation process
separated the serum Cl away from some form of regulation
normally found in serum. In initial experiments, sham
isolation procedures, in which serum was exposed to gel
f'
chromatography, euglobulin fraction, etc., but was left
unfractionated, the Cl failed to spontaneously activate. An
example of this observation was that ultracentrifugation of
serum resulted in a pellet rich in Cl activity, which when
resuspended in the supernatant displayed the properties of
zymogen Cl. However, resuspending the pellet in buffer
yielded Cl activity that had the characteristics of the
fully activated molecule (data not shown). The isolation
process was therefore not the sole cause for the observed
spontaneous activation.
The autoactivation of serum Cl may have an in vivo
corollary in the disease hereditary angioedema (HAE), in
which patients are shown to have low levels of functional
Cl-Inh (Gadek et al., 1980). These individuals experience
episodes of spontaneous activation of their serum Cl
(Gelfand et al., 1979), with serious physiological
consequences. The episodic nature of this disease and the

36
zymogen nature of the Cl found in asymptomatic HAE patients
(Gelfand et al., 1979), raise important questions about the
regulation of serum Cl activity in serum. A naturally
occurring factor (possibly Cl-Inh) in serum that regulates
the activation of native serum Cl would therefore be an
attractive hypothesis.
Other workers (Ziccardi, 1982b) using serum Cl reformed
from isolated Cl subunits have shown that Cl-Inh inhibited
the spontaneous activation of serum Cl to aCl. Those
studies must be viewed with caution because the serum Cl
125
molecules are reformed using modified ( I labeled)
subunits and the reformed serum Cl molecules differed
significantly from serum Cl in a number of physical
characteristics, specifically, sedimentation coeffecients
(16S and 12S for serum Cl and refomed zymogen Cl
respectively)(Colten et al., 1967; Cooper and Ziccardi,
1977), Cl-Inh kinetics (Loos, 1982; Ziccardi and Cooper,
1976), and spontaneous activation kinetics (see Chapter II;
Ziccardi, 1982a). Furthermore, the assays used in those
studies monitored the average physical changes of
populations of Cl subunits during the course of spontaneous
activation. This may or may not directly reflect the
functional changes associated with the activation of serum
Cl to aCl.
In the previous section it was shown, using the Cl
hemolytic titration assay, that crude preparations of
unmodified serum Cl from serum were able to spontaneously

37
convert to aCl. In the experiments described in this
section, various serum fractions, including purified Cl-Inh,
were shown to inhibit this spontaneous autoactivation. This
activity of Cl-Inh to inhibit serum Cl autoactivation was
distinct from its irreversible covalent interaction with
fully activated Cl.
Materials and Methods
Reagents
The reagents and materials used in these studes were
identical to those reported in the previous chapter.
Cl Assays
Titration of Cl hemolytic activity was performed by the
methods described in the previous chapter.
Isolation and Measurement of Cl-Inh
Human Cl-Inh was purified essentially by the method of
Harrison (1984), using DE-52 in place of DEAE ion-exchange
chromatography. Specifically, outdated plasma was converted
to serum by the addition of excess Ca++ and the
supernatant of a 5% polyethylene glycol precipitation was
passed over a lysine-sepharose affinity column. Tne
pass-tnrougn traction was tnen suojectea to ion-exchange
cnromatograpny on DE-b2 equilibrated in 0.1 M KCl and 10 mM
phosphate pH 7.0 with 0.01 M benzamidine. Bound proteins
were eluted with a linear gradient of 0.1 to .25 M KCl in

38
the same phosphate buffer. Functional levels of Cl-Inh in
various fractions were monitored by the method of Gigli,
et.al. (1968). A radioimmunoassay was developed to measure
the antigenic levels of Cl-Inh in the various fractions. In
this assay aliquots of solutions containing unknown amounts
of Cl-Inh were added to Immuno-beads (Bio-Rad) coupled with
partially purified Cl-Inh followed by the addition of
commercially prepared Cl-Inh specific rabbit antiserum
(Pel-Freze). After incubation at 37C for 45 minutes and
extensive washing of the beads in VBS-EDTA, the amount of
rabbit antibodies remaining bound to the beads was
125
determined by the amount of I-Protein A that would
specifically bind to the beads in a 45-minute incubation at
37C. The relative concentration of Cl-Inh in the unknown
was determined by relating the inhibition of the binding of
the rabbit antibodies to the beads caused by the fluid phase
Cl-Inh compared to the standard curve of inhibition
generated by the addition of standard amounts of partially
purified Cl-Inh (see Figure 12). This assay could detect
concentrations of Cl-Inh in solutions of less than 1 pg/ml
(data not shown). Fractions showing Cl-Inh activity were
pooled and applied to a G-150 molecular sieving column.
Fractions containing Cl-Inh activity were pooled and further
purified by hydroxylapatite column chromatography.
Fractions containing Cl-Inh activity were pooled. This
material appeared as a single homogeneous band in SDS-PAGE
(Laemmeli, 1970) (Figure 13).

inhibition of 125i-pa binding
Figure 12. RADIOIMMUNE INHIBITION ASSAY FOR Cl-INH. The
inhibition of rabbit anti-Cl-Inh antibodies adhering to
Cl-Inh coupled^immuno-beads by fluid phase Cl-Inh was
monitored by I-protein A binding. (Error bars denote
the range of duplicate samples). 50% inhibition was
observed with approximately 20 ng of Cl-Inh added.
I

40
Figure 13. SDS-PAGE OF THE VARIOUS STEPS IN THE
PURIFICATION OF Cl-INH. Pictured is the Coomassie stain of
a SDS-PAGE analysis of aliquots from the various stages of
the purification of Cl-Inh from human serum (lanes 1 and 8 -
molecular weight standards, lane 2 serum, lane 3 PEG
supernate, lane 4 lysine-sepharose flow through fraction,
lane 5 DE-52 eluate pool, lane 6 G150 eluate pool, lane
7 hydroxylapatite eluate pool)

41
Human Serum Fractionation
Partially purified serum Cl was prepared by
ion-exchange chromatography described in the previous
chapter. Proteins bound to the ion-exchange column were
eluted by washing the column with equlibration buffer
containing 200 mM NaCl. Other sources of Cl depleted human
serum fractions were produced by eugologulin precipitation
(supernatant of serum dialysed against water) and
ultracentrifugation (supernatant of serum spun for 4 hours
at 100,000 x G).
Results
Inhibition of Spontaneous Activation of Zymogen Cl by
Serum Fractions
To address the role of serum factors in maintaining the
zymogen nature of unfractionated serum Cl, Cl depleted serum
fractions were assayed for their ability to inhibit the
spontaneous activation of partially purified serum Cl (DE-52
pass-thorugh). Human serum was fractionated to remove Cl
activity by euglobulin precipitation (eublobulin
supernatant), ultracentrifugation (supernatant of 100,000 x
G), or ion-exchange chromatography (DE-52 eluate). The
various serum fractions were added to partially purified
serum Cl. The solutions were incubated one hour at 37C
and the ability of each fraction to inhibit the conversion
of serum Cl activity (as shown by a Cl titration slope of

42
1.3) to aCl activity (as shown by a Cl titration slope of
1.0) in subsequent Cl hemolytic titration assays was
measured. As can be seen in Figure 14, at the highest
concentrations tested all three Cl depleted serum fractions
(euglobulin supernatant, ultracentrifuged supernatant, and
DE-52 eluate) were able to inhibit the spontaneous
conversion of serum Cl to aCl. Inhibition of the
spontaneous activation was shown to be dependent on the
concentration of the serum fraction used(l:100 dilutions of
the various serum fractions yielded aCl activity).
Spontaneous serum Cl activation was therefore inhibited by
some factor(s) present in the various serum fractions.
Inhibition of Spontaneous Activation of Zymogen Cl by
Cl-Inh
Human Cl-Inh was purified to a single band in SDS-PAGE
(see Figure 13). Addition of Cl-Inh to partially purified
zymogen Cl (DE-52 pass-through) inhibited the spontaneous
activation of Cl (Figure 15A). This inhibition of the
spontaneous activation of serum Cl by Cl-Inh was
concentration dependent. When the relative amounts of
Cl-Inh in the various dilutions of the various serum
fractions used in Figure 14 were considered, the
concentration of Cl-Inh needed to inhibit by 50% the
activation of the serum Cl was approximately equal to the
Cl-Inh concentration in the crude serum fraction from DE-52
eluate (Figure 14).

RELATIVE Cl CONCENTRATION
Figure 14. EFFECT OF VARIOUS SERUM FRACTIONS ON THE SPONTANEOUS ACTIVATION OF
PARTIALLY PURIFIED SERUM Cl. In separate experiments, various serum fractions
were added to partially purified serum Cl and then incubated one hour at 37C.
The resulting Cl activity was then assayed hemolytically. The amount of Cl
activity contributed by the various serum fractions was less than 20% of the Cl
activity of the mixture. (Panel A euglobulin supernatant of human serum added
to partially purified serum Cl, panel B .2 /M NaCl eluate of the DE-52 column
used to produce the partially purified serum Cl added back to the partially
purified serum Cl, and panel C supernatant of the ultracentrifugation of
diluted human serum added to partially purified serum Cl. In all experiments
the various serum fractions were assayed at 1:10 ( O ) and 1:100 ( ) dilutions
and are graphed with the unfractionated serum control ( )).

-ln(l-y) CZ value!
44
A B
RELATIVE Cl CONCENTRATION
Figure 15. EFFECTS OF PURIFIED Cl-INH ON THE SPONTANEOUS
ACTIVATION OF PARTIALLY PURIFIED SERUM Cl AND ON ACTIVATED
PARTIALLY PURIFIED SERUM Cl. Purified Cl-Inh (at various
dilutions) or buffer was added to partially purified serum
Cl which had been preincubated for 20 minutes at either 4C
(panel A) or 37C (panel B). Following a 20 minute
incubation at 37C the resultant Cl activity was monitored
in the hemolytic assay (dilutions of Cl-Inh used: 1:5
( ), 1:25 ( ), 1:125 ( O ), 1:625 ( A ), buffer ( A*v) ) )
For these studies, the. undiluted preparation of Cl-lTih
contained an equivalent concentration of inhibitor normally
present in human serum (0.1 mg/ml)

45
The action of Cl-Inh in preventing the conversion of
partially purified serum Cl activity to aCl activity in the
preincubation step was not due to its action on aCl. As
mentioned above, addition of Cl-Inh to partially purified
serum Cl solutions, followed by incubation for 20 minutes at
37C, yielded Cl activity similar to that found in serum
(Figure 15A). However, in parallel studies if the serum Cl
was preincubated in the absence of Cl-Inh for 20 minutes at
37C, then further incubation in the presence of Cl-Inh did
not result in the appearance of serum Cl activity (Figure
15B). Rather, under these conditions, Cl-Inh reduced the
aCl activity in a concentration dependent fashion. Similar
results were seen when Cl-Inh was added to purified aCl
(data not shown). The only difference in the experiment
that generated the results of Figure 15A and 15B was that
the serum Cl of 15A was preincubated on ice for 20 minutes
while the serum Cl of 15B was preincubated for 20 minutes at
37 prior to the addition of the Cl-Inh. These results
showed that the action of Cl-Inh in inhibiting the
conversion of partially purified serum Cl to aCl activity
was not due to the interaction of Cl-Inh with aCl alone or
an artefact of the assay system. Furthermore, the amount of
Cl-Inh needed to inhibit the activation of the partially
purified serum Cl was less than that needed to inhibit the
Cl activity after the spontaneous activation (see Figure 15
A and B).

46
Discussion
These studies extend the observations made in the
previous chapter in which crude isolates of Cl from serum,
upon incubation at 37C, spontaneously converted to
activated Cl, as measured functionally in the hemolytic
assay. This spontaneous autoactivation was not observed
with unfractionated serum Cl, i.e. human serum. There are
two potential explanations for why spontaneous Cl activation
does not occur in serum: 1) the isolation process somehow
altered the serum Cl molecule thus allowing it to self
activate, or 2) in the process of isolation, some regulatory
molecule was separated from the serum Cl, thus allowing the
serum Cl molecule to self-activate, a process normally
regulated in serum. The studies reported here support the
second hypothesis that there are regulatory factor(s) found
in serum that inhibit the spontaneous activation of Cl.
Partially purified serum Cl that self-activated when
incubated in buffer failed to activate when admixed with Cl
depleted crude serum fractions.
Due to the relationship of functional Cl-Inh levels and
Cl activation seen in HAE patients and the results from
researchers using reassociated Cl molecules, Cl-Inh was
examined as the most likely regulator of serum Cl
self-activation. It was found that when Cl-Inh was
incubated with partially purified serum Cl, activation could
be prevented. This regulatory activity of Cl-Inh was

47
independent of its irreversible action on aCl. The action
of Cl-Inh is represented schematically in Figure 16.
The regulation of the conversion of zymogen Cl to aCl
by Cl-Inh was concentration dependent. The concentration of
Cl-In'n needed to maintain serum Cl from spontaneous
activation was less that that required to inactivate the
total quantity of activated Cl that could be generated in
the system. That is, if activation was allowed to occur
prior to adding Cl-Inh, then the minimal level previously
shown to inhibit spontaneous activation was insufficient to
inactivate all the aCl. This leads to the hypothesis that
Cl-Inh (Figure 16, Step I) is more effective in regulating
the generation of Cl activity in serum than it is as an
inactivator of fully activated Cl (Figure 16, Step II).
These studies have led to similar conclusions to those of
Ziccardi (1982a).
The ability of a low level of Cl-Inh to maintain serum
Cl in its zymogen form may account for the episodic nature
of the edema in HAE patients. Those patients usually have
10%-15% of the normal level of Cl-Inh. From these studies it
would be predicted that these patients have levels of Cl-Inh
sufficient to prevent continual spontaneous activation of
their serum Cl. However, if an enzyme regulated by Cl-Inh
(Figure 1) was activated in these patients there would be
insufficient Cl-Inh to effectively regulate such an enzyme.

48
Cl-Inh
(I)
Cl-Inh
(II)
V
zymogen Cl X> activated Cl
inactivated Cl
reversible
irreversible
Figure 16. SCHEMATIC OF THE REGULATORY EFFECTS OF Cl-INH ON
Cl ACTIVITY

CHAPTER IV
INACTIVATION OF THE PROTEOLYTIC ACTIVITY OF NGF BY Cl-INH
Introduction
In 1959 Levy and Lepow demonstrated an inhibitor of aCl
in human serum, which they named the Cl esterase inhibitor
(Cl-Inh). The Cl-Inh molecule has since been shown to
interact and inhibit a number of other serum proteases
involved in homeostatic cascade reactions (Travis and
Salvesen, 1983). Other serum serine esterases, e.g.
trypsin, chymotrypsin, etc., are unaffected by Cl-Inh
(Harpel and Cooper, 1975; Travis and Salvesen, 1983). The
nature of the interaction and inhibition involves the
formation of a covalent Cl-Inh-enzyme complex (Travis and
Salvesen, 1983) near the active site on the enzyme (Travis
and Salvesen, 1983). The rate of inactivation of aCl by
Cl-Inh (Sim et al., 1979; Travis and Salvesen, 1983)
compared to the enzymatic activity of aCl in cleaving C4 and
C2 (Patrick et al., 1970) has led to speculation that the
primary function of Cl-Inh is not in regulating aCl activity
(Gigli et al., 1970; Travis and Salvesen, 1983). Other
serum proteases, eg. kallikrein, are more effectively
regulated by Cl-Inh (Gigli et al., 1970). Evidence reported
elsewhere (Ziccardi, 1982b, see Chapter III) suggests that
Cl-Inh has the ability to regulate zymogen Cl activation
49

50
independent of its esterase inhibiting ability. Clearly,
Cl-Inh is a multi-functional molecule and a full
understanding of its regulatory capabilities requires
further investigation.
Recently, nerve growth factor (NGF) was shown to
substitute for aCl in initiating the classical complement
pathway (Boyle and Young, 1982). This ability to cleave and
activate C4 and C2 was dependent on the serine esterase
activity of the gamma subunit of NGF (gamma NGF) (Boyle and
Young, 1982). This activity was inhibited by preincubation
of activated gamma NGF with Cl-Inh (Boyle and Young, 1982).
Nerve growth factor is naturally found as a
multimolecular complex of three polypeptide subunits (alpha,
beta, and gamma chains, in a 2:1:1 molar ratio) (Bradshaw
and Young, 1976; Greene et al., 1969). It was first
described by Levi-Montalcini and Booker (1960) as a soluble
factor that stimulated neurite outgrowth. This nerve growth
inducing activity is mediated by the beta subunit of the
molecule (Levi-Montalcini, 1965). No known biological
activity has been reported for the alpha subunit (Isackson
and Bradshaw, 1984). The gamma subunit in addition to its
role in complement activation has been shown to act as a
plasminogen activator (Orenstein et al., 1978) and topical
administration of enzymatically active NGF has been shown to
enhance the rate of wound healing in mice (Li et al., 1980).
Other researchers have reported that the gamma subunit of
NGF stimulated polymorphonuclear phagocyte chemotaxis i_n

51
vitro (Gee et al., 1983) and .in vivo (Boyle et al.,
1985). From these studies it would appear that gamma NGF
may have important physiological roles in the inflammation
response and wound repair.
A series of cell secreted protease inhibitors (nexins)
have recently been described that bind and inactivate
specific serine esterases (Baker et al., 1980; Knauer et
al., 1983). In binding to the enzyme, the nexin inactivates
the enzyme and leads to efficient clearance via ingestion by
cells expressing receptors for the enzyme-nexin complexes
(Knauer et al., 1983). Two proteins that function as nexins
in complexing with activated gamma NGF have been reported
and partially characterized (Knauer et al., 1982).
Experiments presented in this chapter examine the
interaction of Cl-Inh with gamma NGF. It was found that
purified Cl-Inh was able to inhibit the esterolytic activity
and form covalent complexes with gamma NGF in a time
dependent fashion, with a 1:1 stoichiometry. These results
are similar to those reported for the interactions between
Cl-Inh and other enzymes.
Materials and Methods
Reagents
Cl-Inh was purified to homogeneity by the method of
Harrison (1984) as reported in Chapter III. Purified gamma
NGF from mouse submaxillary salivary glands was provided by
Dr. M. Young.

52
Enzymatic Assay
Various amounts of purified Cl-Inh were incubated with
2.6 pg of gamma NGF and the enzymatic activity of the gamma
NGF was measured by monitoring the cleavage of the synthetic
substrate, N-alpha-benzoyl-l-arginine-p-nitroanilide
(1-BAPNA). The gamma NGF containing solution was added to a
1 mM solution of 1-BAPNA in 10 mM Tris(hydroxymethyl)
aminomethane (Tris) buffer, pH 7.4, and the cleavage of the
substrate was determined by an increase in the 041g of
the solution, using a Varian spectrophotometer, with the
cuvette jacketed at 24C.
Physiochemical Assay
125
Gamma NGF was labeled with I by the
lactoperoxidase method. Radiolabeled gamma NGF and Cl-Inh
were incubated together under various conditions and the
solutions were subjected to SDS-PAGE (Weber and Osborn,
1969). The slab gels were dried and the radiolabeled gamma
NGF was visualized by autoradiography. The relative amounts
of radioactivity migrating at different rates in the gels
were measured by scanning the autoradiographs on a Joyce
Lobel densitometer. Molecular weights were determined from
Coomasse R-250 stained reference wells containing molecular
weight standards (Sigma Chemicals).

53
Results
Inhibition of Gamma NGF Enzymatic Activity by Cl-Inh
Preincubation of active gamma NGF with a ninefold molar
excess Cl-Inh for one hour at 37C effectively inhibited
the ability of the gamma NGF to subsequently cleave the
synthetic substrate 1-BAPNA (see Figure 17). The inhibition
of gamma NGF by Cl-Inh was complete within one hour of
incubation at 37C. The stoichiometry of the gamma
NGF-Cl-Inh interaction was consequently determined following
a one hour preincubation at 37C. The results of adding
various amounts of Cl-Inh to a constant amount gamma NGF,
plotted in Figure 18, showed an approximately 1:1
stoichiometry for the inhibition of gamma NGF by Cl-Inh. In
the next group of experiments the rate of inactivation of
gamma NGF by Cl-Inh was examined. In preliminary
experiments it was shown that addition of 1 mM 1-BAPNA to
gamma NGF prior to addition of Cl-Inh to the mixture
prevented the interaction of Cl-Inh with gamma NGF (data not
shown). This blockage of Cl-Inh activity was due most
probabily to competitive inhibition for the active site of
the gamma NGF. Therefore, the time course experiment was
not affected by the interaction of Cl-Inh and gamma NGF
during the enzyme assay. Using a 1:2 molar ratio of Cl-Inh
to gamma NGF, the rate of inhibition of the gamma NGF enzyme
activity was measured at 24C (see Figure 19). These
results are similar to those reported for the interaction of
Cl-Inh with aCls (Harpel and Cooper, 1975), and suggest the

54
Figure 17. ENZYMATIC ACTIVITY OF NGF AFTER PREINCUBATION
WITH Cl-INH. Cleavage of the synthetic substrate 1-BAPNA,
as measured by the generation of OD^,was monitored
for the enzymatic activity of gamma NGF after incubation for
one hour at 37C in buffer (top tracing) or after the
addition of 9 molaf excess of purified Cl-Inh (lower
tracing).

55
CCI-INffl
C NGFJ
Figure 18. CONCENTRATION DEPENDENCE OF THE INHIBITION OF
THE ENZYMATIC ACTIVITY OF NGF BY Cl-INH. Various molar
equivalents of Cl-Inh were added to a constant amount of
gamma NGF, incubated one hour.at 37C, and then assayed for
remaining enzymatic activity in Cleaving 1-BAPNA. The graph
depicts the amount of gamma NGF activity remaining after
incubation with various molar amounts of Cl-Inh .

56
Figure 19. TIME COURSE OF THE INHIBITION OF THE ENZYMATIC
ACTIVITY OF NGF BY Cl-INH. Gamma NGF was added to 1/2 molar
equivalent of Cl-Inh and incubated at 24C for various
times. The enzymatic activity of the gamma NGF was then
analyzed by monitoring the cleavage of 1-BAPNA
spectrophotometrically. The graph depicts the amount of
gamma NGF activity remaining with time (minutes).

57
interaction of gamma NGF and Cl-Inh requires an enzymatic
reaction.
Complex Formation between Gamma NGF and Cl-Inh
The apparent enzymatic nature of the interaction of
gamma NGF and Cl-Inh and the 1:1 stoichiometry described
above led to the hypothesis that the enzymatic activity of
gamma NGF was inhibited by the formation of a stable complex
with Cl-Inh. Experiments were designed to monitor the
formation of a covalent complex formed between gamma NGF
(26,000 molecular weight) and Cl-Inh (97,000 molecular
weight) using SDS-PAGE analysis. In these experiments,
125
I-gamma NGF was incubated with Cl-Inh for one hour at
37C and the mixture was electrophoresed in SDS-PAGE using
the Weber and Osborn buffer system (phosphate buffer pH
12 5
7.2). Incubation of I-gamma NGF with an 80-fold molar
excess Cl-Inh for one hour at 37C yielded greater than 97%
125
of the I migrating in a high molecular weight complex
of approximately 125,000 Daltons (Figures 20 and 21). In
order to investigate the nature of the bond formed in the
high molecular weight complex, the effects of hydroxylamine
and high pH on the stability of the complex were examined.
Preformed high molecular complexes were formed by the
125
incubation of I-gamma NGF with 1/2 molar equilivant of
Cl-Inh. An aliquot was removed and further incubated for 20
minutes at 37C in .1 M hydroxylamine pH 10.5 and
electrophoresed. The results (see Figure 22) show that the

58
116,000-
97,000-
66,000-
45,000-
29,000-
I 2 3
Figure 20. AUTORADIOGRAPH OF SDS-PAGE ANALYSIS OF
i23I-GAMMA NGF INTERACTION WITH Cl-INH. Radiolabeled
gamma NGF was incubted with various amounts of Cl-Inh (lane
1 none, lane 2 1/2 molar equivalent of cl-Inh, lane 3 -
80 molar excess Cl-Inh) and incubated one hour at 37C,
added to 2% SDS, and electrophoresed.

59
Rf
Figure 21. DENSITOMETRY OF SDS-PAGE OF 125I-NGF INCUBATED
WITH VARIOUS CONCENTRATIONS OF Cl-INH. The autoradiograph
from Figure 20 was scanned with a densitometer. The
relative areas under the curves had been shown in previous
experiments to obey Beer's law, and thus were analysed to
estimate the relative amounts of radioactivity in each peak
(panel A No Cl-Inh added, panel B- 1/2 molar equivalent of
Cl-Inh added, and panel C 80 molar excess Cl-Inh added).
Eectrophoresis proceeded from right to left.

60
Figure 22. AUTORADIOGRAPH OF SDS-PAGE ANALYSIS OF
HYDROXYLAMINE TREATED HIGH MOLECULAR WEIGHT COMPLEXES.
Radiolabeled gamma NGF was incubated with 1/2 molar
equivilant of Cl-Inh and incubated one hour at 37C (lane
1). An aliquot was removed and was made .1 M hydroxylamine
(pH 10.5) and incubated an additional 20 minutes at 37C
(lane 2). The solutions were then subjected to SDS-PAGE
separation and autoradiographed.
*r~

61
high molecular weight complex was labile to treatment with
hydroxylamine at pH 10.5, which is characteristic of an acyl
125
linkage between the Cl-Inh and I-NGF in the high
molecular weight complex.
Discussion
The proteolytic activity of gamma NGF, as measured by
the cleavage of the synthetic substrate 1-BAPNA, has been
shown to be inhibited by Cl-Inh. This inhibition was time
dependent with a stoichiometry of approximately 1:1. The
interaction between gamma NGF and Cl-Inh was shown to result
in the formation of a covalent bonded complex, as judged by
SDS-PAGE. These results would be consistent with a model of
interaction in which the inhibition of the enzymatic
activity of gamma NGF was inhibited by the association of
Cl-Inh near the active site of the molecule, followed by the
cleavage of Cl-Inh and covalent coupling to gamma NGF.
There are a number of predictions associated with such a
model. First, the cleavage of Cl-Inh by gamma NGF should
generate a new amino terminal in the Cl-Inh. Second, a new
Cl-Inh fragment should be produced by the complexing of
gamma NGF and Cl-Inh. Finally, the formation of the
covalent bond should be localized to a specific polypeptide
chain(s) in the gamma NGF.
The bond formed in the interaction of gamma NGF and
Cl-Inh was shown to be susceptible to the action of
hydroxylamine at high pH. These results are similar to

62
those seen with the acyl linkage between gamma NGF and the
protease nexins (Baker et al., 1980) or the characteristic
serpin interaction with their appropriate serine esterase.
These properties suggest that Cl-Inh may have many of the
characteristics of a protease nexin. Further studies are
required to determine if any cell type has specific
receptors for Cl-Inh-gamma NGF complexes that would indicate
that Cl-Inh closely mimicked the biological activity of a
cell-secreted nexin.

CHAPTER V
SUMMARY AND CONCLUSIONS
This investigation has examined protein interactions
important in the activation, regulation, and inactivation of
serum proteases. In addressing these topics, three basic
questions were posed and answered. First, what is the
nature of the spontaneous activation reaction of partially
r'*
purified serum Cl? Second, what is the role of Cl-Inh in
regulating the activation of serum Cl? Finally, wnat is tne
nature or tne interaction or ei-lnn witn tne activated gamma
subunit or NGF?
The process or serum Cl activation was shown to require
intermolecular interactions. This process was not
autocatalytic in that the addition of enzymatically active
Cl did not alter the extent of serum Cl activation. The
studies presented here confirm the work of others (Borsos et
al., 1980; Borsos et al., 1964; Heinz et al. 1984; Loos et
al., 1972) and clearly show that the level of activatable
zymogen Cl is not simply limited by the association of the
Clq subunits with Clr2s2 complexes, as has been
proposed by Ziccardi (1982a). The concentration dependence
of the Cl activation reaction has not been adequately
explained. Borsos et al. (Borsos et al., 1980; Loos et al.,
1972) suggest that heterogeneity in the Clq subunits could
63

64
account for the observed behavior. The specific subunit
interactions needed for activation of macromolecular serum
Cl were not established in this study. This is an important
question that merits further investigation.
In this investigation, the activity of Cl-Inh was shown
to be broader than previously appreciated. The serine
esterase inhibitor activity of Cl-Inh has been well
documented (Harpel and Cooper, 1975). However, the rate of
inactivation of aCl by Cl-Inh is slow (Sim and Porter, 1976)
compared to the activity of aCl on C4 and C2 (Patrick et
al., 1970), and it is thought that the true physiological
role of the serpin activity of Cl-Inh is in inactivating the
enzymatic activity of kallikrein (Gigli et al., 1970). The
results reported in this study present a new role of Cl-Inh
in regulating serum Cl activity which entails the regulation
of the activation of zymogen Cl. This regulation of serum
Cl activation was distinct from the activity of Cl-Inh in
inactivating aCl. In regulating both the activation and
resultant activity of serum Cl, Cl-Inh is able to
effectively control the initiation of the classical
complement pathway. Furthermore, it has been seen by others
(Harpel and Cooper, 1975; Minta and Aziz, 1981) that the
interaction of aCl with Cl-Inh not only inactivates the aClr
and aCls subunits, but also dissociates the complexed
subunits from the Clq subunit. Though enhanced clearance of
complexes of Cl-Inh with aClr or aCls has not been observed,
the dissociation of the aCl macromolecule would enhance the

65
binding of Clq to Clq receptors on phagocytes. Thus, Cl-Inh
may be considered a nexin of sorts, in that by binding to
the aCl macromolecule and dissociating it, the Clq is more
readily cleared by binding to specific cell receptors.
The interaction of Cl-Inh with gamma NGF led to the
inactivation of the enzymatic activity of the gamma NGF.
Associated with the inactivation was the formation of a
gamma NGF-Cl-Inh complex that was resistant to SDS and
susceptible to hydroxylamine treatment. This interaction is
similar to that seen with Cl-Inh and aCls (Harpel and
Cooper, 1975). In binding the aCls, Cl-Inh becomes cleaved
and new carboxy and amino terminals are formed (Salvasen et
al., 1985). By sequencing these new fragments a partial
amino acid sequence of the enzymatic target site on the
Cl-Inh has been reported (Carrell, 1984; Salvesen et al.,
1985). Other than an arginine residue where the serine
esterase cleaves, this target site sequence is unrelated to
any other enzymatic target site reported for other serpins
(Carrell, 1984). The similarities between the target sites
of the various enzymes inhibited by Cl-Inh could provide
information regarding the specificities of both the
proteases and the Cl-Inh inactivation activity.
Additionally, knowledge of the fine structure of the
enzymatic target site of some serpins has led to new
research in producing more effective protease inhibitors by
site specific mutagenesis and genetic recombination
(Rosenberg et al., 1984). The production of a molecule that

66
could more effectively control the activation of the
classical complement pathway would have great benefits in
the treatment of HAE and other disorders resulting in
inappropriate Cl activation.
The interaction of Cl-Inh with the gamma subunit of
nerve growth factor is characteristic of that of most serine
protease inhibitors (Travis and Salvesen, 1983). There is a
1:1 stoichiometric complex formed between Cl-Inh and NGF and
this complex is stabilized by the formation of a covalent
bond between the two proteins. Recently, a new group of
cell secreted protease inhibitors known as protease nexins
(PN) have been described (Baker et al., 1980). Three
distinct protease nexins (PNl, PN2, and PN3) have been
identified (Knauer et al., 1983) and their physiological
function appears to be inhibition of the enzyme activity of
selected proteases by complex formation followed by
enhancement of the specific clearance of PN-protease
complexes by a variety of cells (Knauer et al., 1983). One
of these nexins, PN3, reacts in a highly selective manner
with the gamma subunit of NGF (Knauer et al., 1982). The
Cl-Inh-NGF and the PN3-NGF complexes share a number of
similarities and future studies should be directed towards
determining whether Cl-inactivator-NGF complexes have other
biological properties of PN3-NGF complexes.
In the comparison of the properties of nerve growth
factor and Cl a number of similarities have emerged. First,
both can activate the classical complement pathway (Boyle

67
and Young, 1982; Lepow et al., 1965) and their enzyme
activity can be regulated by the Cl inhibitor protein (Boyle
and Young, 1982; Lepow et al., 1965). Second, both are
produced as zymogens (Lepow et al., 1965; Young and Koroly,
1980). Third, both zymogen NGF and Cl undergo
self-activation reactions (Lepow et al., 1965; Young and
Koroly, 1980; Ziccardi, 1982a). In the case of NGF this is
a classical autocatalytic reaction in which the removal of
the Zn++ associated with the zymogen (Young and Koroly,
1980) allows the NGF to spontaneously activate in a biphasic
reaction (Young and Koroly, 1980). Initially, there is a
low level of intramolecular spontaneous activation (Young
and Koroly, iyU), ronowea oy a rapid classical
autocataiytic reaction, in wnicn zymogen /S NGF serves as a
substrate ot activated 7S NGF to produce more activated NGF
(Young and Koroly, lySD). The autocatalysis of the molecule
can be inhibited by inhibiting the esterolytic activity of
the gamma subunit (Young and Koroly, 1980). One would
predict that addition of Cl-Inh to zymogen 7S NGF would
inhibit the autocatalytic activation (enzymatic activation).
However, because of the results reported for Cl in Chapter
III, the possibility would also exist that Cl-Inh may
inhibit the first phase of the activation (spontaneous
activation) of 7S NGF. The regulation of activation of
other zymogen serum proteases by Cl-Inh warrants further
investigation.

68
The many parallels and similarities between the
structure, function, and regulation of NGF and Cl suggest
that nerve growth factor may play an important physiological
role totally distinct from its trophic effect on the sensory
and sympathetic nervous system. The possibility that NGF
may be part of a heretofore unrecognized system involved in
homeostatic reactions merits closer scrutiny. To date
Cl-Inh has only been recognized to regulate enzymes within
cascade systems, see Figure 1, and this raises the
possibility that NGF is part of some unknown cascade system.
j f.
A number of recent observations indicating that NGF is
chemotactic for polymorphonuclear leukocytes ini vivo
(Boyle and Young, 1982) and iri vitro (Gee et al. 1983)
coupled with the reported ability of the growth factor to
accelerate wound healing (Li et al., 1980) suggest NGF may
be a potential mediator of a variety of normal responses
subsequent to tissue damage.

BIBLIOGRAPHY
Arlaud, G.J., C.L. Villiers, S. Chesne, and M.G. Colomb.
1980. Purified proenzyme Clr: Some characteristics of its
activation and subsequent proteolytic cleavage. Biochem.
et Biophys. Acta 616:116-129.
Baker, J.B., D.A. Low, R.L. Simmer, and D.D. Cunningham.
1980. Protease-nexin: A cellular component that links
thrombin and plasminogen activator and mediates their
binding to cells. Cell 21:37-45.
Borsos, T., M. Loos, R.M. Chapuis, R. Medicus, and J.
Isliker. 1980. A novel way of relating the structure of
Clq to the hemolytic activity of the first component of
complement. Molec. Immunol. 17:1415-1421.
Borsos, T., and M.M. Mayer. 1962. Mechanism of action of
guinea pig complement. In Mechanism of Cell and
Tissue Damage Produced by Immune Reactions Ilnd
International Symposium on Immunopathology. Benno
Schwabe Co., Basel (Switzerland), pp. 13-22.
Borsos, T., and H.J. Rapp. 1963. Chromatographic separation
of the first component of complement and its assay in a
molecular basis. J. Immunol. 91:851-858.
Borsos, T., and H.J. Rapp. 1967. Immune hemolysis: A
simplified method for the preparation of EAC'4 with
guinea pig or human complement. J. Immunol. 99:263-268.
Borsos, T., H.J. Rapp, and U.L. Walz. 1964. Action of the
first component of complement: Activation of C'l to C'la
in the hemolytic system. J. Immunol. 92:108-112.
Boyle, M.D.P., M.J.P. Lawman, A.P. Gee, and M. Young. 1985.
Nerve growth factor: A chemotactic factor for
polymorphonuclear leukocytes in vivo. J. Immunol.
134:564-568.
Boyle, M.D.P., and M. Young. 1982. Nerve growth factor:
Activation of the Classical complement pathway by
specific substitution for component Cl. Proc. Natl. Acad.
Sci. 79:2519-2522.
69

70
Bradshaw, R.A., and M. Young. 1976. Nerve growth factor:
Recent developments and perspectives. Biochem. Pharmacol.
25:1445-1449.
Carrell, R.W.. 1984. Therapy by instant evolution. Nature
312:14.
Cas, F., and M. Loos. 1981. Antibody-independent binding of
the first component of complement (Cl) and its
subcomponent Clq to the S and R forms of Salmonella
minnesota. Infect, and Immunity 31:1138-1144.
Colten, H.R., T. Borsos, and H.J. Rapp. 1967. Efficiency of
the first component of complement (C'l) in the hemolytic
reaction. Science 158:1590-1592.
Cooper, N.R. 1983. Activation and regulation of the first
complement component. Fed. Proc. 42:134-138.
Cooper, N.R., and R.J. Ziccardi. 1977. Reconstitution of Cl
in native, proenzyme form and its use in a quantitative
Cl activation test. J. Immunol. 119:1664-1667.
Cyong, C.-C., S.S. Witkin, B. Rieger, L. Barbarese, R.A.
Good, and N.K. Day. 1982. Antibody-independent complement
activation by myelin via the classical complement
pathway. J. Exp. Med. 155:587-598.
Donaldson, V.H. 1970. Blood coagulation and related plasma
enzymes in inflammation. Ser. Haemat. 3:39-95.
Donaldson, V.H. and R.A. Harrison. 1982. Complexes between
Cl-inhibitor, kallikrein, high molecular weight
kininogen, plasma thromboplastin antecedent, and plasmin
in normal plasma and hereditary angioneurotic edema
plasmas containing dysmorphic Cl-inhibitors: Role of cold
activation. Blood 60:121-129.
Gadek, J.E., S.W. Hosea, J.A. Gelfand, M. Santaella, M.
Wickerhauser, D.C. Triantapyhllopoulos, and M.M. Frank.
1980. Replacement therapy in heredity angioedema:
Successful treatment of acute episodes of angioedema with
partially purified Cl inhibitor. New England J. Med.
302:542-546.
Gee, A.P., M.D.P. Boyle, K. Munger, M.J.P Lawman, and M.
Young. 1983. Nerve growth factor: Stimulation of
polymorphonuclear leukocytes in vitro. Proc. Natl.
Acad. Sci. USA 80:7215-7218.
Gelfand, J.A., G.R. Boss, C.L. Conley, R. Reinhart, and M.M.
Frank. 1979. Aquired Cl esterase inhibitor deficiency and
angioedema: A review. Medicine 58:321-328.

71
Ghebrehiwet, B., B.P. Randazzo, J.T. Dunn, M. Silverberg,
and A.P. Kaplan. 1983. Mechanisms of activation of the
classical pathway of complement by Hageman factor
fragment. J. Clin. Invest. 71:1450-1456.
Gigli, I., J.W. Mason, R.W. Coleman, and K.F. Austen. 1970.
Interaction of plasma kallikrein with the Cl inhibitor.
J. Immonol. 104:574-581.
Gigli,I., S. Ruddy, and K.L. Austen. 1968. The
stoichiometric measurement of the serum inhibitor of the
first component of complement by the inhibition of immune
hemolysis. J. Immunol. 100:1154-1164.
Greene, L.A., E.M. Shooter, and S. Varn. 1969. Subunit
interaction and enzymatic activity of mouse 7S nerve
growth factor. Biochemistry 8:3735-3741.
Harpel, P.C., and N.R. Cooper. 1975. Studies on human plasma
Cl inactivator-enzyme interactions. I. Mechanisms of
interaction with aCls, plasmin, and trypsin. J. Clin.
Invest. 55:593-603.
Harrison, R.A. 1984. Human Cl inhibitor: Improved isolation
and preliminary structural characterization. Biochemistry
22:5001-5007.
Heinz, H.-P., R. Burger, M.D. Golan, and M. Loos. 1984.
Activation of the first component of complement, Cl, by a
monoclonal antibody recognizing the C chain of Clq. J.
Immunol. 132:804-808.
Hoffman, L.G., and N.P. Stevland. 1971. Difference in dose
response between native and activated first component of
guinea pig complement (Cl): Reaction with aCl inactivator
offers no explaination. Immunochemistry 8:499-508.
Isackson, P.J., and R.A. Bradshaw. 1984. The alpha- and
beta-subunit of mouse 7S nerve growth factor is an
inactive serine protease. J. Biol. Chem. 259:5380-5383.
Knauer,D.J., K.M. Scaparro, and D.D. Cunningham. 1982. The
gamma subunit of 7S nerve growth factor binds to cells
via complexes formed with two cell-secreted nexins. J.
Biol. Chem. 257:15098-15104.
Knauer, D.J., J.A. Thompson, and D.D. Cunningham. 1983.
Protease nexins: Cell-secreted proteins that mediate the
binding, internalization, and degradation of regulatory
serine proteases. J. Cell Physiol. 117:385-396.
Laemmli, U.K. 1970. Cleavage of structural proteins during
the assembly of the head of bacteriophage T4. Nature
227:680-685.

72
Lepow, I.H., G.B. Naff, and J. Pensky. 1965. Mechanisms of
activation of C'l and inhibition of C'l esterase. In
CIBA Foundation Symposium; Complement, ed.
Wolstenholme, G.E.W., and Julie Knight. Little Brown and
Co., Boston, MA. pp. 74-98.
Levi-Montalcini,R. 1965. The nerve growth factor: Its mode
of action on sensory and sympathetic nerve cells. Harvey
Lect. 60:217-259.
Levi-Montalcini, R., and B. Booker. 1960. Excessive growth
of sympathetic ganglia evoked by a protein isolated from
mouse salivary gland. Proc. Natl. Acad. Sci. USA
46:373-384.
Levy, L.R. and I.H. Lepow. 1959. Assay and properties of
serum inhibitor of C'l esterase. Proc. Soc. Exp. Biol.
Med. 101:608-611.
Li, A.K.C., M.J. Koroly, M.E. Schattenkerk, R.A. Malt, and
M. Young. 1980. Nerve growth factor: Acceleration of the
rate of wound healing in mice. Proc. Natl. Acad. Sci. USA
77:4379-4381.
Loos, M. 1982. The classical complement pathway: Mechanism
of activation of the first component by antigen-antibody
complexes. Prog, in Allergy 30:135-192.
Loos, M., T. Borsos, and H.J. Rapp. 1972. Activation of the
first component of complement: Evidence for an internal
activation step. J. Immunol. 108:683-688.
Loos, M., and H.-U. Hill. 1977. Different behaviour of
native and purified Clq, a subcomponent of the first
component of complement (Cl). Monogr. in Allergy
12:46-51.
McRae, B.J., T.-Y. Lin, and J.C. Powers. 1981. Mapping the
substrate binding site of human aClr and aCls with
peptide thioesters. J. Biol. Chem. 256:12362-12366.
Minta, J.O., and E. Aziz. 1981. Analysis of the reactive
site peptide bond in aCl-inhibitor by chemical
modification of tyrosyl, lysyl, and arginyl residues: The
essential role of lysyl residues in the functional
activity of aCl-Inh. J. Immunol. 126:250-255.
Orenstein, N.S., H.F. Dvorak, M.H. Blanchard, and M. Young.
1978. Nerve growth factor: A protease that can activate
plasminogen. Proc. Natl. Acad. Sci. USA 75:5497-5500.

73
Patrick, R.A., S.B. Taubman, and I.H. Lepow. 1970. Cleavage
of the fourth component of human complement (C4) by
activated Cls. Immunochemistry 7:217-225.
Pensky, J., L.R. Levy, and I.H. Lepow. 1961. Partial
purification of serum inhibitor of C'1-esterase. J. Biol.
Chem. 236:1674-1679.
Phillips, J.K., R. Snyderman, and S.E. Mergenhagen. 1972.
Activation of complement by endotoxin: A role for gamma2
globulin, Cl, C4 and C2 in the consumption of terminal
complement components by endotoxin-coated erythrocytes.
J. Immunol. 109:334-341.
Rapp, H.J. 1955. A mathematical theory and analysis of
immune hemolysis. Dissertation, Johns Hopkins University,
School of Hygiene and Public Health.
Ratnoff, O.D., and I.H. Lepow. 1957. Some properties of an
esterase derived from preparations of the first component
of complement. J. Exper. Med. 106:327-342.
Reid, K.B.M. 1983. Proteins involved in the activation and
control of the two pathways of human complement. Proc.
Royal Acad. Sci. 11:1-12.
Rosenberg, S., P.J. Barr, R.C. Najarin, and R.A. Hallewell.
1984. Synthesis in yeast of a functional
oxidation-resistant mutant of human alpha.-antitrypsin.
Nature 312:77-80.
Salvesen, G.S., J.J. Catanese, L.F. Kress, and J. Travis.
1985. Primary structure of the reactive site of human
Cl-inhibitor. J. Biol. Chem. 260:2432-2436.
Schapira, M., L.D. Silver, C.F. Scott, A.H. Schaier, L.J.
Prograis, J.G. Curd, and R.W. Coleman. 1983.
Prekallikrein activation and high-molecular-weight
kinigen consumption in hereditary angioedema. New England
J. Med. 308:1050-1053.
Server, A.C., and E.M. Shooter. 1977. Nerve growth factor.
Adv. Prot. Chem. 31:339-409.
Sliverberg,M., J.T. Dunn, L. Garen, and A.P. Kaplan. 1980.
Autoactivation of human Hageman factor: Demonstration
utilizing a synthetic substrate. J. Biol. Chem.
255:7281-7286.
Sim, R.B. 1981. The human complement system serine proteases
aClr and aCls and their proenzymes. Meth. Enzymol.
80:26-42.

74
Sim, R.B., G.J. Arlaud, and M.G. Colomb. 1979. Kinetics of
reaction of human Cl-inhibitor with the human complement
system proteases Clr and Cls. Biochem. et Biophys. Acta
612:433-449.
Sim, R.B., and R.R. Porter. 1976. Isolation and comparison
of the proenzyme and activated forms of the human serum
components Clr and Cls. Biochem. Soc. Trans. 4:127-129.
Travis, J., and G.S. Salvesen. 1983. Human plasma protease
inhibitors. Ann. Rev. Biochem. 52:655-709.
Valet, G., and N.R. Cooper. 1974. Isolation and
characterization of the proenzyme form of the Clr subunit
of the first complement component. J. Immunol.
112:1667-1673.
Varn, S., J. Nomura, and E.M. Shooter. 1968. Reversible
dissociation of the mouse nerve growth factor protein
into different subunits. Biochemistry 7:1297-1303.
Weber K., and M. Osborn. 1969. The reliability of molecular
weight determinations by dodecyl sulfate-polyacrylamide
gel electrophoresis. J. Biol. Chem. 244:4406-4412.58.
Young, M., and M.J. Koroly. 1980. The nerve growth factor
zymogen. Stoichiometry of the active site serine and the
role of Zn(II) in controlling autocatalytic
self-activation. Biochemistry 19:5316-5322.
Ziccardi, R.J. 1982a. A new role for aCl-inhibitor in
homeostasis: Control of activation of the first component
of human complement. J. Immunol. 128:2505-2508.
Ziccardi, R.J. 1982b. Spontaneous activation of the first
component of human complement (Cl) by an intramolecular
autocatalytic mechanism. J. Immunol. 128:2500-2504.
Ziccardi, R.J., and N.R. Cooper. 1976. Physicochemical and
functional characterization of the Clr subunit of the
first complement component. J. Immunol. 116:496-503.

BIOGRAPHICAL SKETCH
Ervin L.W. Faulmann, Jr., was born on February 10,
1949, to Ervin and Leithel Faulmann. He was raised and
educated in southern California and graduated from Alhambra
High School in 1967. He attended California Polytechnic
State University at San Luis Obispo. After a brief period
of working on ranches in Montana, he returned to school and
received a Bachelor of Science degree in microbiology from
Montana State University in 1976. He entered the Department
of Immunology and Medical Microbiology at the University of
Florida College of Medicine for post graduate studies.
During the course of those studies he worked at the
University Medical Center in Jackson, Mississippi, under Dr.
L. William Clem and completed the work on the effects of
temperature on the mitogen responses of catfish leukocytes,
for which he received a Master of Science degree from the
University of Florida. He returned to the Department of
Immunology and Medical Microbiology and and continued his
studies under Dr. Michael D.P. Boyle. On October 24, 1982,
he married Phoebe Castle Hellems. Following the completion
of his dissertation, he intends to continue research in
immunology in hopes of improving our understanding of
molecular interactions associated with disease processes.
75

I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosophy.
Michael D.P. Boy
Associate Professor of
Immunology and Medical
Microbiology
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosophy.
^ m Michael Young^
Professor of Biocm?mistry
and Molecular Biology
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosophy.
/ .
/
/ /D ^
Adrian" P. tiee~ Jj
Assistant Professor of
Immunology and Medical
Microbiology
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Ph^i^osophy.
Medical Microbiology

I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosophy.
Edward M. Hoffmamvi/
Professor of Micjrqoiology
and Cell' Science
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosophy.
'ames B. Flanegan / /
Associate Professor-/
Immunology and Medical
Microbiology
This dissertation was submitted to the Graduate Faculty of
the College of Medicine and to the Graduate School and was
accepted as partial fulfillment of the requirements for the
degree of Doctor of Philosophy.
May, 1985
Dean, Graduate School



9
inhibited by Cl-Inh (Boyle and Young, 1982). To date, gamma
NGF has been reported to mediate a number of activities
associated with inflammation, i.e., enhanced wound healing
(Li et al., 1980), and both i_n vivo and in vitro
leukocyte chemotaxis (Boyle et al., 1985; Gee et al., 1983).
However, gamma NGF is not known to be a member of any
physiological cascade system. Similarly, Cl-Inh is not
known to inhibit proteases not associated with cascade
systems. In order to determine if Cl-Inh was acting as a
serpin (or nexin), the nature of the interaction of gamma
NGF and Cl-Inh seen in the complement assay needed to be
examined. In this study, purified Cl-Inh was shown to
inhibit the enzymatic activity of purified gamma NGF in a
time-dependent manner with 1:1 stoichiometry. The
interaction of gamma NGF with Cl-Inh resulted in the
formation of a covalent bonded enzyme-inhibitor complex.
The properties of Cl-Inh reported in this study expand
our understanding of its physiological role as a serum
protease inhibitor. The results presented demonstrate that
Cl-Inh displays the general characteristics used to define a
protease nexin, in addition to regulating Cl activation and
the enzyme activity of fully activated Cl and nerve growth
factor.


23
Cl activity in a single dilution of partially purified serum
Cl increased, while the relative Cl activity found in serum
or aCl dilutions remained constant (see Figure 8). This
activation of partially purified serum Cl upon incubation at
37C could be seen more dramatically when the various Cl
solutions were titered in the Cl hemolytic assay. In Figure
9 it can be seen that the partially purified serum Cl
expressed an increased titer of Cl activity in the hemolytic
assay and demonstrated the characteristics of aCl.
Unfractionated serum incubated under identical conditions
did not self-activate (data not shown). Similar results
were obtained when Cl was partially purified from serum by
precipitation with 40% saturated ammonium sulphate or by
ultracentrifugation (100,000 x G) (data not shown). These
results suggest that Cl can be partially purified from serum
and remain zymogen; however, once separated from certain
serum proteins, this zymogen Cl undergoes self-activation.
Concentration Dependence of Serum Cl Activation
In the next series of experiments, the effect of the Cl
concentration on the process of spontaneous activation of
the DE-52 separated material was tested. Serum Cl was
isolated as described above, and divided into three pools,
and diluted 1:4, 1:16, and 1:64 respectively. Each pool was
preincubated at 37C for various times before testing at
equal final dilutions for Cl activity. The results
presented in Figure 10 indicate that the rate of spontaneous


33
The differences reported in the various studies will
only be resolved when methods become available to compare
the Cl enzyme activity detected in the hemolytic assay,
synthetic substrate assay, and physicochemical assay can all
be carried out using the same preparation and the same
concentration of zymogen Cl under the same experimental
conditions.
The apparent lack of spontaneous activation seen with
Cl in unfractionated serum will be addressed in the
following section (Chapter III).


BIOGRAPHICAL SKETCH
Ervin L.W. Faulmann, Jr., was born on February 10,
1949, to Ervin and Leithel Faulmann. He was raised and
educated in southern California and graduated from Alhambra
High School in 1967. He attended California Polytechnic
State University at San Luis Obispo. After a brief period
of working on ranches in Montana, he returned to school and
received a Bachelor of Science degree in microbiology from
Montana State University in 1976. He entered the Department
of Immunology and Medical Microbiology at the University of
Florida College of Medicine for post graduate studies.
During the course of those studies he worked at the
University Medical Center in Jackson, Mississippi, under Dr.
L. William Clem and completed the work on the effects of
temperature on the mitogen responses of catfish leukocytes,
for which he received a Master of Science degree from the
University of Florida. He returned to the Department of
Immunology and Medical Microbiology and and continued his
studies under Dr. Michael D.P. Boyle. On October 24, 1982,
he married Phoebe Castle Hellems. Following the completion
of his dissertation, he intends to continue research in
immunology in hopes of improving our understanding of
molecular interactions associated with disease processes.
75


68
The many parallels and similarities between the
structure, function, and regulation of NGF and Cl suggest
that nerve growth factor may play an important physiological
role totally distinct from its trophic effect on the sensory
and sympathetic nervous system. The possibility that NGF
may be part of a heretofore unrecognized system involved in
homeostatic reactions merits closer scrutiny. To date
Cl-Inh has only been recognized to regulate enzymes within
cascade systems, see Figure 1, and this raises the
possibility that NGF is part of some unknown cascade system.
j f.
A number of recent observations indicating that NGF is
chemotactic for polymorphonuclear leukocytes ini vivo
(Boyle and Young, 1982) and iri vitro (Gee et al. 1983)
coupled with the reported ability of the growth factor to
accelerate wound healing (Li et al., 1980) suggest NGF may
be a potential mediator of a variety of normal responses
subsequent to tissue damage.


51
vitro (Gee et al., 1983) and .in vivo (Boyle et al.,
1985). From these studies it would appear that gamma NGF
may have important physiological roles in the inflammation
response and wound repair.
A series of cell secreted protease inhibitors (nexins)
have recently been described that bind and inactivate
specific serine esterases (Baker et al., 1980; Knauer et
al., 1983). In binding to the enzyme, the nexin inactivates
the enzyme and leads to efficient clearance via ingestion by
cells expressing receptors for the enzyme-nexin complexes
(Knauer et al., 1983). Two proteins that function as nexins
in complexing with activated gamma NGF have been reported
and partially characterized (Knauer et al., 1982).
Experiments presented in this chapter examine the
interaction of Cl-Inh with gamma NGF. It was found that
purified Cl-Inh was able to inhibit the esterolytic activity
and form covalent complexes with gamma NGF in a time
dependent fashion, with a 1:1 stoichiometry. These results
are similar to those reported for the interactions between
Cl-Inh and other enzymes.
Materials and Methods
Reagents
Cl-Inh was purified to homogeneity by the method of
Harrison (1984) as reported in Chapter III. Purified gamma
NGF from mouse submaxillary salivary glands was provided by
Dr. M. Young.


y ) Cz Valu]
were removed and added to buffer containing excess C2 at
37C. After an additional 5 minute incubation at 37C,
CEDTA was added and the reaction was allowed to proceed for
60 minutes, after which the amount of lysis that had occured
was determined by 0D4^2*


31
antibodies rather than the polyclonal were used to prepare
the indicator cells.
The differences in aCl and zymogen Cl activity were
used to monitor the activation of Cl either in serum or
following partial purification. It was shown that serum Cl
could be partially purified from other serum proteins
without being fully activated in the process. This
partially purified serum Cl was seen to undergo a
spontaneous self-activation process that was
time-and-concentration-dependent. This activation required
some form of Cl to Cl interaction but activation was not
promoted in a classic autocatalytic fashion. These findings
are distinct from either the results of Ziccardi (1982a) or
of Lepow et al. (1965). The disparity between the findings
reported here and those of Ziccardi may be due to the
characteristics of the assays used to measure Cl activation.
The change in activation rate seen in these studies was
most noticeable at concentrations of Cl below the detection
limits of the physicochemical assay used by Ziccardi (Cooper
and Ziccardi, 1977; Ziccardi, 1982a; Ziccardi and Cooper,
1976). Moreover, due to the one-hit nature of the aCl
molecule, the hemolytic assay has the potential of being
more sensitive for aCl than zymogen Cl. This has been shown
in a similar hemolytic assay system in which IgG sensitizing
antibodies are compared to IgM sensitizing antibodies for
the ability to induce complement mediated lysis. The
fixation of Cl to IgG is a two-hit phenomenon in which one


64
account for the observed behavior. The specific subunit
interactions needed for activation of macromolecular serum
Cl were not established in this study. This is an important
question that merits further investigation.
In this investigation, the activity of Cl-Inh was shown
to be broader than previously appreciated. The serine
esterase inhibitor activity of Cl-Inh has been well
documented (Harpel and Cooper, 1975). However, the rate of
inactivation of aCl by Cl-Inh is slow (Sim and Porter, 1976)
compared to the activity of aCl on C4 and C2 (Patrick et
al., 1970), and it is thought that the true physiological
role of the serpin activity of Cl-Inh is in inactivating the
enzymatic activity of kallikrein (Gigli et al., 1970). The
results reported in this study present a new role of Cl-Inh
in regulating serum Cl activity which entails the regulation
of the activation of zymogen Cl. This regulation of serum
Cl activation was distinct from the activity of Cl-Inh in
inactivating aCl. In regulating both the activation and
resultant activity of serum Cl, Cl-Inh is able to
effectively control the initiation of the classical
complement pathway. Furthermore, it has been seen by others
(Harpel and Cooper, 1975; Minta and Aziz, 1981) that the
interaction of aCl with Cl-Inh not only inactivates the aClr
and aCls subunits, but also dissociates the complexed
subunits from the Clq subunit. Though enhanced clearance of
complexes of Cl-Inh with aClr or aCls has not been observed,
the dissociation of the aCl macromolecule would enhance the


8
inhibition remains unclear (Ziccardi, 1982b). This study
was initiated to resolve the apparent controversy regarding
the spontaneous activation of zymogen Cl and the role Cl-Inh
may play in it. Partially purified serum Cl was shown to
undergo spontaneous activation in a
time-and-concentration-dependent manner. This spontaneous
activation was independent of the enzymatic activity of
exogenous aCl. Serum fractions depleted of Cl and purified
Cl-Inh were shown to inhibit the spontaneous activation of
serum Cl. These studies suggest a model of Cl activation
that is dependent on an intermolecular interaction between
zymogen Cl molecules that is independent of the enzymatic
activity of either the Clr or Cls subunits.
Recently, it has been reported that the high molecular
weight form of NGF can substitute for activated Cl in
activating the classical complement pathway cleaving C4 and
C2 (Boyle and Young, 1982). High molecular weight NGF is a
multisubunit protein composed of three subunits, alpha,
beta, and gamma in the stoichiometric ratio of 2:1:1
respectively (Greene et al., 1969). The alpha subunit has
no known biological function (Isackson and Bradshaw, 1984).
The beta subunit is responsible for the outgrowth of
sympathetic and sensory ganglia from chicken embryo
(Levi-Montalcini, 1965). The gamma subunit is a serine
class enzyme (Orenstein et al., 1978) which is responsible
for complement activation (Boyle and Young, 1982). This
complement activating ability of gamma NGF was shown to be


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
REGULATION OF SERUM PROTEASE ACTIVITY:
ACTION OF Cl-INHIBITOR ON THE ACTIVITIES
OF SERUM Cl AND NERVE GROWTH FACTOR
By
ERVIN L.W. FAULMANN, JR.
MAY 1985
Chairman: Michael D.P. Boyle
Major Department: Immunology and Medical Microbiology
The interactions between serum proteases (serpins) and
serum protease inhibitors are the basis of the regulation of
many homeostatic processes. Evidence from a number of
researchers suggests that protease inhibitors may have
physiological roles in addition to inactivating proteolytic
enzymes.
The first component of the classical complement pathway
(Cl) exists in serum in a precurser zymogen form. During
purification Cl undergoes spontaneous activation yielding
activated Cl (aCl); however the mechanism of this
autoactivation process is unclear.
In this study, by means of the Cl hemolytic assay,
partially purified serum Cl was shown to spontaneously
activate in a time-and-concentration-dependent manner.
Addition of aCl did not alter the level of Cl activation.
iv


41
Human Serum Fractionation
Partially purified serum Cl was prepared by
ion-exchange chromatography described in the previous
chapter. Proteins bound to the ion-exchange column were
eluted by washing the column with equlibration buffer
containing 200 mM NaCl. Other sources of Cl depleted human
serum fractions were produced by eugologulin precipitation
(supernatant of serum dialysed against water) and
ultracentrifugation (supernatant of serum spun for 4 hours
at 100,000 x G).
Results
Inhibition of Spontaneous Activation of Zymogen Cl by
Serum Fractions
To address the role of serum factors in maintaining the
zymogen nature of unfractionated serum Cl, Cl depleted serum
fractions were assayed for their ability to inhibit the
spontaneous activation of partially purified serum Cl (DE-52
pass-thorugh). Human serum was fractionated to remove Cl
activity by euglobulin precipitation (eublobulin
supernatant), ultracentrifugation (supernatant of 100,000 x
G), or ion-exchange chromatography (DE-52 eluate). The
various serum fractions were added to partially purified
serum Cl. The solutions were incubated one hour at 37C
and the ability of each fraction to inhibit the conversion
of serum Cl activity (as shown by a Cl titration slope of


21
Figure 6. EFFECT OF TRYPSIN TREATMENT ON THE HEMOLYTIC
ACTIVITY OF CELL BOUND SERUM Cl AND ACTIVATED Cl. Various
dilutions of serum and activated Cl were incubated with
EAC4b for 20 minutes at 37C in SVB buffer. The cells were
washed and treated with trypsin (1 pg/ml) (closed symbols)
or buffer (open symbols) for 5 minutes at 27C. The cells
were again washed, resuspended in SVB, and assayed for cell
bound Cl activity as described in the Methods. ( $ and o -
serum treated cells, m and activated Cl treated cells)


These results suggest a model of: serum Cl activation that
requires intermolecular interactions between the Cl
molecules independent of the enzymatic activity of aCl
produced. However, addition of purified Cl esterase
inhibitor protein (Cl-Inh), a serpin that inhibits the
enzymatic activity of aCl and a variety of cascade
proteases, inhibited the spontaneous activation of serum Cl
in a concentration-dependent manner. The effect of Cl-Inh
regulating Cl activation was independent of its activity in
inactivating aCl. These results document a new role of
r
Cl-Inh in which it reversibly inhibits the activation of
zymogen Cl as compared to its irreversible inhibition of
aCl.
The role of Cl-Inh in inhibiting another classical
complement pathway activator, the gamma subunit of nerve
growth factor (gamma NGF), was also examined. The enzymatic
activity of gamma NGF was inhibited by Cl-Inh in a
time-dependent manner with 1:1 stoichiometry. The
interaction of Cl-Inh with gamma NGF produced a sodium
dodecyl sulfate (SDS) stable complex that was suceptible to
cleavage by hydroxylamine, thus suggesting a covalent bond
between the protease and protease inhibitor. This reaction
between gamma NGF and Cl-Inh is similar to that between
gamma NGF and a newly described class of cell secreted
protease inhibitors, called nexins, and provides new insight
as to the possible physiological roles of both Cl-Inh and
gamma NGF.
v


40
Figure 13. SDS-PAGE OF THE VARIOUS STEPS IN THE
PURIFICATION OF Cl-INH. Pictured is the Coomassie stain of
a SDS-PAGE analysis of aliquots from the various stages of
the purification of Cl-Inh from human serum (lanes 1 and 8 -
molecular weight standards, lane 2 serum, lane 3 PEG
supernate, lane 4 lysine-sepharose flow through fraction,
lane 5 DE-52 eluate pool, lane 6 G150 eluate pool, lane
7 hydroxylapatite eluate pool)


36
zymogen nature of the Cl found in asymptomatic HAE patients
(Gelfand et al., 1979), raise important questions about the
regulation of serum Cl activity in serum. A naturally
occurring factor (possibly Cl-Inh) in serum that regulates
the activation of native serum Cl would therefore be an
attractive hypothesis.
Other workers (Ziccardi, 1982b) using serum Cl reformed
from isolated Cl subunits have shown that Cl-Inh inhibited
the spontaneous activation of serum Cl to aCl. Those
studies must be viewed with caution because the serum Cl
125
molecules are reformed using modified ( I labeled)
subunits and the reformed serum Cl molecules differed
significantly from serum Cl in a number of physical
characteristics, specifically, sedimentation coeffecients
(16S and 12S for serum Cl and refomed zymogen Cl
respectively)(Colten et al., 1967; Cooper and Ziccardi,
1977), Cl-Inh kinetics (Loos, 1982; Ziccardi and Cooper,
1976), and spontaneous activation kinetics (see Chapter II;
Ziccardi, 1982a). Furthermore, the assays used in those
studies monitored the average physical changes of
populations of Cl subunits during the course of spontaneous
activation. This may or may not directly reflect the
functional changes associated with the activation of serum
Cl to aCl.
In the previous section it was shown, using the Cl
hemolytic titration assay, that crude preparations of
unmodified serum Cl from serum were able to spontaneously


25
RELATIVE Cl CONCENTRATION
Figure 9. HEMOLYTIC ACTIVITY OF PARTIALLY PURIFIED SERUM Cl
AFTER INCUBATION AT 37C. A single dilution (1:4 in VBS)
of partially purified serum Cl was incubated at 37C. At
various times, aliquots were removed and diluted. The
samples were then titered in the Cl hemolytic assay.
( 0, 2, A 5, a 10, A 20, o 60 minutes)


28
Figure 11. EFFECT OF ENZYMATICALLY ACTIVE Cl ON THE
SPONTANEOUS ACTIVATION OF PARTIALLY PURIFIED SERUM Cl.
Activated Cl or buffer was added to partially purified serum
Cl (diluted 1:400 in VBS), incubated 20 minutes at 37C,
and assayed for Cl activity by the hemolytic assay.
( 9 partially purified serum Cl, activated Cl,
A- partially purified serum Cl + activated Cl,
- - additive activity of partially purified serum Cl and
activated Cl separately)


6
altered activity has been shown to be associated with
mutations in the amino acid sequence of the target site on
the serpin recognized by the specific enzyme (Carrell,
1984). During a disease episode, the serum Cl of HAE
patients is fully activated and there is no measurable C4
and C2 (Gelfand et al., 1979). Asymptomatic HAE patients,
however, exibit low but measurable levels of C4 and C2
(Gelfand et al., 1979), hence their Cl is not fully
activated. The low levels of functional Cl-Inh and the
anomalous episodic activation of the serum Cl of HAE
patients appear to be causally related. Clinical
administration of Cl-Inh reverses the disease process and
the patient's C4 and C2 levels are restored (Gadek et al.,
1980 ) .
The main focus of this study was to examine the
activation of serum Cl and determine the role of Cl-Inh in
regulating the activation process. Previous studies have
reported the autocatalytic activation of partially purified
serum Cl, in which aCl reacts enzymatically with zymogen Cl
resulting in the conversion of zymogen Cl to aCl (see Figure
2) (Lepow et al., 1965). In this case, Cl-Inh would
regulate the activation of Cl by inhibiting the enzymatic
activity of aCl. A recent report has shown that purified
zymogen Cl spontaneously activates by a intramolecular event
(concentration independent) that was not autocatalytic
(Ziccardi, 1982a). Furthermore, Cl-Inh was shown to inhibit
this spontaneous activation, though the mechanism of this


2
unregulated there would be serious physiological
consequences. Serum protease inhibitors provide an
important level of regulation by inactivating specific
proteolytic enzymes associated with key amplifying roles in
these cascade systems.
Recent evidence suggests that protease inhibitors may
have physiological roles in addition to the inactivation of
proteolytic enzymes. In 1980, Baker et al. described the
secretion of a protein from normal human fibroblasts that
formed a covalent acyl linkage with urokinase and thrombin.
The resulting proteolytically inactive complexes were
internalized and degraded (Knauer et al., 1983). These
regulating anti-proteases were found to be secreted by a
wide range of normal and transformed cell lines and were
named nexins. Detailed studies using a variety of sources
of nexins and a number of serine class proteases have
indicated the existance of three distinct protease nexins
(PNs). These regulating proteins designated PNl, PN2, and
PN3 share many of the functional characteristics of the
serpins, but appear to be both physicochemically and
antigenically distinct proteins (Knauer et al., 1983). The
three groups, PNl, PN2, and PN3, differ in the enzymes which
they inhibit. The first, PNl, is a protein of molecular
weight of 38,000 Daltons that binds thrombin and urokinase.
Complexes with either of these enzymes bind to the surface
receptors on a variety of cell types and are rapidly
internalized and degraded. In the presence of heparin, cell


RELATIVE Cl CONCENTRATION
Figure 10. EFFECT OF DILUTION ON THE RATE OF SPONTANEOUS
ACTIVATION OF PARTIALLY PURIFIED SERUM Cl. Partially
purified serum Cl was split into three aliquots and diluted
either 1:4 (panel A), 1:16 (panel B), or 1:64 (panel C).
These three solutions were then incubated at 37C and at
various times, aliquots were removed and diluted further
(1:512 final dilution) and stored on ice. At the end of the
incubation all aliquots were assayed for Cl activity in the
hemolytic assay. ( A 0, 20, 60 minutes)


3
binding is enhanced (Knauer et al., 1983). The second, PN2,
is a protein of 93,000 Daltons with selective reactivity for
the epidermal growth factor (EGF)-binding protein (Knauer et
al., 1983). These complexes are also cleared by a variety
of cell types but their uptake is not augmented by addition
of heparin. The last, PN3, is a 31,000 Dalton protein with
selective reactivity for the gamma subunit of nerve growth
factor (gamma NGF) and the resulting gamma NGF-PN3 complexes
are rapidly cleared by a variety of cells. As in
PN2-containing complexes, clearance is not augmented by
heparin. The binding specificities of the PNs are not
absolute and some reactivity of gamma NGF and PNl and PN2
has also been reported (Knauer et al., 1983). The
physiological function of the nexins is not clear. The
current findings suggest that they act to regulate the
biological activities of certain serine proteases by forming
complexes that are rapidly cleared and degraded. At
present, no stimulatory biological activity has been
reported for any PN-protease complex.
Of particular interest to this study is the Cl esterase
inhibitor protein (Cl-Inh) that regulates a variety of
cascade systems associated with the host response to tissue
damage and inflammation (see Figure 1). Human Cl-Inh is a
single-chain glycoprotein with a molecular weight of 116,000
Daltons (Harrison, 1984) (97,000 Daltons in SDS-PAGE (Harpel
and Cooper, 1975)), containing about 35% carbohydrate
(Harrison, 1984). It was first recognized for its ability


67
and Young, 1982; Lepow et al., 1965) and their enzyme
activity can be regulated by the Cl inhibitor protein (Boyle
and Young, 1982; Lepow et al., 1965). Second, both are
produced as zymogens (Lepow et al., 1965; Young and Koroly,
1980). Third, both zymogen NGF and Cl undergo
self-activation reactions (Lepow et al., 1965; Young and
Koroly, 1980; Ziccardi, 1982a). In the case of NGF this is
a classical autocatalytic reaction in which the removal of
the Zn++ associated with the zymogen (Young and Koroly,
1980) allows the NGF to spontaneously activate in a biphasic
reaction (Young and Koroly, 1980). Initially, there is a
low level of intramolecular spontaneous activation (Young
and Koroly, iyU), ronowea oy a rapid classical
autocataiytic reaction, in wnicn zymogen /S NGF serves as a
substrate ot activated 7S NGF to produce more activated NGF
(Young and Koroly, lySD). The autocatalysis of the molecule
can be inhibited by inhibiting the esterolytic activity of
the gamma subunit (Young and Koroly, 1980). One would
predict that addition of Cl-Inh to zymogen 7S NGF would
inhibit the autocatalytic activation (enzymatic activation).
However, because of the results reported for Cl in Chapter
III, the possibility would also exist that Cl-Inh may
inhibit the first phase of the activation (spontaneous
activation) of 7S NGF. The regulation of activation of
other zymogen serum proteases by Cl-Inh warrants further
investigation.


15
Results
Activity of Serum and Activated Cl in the Hemolytic
Assay
In the Cl hemolytic assay, aCl yielded one-hit
characteristics in which there was a 1:1 relationship
between the relative amount of aCl added to the assay and
the number of effective Cl induced lytic sites produced (see
Figure 3). This relationship was observed irrespective of
whether aCl was diluted or the number of indicator cells was
changed (see Figure 4 and Table 1). These findings fully
confirm the previous work of others (Borsos et al., 1980;
Borsos et al., 1964; Loos, 1982).
Serum Cl titrated under identical conditions did not
follow a similar pattern. As shown in Figure 3, the
quantity of aCl activity detected fell off more quickly than
predicted as serum was diluted. This was not due to the
heterogeneous nature of the binding specificities of the
sensitizing antibodies, because indicator cells prepared
with monoclonal antibodies yielded similar results (see
Figure 5). Additionally, the excessive loss of Cl activity
upon diluting serum Cl was not due solely to dissociation
since, as is shown in Figure 4 and Table 1, the total number
of effective Cl molecules detected at any dilution was
dependent upon the indicator cell number used in the assay.
In agreement with earlier studies (Loos et al., 1972), mild
treatment of serum Cl coated EAC4b indicator cells with
trypsin resulted in the appearance of an increased titer of


61
high molecular weight complex was labile to treatment with
hydroxylamine at pH 10.5, which is characteristic of an acyl
125
linkage between the Cl-Inh and I-NGF in the high
molecular weight complex.
Discussion
The proteolytic activity of gamma NGF, as measured by
the cleavage of the synthetic substrate 1-BAPNA, has been
shown to be inhibited by Cl-Inh. This inhibition was time
dependent with a stoichiometry of approximately 1:1. The
interaction between gamma NGF and Cl-Inh was shown to result
in the formation of a covalent bonded complex, as judged by
SDS-PAGE. These results would be consistent with a model of
interaction in which the inhibition of the enzymatic
activity of gamma NGF was inhibited by the association of
Cl-Inh near the active site of the molecule, followed by the
cleavage of Cl-Inh and covalent coupling to gamma NGF.
There are a number of predictions associated with such a
model. First, the cleavage of Cl-Inh by gamma NGF should
generate a new amino terminal in the Cl-Inh. Second, a new
Cl-Inh fragment should be produced by the complexing of
gamma NGF and Cl-Inh. Finally, the formation of the
covalent bond should be localized to a specific polypeptide
chain(s) in the gamma NGF.
The bond formed in the interaction of gamma NGF and
Cl-Inh was shown to be susceptible to the action of
hydroxylamine at high pH. These results are similar to


73
Patrick, R.A., S.B. Taubman, and I.H. Lepow. 1970. Cleavage
of the fourth component of human complement (C4) by
activated Cls. Immunochemistry 7:217-225.
Pensky, J., L.R. Levy, and I.H. Lepow. 1961. Partial
purification of serum inhibitor of C'1-esterase. J. Biol.
Chem. 236:1674-1679.
Phillips, J.K., R. Snyderman, and S.E. Mergenhagen. 1972.
Activation of complement by endotoxin: A role for gamma2
globulin, Cl, C4 and C2 in the consumption of terminal
complement components by endotoxin-coated erythrocytes.
J. Immunol. 109:334-341.
Rapp, H.J. 1955. A mathematical theory and analysis of
immune hemolysis. Dissertation, Johns Hopkins University,
School of Hygiene and Public Health.
Ratnoff, O.D., and I.H. Lepow. 1957. Some properties of an
esterase derived from preparations of the first component
of complement. J. Exper. Med. 106:327-342.
Reid, K.B.M. 1983. Proteins involved in the activation and
control of the two pathways of human complement. Proc.
Royal Acad. Sci. 11:1-12.
Rosenberg, S., P.J. Barr, R.C. Najarin, and R.A. Hallewell.
1984. Synthesis in yeast of a functional
oxidation-resistant mutant of human alpha.-antitrypsin.
Nature 312:77-80.
Salvesen, G.S., J.J. Catanese, L.F. Kress, and J. Travis.
1985. Primary structure of the reactive site of human
Cl-inhibitor. J. Biol. Chem. 260:2432-2436.
Schapira, M., L.D. Silver, C.F. Scott, A.H. Schaier, L.J.
Prograis, J.G. Curd, and R.W. Coleman. 1983.
Prekallikrein activation and high-molecular-weight
kinigen consumption in hereditary angioedema. New England
J. Med. 308:1050-1053.
Server, A.C., and E.M. Shooter. 1977. Nerve growth factor.
Adv. Prot. Chem. 31:339-409.
Sliverberg,M., J.T. Dunn, L. Garen, and A.P. Kaplan. 1980.
Autoactivation of human Hageman factor: Demonstration
utilizing a synthetic substrate. J. Biol. Chem.
255:7281-7286.
Sim, R.B. 1981. The human complement system serine proteases
aClr and aCls and their proenzymes. Meth. Enzymol.
80:26-42.


29
Taken together, the results in Figures 10 and 11 indicate
that Cl activation is neither a concentration independent
autoactivation reaction nor a classical autocatalytic
reaction. This suggests that Cl activation requires the
interaction of more than one Cl molecule but the interaction
is not one that is characteristic of an enzyme-substrate
interaction. These findings are consistent with the
observation that Cl can be demonstrated in self-activatable
and non-self-activating forms (Borsos et al., 1980; Loos et
al., 1972).
r1
Discussion
When assayed under pseudo-first order conditions in
hemolytic assays, the fully activated form of Cl exhibits
all the characteristics of a single hit reaction (Borsos and
Mayer, 1962). The behavior of activated Cl in the hemolytic
assay is summarized below:
1) Activated Cl is 100% efficient in the generation of
lytic sites on the sensitized red cell membrane (Colten et
al., 1967).
2) There is a direct proportionality between the
quantity of Cl added and the number of lytic sites generated
(Colten et al., 1967).
3) The number of effective Cl molecules measured in the
hemolytic assay is independent of the number of indicator
cells used (Loos, 1982).


53
Results
Inhibition of Gamma NGF Enzymatic Activity by Cl-Inh
Preincubation of active gamma NGF with a ninefold molar
excess Cl-Inh for one hour at 37C effectively inhibited
the ability of the gamma NGF to subsequently cleave the
synthetic substrate 1-BAPNA (see Figure 17). The inhibition
of gamma NGF by Cl-Inh was complete within one hour of
incubation at 37C. The stoichiometry of the gamma
NGF-Cl-Inh interaction was consequently determined following
a one hour preincubation at 37C. The results of adding
various amounts of Cl-Inh to a constant amount gamma NGF,
plotted in Figure 18, showed an approximately 1:1
stoichiometry for the inhibition of gamma NGF by Cl-Inh. In
the next group of experiments the rate of inactivation of
gamma NGF by Cl-Inh was examined. In preliminary
experiments it was shown that addition of 1 mM 1-BAPNA to
gamma NGF prior to addition of Cl-Inh to the mixture
prevented the interaction of Cl-Inh with gamma NGF (data not
shown). This blockage of Cl-Inh activity was due most
probabily to competitive inhibition for the active site of
the gamma NGF. Therefore, the time course experiment was
not affected by the interaction of Cl-Inh and gamma NGF
during the enzyme assay. Using a 1:2 molar ratio of Cl-Inh
to gamma NGF, the rate of inhibition of the gamma NGF enzyme
activity was measured at 24C (see Figure 19). These
results are similar to those reported for the interaction of
Cl-Inh with aCls (Harpel and Cooper, 1975), and suggest the


14
review see Loos (1982)). It has been shown from dose
response and kinetic experiments that every aCl molecule is
capable of generating a lytic site (Borsos and Rapp, 1963;
Colten et al., 1967) and that one such site is necessary and
sufficient for cell lysis (Borsos and Mayer, 1962). The
hemolytic assay (described above) provided all the other
components necessary for completing the lysis of all the
cells in the reaction mixture, i.e. pseudo-first order for
aCl activity, and the extent of hemolysis was measured at
end point. The physical conditions, eg., pH, ionic
v -
strength, and temperature, of each reaction step were
carefully controlled to ensure the reproducibility of the
reaction. In meeting these criteria the distribution of aCl
molecules among the cells would conform to the predictions
of the Poisson distribution (Rapp, 1955) and consequently it
was possible to relate the percentage of cells lysed (y) to
the absolute number of hemolytically effective aCl molecules
added to the assay by the relationship
Z = In ( 1 y ) .
In this equation Z represents the average number of lytic
sites per cell (Borsos and Mayer, 1962) and the total number
of hemolytically effective aCl molecules in the system was
determined by multiplying Z times the number of cells in the
assay. Further considerations of the activity of aCl and
serum Cl in the hemolytic assay are dealt with more fully in
the results and discussion sections below.


13
N-2-hydroxlethyl piperazine-N'-2-ethanesulfonic acid (HEPES)
and 0.075 M NaCl was eluted with equlibration buffer at
4C. This procedure depletes the serum of Cl-Inh (see
Chapter III) and yields a preparation of zymogen Cl that
will spontaneously self-activate (see Figure 8). Throughout
the procedures, the isolation and activation state of the Cl
was monitored using the hemolytic assay described below.
Measurement of the Hemolytic Activity of aCl
The assay for aCl was carried out by incubation of
dilutions of a sample containing an unknown amount of aCl
with 1.5xl07 EAC4b in a total volume of .3 ml of SVB
(p=0.065) (Borsos and Rapp, 1967) for 20 minutes at 37C.
The hemolytic reaction was completed by the sequential
addition of excess C2 for 5 minutes at 37C followed by the
addition of excess late acting components (CEDTA) for 1 hour
at 37C. Under these conditions, the reaction is at
endpoint (Borsos and Rapp, 1967). The number of effective
aCl molecules was determined by measuring the extent of
lysis of the EAC4b indicator cells by spectrophotometrically
monitoring the cell free supernatants at 412 nm for the
release of hemoglobin.
Analysis of Cl Activation Status from Hemolytic Activity
Data
The functional activity of serum Cl and fully activated
Cl has been extensively studied and characterized (for


38
the same phosphate buffer. Functional levels of Cl-Inh in
various fractions were monitored by the method of Gigli,
et.al. (1968). A radioimmunoassay was developed to measure
the antigenic levels of Cl-Inh in the various fractions. In
this assay aliquots of solutions containing unknown amounts
of Cl-Inh were added to Immuno-beads (Bio-Rad) coupled with
partially purified Cl-Inh followed by the addition of
commercially prepared Cl-Inh specific rabbit antiserum
(Pel-Freze). After incubation at 37C for 45 minutes and
extensive washing of the beads in VBS-EDTA, the amount of
rabbit antibodies remaining bound to the beads was
125
determined by the amount of I-Protein A that would
specifically bind to the beads in a 45-minute incubation at
37C. The relative concentration of Cl-Inh in the unknown
was determined by relating the inhibition of the binding of
the rabbit antibodies to the beads caused by the fluid phase
Cl-Inh compared to the standard curve of inhibition
generated by the addition of standard amounts of partially
purified Cl-Inh (see Figure 12). This assay could detect
concentrations of Cl-Inh in solutions of less than 1 pg/ml
(data not shown). Fractions showing Cl-Inh activity were
pooled and applied to a G-150 molecular sieving column.
Fractions containing Cl-Inh activity were pooled and further
purified by hydroxylapatite column chromatography.
Fractions containing Cl-Inh activity were pooled. This
material appeared as a single homogeneous band in SDS-PAGE
(Laemmeli, 1970) (Figure 13).


18
Table 1. MEASUREMENT OF SERUM AND ACTIVATED Cl ACTIVITY AS
A FUNCTION OF INDICATOR CELL NUMBER. A fixed dilution of
activated Cl or serum Cl was assayed using various
concentrations of indicator cells (EAC4b). These results
were calculated from the data presented in Figure 4.
Number of EAC4b
Target Cells
1.5 x 107
6.0 x 107
24.0 x 107
Number of Hemolytic Effective
Cl Molecules/ml Measured
Activated Cl Serum Cl
3.00 x 107
2.88 x 107
2.88 x 107
17.0 x 107
13.8 x 107
7.2 x 107


BIBLIOGRAPHY
Arlaud, G.J., C.L. Villiers, S. Chesne, and M.G. Colomb.
1980. Purified proenzyme Clr: Some characteristics of its
activation and subsequent proteolytic cleavage. Biochem.
et Biophys. Acta 616:116-129.
Baker, J.B., D.A. Low, R.L. Simmer, and D.D. Cunningham.
1980. Protease-nexin: A cellular component that links
thrombin and plasminogen activator and mediates their
binding to cells. Cell 21:37-45.
Borsos, T., M. Loos, R.M. Chapuis, R. Medicus, and J.
Isliker. 1980. A novel way of relating the structure of
Clq to the hemolytic activity of the first component of
complement. Molec. Immunol. 17:1415-1421.
Borsos, T., and M.M. Mayer. 1962. Mechanism of action of
guinea pig complement. In Mechanism of Cell and
Tissue Damage Produced by Immune Reactions Ilnd
International Symposium on Immunopathology. Benno
Schwabe Co., Basel (Switzerland), pp. 13-22.
Borsos, T., and H.J. Rapp. 1963. Chromatographic separation
of the first component of complement and its assay in a
molecular basis. J. Immunol. 91:851-858.
Borsos, T., and H.J. Rapp. 1967. Immune hemolysis: A
simplified method for the preparation of EAC'4 with
guinea pig or human complement. J. Immunol. 99:263-268.
Borsos, T., H.J. Rapp, and U.L. Walz. 1964. Action of the
first component of complement: Activation of C'l to C'la
in the hemolytic system. J. Immunol. 92:108-112.
Boyle, M.D.P., M.J.P. Lawman, A.P. Gee, and M. Young. 1985.
Nerve growth factor: A chemotactic factor for
polymorphonuclear leukocytes in vivo. J. Immunol.
134:564-568.
Boyle, M.D.P., and M. Young. 1982. Nerve growth factor:
Activation of the Classical complement pathway by
specific substitution for component Cl. Proc. Natl. Acad.
Sci. 79:2519-2522.
69


inhibition of 125i-pa binding
Figure 12. RADIOIMMUNE INHIBITION ASSAY FOR Cl-INH. The
inhibition of rabbit anti-Cl-Inh antibodies adhering to
Cl-Inh coupled^immuno-beads by fluid phase Cl-Inh was
monitored by I-protein A binding. (Error bars denote
the range of duplicate samples). 50% inhibition was
observed with approximately 20 ng of Cl-Inh added.
I


50
independent of its esterase inhibiting ability. Clearly,
Cl-Inh is a multi-functional molecule and a full
understanding of its regulatory capabilities requires
further investigation.
Recently, nerve growth factor (NGF) was shown to
substitute for aCl in initiating the classical complement
pathway (Boyle and Young, 1982). This ability to cleave and
activate C4 and C2 was dependent on the serine esterase
activity of the gamma subunit of NGF (gamma NGF) (Boyle and
Young, 1982). This activity was inhibited by preincubation
of activated gamma NGF with Cl-Inh (Boyle and Young, 1982).
Nerve growth factor is naturally found as a
multimolecular complex of three polypeptide subunits (alpha,
beta, and gamma chains, in a 2:1:1 molar ratio) (Bradshaw
and Young, 1976; Greene et al., 1969). It was first
described by Levi-Montalcini and Booker (1960) as a soluble
factor that stimulated neurite outgrowth. This nerve growth
inducing activity is mediated by the beta subunit of the
molecule (Levi-Montalcini, 1965). No known biological
activity has been reported for the alpha subunit (Isackson
and Bradshaw, 1984). The gamma subunit in addition to its
role in complement activation has been shown to act as a
plasminogen activator (Orenstein et al., 1978) and topical
administration of enzymatically active NGF has been shown to
enhance the rate of wound healing in mice (Li et al., 1980).
Other researchers have reported that the gamma subunit of
NGF stimulated polymorphonuclear phagocyte chemotaxis i_n


xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EUE3RG5JQ_7ZVJNJ INGEST_TIME 2014-10-03T21:37:39Z PACKAGE AA00025693_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES


7
Figure 2.
ACTIVATION
Z d Slow u
ZCI
%
fast
Cl INA
* aCI
SCHEMATIC OF THE CLASSICAL AUTOCATALYTIC
MODEL.


66
could more effectively control the activation of the
classical complement pathway would have great benefits in
the treatment of HAE and other disorders resulting in
inappropriate Cl activation.
The interaction of Cl-Inh with the gamma subunit of
nerve growth factor is characteristic of that of most serine
protease inhibitors (Travis and Salvesen, 1983). There is a
1:1 stoichiometric complex formed between Cl-Inh and NGF and
this complex is stabilized by the formation of a covalent
bond between the two proteins. Recently, a new group of
cell secreted protease inhibitors known as protease nexins
(PN) have been described (Baker et al., 1980). Three
distinct protease nexins (PNl, PN2, and PN3) have been
identified (Knauer et al., 1983) and their physiological
function appears to be inhibition of the enzyme activity of
selected proteases by complex formation followed by
enhancement of the specific clearance of PN-protease
complexes by a variety of cells (Knauer et al., 1983). One
of these nexins, PN3, reacts in a highly selective manner
with the gamma subunit of NGF (Knauer et al., 1982). The
Cl-Inh-NGF and the PN3-NGF complexes share a number of
similarities and future studies should be directed towards
determining whether Cl-inactivator-NGF complexes have other
biological properties of PN3-NGF complexes.
In the comparison of the properties of nerve growth
factor and Cl a number of similarities have emerged. First,
both can activate the classical complement pathway (Boyle


37
convert to aCl. In the experiments described in this
section, various serum fractions, including purified Cl-Inh,
were shown to inhibit this spontaneous autoactivation. This
activity of Cl-Inh to inhibit serum Cl autoactivation was
distinct from its irreversible covalent interaction with
fully activated Cl.
Materials and Methods
Reagents
The reagents and materials used in these studes were
identical to those reported in the previous chapter.
Cl Assays
Titration of Cl hemolytic activity was performed by the
methods described in the previous chapter.
Isolation and Measurement of Cl-Inh
Human Cl-Inh was purified essentially by the method of
Harrison (1984), using DE-52 in place of DEAE ion-exchange
chromatography. Specifically, outdated plasma was converted
to serum by the addition of excess Ca++ and the
supernatant of a 5% polyethylene glycol precipitation was
passed over a lysine-sepharose affinity column. Tne
pass-tnrougn traction was tnen suojectea to ion-exchange
cnromatograpny on DE-b2 equilibrated in 0.1 M KCl and 10 mM
phosphate pH 7.0 with 0.01 M benzamidine. Bound proteins
were eluted with a linear gradient of 0.1 to .25 M KCl in


CHAPTER I
INTRODUCTION
The activation and control of specific serum proteases
is the basis of many homeostatic processes. These processes
include the clotting system, the fibrinolytic system, the
kinin generating system, and the complement system, all of
which have important roles in inflammation and in the normal
response to tissue damage. Each of these processes entails
the serial interaction of specific serum proteins, in which
the activation of one component can initiate a chain
reaction (cascade reaction) culminating in the generation of
a specific effector molecule, e.g., the activation of
Hageman factor initiates a series of reactions that can
result in the formation of a fibrin clot. Many of the
interactions of these cascades are mediated through the
activation of a proteolytic enzyme from an inactive
precursor (zymogen) molecule. At various points along these
cascade reactions there are specific amplification steps in
which one activated molecule can stimulate the activation of
a large number of subsequent components in the cascade.
During the course of these cascade reactions biologically
active split products are produced, e.g., the anaphylatoxins
of the complement system. It is clear that if such
pharmacologically active split products are produced
1


56
Figure 19. TIME COURSE OF THE INHIBITION OF THE ENZYMATIC
ACTIVITY OF NGF BY Cl-INH. Gamma NGF was added to 1/2 molar
equivalent of Cl-Inh and incubated at 24C for various
times. The enzymatic activity of the gamma NGF was then
analyzed by monitoring the cleavage of 1-BAPNA
spectrophotometrically. The graph depicts the amount of
gamma NGF activity remaining with time (minutes).


65
binding of Clq to Clq receptors on phagocytes. Thus, Cl-Inh
may be considered a nexin of sorts, in that by binding to
the aCl macromolecule and dissociating it, the Clq is more
readily cleared by binding to specific cell receptors.
The interaction of Cl-Inh with gamma NGF led to the
inactivation of the enzymatic activity of the gamma NGF.
Associated with the inactivation was the formation of a
gamma NGF-Cl-Inh complex that was resistant to SDS and
susceptible to hydroxylamine treatment. This interaction is
similar to that seen with Cl-Inh and aCls (Harpel and
Cooper, 1975). In binding the aCls, Cl-Inh becomes cleaved
and new carboxy and amino terminals are formed (Salvasen et
al., 1985). By sequencing these new fragments a partial
amino acid sequence of the enzymatic target site on the
Cl-Inh has been reported (Carrell, 1984; Salvesen et al.,
1985). Other than an arginine residue where the serine
esterase cleaves, this target site sequence is unrelated to
any other enzymatic target site reported for other serpins
(Carrell, 1984). The similarities between the target sites
of the various enzymes inhibited by Cl-Inh could provide
information regarding the specificities of both the
proteases and the Cl-Inh inactivation activity.
Additionally, knowledge of the fine structure of the
enzymatic target site of some serpins has led to new
research in producing more effective protease inhibitors by
site specific mutagenesis and genetic recombination
(Rosenberg et al., 1984). The production of a molecule that


CHAPTER IV
INACTIVATION OF THE PROTEOLYTIC ACTIVITY OF NGF BY Cl-INH
Introduction
In 1959 Levy and Lepow demonstrated an inhibitor of aCl
in human serum, which they named the Cl esterase inhibitor
(Cl-Inh). The Cl-Inh molecule has since been shown to
interact and inhibit a number of other serum proteases
involved in homeostatic cascade reactions (Travis and
Salvesen, 1983). Other serum serine esterases, e.g.
trypsin, chymotrypsin, etc., are unaffected by Cl-Inh
(Harpel and Cooper, 1975; Travis and Salvesen, 1983). The
nature of the interaction and inhibition involves the
formation of a covalent Cl-Inh-enzyme complex (Travis and
Salvesen, 1983) near the active site on the enzyme (Travis
and Salvesen, 1983). The rate of inactivation of aCl by
Cl-Inh (Sim et al., 1979; Travis and Salvesen, 1983)
compared to the enzymatic activity of aCl in cleaving C4 and
C2 (Patrick et al., 1970) has led to speculation that the
primary function of Cl-Inh is not in regulating aCl activity
(Gigli et al., 1970; Travis and Salvesen, 1983). Other
serum proteases, eg. kallikrein, are more effectively
regulated by Cl-Inh (Gigli et al., 1970). Evidence reported
elsewhere (Ziccardi, 1982b, see Chapter III) suggests that
Cl-Inh has the ability to regulate zymogen Cl activation
49


59
Rf
Figure 21. DENSITOMETRY OF SDS-PAGE OF 125I-NGF INCUBATED
WITH VARIOUS CONCENTRATIONS OF Cl-INH. The autoradiograph
from Figure 20 was scanned with a densitometer. The
relative areas under the curves had been shown in previous
experiments to obey Beer's law, and thus were analysed to
estimate the relative amounts of radioactivity in each peak
(panel A No Cl-Inh added, panel B- 1/2 molar equivalent of
Cl-Inh added, and panel C 80 molar excess Cl-Inh added).
Eectrophoresis proceeded from right to left.


I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosophy.
Edward M. Hoffmamvi/
Professor of Micjrqoiology
and Cell' Science
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosophy.
'ames B. Flanegan / /
Associate Professor-/
Immunology and Medical
Microbiology
This dissertation was submitted to the Graduate Faculty of
the College of Medicine and to the Graduate School and was
accepted as partial fulfillment of the requirements for the
degree of Doctor of Philosophy.
May, 1985
Dean, Graduate School


11
1965). The reason for the discrepancy between Lepow's and
Ziccardi's findings has not been established but may be
related to the source of zymogen Cl used and/or the method
chosen for measuring Cl activation.
In the studies presented here, the Cl hemolytic assay
was was shown to be an effective method to discriminate
zymogen from activated Cl and was used to monitor the
activation of serum Cl. Partially purified serum Cl was
shown to spontaneously activate in a concentration-dependent
manner and that the addition of aCl to the zymogen Cl did
not increase the extent of spontaneous activation. These
studies suggest a model of Cl activation that is dependent
on an intermolecular interaction between zymogen Cl
molecules which is independent of the enzymatic activity of
either the aClr or aCls subunits.
Materials and Methods
Reagents
Sheep erythrocytes (E) were collected and washed as
described in Borsos and Rapp (1967). Rabbit antiserum rich
in IgM ant.i-E was produced and depleted of IgG by passage
over a Protein A-Sepharose (Pharmacia, Upsala, Sweden)
affinity column. Ascites fluids containing mouse monoclonal
IgM antibodies specific for E were derived from a hybridoma
(TIB 110) purchased from ATCC, Washington, DC (contributed
by R. Raschlce) Human serum was prepared from outdated
plasma (Civitan Regional Blood Center, Gainesville, FL).


60
Figure 22. AUTORADIOGRAPH OF SDS-PAGE ANALYSIS OF
HYDROXYLAMINE TREATED HIGH MOLECULAR WEIGHT COMPLEXES.
Radiolabeled gamma NGF was incubated with 1/2 molar
equivilant of Cl-Inh and incubated one hour at 37C (lane
1). An aliquot was removed and was made .1 M hydroxylamine
(pH 10.5) and incubated an additional 20 minutes at 37C
(lane 2). The solutions were then subjected to SDS-PAGE
separation and autoradiographed.
*r~