SECOND (COl]NELL) MEDICAL DIVISION
LIBRARY BELLEVUE HOSPITAL
MEDIATORS OF INFLAMMATION IN LEUKOCYTE LYSOSOMES
II. M~CHANIS~ OF AdiOS OF LYSOSO~:AL CATmmc PROTEIN
UPON VASCULAR PERMEABILITY IN THE RAT*
B~¢ AARON JANOFF,:~ PH.D., SONJA SCHAEFER,§ JOAN SCHERER, H A~m
MICHAEL A. BEAN¶
(From the Department of Pathology, New York University Sckool of Medicine,
(Received for publication, June ll, 1965)
Considerable emphasis has been placed on the role of acid-proteases within lyso-
somes of polymorphonuclear leukocytes (PMNL) in the pathogenesis of allergic and
other forms of tissue injury (1, 2). We reported previously (3) that a cationic protein-
containing fraction of PMNL lysosomes, extracted by weak mineral acid and precipi-
tated by ethanol at a 20 per cent (v/v) concentration, produced inflammatory changes
in the microcirculation of rat and rabbit mesentery. However, such cationic protein
fractions of leukocyte granules are reportedly free of acid-proteases (4) and other
hydrolases (3) normally associated with intact leukocyte lysosomes. This apparent
discrepancy led us to suggest alternative mechanisms of action of lysosomal cationic
protein fractions in local tissue injury reactions and to propose that a part, at least,
of the inflammatory changes induced by whole granule preparations might be medi-
ated independently of their protease content.
In the course of these earlier investigations it was found that the cationic
protein fraction, in addition to causing adhesion and emigration of leukocytes,
also produced a marked increase in vascular permeability to circulating Evan's
blue dye when the protein was applied topically onto rat mesentery or when it
was injected intracutaneously into rat skin. However, in vitro tests using iso-
lated smooth muscle preparations failed to reveal any kinin-like, serotonin, or
histamine activity in this extract (3). Further studies have now been carried
out on the mechanism of action of crude extracts of PMNL lysosomes, and
* This work was supported by a grant from the United States Public Health Service
J~ Career Development Award Fellow of the United States Public Health Service (GM-
§ Candidate for the degree of Master of Science. Part of the work reported here has been
submitted in partial fulfillment of the requirements for this degree at the New York Uni-
versity Graduate School of Arts and Science.
 Research trainee of the United States Public Health Service (HF_,-5501).
¶ Research trainee of the United States Public Health Service. Present address: Colorado
University School of Medicine, Denver.
also of the cationic protein isolated from these granules, upon vascular perme-
ability in skin and cremasteric muscle of the rat. The present report will present
evidence which suggests that the early increase in vascular permeability pro-
duced in rat tissues by PMNL lysosomes is caused by vasotropic agents dis-
charged from tissue mast cells and not by protease activity of the granules.
This effect is due to a mastocytolyfic agent present in the cationic protein
fraction of PMNL lysosomes.
Materials and Methods
Preparation of Leukocyte Lysosome Fractions.--Rabbit peritoneal exudate polymorpho-
nuclear leukocytes (PMNL) were collected and their granules isolated by the technique of
Cohn and Hirsch (5). Superuates of frozen-thawed granules and cationic protein-containing
extracts of the granules were prepared by previously described techniques (3) except that in
the present studies dialysis of the fractions was carried out against sodium bicarbonate-
buffered, mammalian-Ringer's balanced salt solution at pH 7.2. Protein content of the various
fractions was determined by the method of Lowry (6), with crystalline bovine serum albumin
as a standard.
General Method of Permeability Tests in Rat Skin.--The method employed to evaluate
permeability effects in rat skin was essentially similar to that described by Miles and Wilhelm
(7). Materials to be tested were injected in a volume of 0.05 ml, intracutaneously, into the
shaved abdominal skin of rats (males, Sprague-Dawley strain, 250 to 350 gm body weight)
anesthetized with sodium pentobarbital (nembutal, Abbott Laboratories, North Chicago,
Ill.). In each of the test animals at least two separate sites were prepared at each dose of
test agent. Either immediately or at varying times after the intracutaneous administration of
test substances, Evan's blue dye was injected intravenously as a 1 per cent solution in iso-
tonic saline and at a dose of 0.4 ml per 100 gm of body weight. Animals were regularly sacri-
ficed 20 minutes after dye injection and the abdominal skin was reflected for examination
of the undersurface. The grade of edema in the injected skin site was estimated according
to the diameter and color intensity (visual observation) of the dye accumulated in the tissues.
General Method of Permeability Tests in Cremaster Muscle of the Rat.--Permeability effects
in rat cremasteric muscle were studied using the carbon-labelling technique described by
Majno (8). Aliquots of 0.05 ml of the materials to be tested were injected subcutaneously
into the scrotal sac just over the mid-ventral surface of the testis. Either immediately or at
varying times after the injection of test substances, carbon suspension (pelikan Cl1-1431a,
Gunther Wagner, Germany) was injected intravenously (0.1 mi per 100 gm body weight).
Animals were sacrificed 1 hour after injecting the carbon suspension, and the testes were
placed briefly in 10 per cent formalin. The muscle was then dissected free of other structures,
stretched, and pinned on dental wax, fixed in formalin for an additional 24 hours, cleared in
xylene, and mounted in glycerine-jelly for low power light microscopic observation. In this
test, vessels whose permeability has been abnormally increased by the subcutaneous injection
of test material appear blackened by carbon particles which have passed between endothelial
cells. Furthermore, the characteristic topography of the microcirculation in rat cremasteric
muscle (8) readily permits identification of the type of vessel (venule, capillary, or arteriole)
most affected by the test agent.
Assay of Mastocytolytic Activity in Vitro.--The effect of leukocyte granule fractions upon
rat mast cells in ~itro was assayed according to the technique described by Norton (9). Briefly,
this involved incubation of pieces of rat mesentery in Ringer-Locke glucose solution buffered
with bicarbonate to pH 7.45 and containing varying concentrations of either leukocyte frac-
A. JANO~F~'~ S. SCHAEFER, J. SCHERER, AND M. A. BEAN
tions or known mastocytolytic agents such as compound 48/80 (Burroughs Wellcome and
Co., Inc., Tuckahoe, New York). Thirty minutes of incubation at room temperature was
routinely employed in all tests; and, incubation in Ringer-Locke alone was always included
as a control for spontaneous mast cell rupture. Following the period of incubation, the tissues
were fixed in 10 per cent formalin containing 0.1 per cent toluidine blue, dehydrated in ace-
tone, cleared in xylene, mounted in permount (Fisher Scientific Co., Fairlawn, New Jersey),
and examined microscopically for the enumeration of intact vs. disrupted mast cells.
Details of methodology which were varied according to the needs of each experiment are
described in the appropriate sections of the Results.
Time Course of Vascular Permeability Change Produced by PMNL Lysosomes in Rats
protein (20 per cent
Reaction/No. of rats in test
30 rain.* 60 mln.*/120 rain.*
+ (skin), fiat, pale-blue, uneven reaction, 5 mm or less in diameter.
++ (cremaster), postcapillary collecting veunles show carbon deposits over area 10 to
15 mm in diameter.
0 (no reaction), no blue dye grossly visible (skin) or carbon deposits microscopically visible
* Elapsed time between intracutaneous (skin) or subcutaneous (cremaster) injection of
test agent and intravenous injection of Evan's blue (skin) or carbon suspension (eremaster).
(skin), raised, intensely blue wheal, 12 to 15 mm diameter.
Time Course of the Permeability Change Produced in Skin and Cremaster
Muscle of Rats by PMNL Lysosomes.--Table I shows that the permeability
changes induced in skin and cremaster muscle of rats by single injections of
PMNL lysosomes (both extracts of frozen-thawed granules and cationic pro-
tein fractions) are acute, transient changes which disappear rapidly.
When animals were injected intravenously with Evan's blue either immedi-
ately before or after intracutaneous application of lysosomal materials, an
intense exudation of plasma-bound dye developed, rapidly reaching a maxi-
mum diameter of 12 to 15 mm in 15 or 20 minutes. However, when dye was
IN'FL.A MMA.TORY MECHANISM
injected 30 or 60 minutes after the lysosomal fractions, only trace reactions
occurred (pale, incomplete staining); and, at 2 and 3 hours after injection of
materials into the skin, no discernible exudation of dye was observed except
at the finely circumscribed point of passage of the needle into the skin (see
Table I). The absence of dye exudation at later intervals was not exclusively
the result of mechanical interference with local blood flow following engorge-
ment of tissue spaces with fluid during the early phase of the reaction. This was
shown by the fact that injection of 2/zg of histamine into the same skin sites
previously treated with lysosome fractions produced typical histamine blueing
reactions at these sites in the later time-groups (2 and 3 hours).
When animals received an intravenous injection of carbon suspension im-
mediately after subcutaneous application of lysosomal materials over the
cremaster muscle (frozen-thawed granules or cationic protein), a widespread
blackening of venules developed in the muscle, in a pattern identical with that
produced in this tissue by similar injections of histamine. However, if carbon
suspension was administered 2 hours after injection of lysosomal materials
over the muscle, no labelling of vessels could be discerned (see Table I). Again,
it could be shown that potential vascular reaction was still present in the
cremaster at this time, since reinjection of the scrotal sac with the standard
dose of histamine just before the administration of carbon produced a normal
venular-labelling pattern in animals that had been treated with lysosomal
materials 2 hours earlier.
Although the doses of lysed granule protein and cationic protein shown in
Table I were the ones most frequently employed in permeability tests, other
experiments have shown that strong permeability changes (2+ and 3+ re-
actions, see Table I) can be produced by as little as 2 #g of protein in the skin
and 10 #g in cremaster muscle.
Mastocytolytic Activity of PMNL Lysosomes in Vitro.--In view of the acute,
transient nature of the vascular permeability change produced in rat tissues by
PMNL lysosomal fractions, it seemed reasonable to suppose that the reaction
might be mediated by vasotropic agents released from tissue mast cells which
had been disrupted by contact with the injected lysosomal material. It was
therefore decided to investigate the effect of PMNL lysosomes upon rat mast
cells in vitro, in order to determine whether the lysosomal fractions employed
in our permeability tests were, in fact, capable of rupturing these cells.
Table II shows the results of these experiments. Both lysosomal cationic
protein (20 per cent ethanol fraction) and extracts of frozen-thawed lysosomes
were compared with a known mast cell-rupturing agent, compound 48/80
(Burroughs Wellcome and Co., Inc., Batch No. NS280). It can be seen that
both of the lysosomal preparations produced a marked mastocytolyfic effect
in these tests, although the activity of the lysosome fractions was less than
one-tenth that of compound 48/80 when these agents were compared on a
A. JANOFF, S. SCHAEFER, J. SCHERER, AND M. A. BEAN
simple weight per weight basis. However, if the lysosomal cationic protein
was compared with compound 48/80 on a molar weight basis (see Discussion),
then the two agents appeared to be more nearly equal in mastocytolytic po-
Effects of PMNL Lysosomes on Rat Mast Cells In Vitro
Percentage disruption of mast cells*
Test material Concentration (/~g) $ EDio§
".P. (lysosomal cationic protein, 20
per cent ethanol fraction)
~.P. in presence of trasylol[[
~.T. (extract of frozen-thawed lyso-
~.T. in presence of trasytolll
C.P. = cationic protein fraction (20 per cent ethanol); F.T. ffi extract of frozen-thawed
* Percentage disruption is based on the number of mast cells showing granule extrusion
out of 1500 cells examined (X640).
~g axe per ml in the case of compound 48/80 and per 0.1 ml in the case of all other
§ Concentration of test agent calculated to produce 50 per cent disruption of mast cells
(estimated from log dose/response plot).
II Trasylol -~ zymofren (Societe Paxisienne d'Expansion Chimique---S.A., Lot No. ES
1402) present in a concentration of 10 kallikrein-inhibiting units per ml.
Preliminary evidence was also obtained which suggested that the action of
PMNL lysosomal cationic protein upon mast cells was not simply due to
the basic groups on these protein molecules, but rather may have been asso-
ciated with some as yet unidentified, specific component of the granules, present
only in the 20 per cent ethanol-precipitated fraction of acid extracts of the
lysosomes (see Discussion).
In view of Archer's recent report (10) that a basic protein fraction derived
from rat eosinophil granules is capable of rupturing rat mast cells in vitro
and is also especially rich in peroxidase activity, the mastocytolytic fraction
of PMNL lysosomal cationic protein was also tested for peroxidase activity in
our experiments. Using an assay system for this enzyme based on the colori-
metric measurement of the rate of oxidation of o-dianisidine, and possessing a
sensitivity sufficient to detect 0.01 ~g of crystalline horseradish peroxidase
(Sigma Chemical Company, St. Louis), it nevertheless proved impossible to
detect any activity of this enzyme in as much as 100/zg of the cationic protein
Finally, it can also be seen from the data in Table II that the mast cell-
rupturing activity of each of the PMNL lysosomal extracts was not inhibited
in the presence of salivary kallikrein-inhibitor, a broad-spectrum proteolytic
enzyme-inhibiting drug (zymofren, trasylol, S.A., Lot No. ES 1402, Societe
Parisienne d'Expansion Chimique, Paris). At the concentration of the drug
employed in these tests (see Table II) the protease-inhibitor produced, by
itself, a slight increase in the percentage disruption of mast cells over con-
Effect of PMNL Lysosomes Upon Vascular Permeability in Rats Pretreated
With a Mast Cell-Depleting Agent or an Antihistamine Drug.--Having made the
observation that PMNL lysosome fractions possessed mastocytolytic activity
in vitro and also produced vascular permeability changes in rat tissues which
closely resembled the known vascular responses to vasotropic agents contained
in rat mast cells, it seemed appropriate to investigate the effect of prior reduc-
tion in numbers of mast cells or prior treatment with an antihistaminic upon
the vascular reactions of rats to these lysosomal materials.
In order to reduce the mast cell population in the skin, animals were given
repeated injections of compound 48/80 according to the scheme proposed by
Riley (11). The agent was administered intraperitoneally twice daily in the
following doses: 100/~g on the 1st day, 200/~g on the 2nd, 300 #g on the 3rd,
and 400 #g on the 4th and last day. Animals were then tested 24 hours after
the final injection of the compound. Other rats were injected intravenously
with the antihistamine drug chlorpheniramine maleate (Schering Corporation,
Bloomfield, New jersey), 6 mg per 100 gm body weight, immediately before
carrying out the permeability tests. The results of these experiments are shown
in Table III. It can be seen that chronic pretreatment with compound 48/80
almost completely suppressed the vascular permeability changes normally
induced in skin by 10/zg of lysosomal cationic protein or by 10/~g of protein
from frozen-thawed granules. In addition, the effects of the cationic protein
(20 #g quantities) in the microcirculation of cremaster muscle were also blocked
by this pretreatment. Injection of an antihistaminic prior to the test also
inhibited the effects of frozen-thawed granule supernate and cationic protein
A. JANOFF, S. SCHAEFER, J. SCHERER, AND M. A. BEAN
upon vessel permeability in rat skin, although the degree of suppression was
slightly less pronounced than in the 48/80 pretreated group. Cremaster tests
were not performed in antihistamine-pretreated animals. Although not shown
in the table, pretreatment with this antihistaminic also markedly reduced the
permeability response in rat skin to doses of lysosomal cationic protein as large
as 50 and 100 #g per test site.
In several of the antihistamine-pretreated rats, representative skin sites
were excised 40 minutes after intracutaneous injection of lysosomal cationic
The Effect of Mast Call Depletion and an Antihistamine Drug upon Vascular Permeability
Changes in Rats Induced by PMNL Lysosomes
Single injection of chlorpheni-
ramine maleate [I
Lysosome fraction *
* C.P. = cationic protein fraction (20 per cent ethanol); F.T. = extract of frozen-thawed
$ Permeability changes were evaluated in all animals during the period of maximal re-
sponse (injection of dye or carbon suspension at zero minutes).
§ Intraperitoneal injections twice dally (doses are stated in accompanying text); animals
tested 24 hours after the last dose.
II 6 mg per 100 gm body weight administered intravenously immediately prior to test.
protein and were formalin-fixed and stained with Harris' hematoxylin-eosin.
This was done in order to compare the leukocyte emigration occurring in re-
sponse to the lysosomal cationic protein (3) in normal rats with that occurring
in animals whose permeability responses had been inhibited by the antihis-
tamine drug. Although the semiquantitative techniques used in this study did
not permit a meaningful comparison of the degrees of inhibition of the two
reactions (permeability response vs. leukotactic response), it was clear that the
latter reaction was also partly inhibited in the skin of antihistamine-pretreated
rats. However, in the rat mesentery, chlorpheniramine maleate did not sup-
press margination of leukocytes caused by the cationic protein. This question
is currently undergoing further study.
The foregoing observations indicate the presence of a mastocytolytic agent
in the lysosomes of rabbit exudate polymorphonuclear leukocytes. This masto-
cytolytic agent is specifically contained in PMNL granules and is not an un-
related cytoplasmic substance contaminating the granule preparation. This
is dearly shown by the fact that the granule-free cytoplasmic fraction of ho-
mogenized leukocytes possesses no demonstrable mast cell-rupturing activity,
even when tested at protein concentrations 10 times greater than that which
produces 50 per cent rupture of mast cells in the case of granule protein (12).
The active principle can be extracted from PMNL granules by weak mineral
acid and is retained in the cationic protein fraction precipitated by ethanol at
a 20 per cent concentration. The mastocytolytic activity present in this frac-
tion of lysosomal basic protein does not appear to be due to non-specific effects
of positive charges. This is shown by the fact that a second basic protein frac-
tion of the granules, precipitated from the acidified extract at 45 per cent con-
centration of ethanol and containing more positively charged species of protein
(as judged from electrophoretic data), is inactive against mast cells in vitro
at concentrations which produce significant disruption of cells in the case of
the 20 per cent ethanol fraction (12). Furthermore, neither arginine-rich nor
lysine-rich preparations of calf-thymus histones produce significant masto-
cytolysis in vitro at concentrations 10 times greater than the ED100 concentra-
tion of the active PMNL basic protein fraction (12). These observations sug-
gest that the activity present in the 20 per cent ethanol fraction of PMNL
lysosomal cationic protein may be due to a highly specific component of this
fraction acting through a mechanism independent of simple charge effects.
Studies currently in progress, designed to further fractionate the active mix-
ture of cationic proteins, may provide additional information concerning this
aspect of the problem.
Previous work (3) has already shown that the active cationic protein frac-
tion is free of a number of the hydrolytic enzymes present in crude extracts of
PMNL lysosomes; and, in the experience of other workers (4) such fractions
are also devoid of catheptic activity. Moreover, the mastocytolyfic effect of
this material in vitro is not suppressed by a broad-spectrum protease inhibitor
(trasylol). These tests were carried out at the highest concentration of the
antiprotease drug which can be employed without causing damage to the mast
cells sufficient to interfere with the assay. Unlike the experience of Archer (10)
with mastocytolytic basic protein obtained from eosinophil granules, the active
PMNL fraction does not contain detectable levels of peroxidase. Earlier work
(3) has also shown that the active PMNL fraction is non-pyrogenic and, on
A. ~'ANOFF, S. SCHAEFER, J. SCHERER, AND M. A. BEAN
this basis, the mastocytolytic agent present in PMNL granules would not ap-
pear to be related to the leukocyte permeability factor ("granulocytic sub-
stance") described by Moses and coworkers (13).
Frimmer has recently extended his studies on the general mechanism of
action of nuclear histones and basic polypeptides upon vascular permeability
to include investigations of artificial membrane systems (14). He suggests a
mechanism of altered permeability in his membrane system based on physico-
chemical changes in acid-mucopolysaccharides caused by their interaction
with basic protein. Thus, there may be additional modes of action of PMNL
lysosomal cationic proteins upon vessel wall permeability in vivo in addition
to the indirect effect mediated through mast cell degranulation. It should be
noted that the permeability-enhancing lysosomal cationic protein fraction was
tested by us for acld-mucopolysaccharases and was found to be free of measur-
able chondroitin sulfatase and hyaluronidase activity (turbidity-reduction
assay method of Tolksdorf et al. (15), modified after Kass and Seastone (16)).
The data presented in this paper show that the cationic protein is less active
than compound 48/80 when the two are compared on a weight per weight basis.
Until the PMNL factor is identified and characterized it will not be possible
to accurately compare its potency with that of compound 48/80 on a molar
weight basis, but reasonable estimates can be made on the grounds of presently
available facts. Since the mastocytolytic principle of PMNL granules is not
dialyzable (see Materials and Methods) it can be assumed to possess a molecu-
lar weight of at least 10,000. On the other hand, the molecular weight of the
active form of compound 48/80 is known. It has been reported (17) that the
trimer (mol wt = about 500) and tetramer (mol wt = about 650) forms of the
compound are active, while polymers of the pentamer class or higher are in-
active. Thus, if the two mastocytolytic principles are compared on the basis
of estimated molar weights, the PMNL principle is highly active with reference
to compound 48/80.
In conclusion, while these studies do not exclude the possible role of lyso-
somal acid proteases in terminal phases of necrotizing tissue injury reactions
such as the Arthus reaction (1, 2) or the dermal Shwartzman reaction (18,
19), the present studies do show that the early phase of vascular injury induced
in rat tissues by extracts of frozen-thawed PMNL lysosomes, or by certain of
the cationic proteins isolated from these granules, is mediated by vasotropic
agents released from disrupted mast cells rather than by proteases present in
the leukocyte granules themselves. It therefore seems reasonable to suggest
that PMN leukocytes entering perivascular tissue spaces during naturally-
occurring tissue injury reactions, release their lysosomal granules and that
these, in turn, may affect local mast cells in a manner analogous to that de-
scribed in the in vitro studies reported here. Such a reaction would constitute
a distinct mechanism by which leukocyte lysosomes play a part in the patho-
genesis of inflammation.
The vascular permeability-increasing action of rabbit PMNL lysosomes has
been studied in skin and cremaster muscle of the rat.
Both an extract of frozen-thawed granules and a cathepsin-free cationic
protein fraction of the granules (which had previously been demonstrated to
cause leukocyte adhesion and emigration in vivo) induce increased vascular
permeability in skin and muscle which resembles that produced by histamine
or histamine-liberators with respect to the timing of the response and the pre-
dominant type of microvessel affected.
Extracts of frozen-thawed lysosomes and the inflammatory lysosomal cationic
protein both cause disruption of rat mesenteric mast cells in vitro, whereas a
granule-free cytoplasmic fraction of PMN leukocytes and a non-inflammatory
cationic protein fraction of the granules do not do so under identical test con-
ditions. The mastocytolytic action of lysosomal materials in vitro is not in-
hibited in the presence of 10 kaUikrein-inhibiting units of trasylol per mi. The
mast cell rupturing fraction of PMNL granules (cationic protein) possesses
no detectable peroxidase activity or acid-mucopolysaccharase activity. When
compared with compound 48/80 on the basis of estimated molecular weight,
the lysosomal cationic protein appears to be at least as active as the latter
compound with respect to in vitro mastocytolytic potency.
Chronic pretreatment of rats with an agent known to reduce tissue mast
cell numbers causes marked suppression of the vascular permeability change
normally induced in skin and muscle by lysosomal extracts and cationic pro-
tein. Similar results are obtained if lysosomal materials are tested in rats
pretreated with an antihistaminic.
These observations are discussed with respect to the mode of action of PMNL
lysosomes in the early and late phases of local tissue-injury reactions.
The authors wish to express their indebtedness to Dr. Benjamin W. Zweifach, whose sug-
gestions at the outset of this work and whose helpful criticisms during its subsequent course
of development provided an invaluable assistance.
1. Movat, H. Z., Uriuhara, T., Macmorine, D. R. L., and Burke, J. S., A perme-
ability factor released from leukocytes after phagoeytosis of immune complexes
and its possible role in the Arthus reaction, Life Sc., 1964, 3, 1025.
2. Burke, J. S., Uriuhara, T., Macmorine, D. R. L., and Movat, H. Z., A perme-
ability factor released from phagocytosing PMN-leukocytes and its inhibition
by protease inhibitors, Life Sc., 1964, 3, 1505.
3. Janoff, A., and Zweifach, B. W., Production of inflammatory changes in the
microcirculation by cationic proteins extracted from lysosomes, dr. Exp. Med.,
1964, 120, 747.
A. JAN0~'F, S. SCHAEI~ER, J. SCHERER, AND M. A. BEAN
4. Golub, E. S., and Spitznagel, J. K., Leukocyte lysosomal basic protein as a
mediator of inflammation, submitted for publication.
5. Cohn, Z. A., and Hirsch, J. G., The isolation and properties of the specific cyto-
plasmic granules of rabbit polymorphonuclear leucocytes, J. F~xp. Meg., 1960,
6. Lowry, O. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J., Protein meas-
urement with the Folin phenol reagent, J. Biol. Chem., 1951, 193, 265.
7. Miles, A. A., and Wilhelm, D. L., Enzyme-like globulins from serum reproducing
the vascular phenomena of inflammation. I. An active permeability factor
and its inhibitor in guinea pig serum, Brit. Y. Exp. Path., 1955, 36, 71.
8. Majno, G., Palade, G. E., and Schoefl, G. I., Studies on inflammation. II. The
site of action of histamine and serotonin along the vascular tree: A topographic
study, J. Biophys. Biochem. Cytol., 1961, 11, 607.
9. Norton, S., Quantitative determination of mast cell fragmentation by compound
48/80, Brit. ]. Pharmacol., 1954, 9, 494.
10. Archer, G. T., and Jackas, M., Disruption of mast cells by a component of eosin-
ophfl granules, Nature, 1965, 9.05, 599.
11. Riley, J. F., and West, G. B., Tissue mast ceils. Studies with a histamine-liberator
of low toxicity (compound 48/80), Y. Path. Bacteriol., 1955, 69, 269.
12. Schaefer, S., and Janoff, A., Effects of leukocyte lysosome fractions on rat mast
cells in vitro. M.S. thesis, to be published.
13. Moses, J. M., Ebert, R. H., Graham, R. C., and Brine, K. L., Pathogenesis of
inflammation. I. The production of an inflammatory substance from rabbit
granulocytes in vitro and its relationship to leucocytic pyrogen, Y. Exp. Meg.,
1954, 120, 57.
14. Frimmer, M., and Buddecke, E., Untersuchungen zum Wirkungsmechanismus
permeationsfSrdernder basicher Polypeptide, Z. Naturforsch., 1%4, 19b, 789.
15. Tolksdorf, S., McCready, M. H., McCullagh, D. R., and Schwenk, E., The
turbidometric assay of hyaluronidase, J. Lab. and Clin. Meg., 1949, 34, 74.
16. Kass, E H., and Seastone, C. V., The role of the mucoid polysaccharide (hya-
luronic acid) in the virulence of group A hemolytic streptococci, Y. Exp. Meg.,
1944, 79, 319.
17. Baltzly, R., Buck, J., de Beer, E., andWebb, F., A family of long-acting depres-
sors, J. Am. Chem. Soc., 1949, 71, 1301.
18. Thomas, L., Possible role of leucocyte granules in the Shwartzman and Arthus
reactions, Proc. Soc. Exp. Biol. and Meg., 1964, 115, 235.
19. Halpern, B. N., Inhibition of the local hemorrhagic Shwartzman reaction by a
polypeptide possessing potent antiprotease activity, Proc. Soc. Exp. Biol. and
Meg., 1964, 115, 273.