- Email: [email protected]
Equine alternative pathway activation by unsensitized rabbit red blood cells
Veterinary Immunology and Immunopathology, 9 (1985) 71--85 Elsevier Science Publishers B.V., Amsterdam -- Printed in The Netherlands
71
EQUINE ALTERNATIVE PATHWAYACTIVATION BY UNSENSITIZED RABBIT RED BLOODCELLS R. WES LEID,I, 2 S. CLAIRE COLEY,3 D.P. BLANCHARDl and L.E. PERRYMANl IDepartment of Veterinary Microblology/Pathology, Washington State University, Pullman, Washington, 99164 (U.S.A.) 2person to whoma l l correspondence is addressed. 3Department of Pathology, Michigan State University, East Lansing, Michigan, 48824 (U.S.A.)
(Accepted 4 September 1984) ABSTRACT Leld, R. Wes, Coley, S. Claire, B1anchard, D.P. and Perryman, L.E., 1985. Equine alternative pathway activation by unsensltlzed rabbit red blood cells. Vet. Immunol. Immunopathol., 9: 71-85. The equine alternative complement pathway has been p a r t i a l l y characterized and compared to the equine classical activation pathway. A dose-dependent lysls of RbRBCwas observed with peak l y t l c values noted within lO minutes at 37°C when rabbit red blood cells (RbRBC) were used as an alternative pathway activator. Sheepred blood cells (SRBC) sensitized with rabbit hemolysin or p a r t i a l l y purified equine IgM antibodies were equally sensitive to lysls. Dilution of the commercial hemolysin by I/5 reduced lysls from 90% to 38% in the presence of constant cell numbers. Hemolysis of SRBCpeaked at lO minutes and the majority of lysls occurred within lO minutes. Dilution of equine sera by as l i t t l e as I/5 decreased hemolytic a c t i v i t y for SRBC to 21.5% from greater than 90% with undiluted sera. The alternative pathway protein, equine factor B, was tested using RbRBC and monitored by Its d i f f e r e n t i a l susceptibility to heat treatment at 50oC. Thls treatment led to almost complete inactivation after a 15-mlnute incubation. An apparent heat-dependent decay of certain classlcal pathway components was also observed after 50oC treatment. This sensitivity was indicated by a reduction in the l y t l c a c t i v i t y for sensitized SRBC. Treatment for 15 minutes at 56oC wlth either RbRBCor SRBC was sufficient to abolish hemolytic a c t i v i t y in a l l equine sera tested. Chelation of cations with 0.04 M EDTA blocked expression of alternative and classical pathway act~vatlon; however, chelation of Ca++ Ions with lO mM EGTA containing l mM Mg++ Ions permitted lysls of the RbRBC but not the SRBC. A dose-related Mg+÷-lon dependence for RbRBC hemolytic a c t i v i t y was observed as the concentration of Mg÷+ was increased to l.O mM. In addition, our results obtained with pre-colostral foal serum strongly suggest that natural antibody to RbRBCwas of l i t t l e importance in the lysls observed with these cells. These results also show that the equine alternative pathway activation may require Ca++ ions. I f Ca÷÷ ions are required, the equine alternative pathway is quite different from any other mammaliancomplement system so far described. Our results suggest that the alternative pathway of activation is of ma~or importance In the equine complement system. Confirmation of this hypothesis requires both purification of the components involved as well as further characterization.
0165-2427/85/$03.30
© 1985 Elsevier Science Publishers B.V.
72 INTRODUCTION Equine complement Is considered nonhemolytlc in the standard sheep erythrocyte-rabblt antibody target system (Muir, 1911, Rice, 1950a,b; Leon and Norden, 1959; Rlce and Crowson, 1950), and is regarded as an excellent conglutlnln reagent (Rice, 1950b; Bordet and Streng, 1909; Hole and Coombs, 1947). The decreased hemolytic a c t i v i t y in equine serum for antlbody-sensltlzed SRBC has been attributed to nonexistent or very low levels of one or more complement components. Barta and Hubbert (IgTB) reported undetectable or low t l t e r s of C4, C2, C3 and C5 in equine sera and postulated the presence of a Cl Inactlvator. Leon and Norden (1959) found equine complement could readily form an EAC142 c e l l , which could be lysed only by the addition of EDTA-treated guinea plg serum, which supplied C3 and the terminal l y t l c components. Antlbody-coated erythrocytes were successfully lysed by equine complement using ox erythrocytes sensitized with cat hemolysin (Muir, 1911) or rabbit erythrocytes sensitized wlth sheep antibodies (Barta, et a l . , 1973; Rlce and Boulanger, 1952; Perryman et a l . , 1971; McGulre, et a l . , 1975, 1976).
I t Is
not clear whether hemolysis with RbRBC Is an antlbody-medlated event or is a result of cell-surface activation as in the case of human and bovine complement (Blanchard and Leld, 19B4; Eearon and Austen, 19BO). Preliminary work reported by Coley and Leld (19B2) suggests that the equine alternative pathway is also activated by the RbRBC. Our assessment of hemolysis of the unsensltlzed rabbit red blood cell presents evidence that equine complement possesses a potent alternative pathway.
We also demonstrate that lysls can be achieved using the standard
SRBC-rabblt antibody target system. MATERIALS AND METHODS
Equine sera, sheep (SRBC) and rabbit (RbRBC) erythrocytes Sera c o l l e c t e d from normal, heal.thy, a d u l t horses were prepared as described by Coley and Leld (1982) and used u n d i l u t e d or d i l u t e d w i t h veronal b u f f e r e d s a l i n e (VBS) c o n t a i n i n g g e l a t i n (0.1%, w / v ) , 0.15 n~1 Ca++ , and 0.5 mM Mg÷+ (GVB++).
Sera were used Immediately a f t e r
thawing; samples were never r e f r o z e n and used. u n s e n s l t l z e d RbRBC were measured by m l c r o t l t r e
Hemagglutlnln a n t i b o d i e s f o r h e m a g g l u t l n a t l o n techniques
but were not d e t e c t a b l e in any of the equine sera used. c o l l e c t e d and t r e a t e d as described by Mayer (1961).
SRBC and RbRBC were
?3 Equine sera were also collected from foals before and within 24 hours after the f i r s t ingestion of mare's colostrum.
These sera were processed
for preservation of complement a c t i v i t y as described for adult horses. Immunoglobulln G and M levels were quantltated by single radial Immunodlffuslon as described by Glaze et al. (1984). Antibodies SRBC were sensitized by the methods of Mayer (1961) using e i t h e r commercially produced rabbit hemolysin (Cordis Laboratories, Miami, Florida) or an equine anti-SRBC IgM preparation.
The rabbit hemolysin was used at a
concentration lO0-fold that recommended by the manufacturers for human or guinea pig complement work; this concentration was necessary to produce optlmal hemolytic conditions. concentration of antibody.
Agglutination was not observed at thls
In a d d i t i o n , the c ells were washed 3x in GVB=
pri o r to use in the hemolytic assays.
We encountered no d i f f i c u l t i e s
washing that might have been caused by heavy antibody s e n s i t i z a t i o n .
in The
equine IgM antibodies were e l i c i t e d and p u r i f i e d by the methods of Dyer and Leid (1983).
B r i e f l y , a horse was injected Intravenously with 30 ml of a 1%
solution of SRBC. Serum was collected from the horse 12 days l a t e r and antibodies obtained a f t e r sequential f r a c t i o n a t i o n on DEAE c e l l u l o s e and Sephacryl S-300 columns.
The chromatographic fractions containlng IgM
antibodies were i d e n t i f i e d by hemagglutinatlon of non-sensitlzed SRBC and by immunoelectrophoresls of the fractions against a polyvalent anti-equine whole serum. Those p o s i t i v e fractions were pooled, concentrated, a11quoted and stored at -70°C.
RbRBCwere used unsensitized.
Assay for hemolytic complement Hemolytlc complement t l t e r s were measured using a modiflcatlon of the methods of Mayer (1961) as described by Coley and Letd (1982).
Briefly,
sensitized SRBC or unsensltlzed RbRBC were suspended In GVB++ at a concentration of 1 x 10]0 c e l l s per ml.
A 100-~1 volume of undiluted
equine sera was added to 100 ~1 of sensitized SRBC or unsensltlzed RbRBC. The tubes were gently mixed and then incubated at 37oC for 60 minutes. A l l reactions were stopped by the addition of 1.8 ml of cold VBS to each tube, followed by c e n t r l f u g a t l o n (300 x g, 10 minutes).
The degree of
hemolysls was determined by reading the absorbance at 414 nm In a G l l f o r d spectrophotometer (Stasar I I , G l l f o r d Inst. Co., Oberlin, Ohio).
Cells in
control tubes were lysed with d l s t l l l e d water to establlsh the 100% ly s ls
74 level while other control tubes included cells with buffer alone to establish background l y t l c levels.
An additional control was required since
undiluted equine sera have considerably more color than sera normally used in complement assays. T h i s control was established by recording the OD414 of the sera from each horse (O.l ml sera ÷ 1.9 ml of VBS) on each experimental date. 50QC and 56oC heat inactivation of equine sera Equine sera were heat inactivated as described by Coley and Leld (1982). After heat treatment, sera were cooled immediately in an Ice-water bath, rewarmed to 37oC, and tested for hemolytic a c t i v i t y with the SRBC and RbRBC systems. Chelation of cations with EDTA and EGTA Chelation of cations in equine sera was done as previously described by Coley and Leld (1982).
B r i e f l y , equine sera were incubated with 40 mM EDTA
in GVB= (GVB without Ca÷+ and Mg÷÷ ions) for 15 minutes prior to the introduction of either the SRBC or RbRBC; the level of hemolysis was then determined.
Sera were also incubated with lO mM EGTA in GVB= containing
l mM Mg++ for 15 minutes to selectively block the classical pathway by the removal of Ca+÷ lons (Fine et a l . , 1972). Sera so treated were then tested for hemolytic a c t i v i t y with the SRBC and RbRBC systems.
MQ++ dependence Equine sera were tested in the standard hemolytic assay with SRBC and RbRBC using a constant Ca++ ion concentration (0.15 mM), while varying the +÷
Mg
concentrations from 0.5 mM to 5.5 mM.
Preparation of equine factor D and RD Equine factor D and reagent D (RD) were prepared according to a modification of the methods of Blanchard and Leld (1984).
B r i e f l y , this
involved the gel f i l t r a t i o n of normal equine sera (NES) on Sephadex G-75 (2.5 x lO0 cm column) equilibrated in VBS+÷.
To remove completely any D
a c t i v i t y from thls peak, the ma~or protein peak was recycled a second time, and, i f necessary, a third time on the same gel f i l t r a t i o n column. The resulting reagent (RD) could not lyse RbRBC unless D was added. The equine
75 D reagent alone was unable to lyse RbRBC. I f equine factor O is p u r i f i e d according to Blanchard and Leld (1984), a slngle protein staining band with an apparent molecular weight of 22,500 can be observed on gradient SDS-slab gels.
Since this procedure involves two other i s o l a t i o n steps, we have used
repeated gel f i l t r a t i o n pure O protein.
as a method for rapid preparation of a f u n c t i o n a l l y
Functlonal antibody proteins were not present In either
factor D preparation.
In a d d i t i o n , functional r e c o n s t l t u t l o n of an equine
RD was not observed I f both equine factor O preparations were used; however, the hlghly p u r i f i e d and f u n c t i o n a l l y p u r i f i e d equine factor O could reconstitute either a bovine or human RD as wet1, I f not better, than the homologous D (81anchard and Leld, 1984).
Normal human D and RD were
prepared as described by Lachmann and Hobart (1978). RESULTS Lysis of unsensltlzed c e l l s Lysis of unsensltlzed SRBC did not occur when these c ells were incubated at 37°C for up to 90 minutes with undiluted or d i l u t e d equine sera. Effect of serum d i l u t i o n s on hemolysis D l l u t l o n of the equine sera resulted In a decrease In hemolysls of the sensitized SRBC. In four separate experlments, the average ly s ls of 90% obtained wlth undiluted equlne serum was reduced to 21.5%, 5.7% and 1.5% by d l t u t l o n of the sera by 1/5, 1/10 and 1/20, respectively (Fig. l a ) .
In
l l g h t of these results, a l l subsequent experiments were carried out using undiluted equine sera. Comparison of rabbit and equine hemolysin The amount of antibody used to sensltlze a standard concentration of SRBC while malntatnlng a constant input of equine complement markedly affected the degree of hemolysls obtained.
Rabblt hemolysin, used at a concentration
of 0.01 ml per 1 x 109 c e l l s , gave optlmal results, y l e l d l n g an average of 90% lysls of sensitized SRBC. This concentration Is lO0-fold higher than that used for guinea ptg and human complement studies.
D t l u t l o n of the
hemolysln by 1/2, 1/3 or 1/5 reduced hemolysis to 73%, 66% and 38%, respectively (averages of four separate experiments, Figs. l a , b , c , d ) ; an equine IgM antl-SRBC antlbody preparation produced s imila r results
76 (Fig. 2).
Since the homologousantibody did not significantly improve
hemolysis, the more easily obtained and standardized rabbit hemolysin was used for a l l subsequent reactions. Kinetics of hemolysis using SRBCand rabbit hemolysin We used undiluted equine sera and rabbit hemolysin to determine the kinetics of hemolyslsby equine complement. The shape of the kinetic curve is similar to that produced wlth human complement, reaching a plateau of 70% of total possible lysls within lO minutes (Fig. 3).
The average total lysls
for 25 individual horses tested under these conditions was 71%. Kinetics of hemolysis using unsensitized rabbit RbRBC Greater total lysls was obtained with this target system than with the sensitized SRBC system. minutes (Fig. 3).
A plateau of up to 98% lysls occurred within lO
The shape of the kinetic curve was similar to that
observed in the SRBC system. 50oC and 56oC heat inactivation of equine sera
Heat inactivation of equine sera was monitored using RbRBC and SRBC. Heat inactivation proceeded rapidly and was essentlally complete after 15 minutes at 50oC (Fig. 4).
In addition, a heat-dependent decay in
hemolytic a c t i v i t y was noted when the sensitized SRBC target cell system was used (Fig. 4).
Treatment of sera for 15 minutes at 56°C t o t a l l y removed
hemolytic a c t i v i t y for SRBCor RbRBC in a l l equine sera tested.
Increased
heat exposure, longer than 15 minutes, often caused protein precipitation and, in some cases, coagulation of the equine sera. Chelation of cations with EDTAand EGTA Equine sera treated with 40 mM EDTA prior to the addition of the target cells did not produce lysls when either sensitized SRBCor unsensltlzed RbRBC were used. Table I Is representative of a single experiment which has been repeated on two other occasions with q u a l i t a t i v e l y , although not quantitatively, similar results. Use of neonatal foal sera resulted in lysis of RbRBC, but to a lesser degree (Fig. 5).
Mg++-EGTAtreatment of either pre-colostral or post-
colostral samples, however, markedly reduced this hemolytic a c t i v i t y .
77 ~90.6
A
B 731 Hemolysin 1:2
Fig. la-d. Effects of d i l u t i o n of equine sera or hemolysin on lysis of sheep red blood cells (SRBC). Lysls Is plotted as a percentage of total lysls obtained by d i s t i l l e d water treatment of the SRBC. F.S. refers to f u l l strength or undiluted reagents
6O
(/J
4O
-J
215
18.1
0
C r-
Hemolysin 1:5
66.6
0
~.
D
Hemolysin 1:3
8o
6o 38.2
40 2(?
fs
1/5 1/10 1 / 2 0
fs.
1/5 1/10 1120
Serum Dilution
100
m
.m
O9 >~ .-I
80
o iii~
Flg. 2. Comparisonbetween lysls of SRBC sensitized wlth rabblt hemolysin or wlth purified equine IgM antl-SRBC antibodies and then developed using whole equine serum as a complement source. Lysls Is plotted as a percentage of total lysls obtained after treatment of SRBCwlth dlst111ed water.
m iZ
m
60
0 t--
40
~
20
o9 :CO t >oO -J 0 ~
-r
o ~i~i~i "'
:jijJii
F.S.
1/5
1/10
Serum Dilution Or)
O) >-J
100
.f>-"
~ m ,
m
80
--e
--I
I-0 I-
60
IZ I,g 0 nUJ O.
40
20
0
J
• •
10
INCUBATION
15
TIME
3'0
$RBC RbRBC
4'5
at
Flg. 3. Kinetics of the lysls of sensitized SRBC (e - e) or unsensltlzed RbRBC (A - ~) by whole equine sera. Lysls Is plotted as a percentage of total cell lysls obtained after treatment of SRBCor RbRBCby d l s t l l l e d water.
37°C
60
.
SREC
.
RbRBC
.-. 0
HEAT
l
5
10
TREATMENT
16
30
45
(mid
60
at 5O’C
Fig. 4. K\netlcs of heat Inactivation at monitored In a RbRBC (A - A) or SRBC (o LYS'IS is plotted as a percentage of total distilled water treatment of the RbRBC or
50°C of equ\ne complement, 0) target cell system. cell lysls obtalned by SRBC.
60 b----o
A-1553.
PresucklerJoMg”-
C_
A-1853,
Presuckle
EGTA
E MO’*-EGTA
W
A-1853,
Post-wckle
wlo
-
A-1853,
Post-suckle
0 Mg’+ - EGTA
Mg**-EGTA
O---G
A- 1354,
Presuckle
wlo
C--.
A-1854,
Presuckle
C Mg”
C---o
A-1354,
Post-suckle
w/o MO**-EgTA
,---.
A-1354,
Post-suckle
ti MO’*-EGTA
MG”-
EGTA
- EGTA
Dilutions of Equine Sera Fig. 5 Hemolysls of RbRBC by sera from two normal equlne foals obtalned both prior to and after colostral transfer of lmmunoglobulln and treated or not treated wlth Hg++-EGTA. Lysls Is plotted as a percentage of total lysls obtalned by dlstllled water treatment of RbRBC.
79 Because Immunoglobulin levels were 26% of normal for IgM and < 2% for IgG in the pre-colostral samples, i t is unlikely that any effective level of natural antibodies to the RbRBC would be present.
In fact, we could not
detect hemagglutlnatlng antibodies to RbRBC In these samples. The mean total IgM and IgG levels in the pre-colostral samples were 128 ~g/ml and < lO0 ug/ml, respectively.
I f natural antibodies were important in the
lysls of the RbRBC, then lysls should have been increased in the postcolostral samples by the transfer of maternal antibodies; this dld not occur (Fig. 5). ÷÷
Mg
dependence
Preliminary tests indicated that when the Mg÷÷ concentration was increased from D.5 mM to l.O mM, the lysls by equine sera using sensitized SRBC or unsensltlzed RbRBC was s i g n i f i c a n t l y increased.
Horses known to have
intermediate levels of hemolytic a c t i v i t y , assessed over a 6-month period, were used as serum sources for thls experiment.
These sera routinely gave
normal total lysls values of between 38 to 45% for RbRBC. These percentages ÷÷
were increased, however, from 51 to 71% when Mg increased.
ion concentrations were
Optimal enhancement was observed at l mM concentrations.
TABLE I Effects of Mg÷+, EDTA and EGTA on lysls of sensitized SRBC and unsensltlzed RbRBC by normal equine sera
Serum Treatment None 40 mM EDTA lO mM EGTA + 1 mM Mg÷+
SRBC1 % lysls
RbRBC % lysls
69.8 0 0
86.2 0 98.1
1All values are expressed as a percentage of the t o t a l hemolysis In d l s t l l l e d water.
80 Equine and human RD and D reconstltutlon We used equine RD and factor D reagents to examinewhether the hemolytic a c t i v i t y for RbRBCobserved in normal equine sera dependedupon natural antibody (Table I I ) .
The combination of the equine RD and D resulted In an
approximately I0% recovery of the ~ n l t l a l hemolytic a c t i v i t y for RbRBC. Suspecting that our separation procedures were causing the problem, we prepared the corresponding human reagents.
Gel f i l t r a t i o n of normal human
sera on SephadexG-75 resulted in a reagent, RD, that had a l l the components of the classical complement system but lacked D of the alternative pathway. This RD could not lyse RbRBCbecause I t lacked one of the alternative pathway components (factor D), but the RD could lyse a sensitized SRBC, as the RD possessed a l l the classical pathway components (Table I I ) .
The
isolated equine D could f u l l y reconstitute a human RD, but the human D could not reconstitute an equlne RD. T h l s lack of reconstltutlon, therefore, appears to originate In the unique characteristics of the equine RD. These experiments have been repeated on two other occasions with q u a l i t a t i v e l y , although not quantitatively, similar results.
A hlghly-purlfled equine D
preparation gave a single protein staining band of 21,500 apparent molecular weight on 7-17% SDS-PAGE gradient gels.
I t was also unable to reconstitute
an equine RD but readily reconstituted both a human and bovine RD (Blanchard and Leld, 19B4). TABLE I I Lysls of RbRBCand sensitized SRBC by human and equine D and RD RbRBCl % lysls Buffer Normal Normal RDEq DEq RDEq + RDHu DHu RDHu + RDHu + RDEq +
Human Serum (NHS2) Equine Serum (NES) DEq DHu DEq DHu
0 lO0 lO0 2.4 0.9 9.7 1.6 0.2 79.0 73.5 3.5
SRBC % lysls 0 lO0 lO0 0.5 0.2 0.7 lO0 O.1 IO0 lO0 0.6
1 A l l values are expressed as a percentage of the t o t a l hemolysis In distilled water. 2Normal human and equine sera were used u n d i l u t e d and 50-pl volumes of RD, O, NHS and NES were added to the e r y t h r o c y t e s .
81 DISCUSSION
The s e n s i t i v i t y of the equine sera to heat treatment at 50oC, as monitored by RbRBC indicator c e l l s , is consistent with i n a c t i v a t i o n of factor B described for human, bovine and guinea pig complement (Gotze and Muller-Eberhard, 1971; Pang and Aston, 1978; Tabel, 1981; Brade, et a l . , 1972).
The results we obtained at 5O°C also suggest an unusual heat
s e n s i t i v i t y of certain key equine classical pathway components, using the SRBC i n d i c a t o r system.
This s e n s i t i v i t y is not a feature of the human or
guinea plg system (Mayer, 1961). We observed marked l y t i c a c t i v i t y of the equine sera, using an unsensltized RbRBC. Our results, comblned with other reports of successful lys i s in a rabbit erythrocyte-sheep antibody system, leads one to question i f the horse has natural antibodies to RbRBC, or i f the equine complement system operates p r i m a r i l y via the a l t e r n a t i v e pathway using this indicator system.
Unsensitlzed RbRBC have been shown to" be an a l t e r n a t i v e pathway
a c t i v a t o r in human and bovine sera (Blanchard and Leld, 1984; Pang and Aston, 1978; Brock et a l . , 1975; P l a t t s - M i l l s and Ishtzaka, 1974; Fearon and Austen, 1977).
The existence of natural equine c i r c u l a t i n g antibodies to
RbRBC has been suggested previously by Rice and Boulanger (1952) and Barta et a l . (1973).
At the time that Rice and Boulanger (1952) conducted t h e i r
experiments, the existence of the a l t e r n a t i v e pathway was not yet demonstrated; therefore, i t was not considered when they interpreted t h e i r data.
Recently, antlbody-mediated enhancement of the a l t e r n a t i v e pathway
has been shown to occur (Perrln et a l . , 1976; P o l h i l l et a l . , 1978).
I f , as
some have suggested, c i r c u l a t i n g antibodies to RbRBC e x is t n a t u r a l l y in the horse (Rice and Boulanger, 1952; Barta et a l . , 1973), then lysis of RbRBC may result from the augmentation of the a l t e r n a t i v e pathway by antibody. Neither Rice and Boulanger (1952) nor Barta et a l . (1973) used a Ca++ specific chelator, such as EGTA, to block classical pathway a c t i v a t i o n yet permit selective expression of the equine a l t e r n a t i v e pathway.
We have not
observed any natural a g g l u t i n a t i n g antibodies to unsensltlzed RbRBC by m i c r o t i t r e hemagglutlnation techniques.
Since the hemolytic a c t i v i t y for
RbRBC is rather l a b i l e , absorption experiments with packed RbRBC in the presence of EDTA cannot be carried out, because
the untreated control sera
decays In a tlme-dependent manner (Leid, unpublished observations). The hemolytic results we obtained using foal sera strongly suggest that natural antibody to the RbRBC plays a small role In hemolysis.
We made a
surprising observation when using the Mg+÷-EGTA treated p r e - c o l o s t r a l foal sera collected p r i o r to the transfer of maternal lmmunoglobulins and other
82 serum proteins.
There was a marked diminution in hemolytic a c t i v i t y when
the Mg+÷-EGTA treated pre-colostral sera were used, when compared to the untreated sera.
Since antibodies to RbRBC have not been detected in these
sera, alternative pathway activation in the horse may require Ca++ at some c r i t i c a l activation step.
T h i s Ca++ requirement is unusual since no other
species has demonstrated a similar cation requirement for expression of the alternative pathway. T h i s hypothesized Ca++ dependence of alternative pathway expression may have led to our varied results with Mg++-EGTA treated adult sera and may have caused the inconsistencies in equine hemolytic a c t i v i t y previously noted by others.
The results we have
descrlbed--heat s u s c e p t i b i l i t y at 50oC, l y t I c a c t i v i t y in the presence of EGTA, and Mg÷÷ dependence--Indlcate that the alternative pathway is activated when the unsensltlzed RbRBC and equine sera are used.
As reported
by Barta et al. (1973), no significant difference was observed In hemolysis of the RbRBC whether i t was sensitized with antibody or not. Our hypothesis that the equine hemolytic a c t i v i t y for RbRBC was primarily due to alternative pathway activation was further supported by our experiments with the equine and human D and RD reagents.
The hemolytic
a c t i v i t y for RbRBC was apparently due to the equine alternative pathway, since lysls did not occur with an equine RD alone.
Although only a I0%
recovery of hemolytic a c t i v i t y was observed upon combination of the equine RD and D, equine D was very active when reconstituting a human RD.
The lack
of lysls of a sensitized SRBC by the equine RD Is puzzling; in contrast, a human RD prepared in a homologous manner w i l l lyse a sensitized SRBC (Table II).
The results with the human RD and the SRBC are to be expected, since
the human RD possesses the classical pathway activation and terminal l y t l c components. The equine RD should have the same characteristics, but clearly does not.
Equine D, however, is active with the human RD and reconstitutes
a human RD.
HumanD, although active with a human RD, does not reconstitute
an equine RD.
These reconstltutlon experiments suggest that either an
equine complement component is extremely l a b i l e and decays during the gel f i l t r a t i o n chromatographic step, or that an additional component, not previously recognized, must be present for reconstltutlon to occur.
We
believe that the former is more l i k e l y , although we are not ruling out the latter possibility.
Our work with the bovine complement system, manipulated
in the same manner, shows that a stable bovine RD and D can be easily prepared in which the equine D is able to substitute f u l l y for i t s bovine counterpart (Blanchard and Leld, 1984).
We suggest that most, i f not a l l ,
previous work using equine complement for lysls of RbRBC was due to alternative pathway activation and was not the result of any antibody-
88 dependent pathway. The r e l a t i v e l y poor lysls of antibody sensitized SRBC suggests either that the horse has a poorly developed classical pathway, or possesses elevated levels of plasma control proteins.
In addition, I t Is
possible that the target cells (SRBC) and buffer systems used were not optimal for demonstration of the equine classical pathway. All of these avenues should be investigated in the future.
Wlth 25 horses giving rise to
an average of 71% hemolytlc a c t i v i t y for the SRBC, I t is clear that almost a l l horses w111 have some l y r i c a c t i v i t y for thls target c e l l .
A note of
caution is appropriate for those who undertake the future characterization of the equine alternative complement system: not a l l horses studied under the conditions we have described w l l l glve marked hemolytic a c t i v i t y for the unsensltlzed RbRBC In Mg++-EGTA. The reasons for such non-reactlvlty may reflect a c r i t i c a l Ca++ dependencefor equine alternative pathway activation, as seen In our results with pre-colostral foal sera.
Thls
unusual cation dependencemay account for the controversy In the literature.
A more careful evaluation of equine antibodies involved In the
hemolysis is called for, as are additional tests on the lyric characteristics of equine complement. The results presented here, however, serve as preliminary evidence for the existence of a domestic animal species In which alternative pathway activation is a major feature of its complement system. ACKNOWLEDGMENTS
This i n v e s t i g a t i o n received support from the f l l a r l a s l s
component of the
UNOP/World Bank/WHO Specla] Program for Research and Training in Tropical Diseases (FIl/T16/181/OS/09) to RWL and the Stanley Adler Research Fund from Washington State U n i v e r s i t y .
Portions of this work were done at both
Michigan State U n l v e r s l t y and Washlngton State U n i v e r s i t y . REFERENCES
Barta, O. and Hubbert, N.L., ]978. Testing of hemolytic complement components in domestic animals. Am. 3. Vet. Res., 39 : 1303-1310. Barta, 0., Barta, V. and Williams, E . I . , 1973. A method for t l t r a t l n g equine hemolytic complement. Z. Immunltaetsforsch., 146 : 114-]22. Blanchard, D.P. and Letd, R.W., 1984. I s o l a t i o n and p a r t l a ] characterization of bovine and equine factor D. Mol. Immunol. (In Press). Bordet, 3. and Streng, O., ]909. Les phenomenes d'adsorptlon et le conglutlne du serum du boeuf. Zb]. Bakt. Hyg. I . Abt. Org., 49 : 260-276. Brade, V., Cook, C.T., Shin, H.S. and Mayer, M.M., 1972. Studies on the properdin system: I s o l a t l o n of a h e a t - l a b l ] e factor from guinea pig serum related to a human g l y c i n e - r i c h beta glycoproteln (GBG or Factor B). 3. Immunol., 109 : 1174-1181.
84 Brock, J.H., Ortega, F. and P1nerio, A., 1975. Bacterlocldal and hemolytic a c t i v i t y in bovine colostrum and serum. Effect of p r o t e o l y t l c enzymes and ethylene glycol tetraacetlc acid (EGTA). Ann. I n s t . Pasteur ( L i l l e ) , 126 : 430-451. Coley, S.C. and Leld, R.W., 1982. Effects of extracts of Onchocerca c e r v l c a l i s from horses on the l y t l c a c t i v i t y of human, rat and equine complement. Clln. Immunol. Immunopathol., 23 : 113-123. Dyer, R.M. and Leld, R.W., 1983. Surface receptors for IgG and complement on the equlne alveolar macrophage. Inflammation, 7 : 183-195. Fearon, D.T. and Austen, K.F., 1980. Current concepts In Immunology. The a l t e r n a t i v e pathway of complement. A system for host resistance to microbial i n f e c t i o n . N, Eng. J. Red., 303 : 259-263. Fearon, D.T. and Austen, K.F., 1977. A c t i v a t i o n of a l t e r n a t i v e complement pathway with rabbit erythrocytes by circumvention of the regulatory action of endogenous control proteins. J. Exp. Med., 146 : 22-33. Fine, P.O., Marney J r . , S.R., Colley, D.G., Sergent, 3.S. and DesPrez, R.M., 1972. C3 shunt a c t i v a t i o n In human serum chelated wlth EGTA. J. Immunol., 109 : 807-809. Glaze, M.B., McGulre, T.C., Schmldt, G.M. and Leld, R. Wes, 1984. Immunoglobulln levels In tears and aqueous humor of horses before and a f t e r diethylcarbamazine (DEC) therapy. Vet. Immunol. Immunopathol., 7: 185-198. Gotze, O. and Muller-Eberhard, H.J., 1971. The C3-actlvator system: An a l t e r n a t i v e pathway of complement a c t i v a t i o n . J. Exp. Red., 134 : 90-I08s. Hole, N.H. and Coombs, R.R.A., 1947. The conglutlnating phenomenon. I I . The technique of the c o n g l u t l n a t i n g complement adsorption test compared wlth the hemolytic complement f i x a t i o n test. J. Hyg. (Camb.), 45 : 490-496. Leon, M.A. and Norden, A. 1959. Kinetics of equine complement. ~. Immunol., B3 : 99-106. Lachmann, P.J. and Hobart, M.J., 1978. Complement technology. In: O.M. Weir ( E d i t o r ) , Handbook of Experimental Immunology, 3rd ed. Blackwe11 S c i e n t i f i c Publications, London, pp. 5A.1-5A.23. Mayer, M.M., 1961. Complement and complement f i x a t i o n . In: E.A. Kabat and M.M. Mayer (Editors), Experimental Immunochemlstry, 2nd ed. Charles C. Thomas, S p r l n g f l e l d , pp. 133-240. McGuire, T.C., Banks, K.L. and Popple, M.J., 1975. Combined Immunodeflclency in horses: Characterization of the lymphocyte defect. Clin. Immunol. Immunopathol., 3 : 555-566. McGulre, T.C., Banks, K.L., Evans, D.R. and Popple, M.3., 1976. Agammaglobullnemla in a horse with evidence of functional T lymphocytes. Am. J. Vet. Res., 3? : 41-46. Muir, R., 1911. On the relationship between the complement and immune bodies of different animals. 3. Path. Bacterlol., 16 : 523-534. Pang, A.S.D. and Aston, W.P., 1978. The alternative complement pathway in bovine serum: The isolation of a serum protein with factor B a c t i v i t y . Immunochem., 15 : 529-534. Perrln, L.H., Joseph, B.S., Cooper, N.R. and Oldstone, M.B.A., 1976. Mechanism of injury of vlrus-lnfected cells by a n t l v l r a l antlbody and complement: Participation of IgG, F(ab') 2 and the alternative pathway. 3. Exp. Med., 143 : 1027-I041. Perryman, L.E., McGulre, T.C., Banks, K.L. and Henson, J.B., 1971. Decreased C3 levels In a chronic virus infection: equine infectious anemia. J. Immunol., I06 : 1074-I078. P l a t t s - M l l l s , T.A.E. and Ishlzaka, K., 1974. Activation of the alternative pathway of human complement by rabbit cells. 3. Immunol., l l 3 : 348-358.
85 P o l h l l l , R.B., Newman, S.L., P r u l t t , K.M. and 3ohnson Jr., R.B., 1978. Kinetic assessment of alternative complement pathway in a hemolytic system. I I . Influence of antibody in alternative pathway activation. 3. Immunol., 121 : 371-376. Rice, C.E., 1950a. The Interchangeabillty of the complement components of different animal species. I. Literature survey. Can. 3. Comp. Med., 14 : 369-379. Rice, C.E., 1950b. The Interchangeabillty of the complement components of different species. I l l . In conglutlnatlon. 3. Immunol., 65 : 499-510. Rice, C.E. and Boulanger, P., 1952. The Interchangeablllty of the complement components of different animal species. IV. In the hemolysis of rabbit erythrocytes sensitized with sheep antibody. J. Immunol., 68 : 197-205. Rice, C.E. and Crowson, N., 1950. The Interchangeablllty of the complement components of different animal species. I I . In hemolysis of sheep erythrocytes sensitized wlth rabbit amboceptor. J. Immunol., 65 : 201-210. Sandberg, A.L. and Osler, A.G., 1971. Dual pathways of complement interaction with guinea pig Immunoglobullns. 3. Immunol., I07 : 1268-1273. Tabel, H., 1981. Alternative pathway of bovine complement. Immunochemlcal studies on factor B-llke serum protein and its conversion product, By2. Can. 3. Comp. Med., 45 : 291-298.
71
EQUINE ALTERNATIVE PATHWAYACTIVATION BY UNSENSITIZED RABBIT RED BLOODCELLS R. WES LEID,I, 2 S. CLAIRE COLEY,3 D.P. BLANCHARDl and L.E. PERRYMANl IDepartment of Veterinary Microblology/Pathology, Washington State University, Pullman, Washington, 99164 (U.S.A.) 2person to whoma l l correspondence is addressed. 3Department of Pathology, Michigan State University, East Lansing, Michigan, 48824 (U.S.A.)
(Accepted 4 September 1984) ABSTRACT Leld, R. Wes, Coley, S. Claire, B1anchard, D.P. and Perryman, L.E., 1985. Equine alternative pathway activation by unsensltlzed rabbit red blood cells. Vet. Immunol. Immunopathol., 9: 71-85. The equine alternative complement pathway has been p a r t i a l l y characterized and compared to the equine classical activation pathway. A dose-dependent lysls of RbRBCwas observed with peak l y t l c values noted within lO minutes at 37°C when rabbit red blood cells (RbRBC) were used as an alternative pathway activator. Sheepred blood cells (SRBC) sensitized with rabbit hemolysin or p a r t i a l l y purified equine IgM antibodies were equally sensitive to lysls. Dilution of the commercial hemolysin by I/5 reduced lysls from 90% to 38% in the presence of constant cell numbers. Hemolysis of SRBCpeaked at lO minutes and the majority of lysls occurred within lO minutes. Dilution of equine sera by as l i t t l e as I/5 decreased hemolytic a c t i v i t y for SRBC to 21.5% from greater than 90% with undiluted sera. The alternative pathway protein, equine factor B, was tested using RbRBC and monitored by Its d i f f e r e n t i a l susceptibility to heat treatment at 50oC. Thls treatment led to almost complete inactivation after a 15-mlnute incubation. An apparent heat-dependent decay of certain classlcal pathway components was also observed after 50oC treatment. This sensitivity was indicated by a reduction in the l y t l c a c t i v i t y for sensitized SRBC. Treatment for 15 minutes at 56oC wlth either RbRBCor SRBC was sufficient to abolish hemolytic a c t i v i t y in a l l equine sera tested. Chelation of cations with 0.04 M EDTA blocked expression of alternative and classical pathway act~vatlon; however, chelation of Ca++ Ions with lO mM EGTA containing l mM Mg++ Ions permitted lysls of the RbRBC but not the SRBC. A dose-related Mg+÷-lon dependence for RbRBC hemolytic a c t i v i t y was observed as the concentration of Mg÷+ was increased to l.O mM. In addition, our results obtained with pre-colostral foal serum strongly suggest that natural antibody to RbRBCwas of l i t t l e importance in the lysls observed with these cells. These results also show that the equine alternative pathway activation may require Ca++ ions. I f Ca÷÷ ions are required, the equine alternative pathway is quite different from any other mammaliancomplement system so far described. Our results suggest that the alternative pathway of activation is of ma~or importance In the equine complement system. Confirmation of this hypothesis requires both purification of the components involved as well as further characterization.
0165-2427/85/$03.30
© 1985 Elsevier Science Publishers B.V.
72 INTRODUCTION Equine complement Is considered nonhemolytlc in the standard sheep erythrocyte-rabblt antibody target system (Muir, 1911, Rice, 1950a,b; Leon and Norden, 1959; Rlce and Crowson, 1950), and is regarded as an excellent conglutlnln reagent (Rice, 1950b; Bordet and Streng, 1909; Hole and Coombs, 1947). The decreased hemolytic a c t i v i t y in equine serum for antlbody-sensltlzed SRBC has been attributed to nonexistent or very low levels of one or more complement components. Barta and Hubbert (IgTB) reported undetectable or low t l t e r s of C4, C2, C3 and C5 in equine sera and postulated the presence of a Cl Inactlvator. Leon and Norden (1959) found equine complement could readily form an EAC142 c e l l , which could be lysed only by the addition of EDTA-treated guinea plg serum, which supplied C3 and the terminal l y t l c components. Antlbody-coated erythrocytes were successfully lysed by equine complement using ox erythrocytes sensitized with cat hemolysin (Muir, 1911) or rabbit erythrocytes sensitized wlth sheep antibodies (Barta, et a l . , 1973; Rlce and Boulanger, 1952; Perryman et a l . , 1971; McGulre, et a l . , 1975, 1976).
I t Is
not clear whether hemolysis with RbRBC Is an antlbody-medlated event or is a result of cell-surface activation as in the case of human and bovine complement (Blanchard and Leld, 19B4; Eearon and Austen, 19BO). Preliminary work reported by Coley and Leld (19B2) suggests that the equine alternative pathway is also activated by the RbRBC. Our assessment of hemolysis of the unsensltlzed rabbit red blood cell presents evidence that equine complement possesses a potent alternative pathway.
We also demonstrate that lysls can be achieved using the standard
SRBC-rabblt antibody target system. MATERIALS AND METHODS
Equine sera, sheep (SRBC) and rabbit (RbRBC) erythrocytes Sera c o l l e c t e d from normal, heal.thy, a d u l t horses were prepared as described by Coley and Leld (1982) and used u n d i l u t e d or d i l u t e d w i t h veronal b u f f e r e d s a l i n e (VBS) c o n t a i n i n g g e l a t i n (0.1%, w / v ) , 0.15 n~1 Ca++ , and 0.5 mM Mg÷+ (GVB++).
Sera were used Immediately a f t e r
thawing; samples were never r e f r o z e n and used. u n s e n s l t l z e d RbRBC were measured by m l c r o t l t r e
Hemagglutlnln a n t i b o d i e s f o r h e m a g g l u t l n a t l o n techniques
but were not d e t e c t a b l e in any of the equine sera used. c o l l e c t e d and t r e a t e d as described by Mayer (1961).
SRBC and RbRBC were
?3 Equine sera were also collected from foals before and within 24 hours after the f i r s t ingestion of mare's colostrum.
These sera were processed
for preservation of complement a c t i v i t y as described for adult horses. Immunoglobulln G and M levels were quantltated by single radial Immunodlffuslon as described by Glaze et al. (1984). Antibodies SRBC were sensitized by the methods of Mayer (1961) using e i t h e r commercially produced rabbit hemolysin (Cordis Laboratories, Miami, Florida) or an equine anti-SRBC IgM preparation.
The rabbit hemolysin was used at a
concentration lO0-fold that recommended by the manufacturers for human or guinea pig complement work; this concentration was necessary to produce optlmal hemolytic conditions. concentration of antibody.
Agglutination was not observed at thls
In a d d i t i o n , the c ells were washed 3x in GVB=
pri o r to use in the hemolytic assays.
We encountered no d i f f i c u l t i e s
washing that might have been caused by heavy antibody s e n s i t i z a t i o n .
in The
equine IgM antibodies were e l i c i t e d and p u r i f i e d by the methods of Dyer and Leid (1983).
B r i e f l y , a horse was injected Intravenously with 30 ml of a 1%
solution of SRBC. Serum was collected from the horse 12 days l a t e r and antibodies obtained a f t e r sequential f r a c t i o n a t i o n on DEAE c e l l u l o s e and Sephacryl S-300 columns.
The chromatographic fractions containlng IgM
antibodies were i d e n t i f i e d by hemagglutinatlon of non-sensitlzed SRBC and by immunoelectrophoresls of the fractions against a polyvalent anti-equine whole serum. Those p o s i t i v e fractions were pooled, concentrated, a11quoted and stored at -70°C.
RbRBCwere used unsensitized.
Assay for hemolytic complement Hemolytlc complement t l t e r s were measured using a modiflcatlon of the methods of Mayer (1961) as described by Coley and Letd (1982).
Briefly,
sensitized SRBC or unsensltlzed RbRBC were suspended In GVB++ at a concentration of 1 x 10]0 c e l l s per ml.
A 100-~1 volume of undiluted
equine sera was added to 100 ~1 of sensitized SRBC or unsensltlzed RbRBC. The tubes were gently mixed and then incubated at 37oC for 60 minutes. A l l reactions were stopped by the addition of 1.8 ml of cold VBS to each tube, followed by c e n t r l f u g a t l o n (300 x g, 10 minutes).
The degree of
hemolysls was determined by reading the absorbance at 414 nm In a G l l f o r d spectrophotometer (Stasar I I , G l l f o r d Inst. Co., Oberlin, Ohio).
Cells in
control tubes were lysed with d l s t l l l e d water to establlsh the 100% ly s ls
74 level while other control tubes included cells with buffer alone to establish background l y t l c levels.
An additional control was required since
undiluted equine sera have considerably more color than sera normally used in complement assays. T h i s control was established by recording the OD414 of the sera from each horse (O.l ml sera ÷ 1.9 ml of VBS) on each experimental date. 50QC and 56oC heat inactivation of equine sera Equine sera were heat inactivated as described by Coley and Leld (1982). After heat treatment, sera were cooled immediately in an Ice-water bath, rewarmed to 37oC, and tested for hemolytic a c t i v i t y with the SRBC and RbRBC systems. Chelation of cations with EDTA and EGTA Chelation of cations in equine sera was done as previously described by Coley and Leld (1982).
B r i e f l y , equine sera were incubated with 40 mM EDTA
in GVB= (GVB without Ca÷+ and Mg÷÷ ions) for 15 minutes prior to the introduction of either the SRBC or RbRBC; the level of hemolysis was then determined.
Sera were also incubated with lO mM EGTA in GVB= containing
l mM Mg++ for 15 minutes to selectively block the classical pathway by the removal of Ca+÷ lons (Fine et a l . , 1972). Sera so treated were then tested for hemolytic a c t i v i t y with the SRBC and RbRBC systems.
MQ++ dependence Equine sera were tested in the standard hemolytic assay with SRBC and RbRBC using a constant Ca++ ion concentration (0.15 mM), while varying the +÷
Mg
concentrations from 0.5 mM to 5.5 mM.
Preparation of equine factor D and RD Equine factor D and reagent D (RD) were prepared according to a modification of the methods of Blanchard and Leld (1984).
B r i e f l y , this
involved the gel f i l t r a t i o n of normal equine sera (NES) on Sephadex G-75 (2.5 x lO0 cm column) equilibrated in VBS+÷.
To remove completely any D
a c t i v i t y from thls peak, the ma~or protein peak was recycled a second time, and, i f necessary, a third time on the same gel f i l t r a t i o n column. The resulting reagent (RD) could not lyse RbRBC unless D was added. The equine
75 D reagent alone was unable to lyse RbRBC. I f equine factor O is p u r i f i e d according to Blanchard and Leld (1984), a slngle protein staining band with an apparent molecular weight of 22,500 can be observed on gradient SDS-slab gels.
Since this procedure involves two other i s o l a t i o n steps, we have used
repeated gel f i l t r a t i o n pure O protein.
as a method for rapid preparation of a f u n c t i o n a l l y
Functlonal antibody proteins were not present In either
factor D preparation.
In a d d i t i o n , functional r e c o n s t l t u t l o n of an equine
RD was not observed I f both equine factor O preparations were used; however, the hlghly p u r i f i e d and f u n c t i o n a l l y p u r i f i e d equine factor O could reconstitute either a bovine or human RD as wet1, I f not better, than the homologous D (81anchard and Leld, 1984).
Normal human D and RD were
prepared as described by Lachmann and Hobart (1978). RESULTS Lysis of unsensltlzed c e l l s Lysis of unsensltlzed SRBC did not occur when these c ells were incubated at 37°C for up to 90 minutes with undiluted or d i l u t e d equine sera. Effect of serum d i l u t i o n s on hemolysis D l l u t l o n of the equine sera resulted In a decrease In hemolysls of the sensitized SRBC. In four separate experlments, the average ly s ls of 90% obtained wlth undiluted equlne serum was reduced to 21.5%, 5.7% and 1.5% by d l t u t l o n of the sera by 1/5, 1/10 and 1/20, respectively (Fig. l a ) .
In
l l g h t of these results, a l l subsequent experiments were carried out using undiluted equine sera. Comparison of rabbit and equine hemolysin The amount of antibody used to sensltlze a standard concentration of SRBC while malntatnlng a constant input of equine complement markedly affected the degree of hemolysls obtained.
Rabblt hemolysin, used at a concentration
of 0.01 ml per 1 x 109 c e l l s , gave optlmal results, y l e l d l n g an average of 90% lysls of sensitized SRBC. This concentration Is lO0-fold higher than that used for guinea ptg and human complement studies.
D t l u t l o n of the
hemolysln by 1/2, 1/3 or 1/5 reduced hemolysis to 73%, 66% and 38%, respectively (averages of four separate experiments, Figs. l a , b , c , d ) ; an equine IgM antl-SRBC antlbody preparation produced s imila r results
76 (Fig. 2).
Since the homologousantibody did not significantly improve
hemolysis, the more easily obtained and standardized rabbit hemolysin was used for a l l subsequent reactions. Kinetics of hemolysis using SRBCand rabbit hemolysin We used undiluted equine sera and rabbit hemolysin to determine the kinetics of hemolyslsby equine complement. The shape of the kinetic curve is similar to that produced wlth human complement, reaching a plateau of 70% of total possible lysls within lO minutes (Fig. 3).
The average total lysls
for 25 individual horses tested under these conditions was 71%. Kinetics of hemolysis using unsensitized rabbit RbRBC Greater total lysls was obtained with this target system than with the sensitized SRBC system. minutes (Fig. 3).
A plateau of up to 98% lysls occurred within lO
The shape of the kinetic curve was similar to that
observed in the SRBC system. 50oC and 56oC heat inactivation of equine sera
Heat inactivation of equine sera was monitored using RbRBC and SRBC. Heat inactivation proceeded rapidly and was essentlally complete after 15 minutes at 50oC (Fig. 4).
In addition, a heat-dependent decay in
hemolytic a c t i v i t y was noted when the sensitized SRBC target cell system was used (Fig. 4).
Treatment of sera for 15 minutes at 56°C t o t a l l y removed
hemolytic a c t i v i t y for SRBCor RbRBC in a l l equine sera tested.
Increased
heat exposure, longer than 15 minutes, often caused protein precipitation and, in some cases, coagulation of the equine sera. Chelation of cations with EDTAand EGTA Equine sera treated with 40 mM EDTA prior to the addition of the target cells did not produce lysls when either sensitized SRBCor unsensltlzed RbRBC were used. Table I Is representative of a single experiment which has been repeated on two other occasions with q u a l i t a t i v e l y , although not quantitatively, similar results. Use of neonatal foal sera resulted in lysis of RbRBC, but to a lesser degree (Fig. 5).
Mg++-EGTAtreatment of either pre-colostral or post-
colostral samples, however, markedly reduced this hemolytic a c t i v i t y .
77 ~90.6
A
B 731 Hemolysin 1:2
Fig. la-d. Effects of d i l u t i o n of equine sera or hemolysin on lysis of sheep red blood cells (SRBC). Lysls Is plotted as a percentage of total lysls obtained by d i s t i l l e d water treatment of the SRBC. F.S. refers to f u l l strength or undiluted reagents
6O
(/J
4O
-J
215
18.1
0
C r-
Hemolysin 1:5
66.6
0
~.
D
Hemolysin 1:3
8o
6o 38.2
40 2(?
fs
1/5 1/10 1 / 2 0
fs.
1/5 1/10 1120
Serum Dilution
100
m
.m
O9 >~ .-I
80
o iii~
Flg. 2. Comparisonbetween lysls of SRBC sensitized wlth rabblt hemolysin or wlth purified equine IgM antl-SRBC antibodies and then developed using whole equine serum as a complement source. Lysls Is plotted as a percentage of total lysls obtained after treatment of SRBCwlth dlst111ed water.
m iZ
m
60
0 t--
40
~
20
o9 :CO t >oO -J 0 ~
-r
o ~i~i~i "'
:jijJii
F.S.
1/5
1/10
Serum Dilution Or)
O) >-J
100
.f>-"
~ m ,
m
80
--e
--I
I-0 I-
60
IZ I,g 0 nUJ O.
40
20
0
J
• •
10
INCUBATION
15
TIME
3'0
$RBC RbRBC
4'5
at
Flg. 3. Kinetics of the lysls of sensitized SRBC (e - e) or unsensltlzed RbRBC (A - ~) by whole equine sera. Lysls Is plotted as a percentage of total cell lysls obtained after treatment of SRBCor RbRBCby d l s t l l l e d water.
37°C
60
.
SREC
.
RbRBC
.-. 0
HEAT
l
5
10
TREATMENT
16
30
45
(mid
60
at 5O’C
Fig. 4. K\netlcs of heat Inactivation at monitored In a RbRBC (A - A) or SRBC (o LYS'IS is plotted as a percentage of total distilled water treatment of the RbRBC or
50°C of equ\ne complement, 0) target cell system. cell lysls obtalned by SRBC.
60 b----o
A-1553.
PresucklerJoMg”-
C_
A-1853,
Presuckle
EGTA
E MO’*-EGTA
W
A-1853,
Post-wckle
wlo
-
A-1853,
Post-suckle
0 Mg’+ - EGTA
Mg**-EGTA
O---G
A- 1354,
Presuckle
wlo
C--.
A-1854,
Presuckle
C Mg”
C---o
A-1354,
Post-suckle
w/o MO**-EgTA
,---.
A-1354,
Post-suckle
ti MO’*-EGTA
MG”-
EGTA
- EGTA
Dilutions of Equine Sera Fig. 5 Hemolysls of RbRBC by sera from two normal equlne foals obtalned both prior to and after colostral transfer of lmmunoglobulln and treated or not treated wlth Hg++-EGTA. Lysls Is plotted as a percentage of total lysls obtalned by dlstllled water treatment of RbRBC.
79 Because Immunoglobulin levels were 26% of normal for IgM and < 2% for IgG in the pre-colostral samples, i t is unlikely that any effective level of natural antibodies to the RbRBC would be present.
In fact, we could not
detect hemagglutlnatlng antibodies to RbRBC In these samples. The mean total IgM and IgG levels in the pre-colostral samples were 128 ~g/ml and < lO0 ug/ml, respectively.
I f natural antibodies were important in the
lysls of the RbRBC, then lysls should have been increased in the postcolostral samples by the transfer of maternal antibodies; this dld not occur (Fig. 5). ÷÷
Mg
dependence
Preliminary tests indicated that when the Mg÷÷ concentration was increased from D.5 mM to l.O mM, the lysls by equine sera using sensitized SRBC or unsensltlzed RbRBC was s i g n i f i c a n t l y increased.
Horses known to have
intermediate levels of hemolytic a c t i v i t y , assessed over a 6-month period, were used as serum sources for thls experiment.
These sera routinely gave
normal total lysls values of between 38 to 45% for RbRBC. These percentages ÷÷
were increased, however, from 51 to 71% when Mg increased.
ion concentrations were
Optimal enhancement was observed at l mM concentrations.
TABLE I Effects of Mg÷+, EDTA and EGTA on lysls of sensitized SRBC and unsensltlzed RbRBC by normal equine sera
Serum Treatment None 40 mM EDTA lO mM EGTA + 1 mM Mg÷+
SRBC1 % lysls
RbRBC % lysls
69.8 0 0
86.2 0 98.1
1All values are expressed as a percentage of the t o t a l hemolysis In d l s t l l l e d water.
80 Equine and human RD and D reconstltutlon We used equine RD and factor D reagents to examinewhether the hemolytic a c t i v i t y for RbRBCobserved in normal equine sera dependedupon natural antibody (Table I I ) .
The combination of the equine RD and D resulted In an
approximately I0% recovery of the ~ n l t l a l hemolytic a c t i v i t y for RbRBC. Suspecting that our separation procedures were causing the problem, we prepared the corresponding human reagents.
Gel f i l t r a t i o n of normal human
sera on SephadexG-75 resulted in a reagent, RD, that had a l l the components of the classical complement system but lacked D of the alternative pathway. This RD could not lyse RbRBCbecause I t lacked one of the alternative pathway components (factor D), but the RD could lyse a sensitized SRBC, as the RD possessed a l l the classical pathway components (Table I I ) .
The
isolated equine D could f u l l y reconstitute a human RD, but the human D could not reconstitute an equlne RD. T h l s lack of reconstltutlon, therefore, appears to originate In the unique characteristics of the equine RD. These experiments have been repeated on two other occasions with q u a l i t a t i v e l y , although not quantitatively, similar results.
A hlghly-purlfled equine D
preparation gave a single protein staining band of 21,500 apparent molecular weight on 7-17% SDS-PAGE gradient gels.
I t was also unable to reconstitute
an equine RD but readily reconstituted both a human and bovine RD (Blanchard and Leld, 19B4). TABLE I I Lysls of RbRBCand sensitized SRBC by human and equine D and RD RbRBCl % lysls Buffer Normal Normal RDEq DEq RDEq + RDHu DHu RDHu + RDHu + RDEq +
Human Serum (NHS2) Equine Serum (NES) DEq DHu DEq DHu
0 lO0 lO0 2.4 0.9 9.7 1.6 0.2 79.0 73.5 3.5
SRBC % lysls 0 lO0 lO0 0.5 0.2 0.7 lO0 O.1 IO0 lO0 0.6
1 A l l values are expressed as a percentage of the t o t a l hemolysis In distilled water. 2Normal human and equine sera were used u n d i l u t e d and 50-pl volumes of RD, O, NHS and NES were added to the e r y t h r o c y t e s .
81 DISCUSSION
The s e n s i t i v i t y of the equine sera to heat treatment at 50oC, as monitored by RbRBC indicator c e l l s , is consistent with i n a c t i v a t i o n of factor B described for human, bovine and guinea pig complement (Gotze and Muller-Eberhard, 1971; Pang and Aston, 1978; Tabel, 1981; Brade, et a l . , 1972).
The results we obtained at 5O°C also suggest an unusual heat
s e n s i t i v i t y of certain key equine classical pathway components, using the SRBC i n d i c a t o r system.
This s e n s i t i v i t y is not a feature of the human or
guinea plg system (Mayer, 1961). We observed marked l y t i c a c t i v i t y of the equine sera, using an unsensltized RbRBC. Our results, comblned with other reports of successful lys i s in a rabbit erythrocyte-sheep antibody system, leads one to question i f the horse has natural antibodies to RbRBC, or i f the equine complement system operates p r i m a r i l y via the a l t e r n a t i v e pathway using this indicator system.
Unsensitlzed RbRBC have been shown to" be an a l t e r n a t i v e pathway
a c t i v a t o r in human and bovine sera (Blanchard and Leld, 1984; Pang and Aston, 1978; Brock et a l . , 1975; P l a t t s - M i l l s and Ishtzaka, 1974; Fearon and Austen, 1977).
The existence of natural equine c i r c u l a t i n g antibodies to
RbRBC has been suggested previously by Rice and Boulanger (1952) and Barta et a l . (1973).
At the time that Rice and Boulanger (1952) conducted t h e i r
experiments, the existence of the a l t e r n a t i v e pathway was not yet demonstrated; therefore, i t was not considered when they interpreted t h e i r data.
Recently, antlbody-mediated enhancement of the a l t e r n a t i v e pathway
has been shown to occur (Perrln et a l . , 1976; P o l h i l l et a l . , 1978).
I f , as
some have suggested, c i r c u l a t i n g antibodies to RbRBC e x is t n a t u r a l l y in the horse (Rice and Boulanger, 1952; Barta et a l . , 1973), then lysis of RbRBC may result from the augmentation of the a l t e r n a t i v e pathway by antibody. Neither Rice and Boulanger (1952) nor Barta et a l . (1973) used a Ca++ specific chelator, such as EGTA, to block classical pathway a c t i v a t i o n yet permit selective expression of the equine a l t e r n a t i v e pathway.
We have not
observed any natural a g g l u t i n a t i n g antibodies to unsensltlzed RbRBC by m i c r o t i t r e hemagglutlnation techniques.
Since the hemolytic a c t i v i t y for
RbRBC is rather l a b i l e , absorption experiments with packed RbRBC in the presence of EDTA cannot be carried out, because
the untreated control sera
decays In a tlme-dependent manner (Leid, unpublished observations). The hemolytic results we obtained using foal sera strongly suggest that natural antibody to the RbRBC plays a small role In hemolysis.
We made a
surprising observation when using the Mg+÷-EGTA treated p r e - c o l o s t r a l foal sera collected p r i o r to the transfer of maternal lmmunoglobulins and other
82 serum proteins.
There was a marked diminution in hemolytic a c t i v i t y when
the Mg+÷-EGTA treated pre-colostral sera were used, when compared to the untreated sera.
Since antibodies to RbRBC have not been detected in these
sera, alternative pathway activation in the horse may require Ca++ at some c r i t i c a l activation step.
T h i s Ca++ requirement is unusual since no other
species has demonstrated a similar cation requirement for expression of the alternative pathway. T h i s hypothesized Ca++ dependence of alternative pathway expression may have led to our varied results with Mg++-EGTA treated adult sera and may have caused the inconsistencies in equine hemolytic a c t i v i t y previously noted by others.
The results we have
descrlbed--heat s u s c e p t i b i l i t y at 50oC, l y t I c a c t i v i t y in the presence of EGTA, and Mg÷÷ dependence--Indlcate that the alternative pathway is activated when the unsensltlzed RbRBC and equine sera are used.
As reported
by Barta et al. (1973), no significant difference was observed In hemolysis of the RbRBC whether i t was sensitized with antibody or not. Our hypothesis that the equine hemolytic a c t i v i t y for RbRBC was primarily due to alternative pathway activation was further supported by our experiments with the equine and human D and RD reagents.
The hemolytic
a c t i v i t y for RbRBC was apparently due to the equine alternative pathway, since lysls did not occur with an equine RD alone.
Although only a I0%
recovery of hemolytic a c t i v i t y was observed upon combination of the equine RD and D, equine D was very active when reconstituting a human RD.
The lack
of lysls of a sensitized SRBC by the equine RD Is puzzling; in contrast, a human RD prepared in a homologous manner w i l l lyse a sensitized SRBC (Table II).
The results with the human RD and the SRBC are to be expected, since
the human RD possesses the classical pathway activation and terminal l y t l c components. The equine RD should have the same characteristics, but clearly does not.
Equine D, however, is active with the human RD and reconstitutes
a human RD.
HumanD, although active with a human RD, does not reconstitute
an equine RD.
These reconstltutlon experiments suggest that either an
equine complement component is extremely l a b i l e and decays during the gel f i l t r a t i o n chromatographic step, or that an additional component, not previously recognized, must be present for reconstltutlon to occur.
We
believe that the former is more l i k e l y , although we are not ruling out the latter possibility.
Our work with the bovine complement system, manipulated
in the same manner, shows that a stable bovine RD and D can be easily prepared in which the equine D is able to substitute f u l l y for i t s bovine counterpart (Blanchard and Leld, 1984).
We suggest that most, i f not a l l ,
previous work using equine complement for lysls of RbRBC was due to alternative pathway activation and was not the result of any antibody-
88 dependent pathway. The r e l a t i v e l y poor lysls of antibody sensitized SRBC suggests either that the horse has a poorly developed classical pathway, or possesses elevated levels of plasma control proteins.
In addition, I t Is
possible that the target cells (SRBC) and buffer systems used were not optimal for demonstration of the equine classical pathway. All of these avenues should be investigated in the future.
Wlth 25 horses giving rise to
an average of 71% hemolytlc a c t i v i t y for the SRBC, I t is clear that almost a l l horses w111 have some l y r i c a c t i v i t y for thls target c e l l .
A note of
caution is appropriate for those who undertake the future characterization of the equine alternative complement system: not a l l horses studied under the conditions we have described w l l l glve marked hemolytic a c t i v i t y for the unsensltlzed RbRBC In Mg++-EGTA. The reasons for such non-reactlvlty may reflect a c r i t i c a l Ca++ dependencefor equine alternative pathway activation, as seen In our results with pre-colostral foal sera.
Thls
unusual cation dependencemay account for the controversy In the literature.
A more careful evaluation of equine antibodies involved In the
hemolysis is called for, as are additional tests on the lyric characteristics of equine complement. The results presented here, however, serve as preliminary evidence for the existence of a domestic animal species In which alternative pathway activation is a major feature of its complement system. ACKNOWLEDGMENTS
This i n v e s t i g a t i o n received support from the f l l a r l a s l s
component of the
UNOP/World Bank/WHO Specla] Program for Research and Training in Tropical Diseases (FIl/T16/181/OS/09) to RWL and the Stanley Adler Research Fund from Washington State U n i v e r s i t y .
Portions of this work were done at both
Michigan State U n l v e r s l t y and Washlngton State U n i v e r s i t y . REFERENCES
Barta, O. and Hubbert, N.L., ]978. Testing of hemolytic complement components in domestic animals. Am. 3. Vet. Res., 39 : 1303-1310. Barta, 0., Barta, V. and Williams, E . I . , 1973. A method for t l t r a t l n g equine hemolytic complement. Z. Immunltaetsforsch., 146 : 114-]22. Blanchard, D.P. and Letd, R.W., 1984. I s o l a t i o n and p a r t l a ] characterization of bovine and equine factor D. Mol. Immunol. (In Press). Bordet, 3. and Streng, O., ]909. Les phenomenes d'adsorptlon et le conglutlne du serum du boeuf. Zb]. Bakt. Hyg. I . Abt. Org., 49 : 260-276. Brade, V., Cook, C.T., Shin, H.S. and Mayer, M.M., 1972. Studies on the properdin system: I s o l a t l o n of a h e a t - l a b l ] e factor from guinea pig serum related to a human g l y c i n e - r i c h beta glycoproteln (GBG or Factor B). 3. Immunol., 109 : 1174-1181.
84 Brock, J.H., Ortega, F. and P1nerio, A., 1975. Bacterlocldal and hemolytic a c t i v i t y in bovine colostrum and serum. Effect of p r o t e o l y t l c enzymes and ethylene glycol tetraacetlc acid (EGTA). Ann. I n s t . Pasteur ( L i l l e ) , 126 : 430-451. Coley, S.C. and Leld, R.W., 1982. Effects of extracts of Onchocerca c e r v l c a l i s from horses on the l y t l c a c t i v i t y of human, rat and equine complement. Clln. Immunol. Immunopathol., 23 : 113-123. Dyer, R.M. and Leld, R.W., 1983. Surface receptors for IgG and complement on the equlne alveolar macrophage. Inflammation, 7 : 183-195. Fearon, D.T. and Austen, K.F., 1980. Current concepts In Immunology. The a l t e r n a t i v e pathway of complement. A system for host resistance to microbial i n f e c t i o n . N, Eng. J. Red., 303 : 259-263. Fearon, D.T. and Austen, K.F., 1977. A c t i v a t i o n of a l t e r n a t i v e complement pathway with rabbit erythrocytes by circumvention of the regulatory action of endogenous control proteins. J. Exp. Med., 146 : 22-33. Fine, P.O., Marney J r . , S.R., Colley, D.G., Sergent, 3.S. and DesPrez, R.M., 1972. C3 shunt a c t i v a t i o n In human serum chelated wlth EGTA. J. Immunol., 109 : 807-809. Glaze, M.B., McGulre, T.C., Schmldt, G.M. and Leld, R. Wes, 1984. Immunoglobulln levels In tears and aqueous humor of horses before and a f t e r diethylcarbamazine (DEC) therapy. Vet. Immunol. Immunopathol., 7: 185-198. Gotze, O. and Muller-Eberhard, H.J., 1971. The C3-actlvator system: An a l t e r n a t i v e pathway of complement a c t i v a t i o n . J. Exp. Red., 134 : 90-I08s. Hole, N.H. and Coombs, R.R.A., 1947. The conglutlnating phenomenon. I I . The technique of the c o n g l u t l n a t i n g complement adsorption test compared wlth the hemolytic complement f i x a t i o n test. J. Hyg. (Camb.), 45 : 490-496. Leon, M.A. and Norden, A. 1959. Kinetics of equine complement. ~. Immunol., B3 : 99-106. Lachmann, P.J. and Hobart, M.J., 1978. Complement technology. In: O.M. Weir ( E d i t o r ) , Handbook of Experimental Immunology, 3rd ed. Blackwe11 S c i e n t i f i c Publications, London, pp. 5A.1-5A.23. Mayer, M.M., 1961. Complement and complement f i x a t i o n . In: E.A. Kabat and M.M. Mayer (Editors), Experimental Immunochemlstry, 2nd ed. Charles C. Thomas, S p r l n g f l e l d , pp. 133-240. McGuire, T.C., Banks, K.L. and Popple, M.J., 1975. Combined Immunodeflclency in horses: Characterization of the lymphocyte defect. Clin. Immunol. Immunopathol., 3 : 555-566. McGulre, T.C., Banks, K.L., Evans, D.R. and Popple, M.3., 1976. Agammaglobullnemla in a horse with evidence of functional T lymphocytes. Am. J. Vet. Res., 3? : 41-46. Muir, R., 1911. On the relationship between the complement and immune bodies of different animals. 3. Path. Bacterlol., 16 : 523-534. Pang, A.S.D. and Aston, W.P., 1978. The alternative complement pathway in bovine serum: The isolation of a serum protein with factor B a c t i v i t y . Immunochem., 15 : 529-534. Perrln, L.H., Joseph, B.S., Cooper, N.R. and Oldstone, M.B.A., 1976. Mechanism of injury of vlrus-lnfected cells by a n t l v l r a l antlbody and complement: Participation of IgG, F(ab') 2 and the alternative pathway. 3. Exp. Med., 143 : 1027-I041. Perryman, L.E., McGulre, T.C., Banks, K.L. and Henson, J.B., 1971. Decreased C3 levels In a chronic virus infection: equine infectious anemia. J. Immunol., I06 : 1074-I078. P l a t t s - M l l l s , T.A.E. and Ishlzaka, K., 1974. Activation of the alternative pathway of human complement by rabbit cells. 3. Immunol., l l 3 : 348-358.
85 P o l h l l l , R.B., Newman, S.L., P r u l t t , K.M. and 3ohnson Jr., R.B., 1978. Kinetic assessment of alternative complement pathway in a hemolytic system. I I . Influence of antibody in alternative pathway activation. 3. Immunol., 121 : 371-376. Rice, C.E., 1950a. The Interchangeabillty of the complement components of different animal species. I. Literature survey. Can. 3. Comp. Med., 14 : 369-379. Rice, C.E., 1950b. The Interchangeabillty of the complement components of different species. I l l . In conglutlnatlon. 3. Immunol., 65 : 499-510. Rice, C.E. and Boulanger, P., 1952. The Interchangeablllty of the complement components of different animal species. IV. In the hemolysis of rabbit erythrocytes sensitized with sheep antibody. J. Immunol., 68 : 197-205. Rice, C.E. and Crowson, N., 1950. The Interchangeablllty of the complement components of different animal species. I I . In hemolysis of sheep erythrocytes sensitized wlth rabbit amboceptor. J. Immunol., 65 : 201-210. Sandberg, A.L. and Osler, A.G., 1971. Dual pathways of complement interaction with guinea pig Immunoglobullns. 3. Immunol., I07 : 1268-1273. Tabel, H., 1981. Alternative pathway of bovine complement. Immunochemlcal studies on factor B-llke serum protein and its conversion product, By2. Can. 3. Comp. Med., 45 : 291-298.