The humoral fluid of Branchiostoma belcheri tsingtauense was examined for the presence of complement-like activity. The humoral fluid showed hemolytic activity for rabbit erythrocytes and those from species representing mammals, birds, amphibians and fish, but not sensitized sheep erythrocytes. There was no relationship between phylogeny of the target erythrocytes and degree of hemolysis. The hemolytic activity was optimally assayed at 20°C, at pH 7.5, and in the presence of 10 mM Mg2 . The hemolytic activity was Mg2 -dependent and heat-sensitive, and was abrogated by treatment with rabbit anti-human C3 serum, zymosan, methylamine, hydrazine, and phenylmethylenesulfonyl fluoride. In addition, Western blotting and titration by turbidimetric immunoassay (TIA) revealed that amphioxus humoral fluid contained C3 component, and its concentration is about 1.17 mg/ml, which is comparable to C3 concentration in human or dog sera. These suggest that the hemolytic activities displayed by amphioxus humoral fluid appear to represent the vertebrate complement system probably operating via the alternative pathway.
INTRODUCTION
Complement is a complex system of serum proteins that plays a pivotal role in vertebrate defense reactions. Activation of the mammalian complement system occurs via three convergent pathways, the classical pathway (CP) which is activated by an antigen-antibody complex and requires Ca2+ and Mg2+ (Ruddy, 1974), the alternative pathway (AP) which is triggered by the third component of complement C3 and requires Mg2+ alone, and the lectin pathway which is activated by mannose-binding lectin (MBL) that interacts with mannose sugars on bacterial cell surfaces (Smith et al., 1999). In contrast to the CP, the AP activity is susceptible to treatment with the mammalian complement inactivators like zymosan (which is a C3-depleting agent), hydrazine (which destroys C3 and C4 components of mammalian complement), and methylamine (which is commonly used in complement analysis to deactivate the thioester sites on complement proteins) (Takada et al., 1978; Gotze and Müller-Eberhard, 1976; Vogel and Müller-Eberhard, 1985; Van Dijk et al., 1980; Suner and Tort, 1994). Phylo-genetically, the relation of the AP and the lectin pathway remains open (Al-Sharif et al., 1998; Zarkadis et al., 2001; Smith et al., 1998), yet the CP may have generated from the alternative and lectin pathways by a genome wide duplication during early period of vertebrate evolution (Ohno, 1970; Zarkadis et al., 2001).
The CP activity can be assayed by determining the amount of serum required to hemolyse a given number of sensitized sheep red blood cells (SRBC), whereas the AP activity is measured in the same manner, but by using rabbit red blood cells (RaRBC) instead of sensitized SRBC. It has been demonstrated that functionally and structurally similar complement components are present in all non-mammalian vertebrates such as birds (Ohta et al., 1983; Kai et al., 1983), reptiles (Vogel et al., 1984; Koppenheffer, 1987), amphibians (Sekizawa et al., 1984) and fish including hag-fish and lamprey, the most primitive class of extant vertebrates (Koppenheffer, 1987; Hanley et al., 1992; Nonaka et al., 1984a, b). It has also been shown that complement components are present in invertebrates such as tunicate and sea urchin (Nonaka et al., 1999; Al-Sharif et al., 1998; Smith et al., 1999).
Amphioxus, a cephalochordate, has long been regarded as a basal lineage of chordates. It lacks free circulating blood cells in the circulation system (Moller and Philpott, 1973; Silva et al., 1995), and thus study of its immune defense system would be unexceptionally intriguing. Previous investigation has showed that both lectin and phenoloxidase are involved in defense reactions of amphioxus (Benedicts and Capalbo, 1981; Li et al., 2000; Mock and Renwrantz, 1987; Wang et al., 2002; Zhang and Li, 2000). However, little is known about the complement system of this animal although from it C3-like and C6-like gene fragments have been cloned (Suzuki et al., 2002).
In this study, we have sought to determine whether the humoral fluid of the amphioxus Branchiostoma belcheri tsingtauense possesses hemolytic activity, and if so, to detect whether it shares characteristics with the vertebrate complement system. It is reported here for the first time that a hemolytic activity is present in amphioxus humoral fluid, which appears to represent the vertebrate complement system probably operating via the AP.
MATERIALS AND METHODS
Reagents
Ethylenediamine tetraacetic acid (EDTA), ethyleneglycol-bis (β-aminoethyl ether)-N, N, N′, N′-tetraacetic acid (EGTA), methylamine hydrochloride, hydrazine hydrate, zymosan, and phenylmethylenesulfonyl fluoride (PMSF) were purchased from Sigma (St Louis, MO), and gelatin and glucose from Amersco (Solon, OH). All the other chemicals used were analytical reagents.
Rabbit anti-human C3 serum (3.44 mg/ml; titer, 1:32) and non-immune rabbit serum were purchased from Sino-American Biotechnology Co. (Beijing), the turbidimetric immunoassay (TIA) kits containing sheep anti-human C3 sera (86 μg/ml) and non-immune sheep sera from Bio-Sun Co. (Fuzhou, Fujian).
The buffers used in the experiments for AP activity assay were 10 mM EGTA-Mg-GVB (pH 7.5) containing 10 mM EGTA, 10 mM Mg2+ and GVB (5 mM sodium barbiturate, 446 mM NaCl and 0.1% gelatin), and 10 mM EDTA-GVB (pH 7.5) containing 10 mM EDTA, 5 mM sodium barbiturate, 446 mM NaCl and 0.1% gelatin. Unless otherwise noted, 10 mM EGTA-Mg-GVB was employed. The buffers used in the experiments for CP activity assay were GVB2+ containing 5 mM sodium barbiturate, 0.5 mM Ca2+, 1.0 mM Mg2+, 446 mM NaCl and 0.1% gelatin, and GGVB2+ containing 5 mM sodium barbiturate, 0.5 mM Ca2+, 1.0 mM Mg2+, 446 mM NaCl, 2.5% glucose and 0.1% gelatin. Unless otherwise noted, GVB2+ was employed. All buffered solutions were stored at 4°C and used within one week.
Preparation of amphioxus humoral fluid
Amphioxus B. belcheri tsingtauense collected from the “amphioxus ground” near Shazikou in the vicinity of Qingdao were reared in laboratory and fed on a diet of single-cell algae daily until use. The seawater was aerated continuously and changed once a day. The humoral fluid was prepared by the method of Wang et al. (2002). Briefly, about one thousand of amphioxus with average of body length of about 4 cm were rinsed with distilled water, wiped out thoroughly with sterilized gauze, and then cut into about 2 mm3 pieces on ice to bleed. After centrifugation at 12 000 g for 30 min at 4°C, the supernatant was collected and stored at −70°C until used.
The protein concentration was determined by the method of Lowry et al. (1951) with bovine serum albumin as standard.
Target erythrocytes
For most experiments target erythrocytes were drawn from the same giant New Zealand male rabbit and prepared for use by washing 3 times with 10 mM EGTA-Mg-GVB with 149 mM NaCl. Erythrocytes were also collected from representative species of major vertebrate classes including human, sheep, rat, chick, pigeon, toad (Bufo gargarizans), grass carp (Ctenopharyngodon idellus), loach (Misgurnus anguillicaudatus) and lip shark (Chiloscyllium plagiosum). All blood samples were each mixed with an equal volume of Alsever's solution immediately after sampling and centrifuged at 500 g for 5 min. The pelleted erythrocytes were washed 3 times with 10 mM EGTA-Mg-GVB with 149 mM NaCl and adjusted spectrophotometrically at 414 nm to a concentration of 2×108 cells/ml, and stored at 4°C.
Alternative complement pathway titer (ACH50) assay
ACH50 assay was performed according to the method of Yano (1992). Pilot experiments showed that dilution of amphioxus humoral fluid resulted in decrease in its hemolytic activity, and when it was diluted 100-fold with 10 mM EGTA-Mg-GVB, the hemolytic activity was no longer detectable. Therefore, 50, 100, 150, 200 and 250 μl of the humoral fluid diluted 20-fold were dispensed into 5 ml test tubes, respectively, and the total volume was made up to 350 μl with 10 mM EGTA-Mg-GVB. To each test tube was 100 μl of RaRBC suspension (2×108 cells/ml) added. A 100% hemolysis control and a cell blank control were included. The tubes were incubated at 20°C for 90 min with frequent shaking and the hemolytic reaction was then terminated by adding cold 3.15 ml 10 mM EDTAGVB (pH7.5). After centrifugation at 1500 g for 5 min, absorbance of the supernatants at 414 nm (OD414) was measured spectrophotometrically. The absorbances including 100% hemolysis control were corrected by subtraction of the absorbance of the cell blank. The ACH50 (units/ml) was calculated according to the method described by Yano (1992).
In addition to RaRBC, the remaining erythrocytes collected were similarly tested to determine the best activators of the AP in the amphioxus.
Classical complement pathway titer (CH50) assay
The CP activity of amphioxus humoral fluid, if any, was assayed by the method of Yano (1992). SRBC were washed three times with 10 mM PBS (pH7.4), and adult lip sharks (750 g each) were injected peritoneally with 1 ml of the erythrocyte suspension (5×108 cells/ml). The fish were successively injected five times at a weekly interval, and bled one week after the final booster injection. For preparation of sensitized SRBC, lip shark antiserum was first heated at 50°C for 30 min to inactivate endogenous complement, diluted 50-fold with 10 mM EDTA-GVB, and mixed with an equal volume of SRBC suspension (1×109 cells/ml EDTA-GVB) with constant swirling. The mixture was incubated at 25°C for 30 min, and the sensitized cells were then washed three times with GGVB2+. After centrifugation at 500 g, the sensitized SRBC was re-suspended in GGVB2+ and adjusted to a concentration of 5×108 cells/ml.
To examine CH50 of amphioxus humoral fluid, lip shark normal serum, grass carp normal serum and amphioxus humoral fluid were diluted with GVB2+ at ratios of 1:50, 1:10 and 1:20, respectively. Different volumes of the diluted sera and amphioxus humoral fluid ranging from 0.4 to 1.0 ml were dispensed into test tubes and the total volume was made up to 1.3 ml with GVB2+. To each tube was 0.2 ml of sensitized SRBC suspension (5×108 cells/ml) added. A 100% hemolysis control and a cell blank control were included. The tubes were incubated at 25°C for 60 min with occasional shaking, and centrifuged at 1000g for 5 min. Absorbance of the supernatants at 541 nm (OD541) was measured spectrophotometrically. The OD541 values including 100% hemolysis control were corrected by subtraction of the absorbance of the cell blank. The calculation of CH50 (units/ml) was according to the method of Yano (1992).
Divalent cation analysis
Pilot experiments showed that addition of 50 μl of 100 mM EDTA (pH 8.5) or 100 mM EGTA (pH 7.5) into 50 μl of amphioxus humoral fluid had little effect on pH value of the humoral fluid. For divalent cation requirement assay, 50 μl of amphioxus humoral fluid was mixed with 50 μl of 100 mM EDTA (pH 8.5) or EGTA (pH 7.5), and incubated at 25°C for 30 min. The mixtures were then diluted 10-fold with 10 mM EGTA-Mg-GVB, and their hemolytic activity was examined, and ACH50 calculated. Meanwhile, 50 μl of amphioxus humoral fluid was mixed with 50 μl of 100 mM EDTA and pre-incubated at 25°C for 30 min. The mixtures were extensively dialyzed against 10 mM EGTA-GVB for 2 hr at 4°C to eliminate EDTA, and mixed again with 100 μl of 80 mM CaCl2 or MgCl2, and their hemolytic activity was similarly examined and ACH50 calculated.
Inactivator assay
A total of 50 μl of amphioxus humoral fluid was mixed with an equal volume of 60 mg/ml zymosan (Day et al., 1970). For control, amphioxus humoral fluid was mixed with 10 mM EGTA-Mg-GVB. The mixtures were incubated at 25°C for 30 min, and centrifuged at 12 000 g for 10 min. The supernatants were pooled, diluted 10-fold with 10 mM EGTA-Mg-GVB, and their hemolytic activity was determined, and ACH50 calculated.
To test the effects of methylamine and hydrazine on the hemolytic activity, 50 μl of amphioxus humoral fluid was mixed with an equal volume of 6.3, 12.5, 25, 50, 100 and 200 mM methylamine or hydrazine prepared in 10 mM EGTA-Mg-GVB (Vogel and Müller-Eberhard, 1985). For control, amphioxus humoral fluid was mixed with 10 mM EGTA-Mg-GVB. The mixtures were pre-incubated at 25°C for 30 min, and diluted 10-fold with 10 mM EGTA-Mg-GVB. After incubation for 90 min at 20°C, their hemolytic activity was determined, and ACH50 calculated.
Similarly, 50 μl of amphioxus humoral fluid was mixed with an equal volume of 1, 2, 5, 10 and 20 mM PMSF prepared in 2-propanol (Leonard et al., 1990). For control, amphioxus humoral fluid was mixed with 2-propanol. The mixtures were pre-incubated at 25°C for 30 min, and diluted 10-fold with 10 mM EGTA-Mg-GVB. After incubation for 90 min at 20°C, their hemolytic activity was determined, and ACH50 calculated.
Anti-human C3 serum treatment assay
The capacity of the rabbit anti-human C3 serum to inhibit the hemolytic activity of amphioxus humoral fluid was analyzed by the method of Nonaka et al. (1999). The rabbit antiserum was inactivated at 56°C for 20 min, and diluted two-fold serially up to 32 times with 10 mM PBS (pH7.4) containing 446 mM NaCl. Aliquots of 50 μl of the serially diluted antiserum were mixed with an equal volume of amphioxus humoral fluid, and the mixtures were incubated at 25°C for 1 hr. After centrifugation at 3000 g for 10 min, the super-natants were diluted 10-fold with 10 mM EGTA-Mg-GVB, and their hemolytic activity was determined. For control, both heat-inactivated non-immune rabbit serum and 10 mM PBS (pH7.4) containing 446 mM NaCl were incubated with amphioxus humoral fluid.
To confirm the presence of C3 component in amphioxus humoral fluid, Western blotting analysis was carried out. Amphioxus humoral fluid and human serum were electrophoresed on 7.5% native PAGE (Davis, 1964). Proteins separated were transferred onto nitrocellulose membrane and immunostained with rabbit anti-human C3 serum (Towbin et al., 1979).
Titration of C3 protein
The protein concentration of amphioxus humoral fluids prepared was about 14.46 mg/ml. The content of C3 component in amphioxus humoral fluid was assayed according to the instructions of the TIA kits. For control, the concentration of C3 in 5 healthy human sera was also assayed. An aliquot of 250 μl of sheep anti-human C3 serum (86 μg/ml) was mixed with 25 μl of amphioxus humoral fluid diluted 5-fold with 10 mM PBS (pH7.4), and the mixture was incubated at 25°C for 200 seconds, and absorbance measured by an automated chemistry analyzer AU 560 (Japan), which had been pre-adjusted to zero with non-immune sheep serum containing in TIA kits. The C3 concentration in human sera was similarly measured.
Optimization of ACH50 assay
To examine the optimum temperature for the hemolytic reaction, seven groups of samples including different volumes of the diluted humoral fluid and 100 μl of RaRBC suspension (2×108 cells/ml) prepared as described above in “Alternative complement pathway titer (ACH50) assay” were incubated at 4°C, 20°C, 25°C, 37°C, 45°C, 50°C and 56°C, respectively, for 90 min (Matsuyama et al., 1988). The hemolytic reaction was terminated by adding a volume of 3.15 ml cold EDTA-GVB (10 mM, pH7.5), and the hemolytic activity then measured and ACH50 calculated.
To test the optimum pH, the buffer 10 mM EGTA-Mg-GVB was readjusted to pH 6.5, 7.0, 7.5, 8.0, 8.5 and 9.0, respectively, and used in the hemolytic activity analysis (Matsuyama et al., 1988). The hemolytic reactions were carried out at 20°C for 90 min, and ACH50 was calculated from the OD414.
To assay the Mg2+ optimum concentration, the concentration of Mg2+ in the buffer 10 mM EGTA-GVB was adjusted to 0, 5, 10 and 15 mM, respectively, and used in the hemolytic activity analysis (Matsuyama et al., 1988). The hemolytic reactions were carried out at 20°C for 90 min, and ACH50 was calculated.
RESULTS
Hemolytic activity
The hemolytic reaction was first tested in the presence of 10 mM EGTA, which was sufficient to prevent any CP activation due to natural antigen-antibody complexes, if any. The hemolytic assay revealed that amphioxus humoral fluid was capable of hemolysing RaRBC suspended in EGTAMg-GVB (Table 1). In contrast, the humoral fluid failed hemolysing sensitized SRBC suspended in GVB2+ though the sensitized SRBC had been proved to be excellent acti-vators for the CH50 titration of grass carp and lip shark sera (Data not shown).
Table 1
Sensitivity of different erythrocyte species to hemolysis by amphioxus humoral fluid.
Target cell phylogeny
Erythrocytes from species representing mammals, birds, amphibians, and fish were tested to determine if target cells from phylogenetically closer animals would be more easily hemolysed. As shown in Table 1, all RBC tested activated the hemolytic activity of amphioxus humoral fluid to some extent, but the highest activity was obtained when RaRBC was used. No evident relationship was observed between the phylogeny of the target cells and degree of hemolysis. Interestingly, however, rat erythrocytes were excellent target cells.
Properties of the hemolytic activity
Chelation with 100 mM EDTA completely depleted the hemolytic activity of amphioxus humoral fluid, while chelation of the humoral fluid in EGTA-Mg did not affect its hemolytic reaction. In addition, the hemolytic activity of the EDTA-treated humoral fluid was partially (32%) restored by adding Mg2+ into it, but not by adding Ca2+ (Fig. 1). This indicates that Mg2+ is required for the hemolytic reaction, while Ca2+ is not.
Incubation of amphioxus humoral fluid with zymosan effectively depleted the hemolytic activity. The hemolytic activity of the humoral fluid was also markedly reduced by incubation with PMSF (Fig. 2).
As shown in Fig. 3, small nucleophiles such as hydra-zine and methylamine were capable of abrogating the hemolytic activity of amphioxus humoral fluid. The effect of methylamine and hydrazine on the hemolytic activity of amphioxus humoral fluid was concentration-dependent. These indicate the presence of at least one methylamineand hydrazine-sensitive component in the hemolytic system of the humoral fluid.
Incubation of rabbit anti-human C3 serum with amphioxus humoral fluid abrogated its hemolytic activity, while incubation of normal rabbit serum or PBS with the humoral fluid did not impair the hemolytic activity (Fig. 4). Western blotting analysis demonstrated that like human serum, amphioxus humoral fluid cross-reacted with rabbit anti-human C3 serum, forming a main band equivalent to C3 and two minor bands resembling C3 α and β chain, respectively(Fig. 5). These show that C3 component is responsible for the hemolytic activity of amphioxus humoral fluid.
Fig. 4
Effects of rabbit anti-human C3 (RAHC3) and non-immune rabbit sera (NRS) on the hemolytic activity of amphioxus humoral fluid. Aliquots of 50 μl of the serially diluted antisera were mixed with an equal volume of amphioxus humoral fluid, and the mixtures were incubated at 25°C for 1 hr. After centrifugation at 3 000 g for 10 min, the supernatants were diluted 10-fold with 10 mM EGTA-Mg-GVB, and their hemolytic activity was determined.
Fig. 5
Western blotting of C3 in amphioxus humoral fluid. Amphioxus humoral fluid and human serum were electrophoresed on 7.5% native PAGE. Proteins separated were transferred onto nitrocellulose membrane and immunostained with rabbit anti-human C3 serum. H, human serum; A, amphioxus humoral fluid.
The content of C3 component in amphioxus humoral fluid was then quantified by TIA. It was found that the concentration of C3 component in amphioxus humoral fluid ranged from 0.79 mg/ml to 1.47 mg/ml with the average of 1.17 mg/ml, which was close to C3 concentration in human sera (0.98 mg/ml).
These results exhibit that the complement-like activity is present in the cephalochordate amphioxus.
Optimal conditions for ACH50 assay
Amphioxus humoral fluid showed the highest hemolytic activity at 20°C, and even at 4°C a considerable amount of hemolysis was observed (Fig. 6). After incubation at 37°C for 90 min, the ACH50 titer decreased to about 50% of the highest value. Heating at 45, 50 and 56°C completely depleted the hemolytic activity. It is clear that the hemolytic activity is heat-labile.
The optimum pH for the hemolytic reaction of amphioxus humoral fluid was 7.5 and at pH values lower than 6.5 or higher than 9, the hemolytic activity was significantly reduced (Fig. 7).
The optimum concentration of Mg2+ for the hemolytic reaction of amphioxus humoral fluid was 10 mM (Fig. 8). Notably, some hemolytic activity was observed at 0 mM Mg2+, which might be attributable to Mg2+ originally present in the humoral fluid.
DISCUSSION
The standard hemolytic analysis method of Yano (1992) has been applied to the assay of hemolytic complement activity in amphioxus humoral fluid in the present work. We have demonstrated for the first time the presence of a hemolytically active complement system in amphioxus humoral fluid. Evidences indicating the complement nature of the hemolytic activity of amphioxus humoral fluid include: (1) ability of the humoral fluid to hemolyse not only RaRBC but also heterologous erythrocytes suspended in EGTA-MgGVB rather than sensitized SRBC suspended in GVB2+; (2) requirement of Mg2+ rather than Ca2+; (3) abrogation of the hemolytic activity by pre-incubation of amphioxus humoral fluid with known inactivators of the mammalian AP such as zymosan, methylamine and hydrazine; (4) depletion of the hemolytic activity by incubation of amphioxus humoral fluid with rabbit anti-human C3 serum, but not incubation with normal rabbit serum; (5) presence of C3 component in the humoral fluid; and (6) inhibition of the hemolytic activity by heat or EDTA. These properties are apparently similar to those observed in the complement system of the vertebrates including fish and mammals (Vogel and Müller-Eber-hard, 1985; Sunyer and Tort, 1995), and indicate that the CP is not present in amphioxus. This is also in line with the hypothesis that the CP appears with the evolutionary emergence of the jawed vertebrates (Smith et al., 1999; Zarkadis et al., 2001; Lambris et al., 1999).
Recently, the mannose-binding lectin-associated serine protease (MASP) (Ji et al., 1997), and C3 (Nonaka et al., 1999) have been identified in the invertebrates like urochor-date, and the origin of lectin-dependent pathway is traced back to the invertebrates (Sekine et al., 2001). Therefore, the possibility cannot be ruled out that the lectin pathway is present in amphioxus. However, the facts that the hemolytic activity displayed by amphioxus humoral fluid does not require Ca2+ and is inhibited by pre-incubation with the known specific inactivators of the mammalian AP, zymosan, methylamine and hydrazine, suggest that the hemolytic activity is probably due to the complement system operating via the AP. Detailed molecular study of the hemolytic activities in amphioxus will aid elucidating the evolutionary relationship between AP and lectin pathway.
In general, the levels of complement are expressed in ACH50 and CH50 titers, which are defined as the reciprocal of the serum dilution causing 50% hemolysis of unsensitized RaRBC and sensitized SRBC, respectively. The ACH50 titers of human, cattle, pig, sheep, dog, guinea pig and rat sera are 18.4, 43.0, 13.6, 15.4, 14.4, 11.9 and 4.5 units/ml, respectively (Vogel and Müller-Eberhard, 1976), and those of porgy, yellowtail, eel and sea bream sera 92, 142, 134 and 170 units/ml, respectively (Matsuyama et al., 1988; Sunyer and Tort, 1995). The ACH50 observed in amphioxus humoral fluid is much higher than that of mammals, but approximates that of fish. This suggests a greater importance for the AP in fish and amphioxus than in higher vertebrates. It is of interest to note that the CP system is only poorly evolved in fish when compared with that of mammals. As amphioxus lacks the CP, it is conceivable that the AP functions as an important non-specific defense mechanism in this animal.
Suzuki et al. (2002) have recently cloned a C3-like gene fragment from amphioxus, which is structurally similar to vertebrate C3. It is revealed here by means of Western blotting and TIA that there exists a C3 component in amphioxus humoral fluid. The presence of the C3 component is further supported by the fact that incubation of rabbit anti-human C3 serum, and C3-depleting and C3-destroying agents with amphioxus humoral fluid abrogates its hemolytic activity. These together show that C3 component is involved in the hemolytic activity of amphioxus humoral fluid, and amphioxus C3 component shares at least partial antigenicity with human C3. The content of C3 component in amphioxus humoral fluid (1.17 mg/ml) is comparable to C3 concentration in human (0.8 to 1.3 mg/ml) and dog (1.24±0.32 mg/ml) sera (Tack and Prahl, 1976; Lucena et al., 1999).
Amphioxus humoral fluid displays hemolytic activity against red blood cells (RBC) from different animals including mammals, birds, amphibians and fish, but there is no clear relationship between the phylogeny of the target cells and degree of hemolysis. RaRBC have been shown the best activators for the AP in higher vertebrates, and RaRBC are also found to be most sensitive to hemolysis by amphioxus humoral fluid, suggesting RaRBC are universally excellent target cells for hemolytic activity assay of both vertebrates and invertebrates. Interestingly, rat RBC are found to be highly sensitive to hemolysis by amphioxus humoral fluid. It is known that the terminal sugars on the glycoproteins on the surface of RBC in addition to some associated serum proteins are important factors to activate the AP. RaRBC are missing sialic acid and therefore make them excellent test cells to assay AP complement activity (Fearon, 1978; Dijk et al., 1985; Sunyer and Tort, 1995). Sheep RBC, on the other hand, have a lot of sialic acid, and this is one aspect of their inability to activate the AP and be lyzed by the terminal pathway. Possibly, rat erythrocytes also lack sialic acid on their cell surface, while erythrocytes from human, chick, pigeon, toad, grass carp, loach and lip shark are all sialic acid-rich cells.
Serine protease has been proved an integral component of the mammalian complement system (Reid and Porter, 1981), sea star complement-like system (Leonard et al., 1990) and arthropod coagulation system (Nakamura and Levin, 1982). The serine protease activity is also an important component of amphioxus hemolytic system since its hemolytic activity is sensitive to treatment of PMSF.
The present study has also established the standard method for ACH50 titration of amphioxus humoral fluid. The activity is optimally assayed at 20°C, at pH 7.5, and in the presence of 10 mM Mg2+. This method will be very useful for further identification of amphioxus hemolytic system and for determination of hemolytic activity in the humoral fluid from other invertebrates.
The optimal pH and Mg2+ concentration for ACH50 titration of amphioxus humoral fluid both resemble those observed in higher vertebrates. The optimum temperature for ACH50 titration of amphioxus humoral fluid is also similar to that (20 to 25°C) for fish (Yano, 1992), but it is considerably lower than that (37°C) for mammals. It is noteworthy that the hemolytic activity of amphioxus humoral fluid remains at 4°C, but it is readily inactivated at 37°C. This may be due to that the amphioxus live in sea water where the ambient temperature varies widely, yet seldom rises as high as 37°C.
Acknowledgments
The authors thank Ms Miao Yu and Dr. Hongyan Li for the assistance in manuscript preparation. This work was supported by Ministry of Science and Technology (MOST) and Natural Science Foundation of China (NSFC).
