Isolation and characterization of acetylcholinesterase and other particulate proteins in the hemolymph of Aplysia californica.

Hemolymph of the marine mollusc, Aplysia californica, contains four large particles: acetylcholinesterase, hemocyanin, a hemagglutinin, and a structure tentatively identified as erythrocurorin. We purified the acetylcholinesterase 20-fold by differential centrifugation and filtration through a column of 4% agarose. The freshly isolated esterase complex was found to have a sedimentation coefficient of 69, but the negatively stained enzyme lacked a definite structure in the electron microscope, and appeared as irregular aggregates of a 60 A subunit. The complex was unstable below pH 5 or during storage at 7 degrees. Under these conditions, enzymatic activity remained essentially unchanged. Treatment of the purified enzyme with trichloroacetic acid, organic solvents, and sodium dodecyl sulfate broke the complex down into two major subunits with molecular weights of about 70,000. Exposure of the enzyme to [3H]diisopropylfluorophosphate resulted in the labeling of one of these subunits. Although similar in specificity, the cholinesterase of the blood differed from the enzyme in Aplysia nervous tissue, which is associated with membrane. Treatment with sodium deoxycholate activated the membrane-associated enzyme but inhibited slightly that of the hemolymph; tyrocidine inhibited the hemolymph enzyme but not the enzyme of nervous tissue; and mild digestion with trypsin released the membrane-bound enzyme in an active, soluble form, but inactivated the enzyme of hemolymph. The other particulates of Aplysia hemolymph were partially characterized. Aplysia hemocyanin was similar in structure to other molluscan hemocyanins. When negatively stained, the unit particle appeared to be a disc with a diameter of 280 A and a width of 45 A. These discs were stacked to form long cylindrical arrays. The purified hemocyanin was found to contain 0.26% copper (dry weight). Using differential centrifugation and gel filtration we also obtained a 9-fold purification of Aplysia hemagglutinin. This particle was 120 A in diameter with a dark staining central core of 40 A consisting of 6 subunits. The particle tentatively identified as erythrocurorin appeared as a structure 200 A in diameter consisting of 5 V-shaped subunits.

hemocyanin, a hemagglutinin, and a structure tentatively identified as erythrocurorin.
We purified the acetylcholinesterase ZO-fold by differential centrifugation and filtration through a column of 4% agarose.
The freshly isolated esterase complex was found to have a sedimentation coefficient of 69, but the negatively stained enzyme lacked a definite structure in the electron microscope, and appeared as irregular aggregates of a 60 A subunit.
The complex was unstable below pH 5 or during storage at 7". Under these conditions, enzymatic activity remained essentially unchanged.
Treatment of the purified enzyme with trichloroacetic acid, organic solvents, and sodium dodecyl sulfate broke the complex down into two major subunits with molecular weights of about 70,000.
Exposure of the enzyme to [3H]diisopropylfluorophosphate resulted in the labeling of one of these subunits.
Although similar in specificity, the cholinesterase of the blood differed from the enzyme in Aplysio nervous tissue, which is associated with membrane. Treatment with sodium deoxycholate activated the membrane-associated enzyme but inhibited slightly that of the hemolymph; tyrocidine inhibited the hemolymph enzyme but not the enzyme of nervous tissue; and mild digestion with trypsin released the membrane-bound enzyme in an active, soluble form, but inactivated the enzyme of hemolymph. The other particulates of Aplysia hemolymph were partially characterized.
Aplysia hemocyanin was similar in structure to other molluscan hemocyanins. When negatively stained, the unit particle appeared to be a disc with a diameter of 280 A and a width of 45 A. These discs were stacked to form long cylindrical arrays. The purified hemocyanin was found to contain 0.26% copper (dry weight). Using differential centrifugation and gel filtration we also obtained a g-fold purification of Aplysia hemagglutinin. This particle was 120 A in diameter with a dark staining central core of 40 A consisting of 6 subunits.
The particle * This investigation was supported by United States Public Health Service Grant NS 09955.
1 Present address, 1Iep:trtmcnt of Physiology, Columbia University College of Physicians and Surgeons, New York 10032.
tentatively identified as erythrocurorin appeared as a structure 200 A in diameter consisting of 5 V-shaped subunits.

Acetylcholinesterase
of nervous tissue is bound to membranes in Aplysia (1,2) as it is in other animals (see Rcfs. 3 and 4 for review).
Giller and Schwartz (2) described an csterase free in the hcmolymph of dplysia with substrate specificities similar to the enzyme from nervous tissue. The spccifiz activity of the acetylcholinestcrase in the blood was about an order of magnitude greater than that in ganglia; an equivalent volume of hemolymph contained about 6 times more activity than did the ~1~11 body of K2, the giant caholinergic nc'uron of the abdominal ganglion.
Esterase of clectroplas can be released from membranes by treatment with tolucne (5) to yield a soluble oligomeric enzyme (G-8). The acetylcholinesterase of ~lplysia hemolymph is unusual since it is a large enzyme complex which is not associated with membranes under physiologiraal conditions.
Most of the protein in Aplysia blood is particulate.
l~;xamination of ncgativelg stained preparations by electron microscopy revealed the presence of four large complexes.
In order to identify the acctylcholinesterase, \vc separated these complexes by diffcrcntial centrifugatioii and gel filtration. l'hc major particulate component was the respiratory protein, hcmocyanin.
This coppercontaining protein is present in the blood of many invertebrates (9). In addition to the acet3lcholincsterase, we also isolated a hcmagglutinin.
l'aulcy et al. (10) have previously reported hemagglutinating activity in ~lplysia blood. n-e have tcntatively ident.ified another particle as erythrocurorin, an iroll-containing protein, which has been described in the hemolymph of other invertcbratcs (11).
of vertebrate blood, was not studied further, but did not result over a cushion of 2 ml of 0.9 M sucrose. This gradient was cenfrom bacterial growth.
No bacteria were seen upon microscopic trifuged at 95,000 X g in the SW 27 rotor for 6 hours. Fractions examinations; addition of toluene or 0.25% sodium azidc was with-were collected dropwise by puncturing the bottom of the tube. out effect. Hemolymph from two animals was never combined.
Ekctron, Microscopy-Negatively stained specimens were pre-The small number of cells (12) and other debris was removed pared either with 2% ammonium molybdate (pH 5.2) or with 1% by centrifugation at 11,500 X g for 15 min. All purification pro-phosphotungstic acid neutralized with NaOH on Formvar-supcedures were carried out at 7". Samples were diluted in saline (0.15 M NaCl in 10 mM Tris-HCl, n1-I 7.6). Acetvlcholincsterase _ was assayed spectrophotometrically using acetylthiocholine (Sigma Chemical Co., St. Louis, MO.) (13) at 22'. A unit of acctylcholinesterase hydrolyzed 1 pmol of acetylcholine per min at 22". Protein was estimated spectrophotometrically (14). Hemocyanin was detcctcd by its ABiO, Purification of acetylcholinesterase was carried out with fresh blood and was completed within 72 hours.
Z?litial UiscorllirLuotls Grad&--Eight milliliters of the freshly isolated and centrifuged blood were layered above two l-ml density steps: 0.6 M sucrose in saline and 0.9 M sucrose in saline. Ten of ported carbon-coated 200-mesh copper grids (18).
The prepared grids were examined with a Siemens Elmiskop 1A electron microscope using a short focal length objective equipped with a decontamination device at an accelerating voltage of 80 kv. The top 8 ml of each of these ten gradients contained most of the hemagglutinin. The 10 0.6 M sucrose layers were combined for Most protein in Aplysia hemolymph is particulate. Using a further purification and characterization of the acetylcholinester-two-step density gradient, we found that the major constituent RSC and of the uarticlc which we have tentativelv identified as was hemocyanin.
Three other particle types were also partially erythrocurorin.
_ The 0.9 M layers contained hemocyanin. separated from each other ( Table I). All four particles were Second L)isco?lti?lUous GradieTlt-The combined 0.6 M sucrose layers from the initial gradients were diluted to 25 ml and placed present in the 0.6 M sucrose layer of the gradient. The distribuover three 5-ml saline layers: the bottom, 0.9 M sucrose; the mid-tion of the particle we have tentatively identified as erythrodle, 0.6 M sucrose; and the upper, 0.45 M sucrose. The gradient curorin was determined only by electron microscopy, since we was centrifuged for 4 hours at 95,000 X g in the SW 27 rotor. The had no other method for assaying the presence of this component. 0.6 hc layer was collected by hand, diluted to 8 ml, and concentrated to a volume of 1 ml by repeating the initial discontinuous This initial centrifugation resulted in fractions enriched in the gradient centrifugation at 105,000 X g for 90 min. three particles which WC could assay. Hemocyanin, the largest Gel Piltratiotl-The concentrated enzyme was applied to a 4% component, was the predominant material in the 0.9 M sucrose Agarose (Bio-Gel A-15m, 100 to 200 mesh, Bio-Rad, Richmond, layer. The hcmagglutinin remained primarily in the super-Calif.) column (1 X 10 cm), which was eluted with saline. natant.
Although both the 0.6 and 0.9 M sucrose layers con-Zsoelcctric Zf'ocusitlg-A gradient was prepared from a less dense solution consisting of 51 ml of water, 4 ml of 10/O Ampholine, pH 3 tained similar proportions of the acetylcholinesterase, the ento 10 (LKB Instruments, Inc., Rockville, Md.), and 5 ml of en-zyme in the 0.6 M layer had the highest specific activity, and was zyme solution (0.6 M sucrose layer of second gradient centrifuga-used as starting material for the further purification of the tion). and a more dense solution consisting of 37 ml of water. 8.5 enzyme.
I I ml of 8% Ampholine (PI-I 3 to lo), and 25 g of sucrose. Focusing was done at 4" for 30 hours with a final potential of 600 volts (15) Purification of Acetylcholinesterase in a 110.ml electrophoresis column (LKG) (16). Usine: differential centrifueation and gel filtration. we obtained Puri$icutiotl ufrd Assay of HemngglutinilL-A l-ml sample of the top layer from the initial discontinuous gradient (see above) was a 20.fold purification of the esterase with a 10% recovery of subjected to gel filtration on 4% Agarose. Hemagglutinin was initial activity (Table II). Loss of enzyme resulted primarily assayed using a modification of the procedure of Pauley et al. (10). from discarding cruder fractions. Enzyme activity was rela-Fresh (7 to 10 davs after collection) chick red blood cells in Alse-tivelv stable. Within a week of collecting the blood. less than were assayed, appropriate amounts of artificial sea water or vier's solution (glow Laboratories, Inc., Rockville, Md.) were sucrose saline were added to controls. The titer end point was the washed three times at, 10" in saline, and adjusted to a final concenhighest dilution which produced agglutination visible to the un-tration of 2'yo; 0.05 ml was added to duplicate serial 2-fold dilutions aided eye. We arbitrarily defined a unit of agglutining activity of samples in 2 ml of 0.15 hl NaCl in IO-ml round-bottom glass test as that amount producing detectable agglutination under these standard assay conditions. tubes. After 1 hour at 22", the tubes were kept for 18 hours at 14". When unfractionated hcmolymph or undialyzed gradient fractions PuriJcatiorL of Hemoc:yu,litz-The 0.9 M sucrose layer of the initial discontinuous gradient contained highly purified hemocyanin. Greater purification was achieved, however, on a discontinuous gradient with several more density layers. After brief centrifugation at 11,500 X g, 20 ml of hemolymph were layered on four sucrose stens each in 5 ml of saline: 0.2 M. 0.6 M. 0.9 M. and 1.2 M. at 105,000 -X g for 90 mjn.
10% of the total activity was lost.
In some experiments we included 0.2% glutaraldehyde in the gradients for a total exposure time We nevertheless carried out of 8 hours. Under these conditions, 80% of the enzymatic activity was lost with no change in the scdimentatlon of the surviving enzyme activity. the purification within 72 hours of collection, since we were concerned that the polymeric structure of the esterase complex might not be stable. Each step of the purification procedure yielded enzymatic activity of which 90 to 100% was sedimented   1A). The Asgo:A260 ratio of the purest fraction was 1.15.

Characterization of PuriJied Acetylcholinesterase
Stability of Esterase Complex-Although cholinesterase activity was almost unchanged after several weeks, the enzyme complex was apparently unstable. When freshly isolated, the blood's cholinesterase activity was 90% sedimented at 105,000 X g for 90 min; about 5 to 7% of the activity could not be sedimented after repeated centrifugation of the supernatants. We have not determined whether this small fraction of the total esterase activity is another enzyme or whether it represents smaller (but active) subunits of the enzyme complex.
After storing unfractionated hemolymph or purified fractions at 7", we found that about 30% of the esterase activity per week remained in the 105,000 X g supernatant.
Treatment with detergents and low pH also altered the sedi-" mentation behavior of the enzyme (see below). Lyophilization (The VirTis Co., Gardner, N. Y.), both in the presence and absence of phosphotungstate and ammonium molybdate, did not affect either the enzymatic activity or the proportion of the enzyme sedimenting at 105,000 X g.
Isoelectric Focusing and Effect of Low pH-Most of the enzymatic activity banded at pH 4.8 (Fig. 3) ; a small proportion had an isoelectric point of 3.6. We are uncertain about the molecular size of the enzyme in either of these bands, however. Treatment of the enzyme for short periods of time at low pH resulted in loss of structure, but not of enzymatic activity.
We incubated the esterase at room temperature in unbuffered saline made pH 3 or 5 with dilute HCl for 1 hour. After neutralization with 1 M phosphate buffer, all of the enzymatic activity was recovered, but only 20 to 30 y0 could be sedimented at 105,000 X g.
Efect of Detergents--The acetylcholinesterase of hemolymph did not appear to be associated with membrane.
Treatment of the enzyme with Triton X-100 or sodium deoxycholate at a detergent to protein ratio of 3.3 : 1 did not decrease substantially the amount of esterase which was sedimented at 105,000 X g (Table III) hemolymph enzyme (Table III). On the contrary, deoxycholate brought about a loss of 24yo in its total activity. Changeaux et al. (21) have shown that the acetylcholinesterase of the electric organ of Electrophorus electricus, when in the mernbrane, was insensitive to tyrocidinc, but that the enzyme in solution after extraction from nervous tissue was inhibited slightly by tyrocidinc.
We obtained similar results. The acetylcholinesterase of Aplysia nervous tissue was not affected (Fig. 4A), but the esterase of hemolymph was inhibited by tyrocidine to a maximum of 3Ooj, (Fig. 4B). In order to obtain this pattern, we found it necessary first to precipitate the purified material with 10% trichloroacetic acid and to treat the precipitate with organic solvents.
None of the sample which had only been heated in the detergent entered the gel. After treatment with trichloroacetic acid and organic solvents, we apparently did not achieve complete disruption of the complex, since some material still remained on top of the Nevertheless, we observed two prominent bands which migrated quite close together to positions relative to the marker proteins which indicated molecular weights of about 70,000. One of these bands corresponded to the only polypeptidc phosphorylated after treatment of the enzyme with [3H]diisopropylfluorophosphate, and presumably represents the catalytic subunit of the esterase (Fig. 5B). This labeling of the esterase was specific since it was reversed by treatment with pyridine-2-aldoxime mcthiodide (Sigma). Reversal was dependent on the concentration of the antagonist. After incubation of the labeled enzyme for 30 min at room temperature at 0.1 mM, 440/c, and at 1 mM, 807', of the radioactivity was removed from the enzyme.
Since the enzyme treated with [31<]diisopropyllluorophosphatc was not subjected to electrophoresis in the same gel as enzyme which was stained with Coomassie blue, we were not able to distinguish which of the two stained bands could be phosphorylated.
Unaccountable differences in preparation of sample or gels presumably might also explain why none of the radioactivity remained on top of the gel (Fig. 5B).

Electron
Jf icroscopy-Analysis of negatively stained preparations of the purified acetylcholincsterase revealed either single or aggregated ringlike structures measuring 60 A in diameter (Fig.  Sil). The lack of a more defined structure was surprising, since the methods of purification and subsequent biochemical characterization suggested that the csterase complex was uniform and should be considerably larger than 60 A. Similar results were obtained after negative staining with either ammonium molybdate or phosphotungstate.

No differences
were seen in preparations of the enzyme fixed in 0.2% glutaraldehyde during the first 8 hours of the purification procedure. Inclusion of 10 mM acetylcholinc chloride in the droplet during the 2-min CXposure to phosphotungstic acid just before examination on the microscope grid did not change the morphology of the enzyme. Addition of 50 PIV~ eserine sulfate, however, appeared to alter the structure slightly.
The unit particles increased to 80 A in diameter, and they tended to mass and form linear arrays (Fig. 6B)

Purification and Characterization of Other Particles in Hemolymph
Hemocyanin-The purified protein was free of all of the other components when examined by electron microscopy (Fig. 7A). Its appearance is quite similar to other molluscan hemocyanins.
The unit particle appears to be a disc 280 A in diameter and 45 A thick.
Within the disc are many spikelike structures surrounding an inner core This symmetrical structure, which is 230 A in diameter, is coated all around by a transparent layer 25 A thick.
This particle can be seen both directly and in profile. Discs stack to form long cylindrical arrays.
Formation of these huge complexes undoubtedly accounts for rapid sedimentation of the hemocyanin. The purified material contained 0.26% copper (dry weight), an amount expected for a molluscan hemocyanin (23). The presence of hemocyanin in the blood of the Californian species is remarkable, however. since analysis of hemolymph from the otherwise closely related Mediterranean species (A. depilans and A. lamacina) has failed to show any hemocyanin (24).
Hemagglutinin-Using differential centrifugation and Agarose gel filtration, we obtained partial purification of a hemagglutinin (Table IV).
Although purified g-fold, the material was contaminated with other protein (Fig. 1B) and electron microscopy revealed the presence of the other particles of hemolymph (Fig.  7B). The hemagglutinin of Aplysia was similar in size and structure to those of other invertebrates (25). The complex is 120 A in diameter containing a dense, circular core of 40 A in diameter surrounded by a transparent ring 40 A thick. This ring appears to be composed of six subunits.
The purified hemagglutinin was contaminated with a small amount of acetylcholinesterase (Fig. 1B). This enzymatic activity was apparently associated with particles smaller than those in the purified enzyme, since it was not entirely excluded during filtration.
Purijkation of Erythrocurorin-The 0.6 M layer from the initial density gradient was examined by electron microscopy and found to contain a particle which we have tentatively identified as erythrocurorin, as well as acetylcholinesterase, some hemagglutinin, and fragments of the hemocyanin. This particle was purified by centrifugation in a continuous sucrose density gradient in which the esterase was found in the center of the gradient, and the hemocyanin at the bottom of the gradient.
A band of protein toward the top contained flower-like particles, which appeared to be uncontaminated (Fig. 7C). The complex is 200 A in diameter, and consists of 10 petals. In some particles it can be seen clearly that each pair of petals is joined toward the center. Thus the particles consist of 5 V-shaped subunits. This configuration is similar to that of erythrocurorin isolated from hemolymph of other invertebrates (11).

DISCUSSION
More than 90% of the acetylcholinesterase activity present in the hemolymph of Aplysia was associated with a 69 S particle which appeared to be a distinct enzyme complex unassociated with membrane.
Two other particles, hemocyanin (23) and erythrocurorin (11) were also isolated, and these are known to have respiratory function in other invertebrates.
Since these particles are abundant in A. californica, it is surprising that Ghiretti et al. (24)  The acetylcholinesterase complex, when isolated, was unstabl and dissociated into smaller active particles.
Exposure to low pH decreased the sedimentability of the enzyme without destroying enzymatic activity.
Isoelectric focusing of a partially purified enzyme preparation yielded a distribution of activity with two isoelectric points, showing that there are at least two possible active forms of the enzyme. Although we have not studied the breakdown of the hemolymph acetylcholinesterase complex in detail, we have also found that it tended to dissociate into smaller units during storage for several weeks at 7". The decrease in sedimentability under these conditions was considerably greater than the total loss of enzymatic activity, and this is another indication that the complex is capable of dissociation into smaller active particles.
In addition to the complex, about 7% of the enzymatic activity exists normally in the hemolymph as smaller particles which were not sedimented at 105,000 x g. This activity may be a different enzyme, but we have not studied its properties further.
Acetylcholinesterases exist in two forms. One form is soluble and is similar to that found in the blood of many mollusks (26) and other invertebrates (3). The other form is associated with membranes, as is the enzyme in all vertebrate tissues ' (3, 4). Although similar in substrate specificity (a), the hemolymph acetylcholinesterase differs in several important respects from the esterase bound to membrane of Aplysia nervous tissue. Thus, treatment with detergents activated the membrane-bound enzyme, but was ineffective on the hemolymph enzyme. Tyrocidine, which inhibited the hemolymph enzyme, did not have any affect on the enzyme associated with membranes of nervous tissue. Also, digestion with trypsin rapidly inactivated the hemolymph enzyme, but released the nervous tissue enzyme in an active, soluble form with a molecular weight of about 250,000.1 Despite these dissimilarities, it is nevertheless possible that the various properties observed only reflect differences in the physical state of the enzyme in the two tissues and that the two enzymes are related.
The source of Aplysia hemolymph acetylcholinesterase is presently unknown, but it does not appear to be associated with cellular elements of the blood.
Perhaps it is secreted by nervous tissue into the hemolymph.
There is some evidence that acetylcholinesterase of electric tissue might exist in the membrane as a large complex, since Bawler (27) extracted an enzyme particle with a molecular weight of 13.4 x 106. Most of the esterase extracted with toluene from the electroplax membrane is an 11 S particle with a molecular weight of approximately 250,000 (6-8). This particle is built up of four subunits, each with an average molecular weight of about 64,000 (4,8,21,29). Toluene extraction possibly involves autolysis with some proteolytic degradation of the complex, however.
In addition to smaller species, a portion of the esterase extracted from Torpedo with ammonium sulfate was in the form of somewhat larger particles (17 to 18 S) (28). In Aplysia we found that the subunit labeled with [3H]diisopropylfluorophosphate has a molecular weight of about 70,000. This subunit appears to correspond to the 60 A particle which we have seen with the electron microscope.
Electron micrographs of the enzyme isolated from electroplax membrane showed 50 to 60 A single units frequently clustered in groups of four, and occasionally in groups of three (21, 29); in the larger complexes, presumably corresponding to the 17 to 18 8 particles, Rieger el al. (29) observed as many as 10 subunits clustered en grappe. Changeaux et al. (21) found that the electroplax enzyme appeared to be more ordered when treated with tyrocidine.
Electron micrographs of the Aplysia esterase were similar to those of the tyrocidinetreated electroplax enzyme, revealing groups of the small particles in ordered clusters.
These clusters were not uniform in size, however, and probably do not reflect accurately the large, complex structure of the enzyme as it most likely exists in solution in the hemolymph.
We do not know the reason why the material on examination by electron microscopy appeared to be smaller and less well ordered than might have been expected from the behavior of the enzyme during centrifugation and gel filtration. We have noted however that eserine appears to stabilize the complex to some degree. In the presence of this inhibitor, which interacts with the active site of the esterase, the enzyme complexes appeared in micrographs to be more ordered and extensive.
Its great abundance in molluscan hemolymph suggests that the acetylcholinesterase might have a function similar to that of soluble pseudocholinesterases and of the acetylcholinesterase of erythrocyte stroma found in vertebrate blood.
Loewi's experiment (30) 011 the frog heart showed that nerve stimulation can release acetylcholine into the blood. The cholinesterases found in both vertebrate and invertebrate blood might function to protect distant nervous and other responsive tissues from the transmitter substance which has escaped into the circulation.