Proc. Nati. Acad. Sci. USAVol. 80, pp. 6723-6727, November 1983Neurobiology

Identification of the probable site of synthesis ofbutyrylcholinesterase in the superior cervicaland ciliary ganglia of the cat

(cholinesterase/propionylcholinesterase/rat/sympathetic nervous system)

Eiji UCHIDA AND GEORGE B. KOELLEDepartment of Pharmacology, Medical School/G3, University of Pennsylvania, Philadelphia, PA 19104

Contributed by George B. Koelle, August 2, 1983

ABSTRACT The source of butyryicholinesterase (acylcholineacylhydrolase, EC 3.1.1.8) in the ganglion cells of the cat superiorcervical and ciliary ganglia has been elusive, inasmuch as the en-zyme is present in high concentrations in the neuropil, where itis confined largely to the dendritic and perikaryonal plasma mem-branes, but appears to be absent from the perikarya. In the pres-ent study, ganglionic butyrylcholinesterase was near-totally inac-tivated by the injection of tetramonoisopropyl pyrophosphoramide(6.0 jmol/kg of body weight) intravenously. During the ensuing72 hr, the regenerating enzyme became detectable by the copperthiocholine histochemical method in the somata of essentially allganglion cells and in the neuropil. Results were similar in pre-ganglionically denervated superior cervical ganglia and in normalciliary ganglia. These findings suggest (i) that butyrylcholinester-ase indeed is synthesized in the ganglion cell perikarya (presum-ably, the rough endoplasmic reticulum) and transported extremelyrapidly to more peripheral cellular sites and (ii) that the synthesisis largely independent of control by any neurotrophic factor pro-vided by the preganglionic axonal terminals. Similar studies wereconducted in the rat. In this species, in contrast to the cat, thesomata of essentially all ganglion cells of the superior cervicalganglion contain various but at least moderate concentrations ofacetylcholinesterase (acetylcholine acetylhydrolase, EC 3.1.1.7)and propionylcholinesterase (acylcholine acylhydrolase, EC 3.1.1.8).After injection of tetramonoisopropyl pyrophosphoramide, pro-pionylcholinesterase reappeared in the ganglion cell somata be-fore its accumulation in the neuropil, as would be expected.

There is a marked species difference between the cat and ratwith regard to the distributions of acetylcholinesterase (Ac-ChoEase; acetylcholine acetylhydrolase, EC 3.1.1.7) and non-specific cholinesterase (acylcholine acylhydrolase, EC 3.1.1.8)in the superior cervical ganglion (SCG) and other autonomicganglia.

In the cat, light microscopic examination by the highly spe-cific copper thiocholine (CuSCho) method has shown thatAcChoEase is present in the perikarya of all ganglion cells ofthe SCG (1); its concentration there is high in a small proportion(<1%; refs. 1 and 2) and moderate or low in the remainder; ahigh concentration of AcChoEase is also present in the neu-ropil. The latter is comprised of preganglionic axons and theirterminals, ganglion cell dendrites, and Schwann sheath cells;these elements are indistinguishable by light microscopy. How-ever, more recent electron microscopic (EM) examination bythe bisthioacetoxy aurate method (3) has clarified the cytolog-ical location of AcChoEase in the neuropil (4, 5). The nonspe-cific cholinesterase (or butyrylcholinesterase; BtChoEase) ofthe cat SCG appears by light microscopy to be confined to the

neuropil and absent from the ganglion cell perikarya (6). It wasconcluded on this basis and from other studies (7, 8) that it islocated in the Schwann sheath cells. However, it was found byEM that most of the BtChoEase is located at the dendritic andperikaryonal membranes of the ganglion cells (4, 5). In view ofthe apparent absence of BtChoEase from the ganglion cell peri-karyonal cytoplasm, which includes the rough endoplasmic re-ticulum (RER), this has raised the question of its site of origin.In studies addressed to this point, it was shown that theBtChoEase of the cat SCG is not derived post-transcriptionallyfrom AcChoEase (9) and is not taken up from the plasma (un-published data).

In this investigation we explored the possibility that Bt-ChoEase is present in the perikaryonal cytoplasm of ganglioncells of the cat SCG and ciliary ganglion (CG) in concentrationsthat are subdetectable by the current CuSCho histochemicalprocedure (10). This has been done by (i) extending the incu-bation time from 2 to 8 hr and (ii) examining the sites of re-generation of BtChoEase from 24 to 72 hr after its inactivationby tetramonoisopropyl pyrophosphoramide (iso-OMPA). Re-sults suggest that BtChoEase is synthesized by the RER of theganglion cells and transported extremely rapidly to their den-dritic and perikaryonal membranes.

In contrast to the cat, the rat and most other species showextremely variable but generally moderate concentrations ofboth AcChoEase and nonspecific cholinesterase [in the rat,propionylcholinesterase (PropChoEase); see ref. 111 in theperikarya of essentially all sympathetic ganglion cells and in theneuropil (12, 13). This suggests that in the rat both enzymes aresynthesized in the RER and are then transported to other sitesin accordance with current concepts (14). We also examined thelocalization of PropChoEase in the SCG of the rat by the sameprocedures as mentioned above for cat ganglia in order to con-firm results obtained with the latter species.

METHODSCats and rats were anesthetized with sodium pentobarbital (50mg/kg of body weight) intraperitoneally. The SCG and CG wereexcised from cats; the SCG was excised from rats. Ganglia werefrozen immediately, and fresh frozen sections were cut with aMinitome-Microtome cryostat (International Equipment) at 10and 20 ,um, placed on slides, and stored in the refrigeratorovernight; in most cases, the 10-gm sections proved to be moresatisfactory. The following day they were stained by the stan-

Abbreviations: AcChoEase, acetylcholinesterase; BtChoEase, butyr-ylcholinesterase; CG, ciliary ganglion; CuSCho, copper thiocholine; EM,electron microscopy; iso-OMPA, tetramonoisopropyl pyrophosphor-amide; PropChoEase, propionylcholinesterase; RER, rough endo-plasmic reticulum; SCG, superior cervical ganglion.

6723

The publication costs of this article were defrayed in part by page chargepayment. This article must therefore be hereby marked "advertise-ment" in accordance with 18 U.S.C. §1734 solely to indicate this fact.

Dow

nloa

ded

by g

uest

on

Dec

embe

r 2,

202

0

6724 Neurobiology: Uchida and Koelle

dard CuSCho method (10) for AcChoEase and BtChoEase (cat)or PropChoEase (rat). Included in the acetylthiocholine me-dium for staining AcChoEase was 10-(a-dimethylaminopro-pionyl)phenothiazine'HCl (designated Astra 1397) at 0.1 mMfor the selective inhibition of BtChoEase or PropChoEase; tothe medium containing butyrylthiocholine for the localizationof BtChoEase or PropChoEase was added 1,5-bis-(4-allyldi-methylammoniumphenyl)pentan-3-one dibromide (designatedB.W. 284) at 3 ,uM to prevent its slow hydrolysis by Ac-ChoEase. Controls with either substrate contained both inhib-itors in the same concentrations. Incubation periods ranged from1 to 8 hr at 370C.

In order to determine the initial sites of regeneration ofBtChoEase after its essentially complete inactivation, a seriesof cats was anesthetized with sodium pentobarbital (35 mg/kg)intraperitoneally, and a 1-cm segment was resected from theright cervical sympathetic trunk. Approximately 1 wk later, theywere reanesthetized, the saphenous vein was exposed, and theywere injected intravenously with iso-OMPA (Ayerst Labora-tories, Rouses Point, NY; 6.0 ,.&mol/kg). At intervals of 24, 48,and 72 hr later, they were again anesthetized, and the SCG andCG were taken as above. The regeneration of PropChoEase inthe rat was studied similarly by injecting iso-OMPA (20.0 /Imol/kg) intraperitoneally 1, 24, 48, and 72 hr prior to removal of theSCG.

RESULTSWhen 10-nim sections of cat SCG were stained for AcChoEaseby incubation for 1 hr, there was marked staining of the neu-ropil. The somata of occasional ganglion cells showed heavystaining, and faint, barely discernible staining could be de-tected in the somata of most of the remainder (Fig. 1A). Withincreasing incubation periods (2, 4, and 8 hr), staining of theganglion cell somata became progressively heavier (Fig. 1 B-D) and, at the longest period, was distinct in all. The nuclei were

faintly stained or unstained. At the same time, the neuropil be-came grossly overstained and exhibited heavy crystalline de-posits. At each of these periods, staining for BtChoEase in theneuropil was likewise marked; however, the ganglion cell so-mata remained essentially unstained (Fig. 1 E-H). Occasionalganglion cell somata appeared at first glance to show stainingfor BtChoEase, especially at the longer incubation periods. Closerinspection suggested that these were cells that were cut nearthe surface, so that the staining was actually associated with theperikaryonal and dendritic membranes, as found by EM (4).

In the cat CG, where all ganglion cells give rise to cholin-ergic postganglionic fibers, the somata of all were markedlystained for AcChoEase at 1 hr (Fig. 2A), and the intensity in-creased progressively at 2, 4 (Fig. 2B), and 8 hr. Here the neu-ropil is represented chiefly by the heavily stained dendritic halosthat surround the individual ganglion cells. Again, staining forBtChoEase was essentially absent from the somata but intensein the neuropil (Fig. 2 C and D).

Control sections of cat SCG and CG, for which both inhib-itors were included with either substrate, remained essentiallyblank after 4- and 8-hr incubation (not shown).

In confirmation of previous findings (12, 13), the patterns ofAcChoEase and PropChoEase staining in the SCG of the ratwere quite different. After a 1-hr incubation, markedly varyingconcentrations of both enzymes were detectable in the somataof essentially all ganglion cells and in the neuropil (Fig. 3 A andC); staining became more intense after 2 (Fig. 3 B and D), 4,and 8 hr (not shown). Again, controls incubated in either sub-strate with both inhibitors remained blank.The intravenous injection of iso-OMPA at 6.0 gmol/kg in

the cat produced essentially complete, selective inactivation ofthe BtChoEase of the SCG and CG; 24 hr later, the BtChoEaseactivity of both ganglia was still less than 1% of controls, andat 72 hr returned to approximately 25% of the control value inthe SCG and slightly less in the CG (15). Measurement of the

FIG. 1. Cat SCG (10-,m sections). (A-D) AcChoEase. (E-F) BtChoEase. Incubation times at 370C: 1 hr (A and E), 2 hr (B and F), 4 hr (C andG), and 8 hr (D and H). (CuSCho; x250.)

Proc. Natl. Acad. Sci. USA 80 (1983)

Dow

nloa

ded

by g

uest

on

Dec

embe

r 2,

202

0

Proc. Natl. Acad. Sci. USA 80 (1983) 6725

B

.....C

FIG. 2. Cat CG. (A and B) AcChoEase. (C-E) BtChoEase. Incu-bation times at 370C: 1 hr (A and C), 4 hr (B and D), 8 hr (E), all 72 hrafter intravenous injection of iso-OMPA (6.0 ,umol/kg). Sections were10 ,an (A-C and E) or 20 pam (D). (CuSCho; x 250.)

rate of regeneration after treatment with sarin (isopropylmeth-ylphosphonofluoridate) has shown that the value for BtCho-Ease in denervated SCG is approximately 60% of the value fornormal SCG at all intervals up to 3 wk (9). In the present study,BtChoEase was undetectable in the normal and denervated SCG24 hr after treatment with iso-OMPA, with 8 hr incubation (Fig.4 A and D). At 48 hr after iso-OMPA injection, trace amountsof BtChoEase were detectable in the somata of most ganglioncells and in the neuropil of both (Fig. 4 and E). By 72 hr,there was definite staining for BtChoEase activity in the somataof practically all neurons of the normal and denervated SCG,and considerable staining occurred in the neuropil (Fig. 4 Cand F). Staining within the ganglion cell somata and in theirsurrounding dendritic halos was somewhat heavier in the CG;this is best illustrated in a 20-,um section incubated for 8 hr (Fig.2E). Controls incubated with Astra 1397 remained unstained.

In the rat, the values for PropChoEase in the SCG after theintraperitoneal injection of 20 tmol of iso-OMPA per kg at 2,24, 48, and 72 hr are 1, 18, 49, and 53% of control values, re-spectively (16). Here, no staining was found 1 hr after iso-OMPAinjection (Fig. 5A). Slight staining was detectable in most gan-glion cell somata at 24 hr (Fig. 5B). At 48 hr (Fig. 5C) and 72

FIG. 3. Rat SCG (10-/im sections). (A andB) AcChoEase. (CandD)PropChoEase. Incubation times at 370: 1 hr (A and C), 2 hr (B andD).(CuSCho; x250.)

hr (Fig. 5D) after iso-OMPA injection, staining increased pro-gressively in the somata and appeared in the neuropil; at thelatter period, there was an extremely high proportion of heavilystained ganglion cells. The relative intensity of staining of gan-glion cell somata to that of the neuropil at both intervals wasconsiderably greater than in the controls (Fig. 3 C and D).

DISCUSSIONThe apparent absence of BtChoEase from the somata of theganglion cells of the cat SCG has been a matter of puzzlement,paricularly after the EM demonstration that its presence in theneuropil is restricted almost exclusively to the ganglion celldendritic and perikaryonal membranes (4). AcChoEase is lo-cated at the same sites as well as at the plasma membranes ofthe preganglionic axons and their terminals (4). The localizationof AcChoEase at the ganglion cell somata (1) has been assumedto account for its source at more peripheral sites in the samecells.

In the present study, the foregoing findings were confirmedand extended. Then it was shown that after near-total inacti-vation of the BtChoEase of the SCG and CG by iso-OMPA, theenzyme became detectable in the ganglion cell somata prac-tically simultaneously with its reappearance in low concentra-tions in the neuropil. The sum of these findings suggests thatBtChoEase is indeed synthesized in the ganglion cell somata(presumably, the RER) and transported extremely rapidly tothe perikaryonal and dendritic membranes. Under normal con-ditions, it is likely that the velocity of this sequence is too rapidto allow the accumulation of histochemically detectable con-centrations of BtChoEase in the somata.The effects of preganglionic denervation of the cat SCG de-

serve mention. Within 72 hr after this procedure, the Ac-ChoEase disappears completely from the neuropil but not from

Neurobiology: Uchida and Koelle

Dow

nloa

ded

by g

uest

on

Dec

embe

r 2,

202

0

6726 Neurobiology: Uchida and Koelle

FIG. 4. Cat SCG. (A-F) BtChoEase. Incubation time at 37TC: 8 hr. (A-C) Normal ganglia. (D-F) Ganglia denervated preganglionically ap-proximately 1 wk prior to removal. Iso-OMPA at 6.0 pmol/kg had been intravenously administered 24 hr (A and D), 48 hr (B and E), and 72 hr(C and F) previously. (CuSCho; x 250.)

4.

..

Ia

I

FIG. 5. Rat SCG. (A-D) PropChoEase. Incubation time at 37TC: 4hr. Iso-OMPA at 20.0 pmol/kg had been intraperitoneally adminis-tered 1 hr (A), 24 hr (B), 48 hr (C), and 72 hr-(D)-previously. (CuSCho;x250.)

the ganglion cell somata, and BtChoEase is lost partially fromthe former site; these changes persist for several weeks (10, 17).Recent findings suggest that these effects are due to the loss ofa neurotrophic factor provided normally by the preganglionicterminals (18). In the present study, there appeared to be nodifference between the appearance of BtChoEase in the so-mata of normal or preganglionically denervated cat SCG neu-rons after iso-OMPA injection. This indicates that the perikary-onal synthesis of BtChoEase, like AcChoEase, is at least partiallyindependent of neurotrophic control.

In the rat SCG, the situation is much less complicated. Here,PropChoEase, like AcChoEase, is present in varying but sig-nificant concentrations in the somata of essentially all ganglioncells. After its inactivation by iso-OMPA, PropChoEase reap-peared first in the ganglion cell somata and subsequently in theneuropil, as would be expected.The bisthioacetoxy aurate method, in contrast to the Cu-

SCho procedure, permits high resolution of the localization ofAcChoEase and BtChoEase, by EM but is relatively nonspe-cific. Accordingly, the positive identification of either enzymeat any site is dependent upon the absence of significant con-centrations of other esterases that are resistant to inhibition byphysostigmine (3). Current results indicate that such enzymesare present in the RER of cat sympathetic ganglion cells, incontrast to their relative absence at the perikaryonal and den-dritic membranes (4); hence, only at the latter sites can Ac-ChoEase and BtChoEase be identified with certainty.

We thank Dr. Gerard A. Ruch, Miss Ursula J. Sanville, and Mrs.Kathleen Kitto Rickard for their assistance in conducting this study. Thisinvestigation -was supported by Research Grant NS-00282-31 from theNational Institute of Neurological and Communicative Disorders andStroke and by a contribution from the Foundation for Vascular-Hy-pertension Research (Philadelphia).

Proc. Natl. Acad. Sci. USA 80 (1983)

Dow

nloa

ded

by g

uest

on

Dec

embe

r 2,

202

0

Neurobiology: Uchida and Koelle

1. Koelle, G. B. (1955)J. Pharmacol. Exp. Ther. 114, 167-184.2. Holmstedt, B. & Sjoqvist, F. (1959) Acta Physiol. Scand. 47, 284-

296.3. Koelle, G. B., Davis, R., Smyrl, E. G. & Fine, A. V. (1974)J. His-

tochem. Cytochem.22, 252-259.4. Davis, R. & Koelle, G. B. (1978)J. Cell Biol. 78, 785-809.5. Davis, R. & Koelle, G. B. (1981)J. Cell Biol. 88, 581-590.6. Koelle, G. B. (1951)J. Pharmacol. Exp. Ther. 103, 153-171.7. Bulbring, E., Philpot, F. J. & Bosanquet, F. D. (1963) Lancet i,

865-866.8. Cavanagh, J. B., Thompson, R. H. & Webster, G. R. (1954) Q. J.

Exp. Physiol. 39, 185-197.9. Koelle, G. B., Ruch, G. A., Rickard, K. K. & Sanville, U. J. (1982)

J. Neurochem. 38, 1695-1698.10. Koelle, G. B.,- Davis, R. & Koelle, W. A. (1974)J. Histochem. Cy-

tochem. 22, 244-251.

Proc. Natl. Acad. Sci. USA 80 (1983) 6727

11. Myers, D. K. (1953) Biochem. J. 55, 67-79.12. Koelle, G. B. (1963) in Cholinesterases and Anticholinesterase

Agents, ed. Koelle, G. B. (Springer, New York), pp. 187-298.13. Eranko, L. (1972) Histochem. J. 4, 545-559.14. Droz, B., Koenig, H. L. & DiGiamberardino, L. (1973) Brain Res.

60, 93-127.15. Koelle, G. B., Koelle, W A. & Smyrl, E. G. (1977)J. Neurochem.

28, 313-319.16. Koelle, G. B., Rickard, K. K. & Smyrl, E. G. (1979) 1. Neuro-

chem. 33, 1159-1164.17. Sawyer, C. H. & Hollinshead, W. H. (1945)1. Neurophysiol. 8,

137-153.18. Koelle, G. B. & Ruch, G. A. (1983) Proc. Natl. Acad. Sci. USA 80,

3106-3110.

Dow

nloa

ded

by g

uest

on

Dec

embe

r 2,

202

0