J. Cell Sci. 19, 621-644 (1975) 621

Printed in Great Britain

CELL SURFACE SACCHARIDES OF

TRYPANOSOMA LEWISI. I. POLYCATION-

INDUCED CELL AGGLUTINATION AND FINE-

STRUCTURE CYTOCHEMISTRY

D. M. DWYER

The Rockefeller University, New York, New York 10021, U.S.A.

SUMMARY

Trypanosoma lewisi bloodstream and culture forms were agglutinated differentially with lowconcentrations of the cationic compounds: ruthenium red, ruthenium violet, Alcian bluechloride, i-hexadecylpyridinium chloride, lanthanum chloride, and cationized ferritin. Thebloodstream form trypanosomes gave the highest agglutination levels with each of the com-pounds tested. Ruthenium red was the most effective inducer of cell agglutination among theseveral cations used. Trypsin-treated bloodstream forms were agglutinated less in the presenceof ruthenium red than untreated controls. Ruthenium red-induced cell agglutination also waslowered with chondroitin sulphate and dextran sulphate, but not with a-D-glucose, a-D-mannoseor with several methyl glycosides. Treatment of the bloodstream trypanosomes with a-amylase,dextranase, or neuraminidase had little effect on agglutination levels obtained with rutheniumred.

Fine-structure cytochemical staining with ruthenium red, ruthenium violet, and Alcianblue-lanthanum nitrate was used to ascertain the presence and distribution of presumptivecarbohydrates in the trypanosome cell surface. The extracellular surface coat of the blood-stream forms stained densely with each of the polycationic dyes. Trypsin treatment removedthe surface coat from bloodstream trypanosomes; however, the surface membranes of theorganisms were stained densely with the several dyes. Similar surface-membrane staining wasobtained with the cationic compounds and the culture forms, which lack a cell surface coat.Cationized ferritin was used at the fine-structure level to visualize the negative surface chargepresent in the cell surface coat and external membrane of the several trypanosome stages.

Results obtained from the agglutination and cytochemistry experiments indicate that com-plex polysaccharides are present in the surface membranes and cell surface coat of T. letvisibloodstream forms. Similar conclusions also pertain to the surface membranes of the T. lewisiculture form trypanosomes. The carbohydrates probably represent glycopeptide and glyco-protein structural components of the surface membrane of this organism.

INTRODUCTION

Trypanosoma lewisi is an extracellular, protozoan blood parasite of rats. Cyclicaldevelopment and transmission of the organism occur in the alimentary tract of the ratflea (Molyneux, 1969). Bloodstream trypanosomes placed into suitable media in vitrotransform and multiply as culture stages morphologically identical with some formsobserved in the vector host (Hoare, 1972). Physiological and structural differencesexist between mammalian bloodstream and insect vector or culture form trypanosomes(Vickerman, 1971). One striking difference is that bloodstream trypanosomes possessa morphologically distinct extracellular surface coat which culture forms lack (Vicker-

622 D. M. Dtcyer

man, 1969). Mammalian trypanosomes are divided in salivarian and stercorariangroups with regard to the sites of development and transmission within the insectvector (Hoare, 1964). Bloodstream salivarian trypanosomes have a uniformly densecompact cell surface coat, whereas a more diffuse fibrous surface coat is character-istic of the Stercoraria, of which T. lewisi is a member (Vickerman, 1971).

Abundant cytochemical data indicate that polysaccharides are present in the surfacemembranes of a vast array of eukaryotic cell types (Rambourg, 1971; Luft, 1971 a, b;Cook & Stoddart, 1973). Cumulative biochemical evidence has substantiated the factthat carbohydrates indeed are present in the surface membranes of various cells(Spiro, 1970; Hughes, 1973; Sharon, 1974). Further, the amassed biochemical dataindicate that carbohydrates are structural and functional constituents of cell surfacesin the form of glycopeptides, glycoproteins, glycolipids, and glycosaminoglycans(Cook& Stoddart, 1973; Oseroff, Robbins& Burger, 1973; Singer, 1974a, b). Indirect(Allsopp & Njogu, 1974; Allsopp, Njogu & Humphreys, 1971; Njogu & Humphreys,1972; Williamson & Desowitz, 1961) and direct (Cross, 1972, 1973; Njogu, 1974;Njogu, Itazi, Enyaru & Abonga, 1974) immunological, immunochemical, and bio-chemical evidence suggest that protein-carbohydrate complexes also exist at thesurfaces of salivarian trypanosomes. Cytochemically, carbohydrates also weredemonstrated at the junctional complex formed between the pellicular membrane andsurface coat of bloodstream forms of Trypanosoma brucei, a salivarian trypanosome(Wright & Hales, 1970; Steiger, 1973; Steiger & Jenni, 1974). To date, however, noreports exist concerning the chemical nature or composition of the surface coat-pellicular membrane complex of any stercorarian trypanosome.

Net surface charge of a cell is a function of the molecular constitution of the cellsurface membrane (Curtis, 1971, 1973)- The mechanism of agglutination of cells withvarious polycations has been postulated as a function of intercellular molecular cross-bridge formation (Curtis, 1973). Recently, cell-induced agglutination with severalpolycations, including ruthenium red, was found to be indicative of cell surfaceglycosaminoglycans (Utsumi & Oda, 1973). Moreover, the cation cetylpyridiniumchloride has been demonstrated as an agent for the precipitation of isolated cellsurface glycoproteins (Terry & Culp, 1974)-

A polycationic derivative of ferritin recently was described as a label for thevisualization of negative charges on cell surfaces at the fine-structure level (Danon,Goldstein, Marikovsky & Skutelsky, 1972). Results of cell electrophoresis and ion-exchange column chromatography studies have indicated that T. lewisi has a high netnegative cell surface charge (Hollingshead, Pethica & Ryley, 1963; Lanham, 1968).

In the light of these foregoing observations, it was of interest to use various poly-cations in agglutination experiments with T. lewisi. Several of the polycations alsowere used as cytochemical stains in order to ascertain the presence and distribu-tion of polysaccharides in the cell surface of bloodstream and culture forms of thisstercorarian trypanosome.

T. lewisi surface saccharides I 623

MATERIALS AND METHODS

Maintenance of trypanosomes

Two strains of Trypanosoma lewisi were used in this study. One, the Taliaferro strain(D'Alesandro, 1972) had been maintained by routine weekly serial in vivo transfers in rats formore than 25 years, and had lost the ability to transform morphologically into culture forms andmultiply as such in vitro. The second, a strain isolated in Taiwan recently was maintained inrats also, and retained the ability to transform and multiply in vitro in a suitable culture medium.(I thank Professor Donald Dusanic, Department of Life Sciences, Indiana State University forproviding this strain of T. lewisi.) The 2 strains were passaged by weekly subinoculations madeintraperitoneally into female albino Sprague-Dawley strain rats (140-160 g, Sprague-DawleyCo., Madison, Wis.) as described by D'Alesandro (1972). Cultures of the Taiwan strain alsowere maintained in vitro at 26 ± 0-5 °C, by serial transfers every 14 days using a diphasic blood-containing medium (Tobie, von Brand & Mehlman, 1950).

Isolation of trypanosomes

Rats infected for 4, 7, and 14 days were bled by cardiac puncture into a small volume(1 part: 10 parts of blood) of cold dextrose-citrate (27 mM D-glucose, 102 mid sodium citratedihydrate) solution (DCS) and this was mixed immediately with an equal volume of the samesolution on ice. The trypanosomes were separated from the blood, using differential and dis-continuous gradient centrifugation methods, as outlined in Fig. 1.

Parasites grown in vitro were chilled on ice, and then centrifuged at 2100 g for 15 min at4 °C. Subsequently the cells were washed 4 times in cold phosphate-buffered saline glucose(129 mM NaH2PO4.H,O, 5-4 mM Na2HPO4, i37mMNaCl, 9-9 mM D-glucose, adjusted at20 °C to pH 745) solution (PBSG) by centrifugation at 2100 g for 15 min at 4 °C.

Enzyme treatment

Isolated and washed bloodstream trypanosomes were treated with trypsin to remove the cellsurface coat. Both crude and crystalline trypsin preparations were employed. Crude trypsin(Difco 1:250, Difco Laboratories, Detroit, Mich.) was dissolved to final concentration of0-25 % (w/v) in PBSG. The solution was filtered through a o-45-/«n Millipore Swinex filteringapparatus (Millipore Corp., Bedford, Mass.) and heated in a waterbath to 37 °C just prior touse. Trypanosome pellets on ice were resuspended in the warm trypsin solutions to a final con-centration of 10s cells/ml, and incubated at 37 °C for 15-20 min with periodic mild agitation.Similarly, crystalline trypsin (Type III, 2x cryst., bovine pancreas, Sigma Chemical Co., StLouis, Mo.) was prepared to a final concentration of 10 mg/ml (11 500 BAEE units of activity/mg) in PBSG, filtered, and warmed to 37 CC. This was used as above; however, the trypanosomeincubation period was reduced to 10 min. Immediately after trypsin treatment, the cell sus-pensions were diluted to 10 times their volume with cold PBSG on ice and centrifuged at2100 g for 15 min at o CC. The pelleted trypanosomes subsequently were washed 4 times in100 times their volume in cold PBSG by centrifugation as above. Control samples were treatedidentically; however, trypsin was omitted during the 37 °C incubation.

Bloodstream forms of T. lewisi also were treated with several glycoside hydrolase enzymes:a-amylase (2x cryst., porcine pancreas, 650 units/mg); dextranase {Penicillium sp., 150 units/mg); and neuraminidase (Clostridiumperfringens, 0-5 units/mg) were purchased from Worthing-ton Biochemical Corp., Freehold, N.J. Solutions of the several enzymes were made to a finalconcentration of 2-0 mg/ml in PBS, filtered, and warmed as described previously. Cells wereresuspended in the various enzyme solutions at a final concentration of io'/ml and incubatedfor 10—20 min at 37 °C. Controls were treated identically except for the omission of the variousenzymes during the incubation procedure. Subsequently, the cell suspensions were chilled,diluted, and washed 4 times in PBSG, as described above.

Agglutination tests

Various polycationic compounds were tested in agglutination reactions with T. lewisi blood-stream and culture forms. Purified ruthenium red and ruthenium violet (see electron micro-

624 D. M. Dwyerscopy section for details) were dissolved in glass-double-distilled HaO, and diluted to thedesired concentrations with PBS, pH 7-45- Solutions containing LaClj (Fisher Chemicals,Pittsburg, Pa.), i-hexadecylpyridinium chloride (Eastman Chemicals, Rochester, N.Y.),Alcian blue chloride (NA 0720, Allied Chemicals, Morristown, N.J.), and cationized ferritin(horse spleen, Miles Laboratories, Kankakee, 111.) were made in PBS.

Previously washed cell suspensions were washed twice in cold PBS and adjusted with PBSto 2 x io8/ml, and this concentration was used in all experiments.

Trypanosome-Blood—DCS Suspension*

P+ PBSG

P+ PBSG

,4200g/10"

. 4200g/1G"

900g/15'

\ /"" 'Duffy coat'

+ DCS

90Og/15'

\ /

4200g/10'

\ /P

'Buffy coat'+ PBSG

ILayer over 20 % Ficoll'

Interface+ PBSG

(pool)+ PBSG

4200g/1G"

Fig. 1. Procedures used for isolating trypanosomes from whole rat blood. Abbrevia-tions: DCS, dextrose citrate solution; PBSG, phosphate-buffered saline glucosesolution; S, supernatant; P, pellet.

• All solutions and cell suspensions were maintained on ice. The centrifugation procedureswere at 0—4 °C.

f Solution containing 20 % (w/v) Ficoll (Pharmacia Fine Chemicals, Piscataway, N J . ) inphosphate-buffered saline (PBS): 8-45 mM Na,HPO4, 1-59 mil NaHsPO4.H2O, 145 HIM NaCl,pH 745 (at 20 °C).

% Final trypanosome pellet was washed 4 times in PBSG by centrifugation at 2100 g for15 min.

Rapid slide agglutination tests were employed and the controls were set in parallel on thesame slide as described previously (Dwyer, 1974a). Various compounds were tested forinhibition of agglutination in the presence of the several polycations. These were substituted inPBS before the addition of the polycation solution.

Similarly, tube agglutination reactions were performed. Equal volumes of the cell suspensionand the various cations were mixed and incubated at 37 °C for 5 min. PBS controls and severalinhibitors also were used in parallel. All reactions were examined macroscopically, and micro-scopically with a phase-contrast microscope. Reactions were scored qualitatively from o (noagglutination) to 4 + (moderately heavy) as illustrated in Table 1 (p. 626). Enzymically treatedbloodstream forms also were used in the agglutination experiments and were prepared asdescribed previously.

T. lewisi surface saccharides I 625

Electron microscopy

Ruthenium red and violet were extracted and purified from crude ruthenium red (K & KLaboratories, Plainview, N.Y.) using an aqueous NH3—NH4C1 extraction method (Luft,1971a). Spectrophotometric absorption maxima were obtained at 533 nm and 734 nm withdilute ammonium acetate solutions of purified ruthenium red and violet respectively. Thesedata were used to calculate the concentration in parts per million (ppm) of the 2 aqueousruthenium stock solutions (Luft, 1971 a). T. letvisi bloodstream and culture forms were fixed insuspension at 21 °C for 1 h in a solution of 2 % (v/v) glutaraldehyde (Polysciences, Warrington,Pa.) in 0 1 M Na cacodylate-HCl with 500 ppm of either ruthenium red or violet, P H 7 4 5 .Subsequently, the cells were washed twice in cold o-i M Na cacodylate-HCl, pH 745 (at4 °C), for 15 min by centrifugation at 2100 g. The resulting pellets were resuspended to 100times their volume in the same buffer and incubated overnight at 4 °C. Following this, the cellswere postfixed in suspension at 21 °C for 1 h in 2 % (w/v) OsO4 (Polysciences) in 0 1 M Nacacodylate-HCl with 500 ppm of ruthenium red or violet, pH 74 . Controls were processedsimilarly; however, the ruthenium compounds were omitted from all solutions. After this, therrypanosomes were washed twice in 0 1 M Na cacodylate-HCl buffer, pH 7-45 (at 4 °C), bycentrifugation at 2100 g for 10 min at 4 °C. The washed pellets were dehydrated through agraded ethanol series terminating in propylene oxide. Samples were infiltrated and embeddedin Epon 812 (Luft, 1961). Sections were cut with a diamond knife using a Porter-Blum ultra-microtome (Model MT-2 Ivan Sorvall, Norwalk, Conn.), mounted on uncoated 300-meshcopper grids and examined at 60 or 80 kV with a Philips 300 electron microscope. All sectionsstudied were untreated with uranyl acetate or lead.

The several stages of T. lewisi also were fixed using methods modified from Shea (1971).Briefly, cells were fixed in suspension for 1 h at 21 °C in a solution containing 3 % (v/v) glutar-aldehyde in 0 1 M Na cacodylate-HCl buffer with 0 5 % (w/v) Alcian blue chloride (AlliedChemicals), pH 7'45- Alcian blue was omitted from control samples. Then the trypanosomeswere washed twice in cold 0 1 M Na cacodylate-HCl buffer, pH 745 (at 4 °C), by centrifugationat 2100 g for 15 min at 4 °C. The washed pellets were postfixed for 1 h at 21 °C in a solutioncontaining 1 % (w/v) OsO4, 1 % (w/v) La(NO3)3.6HaO (Fisher Chemicals) in 0 1 M s-collidine(2,4,6-trimethyl pyridine, Eastman Chemicals)-HCl, pH 7-45. Lanthanum nitrate hexahydratewas omitted from control samples. Subsequently, samples were washed twice in cold 0 1 Mj-collidine-HCl buffer, pH 745 (at 4 °C), by centrifugation at 2100 g for 10 min at 4 °C. Cellswere dehydrated, embedded, sectioned, and observed as described previously.

A method for the visualization of negative surface charge (Danon et al. 1972) was used withthe several T. lewisi forms. Briefly, cells were fixed in suspension at 21 °C for 30 min in 2 % (v/v)glutaraldehyde in PBS, pH 745. Following this, the trypanosomes were washed 3 times at21 °C in PBS by centrifugation at 2100 g for 15 min. After the final wash, a 5 % (v/v, packedtrypano8omes:PBS) suspension of cells was made. Per ml of cell suspension, 0 5 ml of asolution containing 2 4 mg of cationized horse spleen ferritin (Miles) in PBS was added withagitation. This suspension was incubated at 21 °C for 30 min with agitation periodically. Acontrol sample for each experiment was treated as described except that cationized ferritin wasomitted during the 30-min incubation. Other control samples were incubated with 0 8 mg ofhorse spleen ferritin (6x crystallized, Miles Labs.) for 30 min at 21 °C. The concentration ofhorse spleen ferritin used was equivalent in iron content with that of the cationized ferritin.After the 30-min incubation the cells were washed 3 times in PBS at 21 °C by centrifugation at2100 g for 15 min. Samples were postfixed for 1 h at 21 °C in 1 % (w/v) OsO4 in PBS, pH 7-45.Cells were washed twice in PBS, and further processed for electron microscopy as describedpreviously.

RESULTS

Cell viability following enzyme treatment

Trypanosome viability was assessed following each of the several enzyme treat-ments. In each instance, the cells remained > 9 9 % viable (i.e. motile), as assayed by

626 D. M. Dwyer

direct haemocytometer counts. Similar results were obtained with the dye-exclusiontest (i.e. o-oi % trypan blue in PBS). Enzyme-treated Taiwan strain bloodstreamforms also retained the ability to transform into epimastigotes (i.e. culture forms) andmultiply in vitro. Further, the enzyme-treated organisms remained infective for non-immune rat hosts.

Cation-induced cell agglutination

The data given in this section represent the cumulative observations of 4 slide and4 tube agglutination tests performed with each compound and each T. lewisi form in 2separate experiments. A summary of the agglutination results obtained with theseveral T. lewisi forms and the various cationic compounds is given in Table 1.

No spontaneous aggregation of the several parasite forms occurred in the absence ofthe cations during the course of the agglutination experiments. Uniformly negativeresults were obtained with control preparations in each test.

Table 1. Agglutination of trypanosomes with various cationic compounds

Polycation

Ruthenium redRuthenium violetAlcian blue chloride1 -Hexadecylpyridiniumchloride

Lanthanum chlorideCationized ferritin

Finalconcentration*

40 /tM

70 ppm100 fig/m\5OflM

30/iM

80 /ig/ml

NormalfBloodstream

forms

3-4 +2 +2 +

1-2 +

2 +2-3 +

TrypsinizedJ>

Bloodstreamforms

1-2 +1 +

±-1 +1 +

±-1 +1-2 +

Cultureforms

2 +1 +1 +

±-1 +

1 +1 +

Cells were scored: very slight (±) to moderately heavy (4 + ).• The concentrations of the various cations were arrived at by titration. Those given repre-

sent the minimum cation levels which yielded the maximum cell agglutination characteristicfor that compound.

f Trypanosomes which had an intact extracellular surface coat.X Cells treated with trypsin as described in Materials and methods. As ascertained with

electron microscopy, these trypanosomes were morphologically devoid of a cell surface coat.

Among the forms used, normal bloodstream trypanosomes (i.e. organisms withmorphologically intact extracellular surface coat) gave the highest level of agglutina-tion with each of the various polycations. In each experimental series, trypsin-treatedbloodstream forms gave results similar or identical to those obtained with the cultureforms.

Ruthenium red induced the highest level of trypanosome agglutination among thepolycations tested in these experiments. Of the remaining compounds used, cationizedferritin was the most effective for inducing cell agglutination.

Microscopic observations of agglutinated trypanosomes indicated that the organ-isms aggregated randomly. No selective morphological orientation (i.e. neither solelyflagellar nor somatic agglutination) was apparent in the cellular aggregates formed inthe presence of the several polycations.

T. lewisi surface saccharides I 627

Reduction of ruthenium red-induced cell agglutination

Inasmuch as ruthenium red (40 /6M) was the most potent mediator of cell agglutina-tion, various inhibitors were used in an attempt to obviate its action. Results given inthis section represent the cumulative observations of 4 slide and 4 tube agglutinationtests employed with ruthenium red, the T. lewisi bloodstream forms, and each of theseveral inhibitors and enzymes in 2 separate experiments. A summary of the resultsobtained in these experiments is given in Table 2.

Cell agglutination was not inhibited with haptenic inhibitors of some plant lectins,such as a-D-glucose or a-methyl-D-glucoside. Agglutination was inhibited but notabolished in the presence of low concentrations (e.g. 20 /tg/ml) of chondroitin sulphateor dextran sulphate (average molecular weight = 4 x io6). Brief trypsin treatment alsoreduced the level of cell agglutination with ruthenium red. Treatment of the cellswith the glycoside hydrolase enzymes a-amylase, dextranase, and neuraminidase pro-duced only a marginal reduction in the agglutination level of the trypanosomes.

Table 2. Effects of various compounds and enzyme treatment on theruthenium-red-induced agglutination of T. lewisi bloodstream forms

Treatment

Nonea-D-glucosea-D-mannosea-methyl-D-glucosidea-methyl-D-mannosideChondroitin sulphateDextran sulphate (mol. wt 4 x io5)Trypsina-AmylaseDextranaseNeuraminidase

Finalconcentration*

—200 mM200 mM200 mM200 mM

20 /ig/ml20 /tg/ml

1 mg/mlf2 mg/mlf2 mg/mlf2 mg/mlf

Agglutination

3-4 +3-4 +3-4 +3-4 +3-4 +

±-1 +±-1 +1-2 +3 +3 +3 +

Cells scored very slight ( ±) to moderately heavy (4 + ).

• The maximum inhibitory effect, if any, was obtained with the given concentration,f Cells were treated and washed as described in Materials and methods.

Ultrastructure cytochemistry

No morphological differences were observed between bloodstream forms of theTaliaferro and Taiwan strains of T. lewisi in any of the various cytochemistry experi-ments. Further, bloodstream trypanosomes obtained from 4-, 7-, and 14-day infectionsin rats gave identical morphological results within each of the cytochemical stainingprotocols employed. However, for uniformity, the subsequent results pertain tobloodstream forms obtained from animals infected for 7 days.

The surface coat of untreated, control T. lewisi bloodstream forms (i.e. trypo-mastigotes) appeared to blanket the entire external surface of both the pellicular and

628 D. M. Dwyer

flagellar membranes of the organisms. In cross-section, the coat appeared somewhatmore electron-dense than the outer lamina of the surface membrane (Fig. 2). Anirregular, finely grained, fibrillar-like matrix appeared to constitute the surface coat.The coat 'fibrils' were apparently heterogeneous in length, and possibly also inwidth. No morphological diminution was apparent in the surface coat of organismsrepeatedly washed (i.e. up to 12 times) in PBSG after final isolation of the trypano-somes from host blood suspensions. It seems quite evident, from the latter observa-tions, that the surface coat is tightly bound to, or more likely, an integral part of thesurface membrane of T. lewisi bloodstream forms. Moreover, no line of demarcationor electron-lucent space was ever observed between the coat and the outer lamina ofthe surface membrane.

Treatment of T. lewisi bloodstream forms with purified ruthenium red rendered thesurface coat of both the pellicular and flagellar membranes densely stained (Fig. 3).The dense stain reaction product was restricted to the cell surface of treated organ-isms, suggesting that the surface-membrane complex was impenetrable to thisruthenium compound. The outer lamina of the pellicular and flagellar membranesappeared more electron-dense than the inner pellicular membrane lamina of theruthenium-treated cells. A homogeneously granular, electron-dense reaction productwas uniformly deposited in the surface coat of ruthenium red-treated cells. It is ofinterest that the junction formed between the flagellum and the body of the organismwas not as densely stained as the fully exposed portions of the surface coat. Thisobservation might be attributed to the low penetration level of ruthenium red.

A uniformly granular, electron-dense reaction product also was apparent at the cellsurface of the pellicular and flagellar membranes of ruthenium violet-treated organ-isms (Fig. 4). The junction between the flagellum and the body of the treated trypa-nosomes also appeared uniformly stained. However, no intracellular staining wasapparent in these organisms. Both the outer and inner lamina of the cell membraneof ruthenium violet-treated organisms appeared more electron-dense than controlpreparations.

AJcian blue-lanthanum nitrate-treated organisms had a more electron-dense sur-face coat than control preparations (Fig. 5). The cell coat of Alcian blue-treated cellshad a fibrillar-like matrix similar in appearance and distribution to that observedwith controls. The outer lamina of the pellicular membrane of Alcian-treated cellsappeared more electron dense than the inner lamina of this membrane. Further, thejunction between the flagellar and pellicular membranes was of about the same densityas other portions of the cell surface coat.

The pellicular membrane of unstained control bloodstream forms had an averagethickness of 7-2 nm. Similar mean measurements of this membrane were obtainedfrom the cation-stained cells. A comparison of the average surface coat thickness ofcontrol bloodstream trypanosomes and those treated with the various polycationicstains is given in Table 3.

No apparent morphological differences were observed between the surface coat ofT. lewisi bloodstream form controls and unstained preparations of those treated withthe various glycoside hydrolase enzymes (i.e. a-amylase, dextranase, and neuramini-

T. lewisi surface saccharifies I 629

dase). Further, cells treated with these enzymes gave staining reactions with the variouspolycationic dyes identical to those obtained with the non-enzymically treated cells.

The cell surface coat of T. lewisi bloodstream forms was removed after brief treat-ment with either crude or crystalline trypsin. This was the only morphologicalalteration apparent in organisms treated with trypsin. In unstained control prepara-tions, the outer lamina of both the pellicular and flagellar membranes appeared smooth,and no vestige of the cell surface coat was apparent (Fig. 6). Although the flagellumapparently remained attached to the body of the organism no coat matrix was obviousin this junctional complex. The pellicular membrane of trypsin-treated cells had acalculated mean width identical with that obtained from the non-trypsinized controls(i.e. 7-2 nm). However, in trypsinized organisms, both the inner and outer lamina ofthe pellicular and flagellar membranes appeared somewhat more osmophilic than thatobserved with the non-enzymically treated controls (compare insets of Figs. 2 and 6).

Table 3. Average measurements (± S.E.) in nm of the surface coat and surface staindeposits of T. lewisi after various staining procedures*

Surface stain deposit

NoneRuthenium redRuthenium violetAlcian blue-lanthanum nitrate

8-2 ±0-4i3-2±o-79-2 ±o-i9'9±o-3

Surface coat Trypsinized CultureTreatment (blood form) blood form form

o o8-6 ±o-i 7-4 ±0-347 ±0-5 48 ±016-i ±0-2 7-o±o-i

• Cumulative results of 4 separate experiments, in each the several T. leiuisi forms and all ofthe cationic stains were employed. The values given were calculated from the mean of 100cross-sectioned cells of each type per experiment.

A dense reaction product was evident at the external face of both the flagellar andpellicular membranes of trypsinized ruthenium red-treated bloodstream formtrypanosomes (Fig. 7). The stain product obviously was restricted to the cell surface,as no intracellular reactions were apparent in these preparations. This observationsuggests that the cell surface membrane was not disrupted or damaged severely duringtrypsin treatment and subsequent processing of the cells. The ruthenium red reaction-product appeared as a finely-grained, electron-dense, somewhat flocculent matrix. Nodifference was observed in the mean width of the pellicular membrane betweentrypsinized controls and those stained with ruthenium red. The junction between theflagellum and the body of the organism also contained stained matrix material. It isevident from these observations that, aside from the surface coat, the outer laminae ofboth the flagellar and pellicular membranes of T. levrisi bloodstream forms also con-tain heavy concentrations of carbohydrates which are stained with ruthenium red.

Treatment of trypsinized bloodstream forms with purified ruthenium violetresulted in the deposition of a dense reaction product on the outer lamina of both thepellicular and flagellar membranes (Fig. 8). No difference in width of the surface

630 D. M. Dwyer

membrane was observed between trypsinized ruthenium violet-treated organisms andsimilar enzyme-treated controls. A finely grained, electron-dense matrix apparentlyconstituted the surface reaction product in ruthenium violet-treated cells. Qualitativeand quantitative differences in the type and amount of reaction product deposited atthe cell surface were apparent between the ruthenium red and violet preparations(compare Figs. 7 and 8). The junctional complex between the flagellum and thetrypanosome body appeared stained to the same extent as other portions of the surfacemembrane in these organisms.

A rather dense reaction at the surface membranes also was obtained with trypsinizedcells treated with Alcian blue and lanthanum nitrate (Fig. 9). The Alcian blue reactionproduct was finely granular, and somewhat fibrillar-like in appearance. A slightenhancement in electron-density was observed in the outer lamina of the pellicularmembrane. No difference was observed in the mean width of the pellicular membranebetween trypsin-treated controls and the Alcian blue preparations (i.e. 7-2 nm). Nointracellular staining was apparent in the Alcian blue-treated organisms. The cellstreated with Alcian blue and lanthanum nitrate had a quite different appearance withregard to the surface reaction product than that observed with the ruthenium-treated organisms (compare Figs. 7-9). A comparison of the measurements of the cellsurface stain deposits obtained with the trypsin-treated blood forms and the cationicstains is given in Table 3 (p. 629).

It is obvious from the foregoing observations that, even after removal of the surfacecoat with trypsin, the surface membranes of T. lewisi bloodstream forms gave strongpositive staining reactions with the various polycationic compounds. These resultsindicate that polysaccharide moieties are present in the pellicular membrane of thebloodstream form trypanosomes.

No morphologically identifiable surface coat was observed in unstained controlT. lewisi culture forms (Fig. 10). The outer lamina of the pellicular and flagellarmembranes of the control organisms appeared smooth and unadorned. In suchorganisms, the mean width of the pellicular membrane was 7-3 nm. Both the inner andouter lamina of the pellicular membrane appeared only slightly osmophilic in controlpreparations.

T. lewisi culture forms treated with ruthenium red had an electron-dense matrixuniformly deposited at the cell surface (Fig. 11). A finely granular, slightly fibrillar-like material appeared to compose the matrix on the outer lamina of both the pel-licular and flagellar membranes of the ruthenium red-treated cells. The inner laminaof the pellicular membrane of ruthenium red-treated cells also appeared somewhatmore electron-dense than that observed with control preparations. The ruthenium redreaction was restricted to the surface of the treated organisms, as no intracellularstaining was apparent. Results obtained with T. lewisi culture forms and rutheniumred were very similar in appearance to those observed with similarly stained prepara-tions of trypsin-treated bloodstream forms (compare Figs. 11 and 7).

Ruthenium violet-treated T. lewisi culture forms also had a dense staining reactionproduct present on the external face of the pellicular and flagellar membranes (Fig. 12).The stain product appeared as a very dense, and somewhat finely fibrillar, matrix

T. lewisi surface saccharides I 631

closely applied to the outer lamina of the surface membrane. The inner lamina of thepellicular membranes of ruthenium violet-treated cells also appeared more electron-dense than that observed in control organisms. Staining reactions obtained withruthenium violet and the culture forms were similar in appearance to trypsin-treatedbloodstream forms treated with this compound (compare Figs. 12 and 8).

T. lewisi culture forms stained with Alcian blue-lanthanum nitrate had a uniformlydense reaction product distributed on the surface membranes (Fig. 13). The innerlamina of the pellicular membrane of Alcian-treated organisms also appeared moreelectron-dense than that observed in control preparations.

The mean width of the pellicular membrane of culture forms treated with thevarious stains was identical to that obtained from the untreated controls (i.e. 7-3 nm).Measurements of the cationic surface stain deposits obtained from the culture formsare given in Table 3.

Results obtained with the T. lewisi culture forms and the various cationic stainsindicate that complex polyanionic saccharides also are present in the surface of thisdevelopmental stage of the organism. Further, the results obtained with the severaldye compounds and the culture forms appeared very similar to those observedwith the trypsin-treated bloodstream trypanosomes. This suggests that polyanioniccarbohydrates are also functional components of the surface membranes of bothtrypanosome cyclical developmental stages.

Large quantities of cationized ferritin were bound to the surface coat of T. lewisibloodstream forms (Fig. 14). Very few ferritin particles, however, were bound directlyto the outer lamina of the surface membrane of these cells. The particles mostlyappeared interspaced among the fibrillar-like surface coat matrix. Controls, treatedwith non-cationized horse spleen ferritin had no, or at most very few, ferritin particlesattached to the surface coat.

Trypsin-treated bloodstream forms also had cationized ferritin bound to thepellicular and flagellar membranes (Fig. 15). Very few, if any, non-cationized ferritinparticles were observed on the surface membranes of trypsin-treated controls. Thecationized ferritin particles appeared bound directly, or in very close proximity, to theouter lamina of the surface membrane in trypsin-treated organisms. It is of interestthat ferritin particles also were observed in the junction formed between the flagellumand the body of the cell.

Cationized ferritin binding occurred also at the surface membranes of T. lewisiculture forms, and these appeared bound directly to the outer lamina of the surfacemembrane of the organisms (Fig. 16). No ferritin was apparent at the surface ofcontrols treated with non-cationized ferritin.

The mean number of cationized ferritin particles attached to the cell surface of theseveral T. lewisi forms was calculated per 100 linear nm of pellicular membrane.Similarly, counts also were made from tangentially sectioned organisms and thesewere used to calculate the mean number of ferritin particles bound per 100 nm2 of cellsurface area.

A comparison of the amount of cationized ferritin bound to the cell surface of thevarious T. lewisi forms is given in Table 4.

632 D. M. Dzuyer

It is of interest that the bloodstream trypanosomes with intact surface coat had64-7 and 66-2% more ferritin bound per 100 nm2 of surface membrane than thetrypsinized bloodstream and culture forms, respectively (compare Figs. 14-16). Thesesignificant differences in ferritin-binding levels might be attributed to a greater amountof surface area in the bloodstream forms afforded it by the cell surface coat.

Although the trypsinized bloodstream forms bound significantly more ferritin per100 nm2 of membrane (i.e. 4-1 %) than the culture forms, the difference was exceed-ingly small (compare insets of Figs. 15, 16). These results suggest that the saccharidespresent in the surface membranes impart a net negative charge to both T. lewisidevelopmental stages. Further, it seems apparent that the surface coat carbohydratescontribute significantly to the net negative surface charge of T. lewisi bloodstream formtrypanosomes.

Table 4. Mean (± s.E.) number of cationized ferritin particles bound on thepellicular membrane surface of T. lewisi*

Particles/100 nm Particles/100 nms

Form linear surface surface area

BloodstreamTrypsin-treated bloodstreamCulture

17-4 ±O-2IO-5 ±0-4I O I ±O-I

3097 ±0-9109-3 ±0-5104-8 ± 0 7

• Cumulative results of 2 separate experiments, in each the several T. lewisi forms wereemployed. The values given were calculated from 100 cross-sectioned and 100 tangentiallysectioned cells of each type per experiment.

DISCUSSION

Reagents which react with acidic polysaccharides and glycosaminoglycans serve asimportant probes in the analyses of cell surfaces (Hughes, 1973). Polycations arecapable of inducing cell agglutination through a mechanism postulated as multiplecross-bridge formation of molecular constituents of adjacent cell membranes (Curtis,1973). Recently, Utsumi & Oda (1973) also speculated that ruthenium red probablyinduced cell agglutination via cross-bridge formation. Among the polycations testedin the present study, ruthenium red was the most effective inducer of cell agglutina-tion. Ruthenium violet, considered possibly as a polymer and of greater positivevalency than ruthenium red (Luft, 1971a), was much less effective than the lattercompound in mediating T. lewisi agglutination. No explanation can be offered for thisselective difference between the 2 compounds from the present results. Danon et al.(1972) previously reported that several mammalian cell types were agglutinatedwith cationized ferritin, and this polycation also was found an effective mediator ofT. lewisi agglutination in the present study.

Regardless of the polycation tested, bloodstream trypanosomes with intact surfacecoat gave consistently stronger agglutinations than culture forms. Trypsinized blood-stream forms (i.e. cells without surface coat) gave agglutination results with the

T. lewisi surface saccharides I 633

various cations strikingly similar to those obtained with the culture forms. Theseobservations suggest that the trypanosome surface coat matrix constituents are highlyanionic, and that the pellicular membrane of both the bloodstream and culture formspossess qualitatively similar concentrations of polyanionic components.

As expected, attempts to obviate the agglutinogenic activity of ruthenium red on thebloodstream forms with haptenic inhibitors (e.g. a-D-glucose, a-D-mannose or methylglycosides) met with only partial success. However, low concentrations of the anionschondroitin sulphate and dextran sulphate were effective in reducing this type ofagglutination. Similar results were reported by Utsumi & Oda (1973) for the formercompound, ascites hepatoma cells, and ruthenium red.

Pretreatment of bloodstream forms with the several glycoside hydrolase enzymesresulted in only a slight diminution of cell agglutination with ruthenium red. However,a marked decrease in ruthenium red-induced agglutination resulted with trypsinizedblood forms. These results correlate with the fine-structure observation that trypsinremoved the surface coat of blood forms whereas the coat remained morphologicallyintact in cells treated with any of the glycoside hydrolase enzymes. In this regard, ithas been demonstrated by a number of investigators (e.g. Nicolson, 1973; Hynes,1974; Hunt & Brown, 1974; Ruoslahti & Vaheri, 1974; Chiarugi, Vannucchi &Urbano, 1974) that glycopeptides and glycosaminoglycans are released from mam-malian cell surfaces after trypsin treatment.

Current observations of the fine structure and distribution of the surface coat ofintact T. lewisi blood forms are in agreement generally with the preliminary resultsreported by Vickerman (1969). The present results indicated that T. lewisi controlculture forms lacked an extracellular surface coat, and this confirms the findingsreported initially by Vickerman (1969). Further, no surface coat has been reported,to date, for any trypanosome culture form (Vickerman, 1974).

Luft (19710) demonstrated that ruthenium compounds strongly precipitatedcertain polyanions, and these also were shown to stain acid mucopolysaccharidesassociated with proteins as protein-carbohydrate (i.e. glycosaminoglycans) complexes(Luft, 19716; Szubinska & Luft, 1971). Similar conclusions were reported for thestaining affinities and mechanisms of action of Alcian blue and lanthanum by Shea(1971). Results obtained in the present study with the ruthenium compounds, aswell as with Alcian blue and lanthanum nitrate, are in agreement with the abovehypotheses.

The rather uniform appearance of reaction product deposition in the T. lewisisurface coat within each staining protocol suggests that polyanionic carbohydratemoieties are dispersed randomly throughout the coat matrix. These moieties probablyrepresent terminal ligands of constituent surface coat structural glycoproteins. As withthe agglutination results, pretreatment of the bloodstream forms with a-amylase,dextranase, or neuraminidase had no effect on the results obtained with the severalstaining protocols. This suggests that the presumptive surface coat carbohydrateligands probably are not composed of appreciable quantities of repetitive a-1,4 anda-1,6 glucan bonded D-glucose units, or of sialic acid.

The pellicular and flagellar membranes of trypsinized cells were rendered uni-

41 C E L IQ

634 D. M. Dwyer

formly and densely stained following treatment with each of the cationic dyes. Further,fine-structure staining results obtained with the trypsinized blood forms approximatevery closely to those obtained with the culture forms which do not possess an extra-cellular surface coat. This suggests that both T. lewisi life cycle stages possess some-what similar amounts of polyanionic glycosaminoglycan constituents in their respec-tive surface membranes. Cytochemical results resembling those described previouslyfor T. lewisi culture forms also were reported for several insect stages of a salivariantrypanosome species, Trypanosoma brucei (Steiger, 1973; Steiger & Jenni, 1974), aswell as for a distantly related kinetoplastid, Leishmania donovani (Dwyer, Langreth &Dwyer, 1974; Dwyer, 1974 a).

The surface coat of salivarian trypanosomes completely differs morphologicallyfrom that of the Stercoraria, and has been characterized as a compact, uniformly denseextracellular surface layer (Vickerman, 1974). Both direct (Cross, 1972, 1973; Njoguet al. 1974) and indirect (Williamson & Desowitz, 1961; Allsopp et al. 1971; Njogu &Humphreys, 1972; Njogu, 1974; Allsopp & Njogu, 1974) biochemical and immuno-logical data suggest that some carbohydrates are present albeit in small quantities inthe cell surface of some salivarian bloodstream form trypanosomes. Cytochemically, asaccharide-containing layer apparently does exist between the junction of the pel-licular membrane and the extracellular surface coat in blood forms of the salivarianspecies, T. brucei (Wright & Hales, 1970). However, these carbohydrates are probablyassociated with the pellicular membrane outer lamina rather than the surface coat inthis species (Steiger, 1973; Steiger & Jenni, 1974).

A net negative surface charge of T. lewisi blood forms was demonstrated previouslywith cell electrophoresis (Hollingshead et al. 1963) and with ion-exchange chromato-graphy (Lanham, 1968). The present cationized ferritin results indicated that theT. lewisi net negative charge was localized in the extracellular surface coat and surfacemembrane. Further, it seems obvious that the surface coat polyanion constituentssignificantly contribute to the overall net negative surface charge of the T. lewisibloodstream forms. Treatment of the blood forms with the several glycoside hydrolaseenzymes, and specifically neuraminidase, had no gross effect on the amount ofcationized ferritin bound in the cell surface coat. Danon et al. (1972) used the samemethod and reported that neuraminidase-treated erythrocytes bound significantly lesscationized ferritin than control cells. The present results suggest that sialic acid, ifpresent, does not constitute a major component of the T. lewisi cell surface. A similarconclusion was obtained from chemical analysis of certain fractions attributed to thecell surface of the salivarian trypanosome, T. brucei (Allsopp & Njogu, 1974). T. lewisitrypsinized bloodstream forms and culture forms had very similar binding affinitiesfor cationized ferritin, suggesting that the surface membranes of both life cycle stageshad a similar overall negative surface charge.

The results of the present study suggest that a positive correlation exists betweenthe level of cationic dye-induced trypanosome agglutination in vitro and the resultantamount of stain deposited in the cell surface of the several trypanosome forms atthe fine-structure level. Similar conclusions pertain also to the cationized ferritinresults.

T. lezoisi surface saccharides 1 635

The current report has answered only a few questions concerning the cell surface ofthis stercorarian trypanosome species. Further biological and chemical elucidation ofthe surface coat and pellicular membrane of T. lewisi as well as other stercorarianspecies await more detailed experimental studies.

I am grateful to Dr Philip A. D'Alesandro for many stimulating and captious discussionsduring the course of the present study, and to Professor William Trager for his criticalevaluation of this work. I thank Carol Sinatra Wrzosek for her diligent assistance throughout,and Mr John E. Gould and Ms Marika Tershakovec for their technical adroitness in aspectsof this study. I also thank Mrs Diane F. Greene for her skilful preparation of thismanuscript.

This investigation was supported by Grant AI-11126 and in part by Grant AI-11916 fromNIAID, U.S. Public Health Service. A preliminary report (Dwyer, 19746) was presented to theThird International Congress of Parasitology in Munich, Germany, August 1974.

REFERENCES

ALLSOPP, B. A. & NJOGU, A. R. (1974). Monosaccharide composition of the surface glycopro-tein antigens of Trypanosoma brucei. Parasitology 69, 271-281.

ALLSOPP, B. A., Njocu, A. R. & HUMPHREYS, K. C. (1971). Nature and location of Trypano-soma brucei subgroup exoantigen and its relationship to 4S antigen. Expl Parasit. 29,271-284.

CHIARUGI, V. P., VANNUCCHI, S. & URBANO, P. (1974). Exposure of trypsin-removable sul-phated polyanions on the surface of normal and virally transformed BHK 21/C13 cells.Biochim. biophys. Ada 345, 283-293.

COOK, G. M. W. & STODDART, R. W. (1973). Surface Carbohydrates of the Eukaryotic Cell,346 pp. London and New York: Academic Press.

CROSS, G. A. M. (1972). Identification and isolation of surface coat proteins from Trypanosomabrucei. J. Protozool. 19 (Suppl.), 46.

CROSS, G. A. M. (1973). Isolation and purification of a class of soluble surface proteins fromTrypanosoma brucei. Trans. R. Soc. trop. Med. Hyg. 67, 261.

CURTIS, A. S. G. (1971). Intra- and inter-membrane interactions of the cell surface. Sub-Cell.Biochsm. 1, 179-196.

CURTIS, A. S. G. (1973). Cell adhesion. Prog. Biophys. molec. Biol. 27, 315-386.D'ALESANDRO, P. A. (1972). Trypanosoma lewisi: production of exoantigens during infection in

the rat. Expl Parasit. 32, 149-164.DANON, D., GOLDSTEIN, L., MARIKOVSKY, Y. & SKUTELSKY, E. (1972). Use of cationized ferritin

as a label of negative charges on cell surfaces. J. Ultrastruct. Res. 38, 500-510.DWYER, D. M. (1974a). Lectin binding saccharides on a parasitic protozoan. Science, N. Y. 184,

471-473-DWYER, D. M. (1974Z)). The surface coat of Trypanosoma letvisi: a biochemical and fine

structural study. Proc. 3rd int. Congr. Parasit. 1, 24.DWYER, D. M., LANGRETH, S. G. & DWYER, N. K. (1974). Evidence for a polysaccharide sur-

face coat in the developmental stages of Leishmania donovani: a fine structure-cytochemicalstudy. Z. Parasitkde 43, 227-249.

HOARE, C. A. (1964). Morphological and taxonomic studies on mammalian trypanosomes.X. Revision of the systematics. J. Protozool. n , 200-207.

HOARE, C. A. (1972). The Trypanosomes of Mammals, 749 pp. Oxford and Edinburgh: BlackwellScientific Publications.

HOLLINGSHEAD, S., PETHICA, B. A. & RYLEY, J. F. (1963). The electrophoretic behaviour of sometrypanosomes. Biochem.J. 89, 123-127.

HUCHES, R. C. (1973). Glycoproteins as components of cellular membranes. Prog. Biophys.molec. Biol. -2&, 191-268.

636 D. M. Thoyer

HUNT, R. C. & BROWN, J. C. (1974). Surface glycoproteins of mouse L cells. Bioclieinistry,N. Y. 13, 22-28.

HYNES, R. O. (1974). Role of surface alterations in cell transformation: the importance of pro-teases and surface proteins. Cell 1, 147-156.

LANHAM, S. M. (1968). Separation of trypanosomes from the blood of infected rats and miceby anion exchangers. Nature, Lond. 218, 1273-1274.

LUFT, J. H. (1961). Improvement in epoxy resin embedding methods. J. biophys. biochem.Cytol. 9, 409-414.

LUFT, J. H. (1971a). Ruthenium red and violet. I. Chemistry, purification, methods of use forelectron microscopy and mechanism of action. Anat. Rec. 171, 347-368.

LUFT, J. H. (19716). Ruthenium red and violet. II . Fine structural localization in animaltissues. Anat. Rec. 171, 369-416.

MOLYNEUX, D. H. (1969). The fine structure of the epimastigote forms of Trypanosoma letuisiin the rectum of the flea Nosopsyllus fasciatus. Parasitology 59, 55-66.

NICOLSON, G. L. (1973). Anionic sites of human erythrocyte membranes. I. Effect of trypsin,phospholipase C, and pH on the topography of bound positively charged colloidal particles.J. Cell Biol. 57, 373-387.

NJOGU, A. R. (1974). The immunochemistry of the variable antigens of Trypanosoma brucei.Proc. 3rd int. Congr. Parasit. 2, 1094.

NJOGU, A. R. & HUMPHREYS, K. C. (1972). The nature of the 4S antigens of the brucei sub-group trypanosomes. ExpI Parasit. 31, 178-187.

NJOGU, A. R., ITAZI, O. K., ENYARU, J. C. & ABONGA, L. (1974). Direct evidence that the 4S(surface) antigens are located on the outer surface of the Trypanosoma brucei subgroup cellmembrane. Trans. R. Soc. trop. Med. 68, 147-148.

OSEROFF, A. R., ROBBINS, P. W. & BURGER, M. M. (1973). The cell surface membrane: bio-chemical aspects and biophysical probes. A. Rev. Biochem. 42, 647-682.

RAMBOURG, A. (1971). Morphological and histochemical aspects of glycoproteins at the surfaceof animal cells. Int. Rev. Cytol. 31, 57-114.

RUOSLAHTI, E. & VAHERI, A. (1974). Novel human serum protein from fibroblast plasmamembrane. Nature, Lond. 348, 789-791.

SHARON, N. (1974). Glycoproteins. Scient. Am. 230, 78-86.SHEA, S. M. (1971). Lanthanum staining of the surface coat of cells. Its enhancement by the use

of fixatives containing Alcian blue or cetylpyridinium chloride. J. Cell Biol. 51, 611-620.

SINGER, S. J. (1974a). The molecular organization of membranes. A. Rev. Biochem. 43, 805-833-

SINGER, S. J. (19746). Molecular biology of cellular membranes with applications to immuno-logy. Adv. Immun. 19, 1-66.

SPIRO, R. G. (1970). Glycoproteins. A. Rev. Biochem. 39, 599-638.STEIGER, R. F. (1973). On the ultrastructure of Trypanosoma (Trypanozoon) brucei in the course

of its life cycle and some related aspects. Ada Tropica 3, 66-168.STEIGER, R. & JENNI, L. (1974). Cytochemistry of the coat/pellicle complex in Trypanosoma

brucei. Proc. 3rd int. Congr. Parasit. 1, 22-23.SZUBINSKA, B. & LUFT, J. H. (1971). Ruthenium red and violet. I II . Fine structure of the

plasma membrane and extraneous coats in amoebae (A. proteus and Cliaos cliaos). Anat. Rec.171, 417-442.

TERRY, A. H. & CULP, L. A. (1974). Substrate-attached glycoproteins from normal and virus-transformed cells. Biochemistry, N.Y. 13, 414-425.

TOBIE, E. J., VON BRAND, T. & MEHLMAN, B. (1950). Cultural and physiological observations ofTrypanosoma rhodesiense and Trypanosoma gambiense. J. Parasit. 36, 48-54.

UTSUMI, K. & ODA, T. (1973). Ruthenium-red-induced cell agglutination and surface glyco-protein and mucopolysaccharide. J. Cell Sci. 13, 901—911.

VICKERMAN, K. (1969). On the surface coat and flagellar adhesion in trypanosomes. J. Cell Sci.5, 163-193.

VICKERMAN, K. (1971). Morphological and physiological considerations of extracellular bloodprotozoa. In Ecology and Physiology of Parasites (ed. A. M. Fallis), pp. 58—91. Toronto:University of Toronto Press.

T. lewisi surface saccharides I 637

VICKERMAN, K. (1974). The ultrastructure of pathogenic flagellates. In Trypanosomiasis andLeishmaniasis with special reference to Cliagas' Disease (ed. K. Elliott, M. O'Connor & G. E. W.Wolstenholme), pp. 171-198. Amsterdam, London and New York: Elsevier-ExcerptaMedica-North Holland, Associated Scientific Publishers.

WILLIAMSON, J. & DESOWITZ, R. S. (1961). The chemical composition of trypanosomes.I. Protein, amino acid and sugar analysis. Expl Parasit. 11, 161-175.

WRIGHT, K. A. & HALES, H. (1970). Cytochemistry of the pellicle of bloodstream forms ofTrypanosoma {Trypanozoon) brucei.J. Parasit. 56, 671-683.

{Received 30 April 1975)

638 D. M. Dwyer

ABBREVIATIONS ON PLATES

/ flagellum mt microtubulesvi mitochondrion n nucleus

Figs. 2-5. Transverse sections of T. letoisi 7-day bloodstream form trypanosomes.

Fig. 2. Untreated control. The fibrillar-like surface coat is apparent on both thepellicular and flagellar membranes, x 55 300. Inset: note the spatial association of thecoat matrix and the outer lamina of the pellicular membrane, x 152400.

Fig. 3. Ruthenium red-treated organism. The electron-dense reaction product isevident in the surface coat of the pellicular and flagellar membranes, x 46600. Inset:note the thickness of the granular stain product in the surface coat; the outer laminaof the pellicular membrane also appears stained, x 147000.

Fig. 4. Organism treated with purified ruthenium violet. The stain reaction-productis evident in the surface coat of the pellicular and flagellar membranes, x 53400. In-set: the granular nature of the surface coat reaction product is apparent; note thatboth the inner and outer laminae of the cell membrane appear stained, x 140000.

Fig. 5. Cell treated with Alcian blue and lanthanum nitrate. Note the uniformdistribution of the electron-dense, fibrillar-like surface coat matrix, x 51400. Inset:the outer lamina of the pellicular membranes appears more electron-dense than theinner lamina of this membrane; the slightly granular, fibrillar-like nature of thesurface coat matrix is apparent, x 135500.

T. lewisi surface saccharides I 639

f

1.v

mt

n

n\

640 D. M. Dwyer

Figs. 6—9. Transverse sections of trypsin-treated T. letuisi 7-day bloodstream trypano-somes.

Fig. 6. Unstained trypsin-treated control. Note the complete absence of a cell surfacecoat. The flagellum appears to remain in a junctional complex with the body of theorganism even though no coat matrix is obvious, x 59 300. Inset: the outer laminaof the pellicular membrane appears devoid of coat matrix material; note also, thesomewhat enhanced osmophilic appearance of the inner and outer lamina of thismembrane, x 152000.

Fig. 7. Ruthenium red-treated organism. The granular, somewhat flocculent,electron-dense reaction product is evident on the fiagellar and pellicular membranes,x 70000. Inset: note the density and thickness of the ruthenium red deposit on theouter lamina of the pellicular membrane, x 154000.

Fig. 8. Trypanosome treated with purified ruthenium violet. A dense reactionproduct is apparent on the fiagellar and pellicular membranes, x 67800. Inset: theelectron-dense ruthenium violet reaction product appears closely adherent to theouter lamina of the pellicular membrane, x 164400.

Fig. 9. Organism treated with Alcian blue and lanthanum nitrate. A coat-like stainproduct matrix is apparent at the surface of this organism, x 72700. Inset: a finelygranular, fibrillar-like matrix is evident on the outer lamina of the pellicular mem-brane, x 163200.

T. lewisi surface saccharides I 641

f

n

m

1 ^ '

r

f -^

/

\m

\

s * 8

642 D. M. Dwyer

Figs. 10-13. Transverse sections of T. leurisi culture forms.

Fig. 10. Unstained control. No surface coat is evident on the membranes of thisorganism, x 22000. Inset: the outer lamina of the pellicular membrane appearssmooth and unadorned, x 168000.

Fig. 11. Organism treated with ruthenium red. Note the uniformly dense reactionproduct on the cell surface, x 36900. Inset: the finely granular composition of theruthenium red reaction product appears on the outer lamina of the pellicular mem-brane, x 135000.

Fig. 12. Ruthenium violet-treated trypanosome. A dense reaction product isevident on both the pellicular and flagellar membranes, x 61000. Inset: the verydense, slightly fibrillar-like ruthenium violet reaction product is apparent on theouter lamina of the pellicular membrane, x 138500.

Fig. 13. A cell treated with Alcian blue and lanthanum nitrate. The electron-densesurface reaction product is apparent on the cell membrane. X 33500. Inset: note theappearance of the finely granular matrix on the outer lamina of the pellicular mem-brane, x 142800.

T. levxisi surface saccharides I

in

643

\

10 11

f

12

644 D. M. Dwyer

f

f

m

14 15,

• * > .

•ff

16Figs. 14-16. T. lauisi bloodstream and culture forms treated with cationized ferritin.

Fig. 14. Bloodstream form with an intact surface coat; transverse section. Cation-ized ferritin appears bound in the entire cell surface coat. Note the lack of ferritinbinding on the flagellar membrane and the membrane lining the reservoir surroundingthe flagellum. x 50700. Inset: cationized ferritin distributed throughout the pellicularmembrane surface coat, x 152000.

Fig. 15. Trypsin-treated bloodstream form. Ferritin particles are apparent on boththe pellicular and flagellar membranes, x 60900. Inset: note that the cationizedferritin appears directly bound to the outer lamina of the pellicular membrane,x 132000.

Fig. 16. Trypanosome culture form; longitudinal section. Cationized ferritinparticles are apparently bound to the entire pellicular membrane. Note that ferritinis bound also to the flagellar membrane and to the membrane of the reservoir sur-rounding the flagellum. x 31600. Inset, transverse section: cationized ferritin dis-tribution on the pellicular membrane outer lamina, x 142000.