An electron microscopic study of hyalocytes

An electron microscopic study of hyalocytes

Exp. l?ye Reds. (1.965) 4, 249-255 An Electron Microscopic Study of Hyal ytes* GU rCAR D. BLoo tt AND ENDRE A. BAt,AT.S Department of Connective T...

5MB Sizes 5 Downloads 33 Views

Recommend Documents

No documents
Exp. l?ye Reds. (1.965) 4, 249-255

An Electron Microscopic Study of Hyal

ytes*

GU rCAR D. BLoo tt AND ENDRE A. BAt,AT.S

Department of Connective Tissue Re~.arch, Institute of B Sc4enees, R~tina Foundation, Boston,

I al~ M~ical ,U.S.A.

( Rec~ved 3 Ju,,z 1965) The fine structure of the h y a l o o y ~ in the cortical ti~uo layer of the calf vitreous has for electron microscopy wore Mso investigated. Among the latter, vapor fixation followed by dehydration in highly concex~trat~d alcohol solutions was includ~ ia an attempt to stabilize the intracc|lul~r, ~ater-soluble glycosaminoglycans. 1. I n t r o d u c t i o n

The h y a l o c y t e s of tile vitreous of v a r i o u s animM species h a v e been the s u b j e c t of n u m e r o u s cytological a n d histochemicM studies in r e c e n t years. E l e c t m n microscopic r e p o ~ s on this subject are, however, few. I~[ogan (1963) described and s h o w ~ an electron m i c r o ~ p h of a cell in t h e c d ~ i c a l tissue l a y e r of t h e m o n k e y vitreous which he asmlmed t o be a " t r u e vitreous ceU". I n a v e r y recent publication, ,Seaman and S t o r m (1965) described t h e fine s t r u c t u r e of h y a l o c y t e s from domestic fowl.. G~rtner (1965) investigated t h e cells of t h e vitreous of r a t e m b r y o s and one 3-yearold child b y electron microscopic techniques. As a c o n t i n u a t i o n o f previous cytological a n d histochemical studies on t h e cells of t h e vitreous carried o u t in this l a b o r a t o r y , t h e present i n v e s t i ~ t i o n was u n d e r t a k e n to elucidat~ t h e fine s t r u c t u r e of t h e m o s t e x t e n s i v e l y s t u d i e d h y a l o c y t e p o p u ~ t i o n of the v i t r e o u s ~ v i z . , t h a t of t h e calf. A p r e l i m i n a r y s t u d y on this s u b j e c t has been reported b y BaIazs, Mitchell a n d EckI (1964a). T h e few electron mi plm on h y a l o o y t e s ~ u e t u r e h i t h e r t o published differ in c e r t a i n r e s p e c t , ~ do the i n ~ r p r c tatiorm given b y t h e different investigators. The), also d e v i a t e s o m e w h a t from our findings. Since v a r i o u s fixation a n d e m b e d d i n g techniques were used in these o t h e r investigations, i t was considered of interest to include in t h e present s t u d y some observations m a d e with a few such different t e c h n i q u e s on t h e same h y a l o v y t o population. 2. M a t e r i a l s a n d l~iethods

The material for the present investigation consisted of eyes from 7 calves, 10~12 weeks of age. The eyes were enueleated immediately after slaught~er and placed on ice. They were transported to the laboratory where, 35-~50 rain after enucleation, they were opened by an equatoriM incision through ~he sclera, choroid and retina. Small p i ~ of gel from t h e cortical tissue layer of the vitreous were collected from areas contiguous to the retina and located mid-way between the optic disc and the era s e r r a ~ . The st~cimens w e ~ fixed either by immersion in fluid fixatives or by treatment with fixing .vapors. The f o l l o w ; i t fixation fluids were employed: (a) 1% osmium tetroxide in Veronal-aceta~ buffer (ptl 7,3) * This investigation was supported in part by a PHS r ~ r c h grant (NB 03370) from the N ~ t ~ l Instim~ of Neurological Diseae~ and Blindness, U.S. Public Health Se~io0, and in ~ by the ~V~to* hall Foundation, Ino., New York, N.Y., U.S.A. Permanent address: Institute for ~ l l l~_aearch. Karo]L~8~ Ir~titutert, Stockholm, 8 w ~ n . 249 1t

250

G. D. BLOOM AND E. A. BALAZS

(Palade, 1952), fixation time 1-2 hr; (b) 1~o osmium t~gtroxide in Veronal-acetate buffer to which sucrose had been added to a molarity of 0.I46 (pH 7-3), fixation time 1 hr; and (c) 4'~ glutaraldehyde in 0.1 M p h o s p h a ~ buffer (pH 7.4), fixation time 2 hr. After rinsing in 0.2 M sucrose, buffered ~ p H 7-4 with 0.1 ~ phosphate, postfixation was carried out in phosphates-buffered 1% osmium tetroxide in 0.2 M sucrose for a period of 1 hr (Sabatini, Bensch and Barrnett, 1963). For vapor fixation, the vapors of the following fluid fixatives were employed: (d) 1% osmAc acid for 30 ~ n or (e) 4% glutaraldehyde for 30 min, followed by l ~ j osmium tetroxide for an a d ~ t i o n a l 30 mln. All fixations were carried out at 4 ~ C. Specimens fixed according to (a) and (b) were rinsed in Veronal buffer for ~ rain prior to dehydration, whereas those fixed as described in (c) were rinsed in distilled water for 1 hr. Dehydration was carried out in ~sing concentrations of ethanol, starting a t 30%0 for the specimens fixed by immersion and a t 6 0 ~ for those fixed in vapors. Specimens t r e a ~ d as described in (a) were embedded in a methacry-laW mixture (15% methyl and 85% butyl methacryIate), which was h e a t po[ymerized. For all other specimens, embedding was carried out in the epoxy resin Epon 812 according to the method of Finek (1960). For sectioning, an L K B Ultrotome was employed, and before examination the sections were stained with saturated uranyl a c , tate, followed b y lead c i t r a ~ , according to the method of Reynolds (1963). The sections were studied in a Phillps EM 200 electron microscope a t original magnifications of × 3,0~A)-30,C~O0. 3. Calf hyalocy+~es show a fairly regular round, or oval, s h a p e (Plates 1, 3, 5 a n d 6). The cells are b o u n d e d b y a distinct p l a s m a m e m b r a n e with a thickness o f a p p r o x i m a t e l y 90 A. P e r i p h e r a l l y , fine asmic sions, or mierovilli, m a y project from the surface of t h e cells (Plate 1). These sions v a r y somewhat, in size a n d shape. Generally a b o u t 0.3-0-5 ~ in length a n d 0-1-0-2 ~ in width, some protrusions m a y r e a l a l e n ~ h of more ~ a n 2 ~. These structures are more obvious a f t e r o s r n i l ~ ~x_ation t h a n w h e n g l u t a r a l d e h y d e fixation is employed before o treatment. The nucleus of t h e h y a l o c ~ e is generally elongated, m e ~ u r m g 5 - 6 ~ in length a n d 2 - 3 ~ in width, a n d shows an a r outline. I t is sa~rounded b y a double m e m b r a n e , the ipmer p o ~ i o n of which is s o m e w h a t t h i n n e r (~-~ 50-60 tk) t h a n t h e o u t e r ( ~ 60-70 ~). The o u t e r me~mbrane is occasionally s t u d d ~ with ribonucI in ( R N P ) p a i l d e s a n d is also obser¢~d to be c o n ~ o u s here a n d there with t h e endopla~smie retie* ulum. I n osmium-fi_xed spe a thin, b u t variable, space o f t h e order of 100-150 tk sep the m e m b r a n o u s stlalc~res. Nuclear pores are n, a n d t h e of t h e nuclmm is irregularly ~ t r i b u t e ~ l ~ r o u g h o u t t h e nueleoplasm, ~ t h t h e exception of t h e inner aspect of t h e nuclear m e m b r a n e . Here, it f r e q u e n t l y l : o r ~ a dense b a n d of variable t h i c k e n s interI~apted b y areas ~ a c e n t to t h e nuclear pores (Plate 1). Mitochondria are regallarly pre~ent in f a ~ I I ~ b e r s . T h e ~ 1 Stl~CtUre d o ~ n o t differ f[~m t h a t which is t)~pical of these lles in other mammalial~ cells. T h e y t h u s exhibit a surrounding d o u b l e - m e m b r a n e s t r u c t u r e a n d show internal e r e ~ s or m e m b r a n e - b o u n d e d lameUae. A fairly well-developed Golgi a p p a r a t u s is also prese n t in t h e hyMoc;ytes ( P l a ~ s 1 a n d 3). Often located in t h e d o s e vichfity of t h e nucleus, this organelle is characterized b y a series of parallel m e m b r a n e s which correspond to cross sections of flattened ~ c s , or cisternae; associated with t h e rnc~mbranes are ¢lustem of fairly dense vesicles, v a r y i n g in size f r o m 30 to 100 m/~ or more, a n d a few larger vacuoles.

,ia (M:), pm~

~ro~vidcnt~

h.~urfaC~-d endO '6 shows, dt hig!~ex magtiiticatton* a ¢aeuoi~r at,n,vt,

I'f~W: 2, Portion of the c y t o p l ~ m of a calf hyalocytc: Hyalocyte granules are visible at, diiTerent stages of development or of metabolic ucti vity/Not~ their varying appe*~ran~ and, especially, the contra8~ ~,~tween the ligh t, ,almost ~,Inp~ grannie at the apex:of t h e nucleus and t|m intensely S~ined One~ with a complex st~.hatruetm~ (uj~per left). The eytopb~m shows the prx~:aenceof nl!merous vesicular structures, and a fairly long, cytoplasmic protrusipn e a n l m s-c.en (bottom). Elee,tron mierograph. 'Ep0n e m ~ d d i n g , ~,xa ton: ~ m i u n l ~troxide and sucrose. (X 26,00~))

Pt,AWl,:3, (:aft hyalocytc fixed in gh~taraldehyde and postfixcd in ~mium ~"t~xl~|~. N,t~, the' r~,htiv~, paucity of.cyt;oplasmio vesicle, th~ widenc~!si~ecs t~f th~ endoplasmie rt,tieu|um, t|m thiekv~| mi|~ chondria) crist~e, and the increased humor of cyfoplasmie elu~ter~sof RNP t~articb~ a~ ~uq~mt U~ previous ele~tron micmgraphs. ~Dm granules show a finely partlcutat~ matrix in which at,. db,l u . ~ i clcetwn.den~ particles. Epon emlx,dding. Electron microgl~ph. (,.~2q:t,~D}

in)

(b)

(a)

¢e)

if)

(g)

I~I.AT~ 4,(a}~(h) El~,etru~tt mier~raphs o f c a l f h y a l ~ c y | e granuies showlt~g the variation in the appear* a~¢~ ~,f these utr~Jctur(,s. The n~ierog~ph.a ar~ arrang(~ in a hypothetical scqueitce from the ai~np|e atruct~tn~ (a) t#~ the cozllp|ex granule (g). Epo~ embrddi¢~g. Fixation: osmiu~n tetroxidc and auct'ose. Magntti¢~titma: (a} >.49,fgg); (h) and (h) X:~5,(~g); (c) and (f) x34,0~3; (tl) x31,fl(X); (e) ~24,~(1:

PLATS.5. Electron microg~ph of a calf :ed in osmium ~tm×lde ,apor. Fhe hyatoeyte graimI~ ~tand out well and exhibit vario(zs structures. Cyt~plaamic membmm~ am poorly preserved, as a~ most mltochondrud membratles. ~ cstcular straettm, s in~ the eytop|asm ate ~++~n, aithougli in small numbel~. Note the ovemit density of the cyt~ptasmie gwuud suimUmc<+, Epoa e m i l . ding. ( × 16,(~,))

t'l,,~T~: 6. Calf hyalocyte fi,xed in glutar~ldehyde ~apor follo~'ed by osmium ~etmxide vapor. Them is littJe cytoplasmic detail to be seen, as the cytoplasmic, membmao struvture is poorly preserved. Some gt'~vmle membranes show up fairly well, however, and subgranular structures are visible, although the ovet'alt granule del~sit.y is highly deereaset:l. RNP particles stand out Clearly. Note fl~e ~otai absence of cyt,op/asmie mici'ovilli. Epon embedding. ( × 19,~0)

ELECTRON

~ICROSCOP~Y

OF HYALOCYTES

2~51

~ e e n d o p l ~ c retic~um m not cor~spi~uous and is~f bpth a rough and a smoothsurfaced type. The la~er t~pe generally presents itself in ~he form of I ~ l l e l membranes studded with RNP particles, m~c~ting q em~s-.~ction~ ~ecuJar strutting. The smooth, ed reticulum may also show this appearance, but it a p ~ m more often in ~he form of complefMy rounded profiles. ~ u s ~ r ~ , rosette-s, or isolat~t ~RNP particles are also ~equently observed in the ey~pl~..0Cdntr~oles have been ~ e n in numerous hyM6pytes. Present in the hyaloeyte cy~6ptasm~ are numerous ves~icles and. vacuole~ (Plates 1-3). The former generally lie ~ithm the size rahge of 5 0 - 2 ~ m/~,~wheveas the latter range ~n uize from 0.2 to 2-5 ~, The ve~(;~Ies, some of wbAqh ~ e of the "Coated" t y ~ , can be observed at various )]epCns h ~ t h e c~oplasm~ e~s we~ as in:intimate connection ~ t h the cell membrane. ~ome'~;)p~ar to be formed f r o ~ the plasma membrane some vacuoles may contain ~resie~ar struc~yes~'appr ely 50 m~ in diameter, as well as dense particles me~uring around 30.rn.~ (Plate 1). and appearange~ Nur~;oering.an~here from 1 t~fl7 m.a.cross-sectioned cell, the sectjone~ ~ a n ~ e s measure from Q,2 ~ 2.0/z in diarn~er. *£he.predominar~ type of dxtr~emely fine particulate mass of moderate dectron density in ~'hich~appe~.r irreg}llarI~y~cl~spersed, ro~,rnded electron-opaque parti~es, varying in size from.25 to-60 m~ (Plate 4(a)-(c))or n~lbre (Plate 4(b)-(e)). The latter, in some granules, m~y accl~rmlate in increased numbers at ~he periphery, as v
252

G,

D,

BLOOM

AND

E.

A.

BALAZS

sucrose has been added (Plates 1-3). With the former techniqne, the overall density of cytoplasmic and mitochoudrial mernbranes, ribosomes and Golgi vesicles is increased as compared to the same stru s ~ c d in osmium de alone. F:~rthermore, the mitochondzial cristae, or lamellae, appear thicker after glutaraldehyde fixation, and the m a t e r i ~ enclosed by the cris~e-limiting membranes is m denser ghan the interlamellar substance, a feature not pronounced after fixation vAth osmium tetroxide alone. With the latter t~chnique, peripheral raicrovilli and cytop l ~ m i c vesiclSa appear to be fa-z more numerous than after glutara fixation, but ribosomal clusters do not appear as fre(pmntIy. The membrane-bounded cavities of the endopla~smic reticnalum appear to be considerably wider when gl ehyde is employed. There is little difference in the appearance of the hyalocyte granules, and the substructures of the ~ a u u l e s are visible after both fixation techniques. When glutaraldehyde vapor followed by t r e a t m e n t with osmium tetro:cAde v-apor, or the latter vapor alone is employed as fixative, the picture ~ again different (Plates 5 and 6). As compared wi~h fluid fixation, the most pronounced difference lies in the poor preservation of membrane structures. The endoplasmie rc4ieulum, for example, is recognized b y the alignment of ribosomes rather t h a n b y the actual app e of the parallel membranes to which the ribosomes are attached. Vacuolar, vesicldar, ~ a n u l a r and rnito3hondriabmembranes are seen, b u t t h e y are much less distinct t h a n after fluid fixation. The c3~oplasmic ground s u b . a n t e , however, invariably appears denser ~o_ vapor-fixed specimens. A comparison of the two types of vapor fixation reveals t h a t only the ribosomes stand out equally well in both. _AAIother s~uctures, and especially the yte granules, show diminished electron-scattering properties when glutaratdehyde vapor precedes osmium tetroxide vapor in the fixation process. Moreover, with the latter techniq'ae, cytoplasmic vesicles and vacuolar structures appear greatly diminished in number, and microvilli are almost totally abserrt {Plate 6). 4. Dis

I t was shown earlier t h a t certain cellular elements of the vitreous body, the h:~w~locxtes (t~alazs, 1959), form an entity with characteristic cellular features (Szirmai ~md Balazs. 1958; Balazs, Toth, Eckl and Mitchell, 1964b). I n the ligh~ and ~ a s e contrast microscope, these cells have been studied extensively in nnmerous animal species, such as owl monkey, squi~el monkey, cattle, sheep, dog, cat, gldnea pig, rat, mouse, hamster, rabbit and domestic chicken (B~azs et aL, 19641)). The c~lls are characterized primarily b y their content of v a ~ n g numbers of cytoplasraic granules, which have m common such s~uining features as orthoeb~comatic basophilia, PAS positivity, and vital s t a i p ~ g with neutral red. ermore, t/he ~ a n u l e s exhibit an intense red autofluorescence when subjected to ultraviolet 1L~nt. This aut~ofluorcscence rapidly changes to yellow and, finally, to green before fading during continued illumination. Among certa'm other cellular features, the h y a s show amoeboid movement and exhibit phagocytic properties when incubated in vitl'o with coltoidM iroli pe,rtieles, for example (Balazs et al., I964b). ~ t e r fixation and embedding in Epon, the calf ]lyalocytes generally maintained the r o u n d ~ shape so characte~stic of these 6elAswhen observed in the stat~ in the intact vitreous gel b y means of phe~se-cont-ra~ micqroscopy. Only occmsionally did a c e m d n p o t i o n of the c ~ o p l a s m form a shorter or longer sion. I t has previously been sho,,~: by cinephotomicrogvapllic techniques t h a t the cell surface of l i ~ n g h y a l ~ : ~ e s exhibits dynamic changes. These are characterized by undulating movem c n ~ of the cell membrane and formation and disappearance of c y t o ~ a s m i c extru-

E L E C T R O N M I C R O S C O P Y O F HYALOCYTI~S

2~

sions (Balazs et el., 1964b). The presence of fine c)~oplasmie protnlsions, or microv~lll, a t the p e r i p h e ~ of the fixed ceils m a y reflect this activity of the cell ~ u n d a r i e s . Furtiler support for such a supposition is given by the ~ of numerous cytoplasmic v~icles, and even vacuoles, in close p r o b i t y to the plasma membrane, as well r~ l~;the pinocy~tie vesicles seen to arise from the membrane iteel£ I t is a l ~ int~:~ting to note, with respect to the %vtoplasmic mierovilli, ~ a t similar structures have b ~ n shown to be characteristic of another granular connective t i ~ u e celt, the mast cell {Bloom, Friberg, and .~berg, 1955; Bloom, Friberg and l~ar~.son, 1956). A characteristic of calf hyalocy~s, whmh is emphasized by the presenL findings, is their content of cytoplasmic granule structm'~, These gramfles, although varying in u!trastructure, appea~r to form a distinct entity. The number of granules per cell, however, shows large variations. Therefore, in the thin t i ~ u e ~ections employed in elec~on microscopy, cellular sections of certain portions of hyMc~ytes m a y sho,x a tetra lack of cytoplasmic granules. This is undoubtedly why n (1 ~ 3 ) exerts gn,eat caution in interpreting his electron mierograph of a "presumably true vitreous cell" in the eye of a monkey. Although ~he hyalocyte granules v a r y markedly in appearance, no evidence has been obtained for the presence of ~fferent ~ a n u l e populations frc~u the viewpoint of their origin. On the contrary, cerV-~in basic features are very similar in g~mule~ of otherwise ~ e a t ] y var3~ing appearance, wtfich, in turn, suggest~ t h a t we are dealing ~'ith hyalocyte ~ a n u l e s a t various stages of development or of m e , belie activity, or both. Our tip_dings provdde no evidence t h a t the c y t o p l ~ m i c ~ a n u l a r s~uctur(~s are phagocytized pigment ~ a n u l e s ( S ~ m a n and Storm, 1965) or phagocytized inclusion b ~ h e s of unknown material (G~trtner, 1965). T:he fact t h a t hyalocytes have been shown exhibit phagocytic properties under ce[~ain circumstances (Szirmai and Balazs, 1958; Hamburg, 1959; Balazs et ~., I964b; Seaman and Sh~rm, 1965) he~s perhaps overshadowed the f~ct t h a t these cells are basically a type of granular connec/tive tissue cell. The process of granule formation is not clear. From numerous electron micrographs, however, it appears ~hat the variations in the fine str~acturc of the granule m a y simply represent different stages in a cycle of events taking place in the granules. A hypothetical sequence of these e~.~ents is depic~d in Plate 4(a)-(h). These electron rnicro~aphs of hyalocyte granules taken from different cells are arranged in sequence from the simplest ~ the most comple.," structure. In cytoplasmic vacuolar structures, a finely particulate m a t t e r aceumule ~ until a spherical gramlle, completely filled -with this matrix substance, is formed. Dense particles of sma~ dimensions appear in this matrLx, often locMized a~ the inner surface of the gran~e-delimit~ing membrane, The dense particles increase in size and number and sometimes fill ~he granule, lt'ithln the l n a t r ~ , other subu~At~ form. They appear as rounded areas, oft~zn with the dense particles located peripherally. These subgranular areas a ~ separat~
~4

G,

D,

BLOOM

AND

E.

A,

BALAZS

These observations are in no w a y conclusive, b u t t h e y d o s u g g e s t the l=~)ssibil:rty t h a t such v ~ u o l a r structures m a y represent early s a g e s of ~ a n u l e formation. I t is interesting to note, in this respect, t h a t Seaman and S ~ ( 1 9 6 5 ) r e ~ e d vacuolar s t r u c t u r ~ in fowl h y a l ~ y t e s which weremore or less filled with amorphous substance. They observed, moreover, a connection between such a vacuole and the endoplaarnic reticLfium and suggested t h a t these vacuoles represent ~ a t e d cist~rnae of the endoplasmie retleulum in which s e e r e ~ r y products accumulate. Their observations did not shed any light on the possible role of the Golgi apparatus m such secretory process. This organelle was, furthermore, d~eribed b y them as small at~d poorly developed in the fowl hyalocyt~. The Oolgi zone in the ~ l f hyalocyte appear~ ~'~irly well developed, with large numbers of vesicles in and a r o ~ d this region, i n addition, vacuoles in the vicinity of the Golgi apparatus frequently contain vesicular structures together with the amorphous material. At present, however, it is not po~ible ~ define, wiCn any d e ~ e e of ce~ainty, the origin of these vesicles. The poor preservation of hyalocytes after fixation in osmium tetroxide and embedding in methacryIate is undoubtedly clue to damage c a ~ e d b y polymerization of this pl~tic. ~ine cytoplasmic d e v i l was invariably destroyed to a ~ e a t extent. This polymerization effect is now well estabhshed (see Pease, 1964, for renew)..dznong the published reports on hyalocyte stm~ct~e, only in t h a t of Seaman and Storm (1965) has m e t h a c r y l a ~ been employed as an embedding meditate I t ~s r~'~t possible, however, to compare their r e s ~ with ours, as those authors also use~.~ ).~[araglas, and nowhere in their paper do t h e y state which emberl~dAng medium was employed for the cells shown in their electron micro~aphs. Embedding in epoxy resin ( E ~ n 812) gave consistently satisfactory results, which, as was expected, varied only with the iL~ation procedure employed. The f i n i n g t h a t the overall density, of cellular membranes, mitochondxial cristae, ribosomes and Golgi vesicles increases when glutaraldehyde ~xation precedes osmium tetroxide treatment is in g o ~ agreement with the results of Sabatini et al. (1963), who introduced this mode of fixation for electron n~n~croscopy. The slightly widened cavities of the endoplasmic retictflum wi~h aldchyde iL~atives is pointed out by Pease (1964). I t is interesting ~ note t h e increased nu~]ber of cytoplasmic vesicular structures apparent after osmium ~ t r o x i d e ~ a t i o n alone. A satisfaetoD~ explanation for this cannot be given a t this t ~ e o The in~rcellular matrix surroundfiug the hyalocytes in the cortical tissue layer is rich in hyalurondc acid. F ' u ~ h e ~ o r e , the presence of hyaluronic acid or other carbohydrate-containing raacromolec~e~ in the hyaIocytes is also likely. The ~ a n ~ e s of these c e ~ exhibi~ strong PAS positivity and b a ~ p h ~ e properti~. I t has also been shown t h a t when living caff hyalocytes are incubated in vitro with [~C] glucose, an incorporation of the isotope into the hyaluron~ic acid fraction of high molecular weight ~occurs (Balazs, Sundblad and Toth, 1958). This was interpre~d to i n , c a r e t h a t the h y a l o c y ~ s are inst~umen~l in the synthesis of t ~ s glycosaminoglycan in the vitreous. GlyeosaminogIycans, especially hyaluron~ic acid, are soluble in the aqueous fixatives conventionaUy used for" electron m i c r o c o p y , as well as in the dehydrating agent.s alcohol or acetone in ~ n e e n t r a t i o n s less than 60-70%. In an effort ~ prevent lo'xs of the polysaccharide material from the hyalocytes, vapor ~ a t i o n , followed by dehydration in a series of ethanol concen~ations, sta~ing at 70%, was attempted. Neither after fixation ~ t h gl~taraldehyde and osmium tetrorAde vapor, nor after the latter alone, did the fine structure of the cell stand out in the ~ m e manner as it does after L~rnersion ~ a t i o n . Vapor fi~xation thus dad not add struc~.ural detail to the

ELECTRON

MICROSCOPY

O F HYALOC~'T~S

hyalooyte piece. It is unquestionable, however, that the cytopl~.j¢ ~ m d ~ b . s~flce was much deri~r after vapor fixation, indicating that c~|lu~r ~t~i-~an~ moved by fluid fixation are mainmin~:l when fixation by vapor and deh~atioa starting at high ab~_~ol eon~nt~tiozm are ~pIoyc~l. Th~ imlm~ant o ~ r , ~ a suggests that the ~t~er ee%h~/que m ~ prove ~ ~ an ~mF.o~ t addition ~ the conventional h-nmersion"~e~hmques, especially i ~ f a r as subcellular the ACKNOWLEDGMENTS

The ~ehnical assistan~of Mrs. Anja Mitchell and ~Lr~. Marga~tha J a ~ n aokowledged.

i~ gratefu~y

REFERENCES Balazs, E. A. (1959). A m X V I H Co~w.ilium Optha m, ,1958, I L I ~ 5 . Balazs, E. A., Toth, L. Z. J., Eokl, E. A. and Mitolmil. A. P. (1964b). Exp. E#~ Re,. 3, 57. Balazs, E. A., Mitchell; A. and Eckl, E. A. (1964a). Iave~. OphthaL 3, 473. Balaza, E. A., Sundblad, L. and Toth, L. Z. J. (1958). Fed. Proc. 17, l ~ . Bloom, G., :~iberg, U., Lar~gn, B: and AL~rg, B. (1955). Acta Palhd. Micrab~. S~ad. 37, 163. BI~m, G., Friberg, U. and Lar'~son, B. (1956). Nord. Veteritmermed. 8, 43, ~%nck, H..(1960). d, biophya, ~iochem. CytoL 7, 27. G~nner, J. (1965). Z. Z~for~c,~.(in press). Hamburg, A. (1959). Opht~,almoloffica, BazeZ 138, 81. Hogan, :SL J. (I963). I~veat. O p h i r . 2, 418. Palade, G. E, (1952). J. expl. Med. 95, 285. Pease, D. C, (1964). Hi6toS~9ic~l Tez'Aniqu~for El~.r~rol~ . Aaademio P ~ , New Y o & Reynolds, E. S. (1963). J. Cdl Biol. 17, 208. SabatLai, D. D , Bonaeh, K. and Barrnet% R. J. (1963}. J. Cell Biol. 17, 19. Seaman, A. R. and Storm, H. K. (I965). Exp. Eye Re~. 4, I3. Szirmai, J. A. and Bala~, E, A. (1958}. Arch. Opi~llv~. 59, 34.