Toxlroo VoL 29, No . 6, pp . 569-380, 1991 . Printed io Orat Hriuin.
00"I-0101/91 53.00+ .00 ~ 1991 Paymon l~ pb
MORPHOMETRIC STUDIES ON VENOM SECRETORY CELLS FROM BOTHROPS JARARACUSSU (JARARACUÇU) BEFORE AND AFTER VENOM EXTRACTION* S. M.
CARNSIRO,' V . R . PINTO,' C. JARED,~ L. F . P. FARirA3 AND A. SESSO3
A. B. M.
LuL.A,'
'Seçâo de Venmos, and ~o de Microscopie Eletr6nica, Instituto Butantan, Av . Vital Bratil, 1500, CEP 05504, Sao Paulo, 3P, Bread, and 3Laborat6rio de Patologia Molecular, Departamento de Patologia, Faculdade de Medicine da Univeraidade de Sâo Paulo, Sâo Paulo, Bread. (Received 28 June 1990; accepted 30 October 1990)
S. M. CARNEUto, V . R . Plrrro, C. JARED, L . A. B. M. Lur.A, F . P. FARIw and A. SESSO. Morphometric studies on venom secretory cells from Bothrops jararacussu (Jararacuçu) before and after venom extraction. Toxicon 29, 569580, 1991 .-A comparative morphometrical analysis was carried out on secretory cells from Bothrops jararacussu venom glands, before manual extraction of the venom (milking) and 4 and 8 days after milking. At the 8th day after milking, the cytoplasmic volume increased by 160%. The rough endoplasmic reticulum (RER) volume density increase, up to the 8th day after milking, is mainly due to widening of the infra-ßcisternal space. The total volume and membrane surface of the RER, Golgi apparatus and subcomponents, secretory vesicles and mitochondria, increased during the experimental period while the volume and surface densities of these organelles, with the exception of the RER, did not vary. The numerical density of Golgiassociated microvesicles per Golgi volume unit also increased. The greatest relative increments in these parameters occurred within the first 4 days. These results are compatible with an increased rate of membrane synthesis and transport in the milked glands and suggest that the membrane biogenesis, degradation and circulation that takes place in the first week after milking is achieved through coordinated cellular mechanisms that maintain the rate between total membrane surface and total cytoplasmic volume unaltered. INTRODUCTION
Tim vEr1oM glands of viperid snakes are stimulated for a new cycle of venom synthesis and secretion after discharge of the venom stored in the gland lumina . Biting may promote a partial discharge while a more complete extraction can be achieved manually by milking (KOCHVA et al., 1975 ; KOCHVA, 1987). Concurrent with an increase in the synthesis of RNA (ROTBNBERG et al., 1971 ; Da LuccA and 1MAIZLTMI, 1972) and pmtein (ORON and BDOtAx, 1973 ; DE LuccA et al., 1974) 'Part of these findings was presented at the I2th ColGquio da Sociedade Brasileira de Microscopia EletrSnica, Caxambu, Brasil, Setembro, 1989.
570
S. M. CARNEIRO et al.
FYa. 1. Ux~n vetvo~ awvn of Bothrops Jararacussu . Note narrow RER cisternae in the secretory cells (Sc); Golgi area (G); vacuolated vesicles horizontal cell (H); lumen (L); connective tissue (C). Bar: 2 pm.
M;
there is a large increase (up to five times) in the height of the secretory epithelium until the 4th to 8th day after manual extraction . The secretory cells which appear flattened or cuboidal in the replete gland, gradually become columnar (KocxvA and GArrs, 1967 ; Bl?PiSHnuL et al., 1971 ; ORON and Bl)OLAH, 1973); the intracisternal spaces of the rough endoplasmic reticulum (RER) widen (B>?rr-SHAUL et al., 1971) and the Golgi apparatus becomes more conspicuous (ORON and BDOLAH, 1973). In milked glands, the venom transport to the lumen is greater than that of resting glands (Kocxvn and GANS, 1967; ORON and BDOL,~I-t, 1978). In order to analyze the configurational changes occurring in the cellular compartments involved in the synthesis, condensation and transport of venom secretory proteins, after venom extraction, the volume and total membrane surface of these compartments were morphometrically evaluated in venom gland secretory cells from Bothrops jararacussu at the resting stage (unmilked and not fed) and 4 and 8 days after milking . MATERIALS AND METHODS Snakes acrd venom glands Eleven Bothrops Jararacrctsu snakes, 2.5 months old, not fed since their birth, weighing 7.1-11 .8 g, 27-30 cm
long and maintained at 226°C were used . All were from the same litter . Venom glands from three unmilked snakes, from four snakes milked 4 days previously and from four snakes millced 8 days previously were studied. Venom glands were removed after killing the snakes by decapitation . Pieces of about 1 mm3 were obtained from the median part of the main venom gland; they were food using 1.5%
Morphometry of B. Jararacycssu Venom Cells
57l
FYe. 2. U~n~ren vexoit cr wn of Bothrops jararacussu. Stacked narrow RER cisternae (R) in the basolateral cytoplasm of two secretory cells. Plasmalemma foldinge (P); Golgi area (G); mitochondria (m). Arrow points to a thin cytoplasmic projection of a horizontal cell . Bar. 0.5 pm . glutaraldehyde and 1 .0% paraformaldehyde in 0.08 M cacodylate buffer, (pH 7 .3) for 2hr and 1% osmium tetroxide in the same buffer, for 2 hr at 4°C. The specimens were embedded in Spurr's medium (Sruwe, 1969).
Nuclear acrd cytoplasmic voGr»te measurements Semithin (0 .51an) sections were examined in a light microscope with an attached camera lucida equipment. The nuclei of the projected secretory cells had their contours outlined over a magnetic card with an appropriate cursor. The area measurements thus obtained were transferred to a computer system to be processed. For each gland, 300 nuclear areas were measured. The mean nuclear volume was estimated in two ways: (a) by the BACH (1%3) procedure which is applicable to spherical objects; (b) with aid of the Lrrro~ea-VottwEare (1970) formula, suitable for elliptical objects. The cytoplasmic volume was obtained from the nuclear volume (Vn) and the nuclear volume density in the aecretory cell (Vvn) estimated by the point counting method (W~eEr., 1979) for semithin sections and corrected for the Holmes effect as indicated by At~tr~ and Dtnvrm .t (1982) .
Evaluation of morphometrical parameters in subcellular structwrs Twenty electron micrographs were randomly taken from each gland at 2600 x on 35 mm negative film and analysed on 18,200 x prints. Whenever Golgi stacked saecules were.seen, one or more exposures were taken at a magnification of 5500 and studied in micrographs at a magnification of 38,500. A test system of the Fs~eeW~m. type (1%7) was applied to each of the prints and the following structures were measured: rough endoplasmic reticulum (RER), Golgi apparatus (on the 38,500 magnification prints), mitochondria, secretary vesicles, basolateral and apical plasmalemma, and "other structures" ouch as lysosomo-like profiles, multivesicular bodies, phagosomea, etc. The cytoplasmic volume density (Vv), the surface density (Sv), the surface to volume ratio (s/v) and the numerical density of Golgi microvesicles (Nmv) were estimated as described by C~ttt~eo and Ses'o (1987). The relative standard error (e) associated with volume density (Vv) estimations were calculated by the formula: e~
1-Vv/P . Vv
where P is the total number of points counted over the measured cell cytoplasm. The actual volume (V), the total membrane surface (S) of each subcellular compartment, and the cytoplasmic numerical density of Golgi microvesicles were obtained from the cytoplasmic volume and Vv, Sv and Nmv measurements, respectively. The mean and standard error of the mean are expressed on the tables . Differences between experimental groups were detected with the aid of the Dtnvcex (1955) test.
572
S. M. CARNEIRO et al.
FIG. 3. EIGHT DAYS FOLLOWING MILIUNG OF BOtlllOpS farOaCIttJtt VENOM GLANA. Note the different RER cisternae (R) dilatation patterns of two adjacent celle. Phlsmalemma (arrows); Golgi area with vesicle M; mitochondria (m). Bar: 0.5 ~.
RESULTS
Morphological observations
Under the light microscope, the secretory epithelium of the unmilked glands is predominantly composed of cuboidal or flat secretory cells, about 6 pm height . Occasionally, higher columnar cells are noticed in some tubular transactions . Secretory vesicles or vacuoles are observed mainly at the apical region of the cell, or replenishing some cells. On the 4th and 8th days after venom extraction, the height of the secretory cells has increased; they now become columnar shaped, c. 30 ~m height. Secretory vesicles are noticed at the supra nuclear region . The cytoplasm has a vesiculated appearance, mainly at the apical region . The tubular lamina are narrower and less regular than those of the unmilked glands . A few dense chromophilic granules and some metachromatic inclusions can be observed in both types of glands. At the electron microscopic level the cytoplasm of the unmilked glands is filled with narrow RER cisternae (Figs 1 and 2). Secretory vesicles are not commonly seen at the tracts side of the Golgi stack; nevertheless, secretory vesicles and translucid vacuoles presenting degenerating signals, such as myelinated figtu~es, can be observed either isolated or filling the cytoplasm of some secretory cells. In general, distinct condensed electron dense secretory granules are not seen . Occasionally, lipidic inclusions are observed. Near the apical plasmalemma ,microvilli, at the luminal side, microvesicle-like profiles about 7100 nm diameter are noticed. Basolateral plasmalemma foldings are commonly seen (Fig 2). On the 4th and 8th days after extraction of venom, the morphological features of the secretory cells have similar characteristics but they differ from those from unmilked glands mainly in regard to the expansion of their RER membranes and the widening of the RER intracisternal space (Figs 3, 4). The distribution of the ribosomes on the dilated cisternae is not regular. Near the Golgi apparatus cis side, microvesicles pinching off from the transitional endoplasmic reticulum can be observed. Secretory vesicles and dilated
Morphometry of B. jaraiâcuwu Venom Cells
573
FiO . 4. EIORT DAYS AOLLOWINO lat .CR~iß OF BOrlVOpS J¢I~QCLSTY VENOl1 GLAND. Cohmmar cells presenting supra nuclear Golgi area (G); secretory vesicles (S) ; and dilated RER cisternae (R). Horizontal cell (ü); connective tissue (C); lumen (L) . Bar: 2 gym .
cisternae of the RER, which sometimes exhibit extensive areas devoid of ribosomes, can be observed at the inner side of the well-developed semicircular Golgi apparatus. Microvesicles about 60-80 nm diameter are numerous and multivesicular bodies occur at a greater frequency than in the unmilked glands (Fig . 5). At the luminal plasmalemma, microvilli and associated microvesicle-like profiles are also seen . These structures are more easily observed in regions where the plasmalemma of two cells are facing each other, apparently trapping the microvesicles between them. Exocytotic figures were not observed in these glands. In order to simplify the presentation of morphometric data, possibly degenerating or sutophagic vesicles, translucid vacuoles, multivesicular bodies, lysosomelike granules and a few lipidic inclusions were grouped under the designation "other structures". The morphological features described are the most frequently observed, but intermediate characteristics ca,n also be seen in all the glands studied.
574
S . M. CARNEIRO et at.
FIO . S . FOUR DAYS POLLOWINO ML1C1N0 OF BOtIVOpS JgrarQCUS4Y VENOI! OLAND. Note dilated cisternae (C) at the cir side of the Golgi apparatus . Transitional RER (arrows) . Multivesicular bodice (B). The RER cisternae (R) enclosed in the tan's Golgi side exhibits membrane areas devoid of ribosomes. Har. 0 .5 pm .
Quantitative results
In the glands examined, the secretory cell nuclear shape was not uniform. About 70% of the nuclear transections (part of them pertaining to isodiametric cell profiles) were TABLS 1 . NUCLFAA VOLUlIBS, NUCLSAR VOLUl18 DElVaITY AI~ COARHBPONDINO CYTOPLASiOC UNIDI~ AI~ FROY èla~ VBriD!( OLAI~ 88CASPORY CHLIs
VOLUI®3 FROI(
DayB Bftel VenOm extraCtiOII 0 Nuclear volume density (~'/~~ Nuclear volume (~m~ (LV) Nuclear volume (pmt (B) Cytoplasmic volume (pmt (LV) Cytophunnic volume (tcm~ (H)
0.167 t O .OOSt (0 .065)' 48 t 4 49 t 8 242 f 25 246 t 36
4 0.11 S t 0 .005" (0 .070) 74 f 8" 80 f 14" 572 f 68rr 614 t 100"
8 0.094 f O.OOSr',$ (0 .072) 61 t 3 67 f 7 588 f 28rr 640 f 50r'
Results are shown as mean f S .E., n ~ 3 for 0 day and 4 for four and eight days . Nuclear volumes estimated by the Lindberg-Vorwerk (LV) and by the Bach (B) procedures . 'Relative standard eaor associated with density volume estimates . ''Significantly different (p < 0 .05) from wntrol (0) . tThese values were corrected for the Hohnes effect as indicated by A~er~ and Durnvn.L (1982). Noa corrected data are 7r/o higher . $Significantly different (p < 0 .05) from 4 days .
Morphometry of B. Jararncuasu Venom Cells TABLE
2.
VoLUSm
375
nws1~ (~an 3/Nm~ o~ crroru.9ac oo~or~rts r~ vivtm~ ~n m~ tm.s~ vartor OLAF ~CASrOAY ®.is Days Si1Et Venom extraetlOn
C~toplaemic component Rough endoplasmic retiwlum Sxretory vesicles Golgi ciaternae Golgi apparatus associated microvesicles Golgi apparatus Mitochondria "Other structures" Apical miaovilli aad associated microvesicka Cellular matrix
0
4
8
0.173 f0.024 (0.045)" 0.08910 .037 (0.075) 0.034 f0.004 (0 .111) 0.003 t0.007 (0 .291) 0.107 t0.017 (0.060) 0.049 t0.003 (0.090) 0.01St0.003 (0 .181) 0.020 f0 .004 (0 .147) 0.540 f0.025 (0 .019)
0.281 f0.017"' (0.030) 0.030 t0.004 (0.082) 0.043 f0.004 (0.087) 0.007 t0.002 (0.243) 0.123 t0.011 (0.050) 0.036 t0.004 (0.077) 0.009 f0.002 (0.212) 0.007 t0.002 (0.230) 0.47310 .020"" (O.Ol9)
0.338 t0 .008 "',t (0.025) 0.06410.006 (0.069) 0.033 f0.003 (0.986) 0.006 f 0.001 (0.232) 0.10010.003 (0.054) 0.037 f0.003 (0.073) 0.007 f0.018 (0.228) 0.009 f0.002 (0.190) 0.42410.005 "" (0.020)
Results are shown as mean f S.E., n = 3. fot 0 day and 4 for four and eight days. "Relative standard error associated with density volume estimates. ""Significantly different (p < 0.05) from control (0). tSignificantly different (p < 0.05) from 4 says .
round in shape. The remaining profiles, mainly seen in columnar cells, appeared to be elliptical . It was therefore decided to evaluate nuclear volumes using two procedures. The BACH (1963) method enables the transformation of the distribution of the radii of circles measured in the nuclear transections, into distribution of radii of spheres, correcting for the effects of section thickness and for the lost small nuclear caps. The LINDH$RCr VORWERR (1970) approach uses the formula
TABLE
3.
TOTAL voL~
(pm~ oF
crrorLASiac
ooseor~rts
Days after venom extraction Cytoplasmic component Rough endoplasmic reticulum Sectetory vesicles Golgi cisteroae Golgi asso~iated microvesiclea Total (iolgi apparatus Mitochondria "Other structures" Api~al microvilli and associated microvesicles Cellular matrix
0
4
8
44 f 14 20 f 6 9f2 1 f 0.4 27 f 8 12 f 2 3 f1 Sf I
171 f 27"" 30 f4 27 f4"" 4f 1 "" 73 f 7"" 34 t 6"" 61 2 Sf 2
215 f ll' " 42 f 7"" 21 1 3"" 41 1 "" 67 f T" 37 t 6"" 511 6f1
131 f l9
293f39""
272 f 22""
Results are shown as mean f S.E., n - 3 for 0 day and 4 for four and eight days. ""Significantly different (p < 0.05) from control (0).
S. M. CARNEIRO et d.
S76
TABLE 4. SUAFAr~ TO vOLl1YE ItAT10 ~r//tm~ OF CY1'OPLASIlIC C0IBONEN79 Days 81~Cr VCDOm e7rtraCtiOn GjRoplaamic component Rough endoplasmic reticulum Secretory vesicles Golgi cisteraae Golgi associated microvesicles Mitochondria "Other structures"
0
4
8
45.914.5 4.210.2 55.613.E 88.312.0 12.910.9 L710.6
22.911 .3"" 4.510.3 50.114.5 9L2114.5 (0.810.6"" 3.511 .0
19 .611 .0 "' 3.210 .2 51 .017 .E 88 .91(0.9 9 .210 .2'" 6.213 .3
Results are shown as mean 1 S.E., n = 3 for 0 day and 4 for four and eight days. '"Significantly different (p < 0.05) from control (0).
TABLE 5. $URFACE aErvsrrv (pm'lPm') oF crroPLAS~rc oot~oHErvTs Days after venom extraction Cj~tophismic component Rough endophrsmic rtticulum Secretory vesicles Golgi cisternae Golgi apparatus associated microvesicles Mitochondria Apical rnicrovilli and associated microvesicles "Other structures" liasolateral plasmalemma Apical plaiaDalemma
0
4
8
7.7710.66 0.3610.13 1 .9110.35
6.4010.30"' 0.2210.02 2.2310.27
6.5910.25 0.0210.01 1.6110.13
0.4510.05 0.6410.07
0.5410.03 0.0610 .02
0.5510.08 0.5310.03
0.5910.16 0.0210.01 0.8010.08 0.3210.03
0.2310.07 0.0410.01 0.88 t0.07 0.1410.02
0.3310.02 0.0210.01 0.5310.08 0.1810.03
Results are shown as mean 1 S.E., e = 3 for 0 day and 4 for four and eight days . ""Sigai&antly different (p < 0.05) from control (0).
TABLE 6. TOTAL è®lBRANE SURFACE (/tm~ OF CYIY)PLAS~UC COLlPONENTS Days after venom extraction Cytoplasmic components Rough endoplasmic reticulum Secretory vesicles Golgi cisternae (1) Golgi associated microvesicles (2) Total Golgi apparatus (1)+(2) Mitochondria "Other structures" Basolateral phrsmalearrna Apical plasmalemma Apical microvilG and associated microvesicles Total measurod membranes
0
4
8
19481467 84124 4941163 114130 6081194 157129 51 I 193127 7016
39001608 "" 134116 I3I71I59 "" 330153" 16471196"" 363148 "" 23112 3391106 72110
42261387 "" 133120 10361128'" 349137"" 13861152 "" 338128"" 1418 347179 99111
146143 32181739
140153 68321905""
212120 67701629 ""
Results are shown as mean 1 S.E. n ~ 3 for 0 day and 4 for four and eight days. ""Significantly different (p < 0.05) from control (0).
Morphometry
of B. jararacussu
57 7
Venom Celle
TABLE 7. NU1®IICAL DENSITY (N) OF GOLCiI A390CIATED MICROV~IICLES PER VOLU1~fE UNrr (~~ OF GOLCiI APPARATUS AND TOTAL NUMBER (N!) OF CiOLöI ASSOCIATED 1fICAOVESICLO IN Tlni SECREPORY CELL CYTOPLASM liHye Sttei YCIIOm eRtraChOII
N Nt
l47f30 4410 t 1748
Reeulb are shown ae mean t S.E. n = 3 for 0 day and "SignißcantlY different (p < 0.05) from control (0).
179t8 I3,160 t 1466" 4 for
215t 13" 13,680 t 1340"
four and eight days .
where v is the mean nuclear volume, s is the mean profile area and ß is a conformational coefl'tcient depending on the ratio (e), between the major (a) and the minor (b) axes measured on elliptical profiles. The mean nuclear volume and derived cytoplasmic volume, obtained by both approaches, are very similar (Table 1). To calculate the stereological parameters involving the cytoplasmic volume, such as the total membrane surface etc., the data obtained using the BACH procedure was utilized . Four and eight days after milking, the cytoplasmic volume of the venom gland secretory cell is, respectively, 150% (368 pm', p < 0.05) and 160% (394 Ealt', p < 0.05) greater than in the unmilked gland. The nuclear volume increases by 63% (31 pm', p < 0.05) on the 4th day as compared with unmilked glands (Table 1). Organelle and cellular matrix volume densities in the cytoplasm are shown in Table 2. The increase in the RER volume density and total volume (Table 3) after venom extraction is mainly due to the widening of the intracisternal space. Four and eight days after milking, the RER surface to volume ratio (Table 4) declined significantly and the RER volume was respectively 289% (127 pm', p < 0.05) and 389% (171 pm', p < 0.05) greater than in the unmilked gland. This partly accounts for the observed increase in cytoplasmic volume . Reduction in RER surface density (Table 5) of 18% (1 .37, p < 0.05) is observed at 4 days after milking, in spite of an increase of 100% (1952 pmt, p < 0.05) in its membrane surface in the same interval . On the 8th day after milking, the RER surface density difference is not significant while its membrane surface increased by 117% (2278 pmZ, p < 0.05). The volume densities (Table 2) and the surface densities (Table 5) of the Golgi apparatus and sub-components (cisternae and Golgi-associated microvesicles) and of mitochondria did not vary significantly over the experimental period, while the total volume (Table 3) and total surface membrane (Table 6) were significantly increased after milking . The greatest relative increments in the values of volume and membrane surface occurred within the first four days after milking; between the 4th and the 8th days these stereological parameters did not vary significantly . The Golgi apparatus (cisternae plus microvesicles) volume increased by 170% (46 pmZ, p < 0.05) on the 4th day after milking. This volume rise is not related to widening of Golgi cisternae whose surface to volume ratio does not vary significantly after milking (Table 4). During this period, the Golgi apparatus membrane surface increment (171 %, 1039 ~rll', p < 0.05) was mainly due to the cisternal sub-compartment which increased by 167% (823 ,um', P < 0.05). The cytoplasmic numerical density of the Golgi-associated microvesicles increased by 198% (8750, p < 0.05) in the initial 4 days after venom extraction while at 8 days this
578
S. M. CARNEIRO et at.
parameter appeared to stabilize. An increase of 22% was observed in the number of microvesicles per Golgi volume unit 4 days after extraction ; at 8 days, the increment in this parameter was 46% (68, p < 0.05) (Table 7). Secretory vesicles represent ~9% of the cytoplasmic volume and their surface to volume ratio did not vary significantly during the experiment . In the unmilked gland of one snake there were several cells entirely loaded with secretory vesicles, some of them having degenerating features such as myelinated figures and disrupted membranes. This fact probably contributed to the highest numerical values of volume and surface densities in this group. The volume increment of secretory vesicles of 110% (22 pm', p < 0.05) was observed on the 8th day; the increments of about 60% on the membrane surface after milking were not significant. The progressive reduction in the mitochondria surface to volume ratio parallels the slight swelling of these organelles in the milked glands . The sum of the surfaces of all measured plasma and endomembranes increased by 112% (3614 pm~, p < 0.05) in the initial 4 days; on the 8th day it appeared to have stabilized . The cellular matrix, bordered by the dilated RER cisternae decreased (p < 0.05) in volume density 4 and 8 days after milking, while the total volume increased by 125% (164 pm', p < 0.05) and 108% (141 ltm', p < 0.0~ at these times. DISCUSSION
Bothropsjararacussu venom secretory cells are ultrastructurally similar to those of other Viperidae: Vipers palaestinae (BBN-SHAUL et al., 1971 ; ORON and BDOLAI-I, 1978), Vipers antmodytes (ORON and BDOLAH, 1973) and Crotales durisses terr~cus (WARSHAWSKY et al., 1973). The most noteworthy morphological feature, together with the cytoplasmic volume increase on the 4th and 8th days after venom extraction, is the expansion of the RER membrane surface with concomitant widening of the intracisternal space. The RER volume density increased up to 34% on the 8th day after milking, while its surface to volume ratio decreased gradually from 45 .9 at the resting stage to 19.6 on the 8th day after milking, revealing the widening of this organelle during this period . The RER volume density values obtained at the resting stage (17%) and 4 days after milking (28%) are quite different from those of Vipers palaestinae (ORON and BDOLAH, 1978) at the same stages, 82% and 86% respectively. This difference seems to be due to the fact that in this study the cellular matrix and the RER volume density variations were measured separately . The matrix associated with the RER membranes contributed greatly to the increase in cytoplasmic volume after milking. One of the resting glands possessed an unusually large number of secretory cells loaded with secretory vesicles, a fact that was not reported by other authors in similar studies. Such accumulation of secretory vesicles might explain the relatively elevated mean volume as well as the wide standard error observed for these structures . Notwithstanding, in the species Bothrops jararaca, the embryos secretory cells were filled with moderately electron dense vesicles (unpublished observations). Possibly, in the resting glands of juvenile B. jararacussu snakes, not yet stimulated by feeding or milking, the observed stored secretory vesicles were originated in early fetal life. The RER surface membrane, which represents 57-fi2% of the total cytoplasmic membranes measured, doubled in value (1952 ~m~ on the 4th day after milking. In the unmilked glands, the total Golgi surface membrane corresponds to 19% of the cytoplasmic membranes. Mainly due to the cisternal compartment enlargement, this para-
Morphometry of B. jararacussu Venom Cells
579
meter increased by 171 %, four days after milking, corresponding to 24% of total membranes measured . Although an alternative pathway of venom secretion was suggested for V. palaestinae, as an emergency mechanism, involving the release of secretory product directly from the RER, bypassing the Golgi apparatus and secretory vesicles (BEtv-St-twuL et al., 1971), electron microscopic radioautographic studies on C. durissus terrificus (W~esxwwsxY et al., 1973) and on V. palaestinae (Ottox and BDOLAH, 1978), showed that the synthesis and intracellular transport of venom proteins are similar to those described for mammalian exocrine glands (PALADE, 1975). The question whether venom gland secretion is of the constitutive or regulated type, or both, remains to be established. Regulated secretory cells are characterized in part by storage of secretion in granules to be released after an appropriate stimulus (cf. BURGE44 and KELLY, 1987). Venom gland cells possess a secretion storage compartment represented by the secretory vesicles . However, the appropriate mechanism that controls the synthesis and exocytosis of the secretory products is not clearly known. It is supposed that the amount of the stored venom in the tubules and gland lumina participate in the regulation of the secretory process (cf. KocHVn, 1987) and that it is independent of nervous supply (cf. BDOLAH, 1979). While the protein synthesis declines beyond the first week after milking, the secretion of the venom continues until the gland lumen becomes replete, up to the 30th~Oth days after venom extraction (OROx and BDOLAH, 1973; Ds Luccn et al., 1974). When the cell environment is modified, as in cell culture, synthesis and secretion can be maintained for over 7 months (SELIS et al., 1989). In the venom gland cells, as in other exocrine cells the coexistence of the regulated and the constitutive secretory pathways is plausible (cf. BußcESS and KELLY, 1987). The actual amount of membrane surface of a given cell compartment is the result of the balance between various factors such as the ratio between membrane biogenesis and membrane degradation. However, the intensities of membrane transport to and from the compartments also affects the overall picture. The present observations demonstrated an increase in membrane degradation as indicated by lysosomal or autophagic vacuole formation occurring up to 4 days after milking (Table 6, "other structures") . The expansion of the RER and Golgi cisternae membrane surface and the increase in the number of transporting microvesicles occurring on the 4th and 8th days after milking are compatible with the view that membrane transport from RER to Golgi and to apical plasmalemma increased as a consequence of the enhanced venom synthesis (Oitox and BnoL~x, 1978). Up to the 8th day after venom extraction, the sum of total membranes increased by 110% (3552 ~m~, concomitant to cytoplasmic volume increase. Afterwards, it has been shown that, until the 30th-60th day, the cell height progressively decreases returning to the morphological resting condition (BEx-St-uut et al., 1971). Since membrane biogenesis occurred during the most active synthetic stage, membrane elimination should be expected during the venom accumulation period . This may be achieved in different ways : some part of the membrane excess must be degraded into lysosomes or part must be secreted as microvesicular secretion, as suggested by BEnunonv et al. (1986) in pancreatic acinar cells. Actually, the presence of numerous microvesicle-like profiles in the gland lumen close to the apical plasmalemma seems to be a constant in the venom secretory cells, although their origin and function are unknown (WnttsHawsxY et al., 1974). The membrane surface of the cytoplasmic components related to synthesis and transport of the venom proteins and the membranes of the mitochondrial compartment increased 4 and S days after venom extraction while the surface density did not increase
580
S. M. CARNEIRO et al.
significantly. These results suggest that membrane biogenesis, degradation and circulation that takes place in the first week after venom extraction is achieved through coordinated cellular mechanisms that maintain the ratio between total membrane surface and total cytoplasmic volume unaltered . Aabww/s~gnkntr-The authors sham Mr Biffa ~s and Mr JocutNt J. A. Rooatatms for technical aid and Dr R. D. G. Tt~tat+vx for helpful wggestions . This work was supported in part by: Fuadaçäo de Amparo à Pesquisa do Estado de Sâo Paulo (FAPESP) 89/3805-9; Finaaciadora de Estudw e Projetos - Fundo National de Desenvolvimento Cientffioo e Tecnolbgioo/Programa 3etorial de Mic~asoopie Ektrbaica (FIIdE~ FNDCT/PSME) and Comelho National de Desenvolvimento Cientißco e TecaolGgioo (CNPc~ 40.0055/89 .8 . V. R. Paano and L. A. B. M. Ltn.~ have fellowships from Fundaçilo de Desenvolvimento IWministrativo . REFERENCES
At~txe, W. A. and DIJNNIId ., M. S. (1982) Morphometry. London : Edward Arnold (Publishers Ltd). B~cx, G. (1%3) Über die Hestimmung von aharakterietischen GrBsaen sins Kugelverteilung aus der Verteilung der Schnittkreise. Z. wiu . Mikrark. 6i, 285-291. Booutt, A. (1979) The venom glands of snakes and venom saxetion. In : HandboaE of Ezperünentai Pharmacology S2, pp. 41-57 (C. Y. Lee, Ed.). Berlin : Springer . Be~unaa~r, A. R., Gtwr>Dtx, G., V~ct~te~u, A., Sr-JB~ua, P. and C~a~uve, C. (1986) Detection and characteriation of microveaicles in the acinar hmien and in juice of unstimulated rat panctras . J. Htrtochem. Cytochem. 34, 1079-1084. Bert-Sxeui, Y., L.m~az-t, S. H. and KocRVe, E. (1971) Ultrastructutal aspects of secretion in the venom glands of Vipers palaestinoe. In: Toxins oj .lnftrwl artd PJmtt th(gin, pp. 87-107 (De Vtua, A. and Kocttve, E. Eda). London: Got+don and Breach . Bvttaeas, T. L. and Ket.tY, R. B. (1987) Constitutive and regulated secretion of proteins. Mn. Rev. Cell Blot. 3, 243-293 . Ceaxemo, S. M. and Swan, A. (1987) Morphometric evaluation otzymogea granule membrane transfer to Golgi ciaternae following exocytods in pancreatic acinar cells from suckling newborn rats. J. Submicrarc . Cytol. 19, 19-33. De Lucre, F. L. and I~r~r~nn , M. T. (1972) Synthesis of ribonucleic acid in the venom gland of Crota6o durLxtus terr(ftcus (Ophidia, Reptilia) after manual extraction of the venom. Bioehem. J. 130, 335-342. De Luoc~, F. L., Barmen, A., Ktx~tve, E., ROIiOCiaID, A. M. and V~t~et, V. (1974) Protein aynthesir and morphological changes in the saxetory epithelium of the venom gland of Crotahe daissLS zerrfeav at different times aRer manual extraction of venom. Toxkat 12, 361-368. Dutvcetv, D. B. (1955) Multiple range and multiple F tests. Biometrics 11, 1~2. FaBette, R. H. and W~ E. R. (1%7) Steroological techniques in microscopy. J. R. Microsc. Sot. 87, 25-34. Kocxve, E. (1987) The origin of snakes and evolution of the venom apparatus. Taxicort 2S, 65-106. KocHVe, E. and Gruas, C. (l%7) The atnteturo of thevenom gland and secretion of venom in viperid snakes. Ice: dtwreal Toxins, pp. 195-203 (Russ~.t., F. E. end Seuxuens, P. R., Eds). Oxford : Pergamon Pteaa . Kocttvt., E., Ottort, U., Bnotwx, A. and At t.ox, N. (1975) Regulation of venom secretion and injection in viperid wakes. Toxkon 13, 104. Lurneeaa, L. G. and VtMtweax, P. (1970) On calculating volumes of transsa.Ted bodies from two-dimensional micrographs. Lab . Ingest. 23, 313-317. Oaorr, U and Bnotwtt, A. (1973) Regulation of protein synthesis in the venom gland of viperid snakes. J. Cell Bloc. 36, 177-190. OaoN, U. and Hnot.ex, A. (1978) Intracelluhu transport of prot~ in active and resting sa :retory ceW of the venom gland of Vlpero paiaestbtae. I Cdi Bioi. 78, 488-502. Ptuue, G. (1975) Intracellular aspects of the process of protein synthesis. Scietroe, N. Y. 189, 347-358. Rotstvaeaa, D., B~~t, E. S. and Knows, E. (1971) Studies on ribonucleic acid synthesis in the venom glands of Vipers paiaestinae (Ophidie, Reptilia). Bloehem . J. 121, 609-612. S~.ts, P. G., Bosoms., M. and Tr~arox, R. D. G. (1989) Venom production in snake venom gLmd cells cultured in vitro. Toxirnn 27, 1245-1249. Stntaa, A. R. (1969) A low-viscosity epoxy rain embedding medium for electron microscopy. J. Ultrastruct. Res. 26, 313. Weastt~wsw, H., Barmen, A., Gortçnt vm, R. P., Ver~* V. sad De Luoc~, F. L. (1973) Flame structure of the venom gland epithelium of the South American rattlesnake and radiosutographic studies of protein formation by the secretory cells Am . J. gnat . 138, 79-120. WBta81 E. R. (1979) Stereologlaal Methods I. Practical Methods jor Blologieo% Morphontetry. London, New York : Academic Press.
00"I-0101/91 53.00+ .00 ~ 1991 Paymon l~ pb
MORPHOMETRIC STUDIES ON VENOM SECRETORY CELLS FROM BOTHROPS JARARACUSSU (JARARACUÇU) BEFORE AND AFTER VENOM EXTRACTION* S. M.
CARNSIRO,' V . R . PINTO,' C. JARED,~ L. F . P. FARirA3 AND A. SESSO3
A. B. M.
LuL.A,'
'Seçâo de Venmos, and ~o de Microscopie Eletr6nica, Instituto Butantan, Av . Vital Bratil, 1500, CEP 05504, Sao Paulo, 3P, Bread, and 3Laborat6rio de Patologia Molecular, Departamento de Patologia, Faculdade de Medicine da Univeraidade de Sâo Paulo, Sâo Paulo, Bread. (Received 28 June 1990; accepted 30 October 1990)
S. M. CARNEUto, V . R . Plrrro, C. JARED, L . A. B. M. Lur.A, F . P. FARIw and A. SESSO. Morphometric studies on venom secretory cells from Bothrops jararacussu (Jararacuçu) before and after venom extraction. Toxicon 29, 569580, 1991 .-A comparative morphometrical analysis was carried out on secretory cells from Bothrops jararacussu venom glands, before manual extraction of the venom (milking) and 4 and 8 days after milking. At the 8th day after milking, the cytoplasmic volume increased by 160%. The rough endoplasmic reticulum (RER) volume density increase, up to the 8th day after milking, is mainly due to widening of the infra-ßcisternal space. The total volume and membrane surface of the RER, Golgi apparatus and subcomponents, secretory vesicles and mitochondria, increased during the experimental period while the volume and surface densities of these organelles, with the exception of the RER, did not vary. The numerical density of Golgiassociated microvesicles per Golgi volume unit also increased. The greatest relative increments in these parameters occurred within the first 4 days. These results are compatible with an increased rate of membrane synthesis and transport in the milked glands and suggest that the membrane biogenesis, degradation and circulation that takes place in the first week after milking is achieved through coordinated cellular mechanisms that maintain the rate between total membrane surface and total cytoplasmic volume unaltered. INTRODUCTION
Tim vEr1oM glands of viperid snakes are stimulated for a new cycle of venom synthesis and secretion after discharge of the venom stored in the gland lumina . Biting may promote a partial discharge while a more complete extraction can be achieved manually by milking (KOCHVA et al., 1975 ; KOCHVA, 1987). Concurrent with an increase in the synthesis of RNA (ROTBNBERG et al., 1971 ; Da LuccA and 1MAIZLTMI, 1972) and pmtein (ORON and BDOtAx, 1973 ; DE LuccA et al., 1974) 'Part of these findings was presented at the I2th ColGquio da Sociedade Brasileira de Microscopia EletrSnica, Caxambu, Brasil, Setembro, 1989.
570
S. M. CARNEIRO et al.
FYa. 1. Ux~n vetvo~ awvn of Bothrops Jararacussu . Note narrow RER cisternae in the secretory cells (Sc); Golgi area (G); vacuolated vesicles horizontal cell (H); lumen (L); connective tissue (C). Bar: 2 pm.
M;
there is a large increase (up to five times) in the height of the secretory epithelium until the 4th to 8th day after manual extraction . The secretory cells which appear flattened or cuboidal in the replete gland, gradually become columnar (KocxvA and GArrs, 1967 ; Bl?PiSHnuL et al., 1971 ; ORON and Bl)OLAH, 1973); the intracisternal spaces of the rough endoplasmic reticulum (RER) widen (B>?rr-SHAUL et al., 1971) and the Golgi apparatus becomes more conspicuous (ORON and BDOLAH, 1973). In milked glands, the venom transport to the lumen is greater than that of resting glands (Kocxvn and GANS, 1967; ORON and BDOL,~I-t, 1978). In order to analyze the configurational changes occurring in the cellular compartments involved in the synthesis, condensation and transport of venom secretory proteins, after venom extraction, the volume and total membrane surface of these compartments were morphometrically evaluated in venom gland secretory cells from Bothrops jararacussu at the resting stage (unmilked and not fed) and 4 and 8 days after milking . MATERIALS AND METHODS Snakes acrd venom glands Eleven Bothrops Jararacrctsu snakes, 2.5 months old, not fed since their birth, weighing 7.1-11 .8 g, 27-30 cm
long and maintained at 226°C were used . All were from the same litter . Venom glands from three unmilked snakes, from four snakes milked 4 days previously and from four snakes millced 8 days previously were studied. Venom glands were removed after killing the snakes by decapitation . Pieces of about 1 mm3 were obtained from the median part of the main venom gland; they were food using 1.5%
Morphometry of B. Jararacycssu Venom Cells
57l
FYe. 2. U~n~ren vexoit cr wn of Bothrops jararacussu. Stacked narrow RER cisternae (R) in the basolateral cytoplasm of two secretory cells. Plasmalemma foldinge (P); Golgi area (G); mitochondria (m). Arrow points to a thin cytoplasmic projection of a horizontal cell . Bar. 0.5 pm . glutaraldehyde and 1 .0% paraformaldehyde in 0.08 M cacodylate buffer, (pH 7 .3) for 2hr and 1% osmium tetroxide in the same buffer, for 2 hr at 4°C. The specimens were embedded in Spurr's medium (Sruwe, 1969).
Nuclear acrd cytoplasmic voGr»te measurements Semithin (0 .51an) sections were examined in a light microscope with an attached camera lucida equipment. The nuclei of the projected secretory cells had their contours outlined over a magnetic card with an appropriate cursor. The area measurements thus obtained were transferred to a computer system to be processed. For each gland, 300 nuclear areas were measured. The mean nuclear volume was estimated in two ways: (a) by the BACH (1%3) procedure which is applicable to spherical objects; (b) with aid of the Lrrro~ea-VottwEare (1970) formula, suitable for elliptical objects. The cytoplasmic volume was obtained from the nuclear volume (Vn) and the nuclear volume density in the aecretory cell (Vvn) estimated by the point counting method (W~eEr., 1979) for semithin sections and corrected for the Holmes effect as indicated by At~tr~ and Dtnvrm .t (1982) .
Evaluation of morphometrical parameters in subcellular structwrs Twenty electron micrographs were randomly taken from each gland at 2600 x on 35 mm negative film and analysed on 18,200 x prints. Whenever Golgi stacked saecules were.seen, one or more exposures were taken at a magnification of 5500 and studied in micrographs at a magnification of 38,500. A test system of the Fs~eeW~m. type (1%7) was applied to each of the prints and the following structures were measured: rough endoplasmic reticulum (RER), Golgi apparatus (on the 38,500 magnification prints), mitochondria, secretary vesicles, basolateral and apical plasmalemma, and "other structures" ouch as lysosomo-like profiles, multivesicular bodies, phagosomea, etc. The cytoplasmic volume density (Vv), the surface density (Sv), the surface to volume ratio (s/v) and the numerical density of Golgi microvesicles (Nmv) were estimated as described by C~ttt~eo and Ses'o (1987). The relative standard error (e) associated with volume density (Vv) estimations were calculated by the formula: e~
1-Vv/P . Vv
where P is the total number of points counted over the measured cell cytoplasm. The actual volume (V), the total membrane surface (S) of each subcellular compartment, and the cytoplasmic numerical density of Golgi microvesicles were obtained from the cytoplasmic volume and Vv, Sv and Nmv measurements, respectively. The mean and standard error of the mean are expressed on the tables . Differences between experimental groups were detected with the aid of the Dtnvcex (1955) test.
572
S. M. CARNEIRO et al.
FIG. 3. EIGHT DAYS FOLLOWING MILIUNG OF BOtlllOpS farOaCIttJtt VENOM GLANA. Note the different RER cisternae (R) dilatation patterns of two adjacent celle. Phlsmalemma (arrows); Golgi area with vesicle M; mitochondria (m). Bar: 0.5 ~.
RESULTS
Morphological observations
Under the light microscope, the secretory epithelium of the unmilked glands is predominantly composed of cuboidal or flat secretory cells, about 6 pm height . Occasionally, higher columnar cells are noticed in some tubular transactions . Secretory vesicles or vacuoles are observed mainly at the apical region of the cell, or replenishing some cells. On the 4th and 8th days after venom extraction, the height of the secretory cells has increased; they now become columnar shaped, c. 30 ~m height. Secretory vesicles are noticed at the supra nuclear region . The cytoplasm has a vesiculated appearance, mainly at the apical region . The tubular lamina are narrower and less regular than those of the unmilked glands . A few dense chromophilic granules and some metachromatic inclusions can be observed in both types of glands. At the electron microscopic level the cytoplasm of the unmilked glands is filled with narrow RER cisternae (Figs 1 and 2). Secretory vesicles are not commonly seen at the tracts side of the Golgi stack; nevertheless, secretory vesicles and translucid vacuoles presenting degenerating signals, such as myelinated figtu~es, can be observed either isolated or filling the cytoplasm of some secretory cells. In general, distinct condensed electron dense secretory granules are not seen . Occasionally, lipidic inclusions are observed. Near the apical plasmalemma ,microvilli, at the luminal side, microvesicle-like profiles about 7100 nm diameter are noticed. Basolateral plasmalemma foldings are commonly seen (Fig 2). On the 4th and 8th days after extraction of venom, the morphological features of the secretory cells have similar characteristics but they differ from those from unmilked glands mainly in regard to the expansion of their RER membranes and the widening of the RER intracisternal space (Figs 3, 4). The distribution of the ribosomes on the dilated cisternae is not regular. Near the Golgi apparatus cis side, microvesicles pinching off from the transitional endoplasmic reticulum can be observed. Secretory vesicles and dilated
Morphometry of B. jaraiâcuwu Venom Cells
573
FiO . 4. EIORT DAYS AOLLOWINO lat .CR~iß OF BOrlVOpS J¢I~QCLSTY VENOl1 GLAND. Cohmmar cells presenting supra nuclear Golgi area (G); secretory vesicles (S) ; and dilated RER cisternae (R). Horizontal cell (ü); connective tissue (C); lumen (L) . Bar: 2 gym .
cisternae of the RER, which sometimes exhibit extensive areas devoid of ribosomes, can be observed at the inner side of the well-developed semicircular Golgi apparatus. Microvesicles about 60-80 nm diameter are numerous and multivesicular bodies occur at a greater frequency than in the unmilked glands (Fig . 5). At the luminal plasmalemma, microvilli and associated microvesicle-like profiles are also seen . These structures are more easily observed in regions where the plasmalemma of two cells are facing each other, apparently trapping the microvesicles between them. Exocytotic figures were not observed in these glands. In order to simplify the presentation of morphometric data, possibly degenerating or sutophagic vesicles, translucid vacuoles, multivesicular bodies, lysosomelike granules and a few lipidic inclusions were grouped under the designation "other structures". The morphological features described are the most frequently observed, but intermediate characteristics ca,n also be seen in all the glands studied.
574
S . M. CARNEIRO et at.
FIO . S . FOUR DAYS POLLOWINO ML1C1N0 OF BOtIVOpS JgrarQCUS4Y VENOI! OLAND. Note dilated cisternae (C) at the cir side of the Golgi apparatus . Transitional RER (arrows) . Multivesicular bodice (B). The RER cisternae (R) enclosed in the tan's Golgi side exhibits membrane areas devoid of ribosomes. Har. 0 .5 pm .
Quantitative results
In the glands examined, the secretory cell nuclear shape was not uniform. About 70% of the nuclear transections (part of them pertaining to isodiametric cell profiles) were TABLS 1 . NUCLFAA VOLUlIBS, NUCLSAR VOLUl18 DElVaITY AI~ COARHBPONDINO CYTOPLASiOC UNIDI~ AI~ FROY èla~ VBriD!( OLAI~ 88CASPORY CHLIs
VOLUI®3 FROI(
DayB Bftel VenOm extraCtiOII 0 Nuclear volume density (~'/~~ Nuclear volume (~m~ (LV) Nuclear volume (pmt (B) Cytoplasmic volume (pmt (LV) Cytophunnic volume (tcm~ (H)
0.167 t O .OOSt (0 .065)' 48 t 4 49 t 8 242 f 25 246 t 36
4 0.11 S t 0 .005" (0 .070) 74 f 8" 80 f 14" 572 f 68rr 614 t 100"
8 0.094 f O.OOSr',$ (0 .072) 61 t 3 67 f 7 588 f 28rr 640 f 50r'
Results are shown as mean f S .E., n ~ 3 for 0 day and 4 for four and eight days . Nuclear volumes estimated by the Lindberg-Vorwerk (LV) and by the Bach (B) procedures . 'Relative standard eaor associated with density volume estimates . ''Significantly different (p < 0 .05) from wntrol (0) . tThese values were corrected for the Hohnes effect as indicated by A~er~ and Durnvn.L (1982). Noa corrected data are 7r/o higher . $Significantly different (p < 0 .05) from 4 days .
Morphometry of B. Jararncuasu Venom Cells TABLE
2.
VoLUSm
375
nws1~ (~an 3/Nm~ o~ crroru.9ac oo~or~rts r~ vivtm~ ~n m~ tm.s~ vartor OLAF ~CASrOAY ®.is Days Si1Et Venom extraetlOn
C~toplaemic component Rough endoplasmic retiwlum Sxretory vesicles Golgi ciaternae Golgi apparatus associated microvesicles Golgi apparatus Mitochondria "Other structures" Apical miaovilli aad associated microvesicka Cellular matrix
0
4
8
0.173 f0.024 (0.045)" 0.08910 .037 (0.075) 0.034 f0.004 (0 .111) 0.003 t0.007 (0 .291) 0.107 t0.017 (0.060) 0.049 t0.003 (0.090) 0.01St0.003 (0 .181) 0.020 f0 .004 (0 .147) 0.540 f0.025 (0 .019)
0.281 f0.017"' (0.030) 0.030 t0.004 (0.082) 0.043 f0.004 (0.087) 0.007 t0.002 (0.243) 0.123 t0.011 (0.050) 0.036 t0.004 (0.077) 0.009 f0.002 (0.212) 0.007 t0.002 (0.230) 0.47310 .020"" (O.Ol9)
0.338 t0 .008 "',t (0.025) 0.06410.006 (0.069) 0.033 f0.003 (0.986) 0.006 f 0.001 (0.232) 0.10010.003 (0.054) 0.037 f0.003 (0.073) 0.007 f0.018 (0.228) 0.009 f0.002 (0.190) 0.42410.005 "" (0.020)
Results are shown as mean f S.E., n = 3. fot 0 day and 4 for four and eight days. "Relative standard error associated with density volume estimates. ""Significantly different (p < 0.05) from control (0). tSignificantly different (p < 0.05) from 4 says .
round in shape. The remaining profiles, mainly seen in columnar cells, appeared to be elliptical . It was therefore decided to evaluate nuclear volumes using two procedures. The BACH (1963) method enables the transformation of the distribution of the radii of circles measured in the nuclear transections, into distribution of radii of spheres, correcting for the effects of section thickness and for the lost small nuclear caps. The LINDH$RCr VORWERR (1970) approach uses the formula
TABLE
3.
TOTAL voL~
(pm~ oF
crrorLASiac
ooseor~rts
Days after venom extraction Cytoplasmic component Rough endoplasmic reticulum Sectetory vesicles Golgi cisteroae Golgi asso~iated microvesiclea Total (iolgi apparatus Mitochondria "Other structures" Api~al microvilli and associated microvesicles Cellular matrix
0
4
8
44 f 14 20 f 6 9f2 1 f 0.4 27 f 8 12 f 2 3 f1 Sf I
171 f 27"" 30 f4 27 f4"" 4f 1 "" 73 f 7"" 34 t 6"" 61 2 Sf 2
215 f ll' " 42 f 7"" 21 1 3"" 41 1 "" 67 f T" 37 t 6"" 511 6f1
131 f l9
293f39""
272 f 22""
Results are shown as mean f S.E., n - 3 for 0 day and 4 for four and eight days. ""Significantly different (p < 0.05) from control (0).
S. M. CARNEIRO et d.
S76
TABLE 4. SUAFAr~ TO vOLl1YE ItAT10 ~r//tm~ OF CY1'OPLASIlIC C0IBONEN79 Days 81~Cr VCDOm e7rtraCtiOn GjRoplaamic component Rough endoplasmic reticulum Secretory vesicles Golgi cisteraae Golgi associated microvesicles Mitochondria "Other structures"
0
4
8
45.914.5 4.210.2 55.613.E 88.312.0 12.910.9 L710.6
22.911 .3"" 4.510.3 50.114.5 9L2114.5 (0.810.6"" 3.511 .0
19 .611 .0 "' 3.210 .2 51 .017 .E 88 .91(0.9 9 .210 .2'" 6.213 .3
Results are shown as mean 1 S.E., n = 3 for 0 day and 4 for four and eight days. '"Significantly different (p < 0.05) from control (0).
TABLE 5. $URFACE aErvsrrv (pm'lPm') oF crroPLAS~rc oot~oHErvTs Days after venom extraction Cj~tophismic component Rough endophrsmic rtticulum Secretory vesicles Golgi cisternae Golgi apparatus associated microvesicles Mitochondria Apical rnicrovilli and associated microvesicles "Other structures" liasolateral plasmalemma Apical plaiaDalemma
0
4
8
7.7710.66 0.3610.13 1 .9110.35
6.4010.30"' 0.2210.02 2.2310.27
6.5910.25 0.0210.01 1.6110.13
0.4510.05 0.6410.07
0.5410.03 0.0610 .02
0.5510.08 0.5310.03
0.5910.16 0.0210.01 0.8010.08 0.3210.03
0.2310.07 0.0410.01 0.88 t0.07 0.1410.02
0.3310.02 0.0210.01 0.5310.08 0.1810.03
Results are shown as mean 1 S.E., e = 3 for 0 day and 4 for four and eight days . ""Sigai&antly different (p < 0.05) from control (0).
TABLE 6. TOTAL è®lBRANE SURFACE (/tm~ OF CYIY)PLAS~UC COLlPONENTS Days after venom extraction Cytoplasmic components Rough endoplasmic reticulum Secretory vesicles Golgi cisternae (1) Golgi associated microvesicles (2) Total Golgi apparatus (1)+(2) Mitochondria "Other structures" Basolateral phrsmalearrna Apical plasmalemma Apical microvilG and associated microvesicles Total measurod membranes
0
4
8
19481467 84124 4941163 114130 6081194 157129 51 I 193127 7016
39001608 "" 134116 I3I71I59 "" 330153" 16471196"" 363148 "" 23112 3391106 72110
42261387 "" 133120 10361128'" 349137"" 13861152 "" 338128"" 1418 347179 99111
146143 32181739
140153 68321905""
212120 67701629 ""
Results are shown as mean 1 S.E. n ~ 3 for 0 day and 4 for four and eight days. ""Significantly different (p < 0.05) from control (0).
Morphometry
of B. jararacussu
57 7
Venom Celle
TABLE 7. NU1®IICAL DENSITY (N) OF GOLCiI A390CIATED MICROV~IICLES PER VOLU1~fE UNrr (~~ OF GOLCiI APPARATUS AND TOTAL NUMBER (N!) OF CiOLöI ASSOCIATED 1fICAOVESICLO IN Tlni SECREPORY CELL CYTOPLASM liHye Sttei YCIIOm eRtraChOII
N Nt
l47f30 4410 t 1748
Reeulb are shown ae mean t S.E. n = 3 for 0 day and "SignißcantlY different (p < 0.05) from control (0).
179t8 I3,160 t 1466" 4 for
215t 13" 13,680 t 1340"
four and eight days .
where v is the mean nuclear volume, s is the mean profile area and ß is a conformational coefl'tcient depending on the ratio (e), between the major (a) and the minor (b) axes measured on elliptical profiles. The mean nuclear volume and derived cytoplasmic volume, obtained by both approaches, are very similar (Table 1). To calculate the stereological parameters involving the cytoplasmic volume, such as the total membrane surface etc., the data obtained using the BACH procedure was utilized . Four and eight days after milking, the cytoplasmic volume of the venom gland secretory cell is, respectively, 150% (368 pm', p < 0.05) and 160% (394 Ealt', p < 0.05) greater than in the unmilked gland. The nuclear volume increases by 63% (31 pm', p < 0.05) on the 4th day as compared with unmilked glands (Table 1). Organelle and cellular matrix volume densities in the cytoplasm are shown in Table 2. The increase in the RER volume density and total volume (Table 3) after venom extraction is mainly due to the widening of the intracisternal space. Four and eight days after milking, the RER surface to volume ratio (Table 4) declined significantly and the RER volume was respectively 289% (127 pm', p < 0.05) and 389% (171 pm', p < 0.05) greater than in the unmilked gland. This partly accounts for the observed increase in cytoplasmic volume . Reduction in RER surface density (Table 5) of 18% (1 .37, p < 0.05) is observed at 4 days after milking, in spite of an increase of 100% (1952 pmt, p < 0.05) in its membrane surface in the same interval . On the 8th day after milking, the RER surface density difference is not significant while its membrane surface increased by 117% (2278 pmZ, p < 0.05). The volume densities (Table 2) and the surface densities (Table 5) of the Golgi apparatus and sub-components (cisternae and Golgi-associated microvesicles) and of mitochondria did not vary significantly over the experimental period, while the total volume (Table 3) and total surface membrane (Table 6) were significantly increased after milking . The greatest relative increments in the values of volume and membrane surface occurred within the first four days after milking; between the 4th and the 8th days these stereological parameters did not vary significantly . The Golgi apparatus (cisternae plus microvesicles) volume increased by 170% (46 pmZ, p < 0.05) on the 4th day after milking. This volume rise is not related to widening of Golgi cisternae whose surface to volume ratio does not vary significantly after milking (Table 4). During this period, the Golgi apparatus membrane surface increment (171 %, 1039 ~rll', p < 0.05) was mainly due to the cisternal sub-compartment which increased by 167% (823 ,um', P < 0.05). The cytoplasmic numerical density of the Golgi-associated microvesicles increased by 198% (8750, p < 0.05) in the initial 4 days after venom extraction while at 8 days this
578
S. M. CARNEIRO et at.
parameter appeared to stabilize. An increase of 22% was observed in the number of microvesicles per Golgi volume unit 4 days after extraction ; at 8 days, the increment in this parameter was 46% (68, p < 0.05) (Table 7). Secretory vesicles represent ~9% of the cytoplasmic volume and their surface to volume ratio did not vary significantly during the experiment . In the unmilked gland of one snake there were several cells entirely loaded with secretory vesicles, some of them having degenerating features such as myelinated figures and disrupted membranes. This fact probably contributed to the highest numerical values of volume and surface densities in this group. The volume increment of secretory vesicles of 110% (22 pm', p < 0.05) was observed on the 8th day; the increments of about 60% on the membrane surface after milking were not significant. The progressive reduction in the mitochondria surface to volume ratio parallels the slight swelling of these organelles in the milked glands . The sum of the surfaces of all measured plasma and endomembranes increased by 112% (3614 pm~, p < 0.05) in the initial 4 days; on the 8th day it appeared to have stabilized . The cellular matrix, bordered by the dilated RER cisternae decreased (p < 0.05) in volume density 4 and 8 days after milking, while the total volume increased by 125% (164 pm', p < 0.05) and 108% (141 ltm', p < 0.0~ at these times. DISCUSSION
Bothropsjararacussu venom secretory cells are ultrastructurally similar to those of other Viperidae: Vipers palaestinae (BBN-SHAUL et al., 1971 ; ORON and BDOLAI-I, 1978), Vipers antmodytes (ORON and BDOLAH, 1973) and Crotales durisses terr~cus (WARSHAWSKY et al., 1973). The most noteworthy morphological feature, together with the cytoplasmic volume increase on the 4th and 8th days after venom extraction, is the expansion of the RER membrane surface with concomitant widening of the intracisternal space. The RER volume density increased up to 34% on the 8th day after milking, while its surface to volume ratio decreased gradually from 45 .9 at the resting stage to 19.6 on the 8th day after milking, revealing the widening of this organelle during this period . The RER volume density values obtained at the resting stage (17%) and 4 days after milking (28%) are quite different from those of Vipers palaestinae (ORON and BDOLAH, 1978) at the same stages, 82% and 86% respectively. This difference seems to be due to the fact that in this study the cellular matrix and the RER volume density variations were measured separately . The matrix associated with the RER membranes contributed greatly to the increase in cytoplasmic volume after milking. One of the resting glands possessed an unusually large number of secretory cells loaded with secretory vesicles, a fact that was not reported by other authors in similar studies. Such accumulation of secretory vesicles might explain the relatively elevated mean volume as well as the wide standard error observed for these structures . Notwithstanding, in the species Bothrops jararaca, the embryos secretory cells were filled with moderately electron dense vesicles (unpublished observations). Possibly, in the resting glands of juvenile B. jararacussu snakes, not yet stimulated by feeding or milking, the observed stored secretory vesicles were originated in early fetal life. The RER surface membrane, which represents 57-fi2% of the total cytoplasmic membranes measured, doubled in value (1952 ~m~ on the 4th day after milking. In the unmilked glands, the total Golgi surface membrane corresponds to 19% of the cytoplasmic membranes. Mainly due to the cisternal compartment enlargement, this para-
Morphometry of B. jararacussu Venom Cells
579
meter increased by 171 %, four days after milking, corresponding to 24% of total membranes measured . Although an alternative pathway of venom secretion was suggested for V. palaestinae, as an emergency mechanism, involving the release of secretory product directly from the RER, bypassing the Golgi apparatus and secretory vesicles (BEtv-St-twuL et al., 1971), electron microscopic radioautographic studies on C. durissus terrificus (W~esxwwsxY et al., 1973) and on V. palaestinae (Ottox and BDOLAH, 1978), showed that the synthesis and intracellular transport of venom proteins are similar to those described for mammalian exocrine glands (PALADE, 1975). The question whether venom gland secretion is of the constitutive or regulated type, or both, remains to be established. Regulated secretory cells are characterized in part by storage of secretion in granules to be released after an appropriate stimulus (cf. BURGE44 and KELLY, 1987). Venom gland cells possess a secretion storage compartment represented by the secretory vesicles . However, the appropriate mechanism that controls the synthesis and exocytosis of the secretory products is not clearly known. It is supposed that the amount of the stored venom in the tubules and gland lumina participate in the regulation of the secretory process (cf. KocHVn, 1987) and that it is independent of nervous supply (cf. BDOLAH, 1979). While the protein synthesis declines beyond the first week after milking, the secretion of the venom continues until the gland lumen becomes replete, up to the 30th~Oth days after venom extraction (OROx and BDOLAH, 1973; Ds Luccn et al., 1974). When the cell environment is modified, as in cell culture, synthesis and secretion can be maintained for over 7 months (SELIS et al., 1989). In the venom gland cells, as in other exocrine cells the coexistence of the regulated and the constitutive secretory pathways is plausible (cf. BußcESS and KELLY, 1987). The actual amount of membrane surface of a given cell compartment is the result of the balance between various factors such as the ratio between membrane biogenesis and membrane degradation. However, the intensities of membrane transport to and from the compartments also affects the overall picture. The present observations demonstrated an increase in membrane degradation as indicated by lysosomal or autophagic vacuole formation occurring up to 4 days after milking (Table 6, "other structures") . The expansion of the RER and Golgi cisternae membrane surface and the increase in the number of transporting microvesicles occurring on the 4th and 8th days after milking are compatible with the view that membrane transport from RER to Golgi and to apical plasmalemma increased as a consequence of the enhanced venom synthesis (Oitox and BnoL~x, 1978). Up to the 8th day after venom extraction, the sum of total membranes increased by 110% (3552 ~m~, concomitant to cytoplasmic volume increase. Afterwards, it has been shown that, until the 30th-60th day, the cell height progressively decreases returning to the morphological resting condition (BEx-St-uut et al., 1971). Since membrane biogenesis occurred during the most active synthetic stage, membrane elimination should be expected during the venom accumulation period . This may be achieved in different ways : some part of the membrane excess must be degraded into lysosomes or part must be secreted as microvesicular secretion, as suggested by BEnunonv et al. (1986) in pancreatic acinar cells. Actually, the presence of numerous microvesicle-like profiles in the gland lumen close to the apical plasmalemma seems to be a constant in the venom secretory cells, although their origin and function are unknown (WnttsHawsxY et al., 1974). The membrane surface of the cytoplasmic components related to synthesis and transport of the venom proteins and the membranes of the mitochondrial compartment increased 4 and S days after venom extraction while the surface density did not increase
580
S. M. CARNEIRO et al.
significantly. These results suggest that membrane biogenesis, degradation and circulation that takes place in the first week after venom extraction is achieved through coordinated cellular mechanisms that maintain the ratio between total membrane surface and total cytoplasmic volume unaltered . Aabww/s~gnkntr-The authors sham Mr Biffa ~s and Mr JocutNt J. A. Rooatatms for technical aid and Dr R. D. G. Tt~tat+vx for helpful wggestions . This work was supported in part by: Fuadaçäo de Amparo à Pesquisa do Estado de Sâo Paulo (FAPESP) 89/3805-9; Finaaciadora de Estudw e Projetos - Fundo National de Desenvolvimento Cientffioo e Tecnolbgioo/Programa 3etorial de Mic~asoopie Ektrbaica (FIIdE~ FNDCT/PSME) and Comelho National de Desenvolvimento Cientißco e TecaolGgioo (CNPc~ 40.0055/89 .8 . V. R. Paano and L. A. B. M. Ltn.~ have fellowships from Fundaçilo de Desenvolvimento IWministrativo . REFERENCES
At~txe, W. A. and DIJNNIId ., M. S. (1982) Morphometry. London : Edward Arnold (Publishers Ltd). B~cx, G. (1%3) Über die Hestimmung von aharakterietischen GrBsaen sins Kugelverteilung aus der Verteilung der Schnittkreise. Z. wiu . Mikrark. 6i, 285-291. Booutt, A. (1979) The venom glands of snakes and venom saxetion. In : HandboaE of Ezperünentai Pharmacology S2, pp. 41-57 (C. Y. Lee, Ed.). Berlin : Springer . Be~unaa~r, A. R., Gtwr>Dtx, G., V~ct~te~u, A., Sr-JB~ua, P. and C~a~uve, C. (1986) Detection and characteriation of microveaicles in the acinar hmien and in juice of unstimulated rat panctras . J. Htrtochem. Cytochem. 34, 1079-1084. Bert-Sxeui, Y., L.m~az-t, S. H. and KocRVe, E. (1971) Ultrastructutal aspects of secretion in the venom glands of Vipers palaestinoe. In: Toxins oj .lnftrwl artd PJmtt th(gin, pp. 87-107 (De Vtua, A. and Kocttve, E. Eda). London: Got+don and Breach . Bvttaeas, T. L. and Ket.tY, R. B. (1987) Constitutive and regulated secretion of proteins. Mn. Rev. Cell Blot. 3, 243-293 . Ceaxemo, S. M. and Swan, A. (1987) Morphometric evaluation otzymogea granule membrane transfer to Golgi ciaternae following exocytods in pancreatic acinar cells from suckling newborn rats. J. Submicrarc . Cytol. 19, 19-33. De Lucre, F. L. and I~r~r~nn , M. T. (1972) Synthesis of ribonucleic acid in the venom gland of Crota6o durLxtus terr(ftcus (Ophidia, Reptilia) after manual extraction of the venom. Bioehem. J. 130, 335-342. De Luoc~, F. L., Barmen, A., Ktx~tve, E., ROIiOCiaID, A. M. and V~t~et, V. (1974) Protein aynthesir and morphological changes in the saxetory epithelium of the venom gland of Crotahe daissLS zerrfeav at different times aRer manual extraction of venom. Toxkat 12, 361-368. Dutvcetv, D. B. (1955) Multiple range and multiple F tests. Biometrics 11, 1~2. FaBette, R. H. and W~ E. R. (1%7) Steroological techniques in microscopy. J. R. Microsc. Sot. 87, 25-34. Kocxve, E. (1987) The origin of snakes and evolution of the venom apparatus. Taxicort 2S, 65-106. KocHVe, E. and Gruas, C. (l%7) The atnteturo of thevenom gland and secretion of venom in viperid snakes. Ice: dtwreal Toxins, pp. 195-203 (Russ~.t., F. E. end Seuxuens, P. R., Eds). Oxford : Pergamon Pteaa . Kocttvt., E., Ottort, U., Bnotwx, A. and At t.ox, N. (1975) Regulation of venom secretion and injection in viperid wakes. Toxkon 13, 104. Lurneeaa, L. G. and VtMtweax, P. (1970) On calculating volumes of transsa.Ted bodies from two-dimensional micrographs. Lab . Ingest. 23, 313-317. Oaorr, U and Bnotwtt, A. (1973) Regulation of protein synthesis in the venom gland of viperid snakes. J. Cell Bloc. 36, 177-190. OaoN, U. and Hnot.ex, A. (1978) Intracelluhu transport of prot~ in active and resting sa :retory ceW of the venom gland of Vlpero paiaestbtae. I Cdi Bioi. 78, 488-502. Ptuue, G. (1975) Intracellular aspects of the process of protein synthesis. Scietroe, N. Y. 189, 347-358. Rotstvaeaa, D., B~~t, E. S. and Knows, E. (1971) Studies on ribonucleic acid synthesis in the venom glands of Vipers paiaestinae (Ophidie, Reptilia). Bloehem . J. 121, 609-612. S~.ts, P. G., Bosoms., M. and Tr~arox, R. D. G. (1989) Venom production in snake venom gLmd cells cultured in vitro. Toxirnn 27, 1245-1249. Stntaa, A. R. (1969) A low-viscosity epoxy rain embedding medium for electron microscopy. J. Ultrastruct. Res. 26, 313. Weastt~wsw, H., Barmen, A., Gortçnt vm, R. P., Ver~* V. sad De Luoc~, F. L. (1973) Flame structure of the venom gland epithelium of the South American rattlesnake and radiosutographic studies of protein formation by the secretory cells Am . J. gnat . 138, 79-120. WBta81 E. R. (1979) Stereologlaal Methods I. Practical Methods jor Blologieo% Morphontetry. London, New York : Academic Press.
