- Email: [email protected]
Preparation of porcine thyroid follicles with preserved polarity: functional and morphological properties in comparison to inside-out follicles
Molecular and Cellular Endocrinology, 40 (1985) 9-16 Elsevier Scientific Publishers Ireland, Ltd.
9
MCE 01273
Preparation of porcine thyroid follicles with preserved polarity: functional and morphological properties in comparison to inside-out follicles * R. G g r t n e r , W. Greil, D. Sti~bner, W. P e r m a n e t t e r l, K. H o r n a n d C . R . P i c k a r d t Medizinische Klinik lnnenstadt and I Pathologisches Institut, Universiti~t M~nchen (F.R.G.)
(Received 20 September 1"984; accepted 19 November 1984)
Keywords: isolation of thyroid follicles; cell polarity; cyclic AMP, thyrotropin
Summary One of the main problems in establishing isolated thyroid follicles in vitro is their tendency to form inside-out follicles. The reason for this change in polarity is unknown. We describe here a method for the preparation of stable thyroid follicles with preserved polarity for at least 6 days. Isolated follicles were obtained by infusion of collagenase (1.5 mg/ml) dissolved in minimal essential medium into the artery of intact thyroid glands. The morphological and functional properties of these follicles were compared to inside-out follicles. These inside-out follicles were obtained by digestion of minced thyroid tissue in a collagenase (1 mg/ml) solution. The polarity of follicles was proved by morphological criteria. Follicles with preserved polarity did not change polarity for at least 6 days in the presence of 1% or 5% fetal calf serum. As the culture conditions for both preparation methods were identical, we conclude that the preparation method rather than the culture condition is responsible for the preservation of cell polarity of isolated thyroid follicles in our system. Increases in cyclic AMP levels induced either by bovine thyrotropin (10 -3 U/ml) or by isoproterenol (10 -6 M) as well as iodide uptake and organification were rapid and significant in right-side-right follicles but not in inside-out follicles. Therefore the TSH receptor and the fl-adrenergic receptor appear to be exclusively located at the basal membrane of follicular cells. In addition, iodide uptake apparently is unidirectional.
About 20 years ago, Kerkof et al. (1964) first succeeded in inducing follicle-like structures in thyroid cell culture during TSH stimulation. Thereafter various methods have been reported for the isolation of thyroid follicles from different species such as rat, pig and man (Mauchamp et al., 1979; Nitsch and Wollman, 1980b; Denef et al., 1980; Herzog and Miller, 1981; Karlsson et al., * Supported by the Deutsche Forschungsgemeinschaft. A preliminary report of this work was presented at the 14th Annual Meeting of the European Thyroid Association, Rotterdam, 1984.
1982; Hanafusa et al., 1982). One of the main difficulties in the preparation and cultivation of thyroid follicles in vitro is their tendency to form follicles with inverted polarity (inside-out follicles). A change from normal to inverted polarity appears to occur when the serum concentration in the culture medium is raised from 0.5% to 5% fetal calf serum in cultured rat follicles (Nitsch and Wollman, 1980a) or in human follicles (Hanafusa et al., 1982). In addition, isolated cells from hog thyroids cultured in plastic dishes not treated for tissue culture appear to reorganize into 'hollow spheres', which have the morphological character-
0303-7207/85/$03.30 © 1985 Elsevier Scientific Publishers Ireland, Ltd.
10 istics of follicles with inverted polarity (Mauchamp et al., 1979). Inversion of cell polarity in isolated thyroid follicles seems to be dependent on the preparation method and culture conditions, but the reason for this phenomenon is still unclear. Because the main function of the thyroid gland, namely synthesis, storage and secretion of thyroid hormones, apparently is closely related to the morphological properties of the follicular structure, preparation and cultivation conditions have to be well defined for revealing stable follicles with definite polarity for functional investigations in vitro. We describe here a method for preparing and cultivating porcine thyroid follicles with a stable preserved polarity (right-side-right follicles). Some characteristic functional properties of these rightside-right follicles, iodide uptake and organification, thyroid hormone secretion, as well as increase in intracellular cyclic AMP (cAMP) levels induced by TSH and isoproterenol, were compared with the properties of the inside-out follicles, which were prepared by the method described by Herzog and Miller (1981). Right-side-right follicles showed a rapid response on cAMP formation with low doses of bTSH and showed iodine uptake and organification. In contrast, inside-out follicles responded to bTSH with a delayed and reduced increase in cAMP levels and did not trap iodine. The results strongly suggest that the TSH and fl-adrenergic receptors are located at the basal membrane of the thyroid cells and that iodide uptake is unidirectional from the basal to the apical membrane. Furthermore, the preparation method described here for porcine follicles with preserved polarity resulted in a primary culture of thyroid follicles with unchanged polarity over a culture period for at least 6 days.
obtained from Sigma Chemie (Taufkirchen, F.R.G.). Cyclic adenosine monophosphate (cAMP) radioimmunoassay (RIA) kit was obtained from Becton and Dickinson (Heidelberg, F.R.G.). Petriperm culture dishes were obtained from Heraeus (Hanau, F.R.G.). All other substances were from Merck (Darmstadt, F.R.G.).
Preparation of thyroid follicles with preserved polarity (right-side-right follicles) Pig thyroid glands were obtained from a local slaughterhouse immediately after sacrifice. The time interval between death and preparation of the thyroid glands was less than 1 h. The main thyroid arteries and veins, which are located at the caudal pole, were prepared, cannulated and infused with collagenase (1.5 m g / m l MEM). About 5-10 ml of collagenase solution, depending on the weight of the thyroid glands, were infused into the artery, and the main veins were occluded, so that the capillary system of the gland was filled with collagenase solution. After incubation for 1 h at 37°C the infused glands had a smooth consistency. The capsule and surrounding connective tissue was removed and the total glands carefully dissected into small pieces. Follicles were liberated by gently shaking in MEM. The follicles were then washed by repeated sedimentation for 2.5 min in 12 ml glass tubes, where most of the closed follicles sedimented, whereas broken follicles and single cells remained in the supernatant. Finally, the isolated follicles were washed under sterile conditions. The yield of closed follicles isolated from 10 thyroid glands was in the range of 1.5 2.5 ml follicle sediment.
Preparation of inside-out follicles Materials and methods
Materials Minimal essential medium (MEM), collagenase
(Clost. histolyticum), streptomycin, penicillin and amphotericin B were obtained from Boehringer (Mannheim, F.R.G.). Fetal calf serum (FCS) was obtained from Serva (Heidelberg, F.R.G.). Soybean trypsin inhibitor, bovine TSH (bTSH), 3-isobutyl1-methylxanthine (IBMX), isoproterenol, epinephrine, norepinephrine and propranolol were
Preparation of porcine thyroid follicles with inverted polarity followed the method described by Herzog and Miller (1981). In brief, thyroid glands were minced with razor blades into small pieces, connective tissue was mechanically removed, and the tissue fragments were incubated with collagenase (1 m g / m l MEM) for 30 min at 37°C. Follicle segments were obtained by disrupting the digested tissue with pasteur pipettes of decreasing diameters (1500-600 /~m). The segments were then filtered through nylon gauze (250
11 and 200/~m) to remove collagen. The follicle segments were then washed as described above.
Culture conditions Follicles and follicle segments were incubated under identical conditions. Unless otherwise indicated, culture medium contained MEM with 1% FCS, penicillin (100 U/ml), streptomycin (100 ~g/ml), amphotericin B (0.5 ~ g / m l ) and soybean trypsin inhibitor (0.1 mg/ml). Culture dishes were supplied with a hydrophobic membrane (Petriperm, Heraeus) to prevent attachment of the follicles to the bottom of the dishes. Incubation was performed for 24 h (resting period) in a humidified atmosphere with 5% CO 2 at 37°C before further functional investigations.
Morphological techniques Isolated follicles were fixed for 1 h in 6.25% (v/v) glutaraldehyde in a 0.1 M cacodylate buffer, pH 7.4, rinsed with 0.2 M sucrose and postfixed with 2% OsO 4. Fixed follicles were embedded in Epon and sections (1-2 /zm) were cut with a Reichart Ultramicrotom. Staining was done with Azur-II-Methylene blue.
Iodine uptake and organification After a resting period of 24 h, follicles were washed twice with MEM, resuspended, and aliquots (400 /~1, corresponding to about 10 ~g DNA) were incubated in siliconized plastic tubes for 2 min. The reaction was started by adding KI (0.1 /~M) and 0.1 /~Ci 125I (final concentrations). The reaction was stopped by centrifugation (15 sec) and washing the cells with ice-cold medium containing KI (0.1 /~M). The pellet was counted for radioactivity (iodide uptake), then precipitated with 10% trichloroacetic acid (TCA), centrifuged again, and the pellet was again counted for radioactivity (iodide organification).
Determination of DNA Follicles were incubated with proteinase K (200 /lg/ml) in 10 mM Tris-HC1, p H 7.0, containing 100 mM NaCI, 10 mM EDTA and 0.1% SDS (w/v) for 24 h. DNA content was measured by the method of Burton (1956) using calf thymus DNA as a standard.
Determination of cAMP Intra- a n d / o r extracellular cAMP levels were determined after stimulation of the follicles with bTSH or catecholamines. The suspension of follicles (450/~1) in MEM containing 1% FCS with or without IBMX was preincubated for 2 min in small siliconized plastic tubes. TSH or catecholamines at various concentrations were then added. After incubation at 37°C the follicles were centrifuged (30 sec, 10000 x g) and 500/~1 of 6% TCA were added to the pellet. The extracts were then shaken 3 times with water-saturated ether, evaporated under a nitrogen atmosphere at 65°C and dissolved in water, cAMP was measured either in the cell extracts (intracellular cAMP) or directly in the medium (extracellular cAMP) by a commercial RIA kit.
Determination of thyroxine secretion Thyroxine (T4) liberated into the medium was measured by a double-antibody radioimmunoassay, as described previously (G~irtner et al., 1980). The lower limit of detection was about 1 nmole/ml. Results
Morphological characteristics The method described here for preparation of porcine thyroid follicles with preserved polarity resulted in a suspension of follicles (Fig. 1A). Most of them had a closed structure, and only some fragments were disrupted out of the follicles. All the smaller segments were removed by the sedimentation procedure. The size of most follicles was in the range of 100-200/xm in diameter; a few of them were smaller. After an incubation period of 24 h in MEM containing 1% FCS, all the follicles were closed and showed some aggregation. The nuclei were located on the outer side of the follicular cells (Fig. 1B). In addition, microvilli were found exclusively at the luminal membrane indicating right-side-right polarity (Fig. 1B, C). When follicles were fixed immediately after isolation, about 20% of the closed follicles contained colloid, as demonstrated by the staining procedure. After incubation for 24 h, the number of colloid-containing follicles increased to more than 50%. By immunohistochemical investigation using pig thyroglobulin antibody, we were able to dem-
12
Fig. 1. Morphology of isolated porcine thyroid follicles. (A) Freshly prepared porcine thyroid follicles in suspension representing a mainly intact follicular structure (× 100). (B) Section through an isolated right-side-right follicle after 1 day in culture. Nuclei are located near the membrane, facing the medium (x640). (C) Under higher magnification, microvilli are seen, facing the follicular lumen (x 1000). (D) Section of an inside-out follicle after 1 day in culture. In contrast to right-side-right follicles, the nuclei are located near the membrane, facing the follicular lumen; microvilli are facing the medium (× 640).
onstrate that this protein was thyroglobulin (data not shown). The polarity of the follicles did not change, even during an incubation period of 6 days in 1% or 5% FCS. The isolation procedure described by Herzog and Miller (1981) resulted in follicle segments. After an incubation period of 20 h in M E M containing 1% FCS, all follicles were closed and showed an inverted polarity with the nuclei located close to the luminal m e m b r a n e and the microvilli facing the medium (Fig. 1D). Some of the follicles had a bilayer of cells or segments enclosed. After incubation for 6 days the follicles were enlarged and the cells had a very small cytoplasm. N o protein could be detected inside the lumen of these follicles.
Functional characteristics Iodide uptake and organification.
Iodide uptake of right-side-right follicles was very rapid. 10% and 30% of the activity was found within the follicles after 2 and 10 min, respectively (Fig. 2A). N o difference was found in iodide uptake with or without b T S H (1 m U / m l ) . Organification of iodide, however, depended on TSH stimulation. 50% of the a m o u n t of iodide which had been trapped by the follicle cells was b o u n d to protein within 2 min during TSH stimulation in rightside-right follicles, compared to 25% without TSH stimulation (Fig. 2B). In contrast to right-side-right follicles, no significant iodide uptake was observed in inside-out follicles even after 10 min (Fig. 2A).
13
301
A Right-side-right-
75
BT~ (rig/dish) /~bTSH
foLlicles
{lOOmU/ml} bTSH (1 mU/mL)
50
zo' "N
25
~bTSH
0
fotlic~
t
0 I
0
t
0
Z
5
I
min.
10
B
60 FoLLicLes e~w
40
~
20
o,i
Medium
I
I
I
0
2
5
I
min. 10
Fig. 2. Iodide uptake (A) and organification (B) of isolated follicles with preserved (right-side-right) and inverted (insideout) polarity. Iodide uptake is expressed as percentage of total activity of 125I. Organification of iodide is expressed as perventage of TCA-precipitable material compared to the total amount of 125I trapped by the follicles at the times indicated.
Thyroid hormone release. During TSH stimulation, 40 ng T4 per dish were released into the incubation medium after 14 h from right-side-right follicles; only 5 ng T4 per dish were released in control experiments without TSH (Fig. 3). Inside-out follicles did not release thyroid hormones into the medium. However, they pro-
I
I
17 14
;
ZO
;
~=
24 h
Fig. 3. Release of T4 into the medium by suspended right-sideright follicles after stimulation with bTSH over 24 h.
duced T4 and T3 when porcine thyroglobulin (Tg) (5 m g / m l ) was added to the incubation medium. 15 ng T4 per dish could be detected after 24 h incubation with 5 m g / m l Tg within the lumen of inside-out follicles after disrupting follicles by adding 0.1 mM EDTA to the medium (results not shown). Changes of cAMP levels induced by b TSH. The formation of cAMP appeared to be dose-dependent to bTSH in right-side-right follicles (Fig. 4). A significant increase in cAMP levels was found with 0.1 m U / m l bTSH. Under these conditions (incubation for 1 h with 1 mM IBMX), a 10-20fold increase of total cAMP (intra- and extracellular cAMP) was found with 1 m U / m l or 100 m U / m l bTSH, respectively, compared to the control value without bTSH. When isolated right-side-right follicles were preincubated in the presence of I/~M KI for 24 h, the increases in cAMP levels were reduced by more than 50% after stimulation with bTSH (10 m U / m l ) , as compared to controls without iodide during the preincubation period (data not shown). A comparison of right-side-right and inside-out follicles on cAMP levels after stimulation with bTSH (100 m U / m l ) is shown in Fig. 5. In insideout follicles only 5 pmoles cAMP per /zg DNA were found after 30 min in comparison to 25 pmoles cAMP per /~g DNA in right-side-right follicles (Fig. 5).
14
pMolcAMP II jug ONA I0
pMolcAMP 30 jug ONA
1.:,; :::::
i',ii
9
Rig ht - side - right-
::iii:i:E:
~
:.;:!::
B
70
fo[licIes
b
TSH
/
(I00mUlrn[)
iiE!i !!iii
7 @
/
L
5
Inside - out - fo{licIes
iiii
4
1::1,
3
iiiii! ,,.-.:,
2
:i:i:
[
I0
~
b TSH • ( 100mUlml)
::::: ?. ii
1;:?,
l
:i:i:i
0
0
iilli:iii
m
,,
10.7 10.6 I0s 10.4 10.3"10I bTSH ( u/ml )
0
Fig. 4. Dose-response curve of bTSH on cAMP formation in isolated fight-side-fight follicles. Follicles were incubated in M E M / I % FCS, containing 1 m M IBMX, for 1 h at 37°C. c A M P levels are the total of intra- and extracellular cAMP (mean + range).
pMolcAMP
jug DNA
1,4
T
1,2
~--
--~.~H
1,0 ~
( 1 mU/mt)
i
0,6
0.4
r
~
0,2 0
lsoproterenol ( 1 juM
~ I
I
!
0,51 Z
I
.5
Control
]~' ¢
rain lO
Fig. 6. Time-course of intracellular cAMP levels in right-sideright follicles after stimulation with bTSH (1 m U / m l , × ) or isoproterenol (1 /~M, O). Conditions for incubation were: M E M / I % FCS with 0.5 m M IBMX.
~'
"
!
I
I
I
0
30
60
~'b TSH
min
Fig. 5. Effects of bTSH (100 m U / m l ) on cAMP levels of right-side-right (C)) and inside-out follicles (O). Follicles were preincubated in M E M / I % FCS with 1 m M IBMX for 2 min at 37°C and the intracellular levels of cAMP were determined 30 and 60 min after addition of bTSH.
Short-term incubation of right-side-right follicles with bTSH (1 m U / m l ) in the presence of IBMX (0.5 mM) induced a significant increase in cAMP levels from 0.3 pmoles cAMP per ~g D N A to 0.75 pmoles cAMP per ~g D N A within 30 sec. A maximum of 1.25 pmoles cAMP per #g D N A was obtained after 2 min. The cAMP levels remained at this high level for about 5 min and decreased slightly after 10 min (Fig. 6). In the absence of IBMX, cAMP levels were increased by bTSH (1 m U / m l ) from 0.2 pmoles cAMP per/~g D N A to 0.32 p m o l e s / # g D N A within 5 rain (data not shown). In short-term experiments with insideout follicles no increase in cAMP levels was found.
Changes of cAMP levels induced by catecholamines. Isoproterenol increased intracellular c A M P levels of right-side-right follicles in a dosedependent manner (Fig. 7). A significant increase in cAMP level was found with 0.1 ~M isoproterenol after a 2 min incubation. The cAMP concentration increased from 0.38 pmoles cAMP per
15 pMot c AMP
Hg ONA 1.~ 1.2 10 0.8
0,6 0.~ 0,2 0
n
1mU/ml TSH Co
licles in MEM containing either 1% or 5% FCS did not change the polarity during a 6 day culture. This is in contrast to the findings of Hanafusa et al. (1982) and Nitsch and Wollman (1980), who demonstrated a change of polarity from rightside-right to inside-out by raising the FCS concentration from 0.5% to 5% FCS, In addition, Mauchamp et al. (1979) demonstrated the influence of culture conditions on cell polarity in hog thyroid cells. One explanation for this discrepancy may be the maintenance of the basal lamina at the follicle cells prepared by our method. The absence of a basal lamina in follicles which were prepared by mincing thyroid lobes and subsequent treatment with collagenase has been demonstrated by Herzog and Miller (1981) and also by Garbi and Wollman (1982). It is known from other cell systems that extracellular matrix (basal lamina) is very important not only for preserving the polarity of the cells but also for maintaining differentiation and growth potency (Rojkind et al., 1980). The comparison of functional properties clearly demonstrates that right-side-right follicles respond very rapidly to TSH stimulation. Although a crude bTSH preparation was used in all experiments, the sensitivity of these isolated follicles in terms of cAMP formation was relatively low (100 t~U/ml). Inside-out follicles, however, showed a delayed and reduced response even with high amounts of bTSH in comparison to right-side-right follicles. In addition, rapid iodide uptake and organification were demonstrated only in righ-side-right follicles, indicating that iodide uptake is unidirectional from the basal to the apical membrane. These data represent indirect evidence that TSH receptor, as well as the iodide 'pump', are located exclusively at the basolateral part of the plasma membrane. This is in agreement with the findings of Chambard et al. (1983). As inside-out follicles, however, respond to TSH in a diminished manner in comparison to right-side, right follicles, it may be assumed that TSH is transported by transcytosis (Herzog, i983) from apical to basolateral plasma membrane, reaching in this way its own receptor. Catecholamines are known to stimulate the adenylate cyclase system in thyroid cells by /3adrenergic receptors. It has been demonstrated that/~-adrenergic substances such as isoproterenol
nNnl,nHnn ~m o.lo 1.o lo loo 1~,4 ~O#M UJM I~oM ljuM ] E NE Isoproterenot(~N)" 1(~H P
Fig. 7. Dose-response curve of intracellular cAMP formation after stimulation for 2 min with catecholamines (isoproterenol, I; epinephrine, E; norepinephrine, NE) or bTSH (1 mU/ml). The increase in cAMP level induced by isoproterenol could be blocked by preincubation (1 min) with 10/~M propranolol (P).
/~g DNA to 0.72 pmoles cAMP per #g DNA when the isoproterenol concentration was increased from 0.1 to 100/~M. This rise in the cAMP concentration could be blocked by the addition of propranolol (10 ~M). Epinephrine and norepinephrine induced only a slight increase in cAMP levels, which was significantly lower than the increase in cAMP levels with isoproterenol. Inside-out follicles did not respond to isoproterenol, epinephrine and norepinephrine with a change in cAMP levels. The time-course of cAMP concentration in right-side-right follicles during stimulation with isoproterenol was different from that of TSH (Fig. 6). After stimulation with isoproterenol (1 #M) the cAMP levels reached a maximum after 1 min and then decreased to control values within 10 min, whereas with bTSH (1 m U / m l ) the curve showed a plateau between 2 and 5 min and decreased slightly after 10 min. Discussion
We describe here a new preparation procedure for thyroid follicles with preserved polarity. We demonstrate that incubation of right-side-right fol-
16
stimulate cAMP levels in dog thyroid slices (Dumont et al., 1978), in thyroid membrane preparations (Marshall et al., 1975) and in human thyroid cells (Toccafondi et al., 1983). We therefore investigated the responsiveness of right-sideright follicles to isoproterenol in comparison to inside-out follicles. Because only in right-side-right follicles was a response of cAMP accumulation found, these findings strongly suggest that the /3-adrenergic receptor, too, is located only at the basolateral part of the plasma membrane of follicle cells. In summary, these functional characteristics of porcine thyroid follicles with preserved polarity in comparison to inside-out follicles reveal indirect evidence that TSH and fl-adrenergic receptors are located at the basolateral part of the plasma membrane, and that the iodide uptake also occurs only at this part of the plasma membrane. The method described here for the preparation of thyroid follicles with preserved polarity reveals stable follicles which do not change their polarity and may be an important tool for further investigation of thyroid follicle function and growth. As the incubation conditions for inside-out follicles and right-side-right follicles were identical, we conclude that in our system the preparation method and probably the preservation of the basal lamina is more important for the maintenance of cell polarity than the culture conditions.
Acknowledgements We wish to thank Mrs. R. Buric and P. Rank for their excellent technical assistance and Miss E. Marton for preparation of the manuscript. We thank Dr. V. Herzog (Institut f~r Zellbiologie,
Mianchen) for his help in the preparation of insideout follicles. We also thank Dr. R. Gerzer for helpful suggestions during the preparation of this manuscript.
References Burton, K. (1956) Biochem. J. 62, 315-323. Chambard, M., Verrier, B., Gabrion, J. and Mauchamp, J. (1983) J. Cell Biol. 96, 1172-1177. Denef, J.-F., BjOrkman, U. and Ekholm, R. (1980) J. Ultrastruct. Res. 71, 185-202. Dumont, J.E., Boeynaems, J.M., Decoster, C., Erneux, C., Lamy, F., Lecocq, R., Mockel, J., Unger, J. and van Sande, J. (1978) Adv. Cyclic Nucleotide Res. 9, 723-734. Fayet, G., Michel-Brchet, M. and Lissitzky, S. (1971) Eur. J. Biochem. 24, 100-111. Filetti, S. and Rapoport, B. (1983) Endocrinology 113, 1608-1615. Garbi, C. and Wollman, S.H. (1982) J. Cell Biol. 94, 489-492. Gartner, R., Kewenig, M., Horn, K. and Scriba, P.C. (1980) J. Clin. Chem. Clin. Biochem. 18, 571-577. Hanafusa, T., Pujol-Burrell, R., Chiovato, L., Hammond, L.J. and Bottazzo, G.F. (1982) Ann. Endocrinol. 43, 30. Herzog, V. (1983) J. Cell Biol. 97, 607-617. Herzog, V. and Miller, F. (1981) Eur. J. Cell Biol. 24, 74-84. Karlsson, F.A., Westermark, K. and Westermark, B. (1982) Mol. Cell. Endocrinol. 28, 99-112. Kerkof, P.R., Long, P.J. and Chaikoff, I.L. (1964) Endocrinology 74, 170-179. Marshall, N.J., von Borcke, S. and Malan, P.G. (1975) Endocrinology 96, 1520-1524. Mauchamp, J., Margotat, A., Chambard, M., Charrier, B., Remy, L. and Michel-Brchet, M. (1979) Cell Tissue Res. 204, 417-430. Nitsch, L. and Wollman, S.H. (1980a) J. Cell Biol. 86, 875-880. Nitsch, L. and Wollman, S.H. (1980b) Proc. Natl. Acad. Sci. (U.S.A.) 77, 472-476. Rojkind, M., Gatmaitan, Z., Mackensen, S., Giambrone, M.-A., Ponce, P. and Reid, L.M. (1980) J. Cell Biol. 87, 255-263. Toccafondi, R.S., Brandi, M.L., Rotella, C.M. and Zonefrati, R. (1983) Acta Endocrinol. (Kbh.), 102, 62-67.
9
MCE 01273
Preparation of porcine thyroid follicles with preserved polarity: functional and morphological properties in comparison to inside-out follicles * R. G g r t n e r , W. Greil, D. Sti~bner, W. P e r m a n e t t e r l, K. H o r n a n d C . R . P i c k a r d t Medizinische Klinik lnnenstadt and I Pathologisches Institut, Universiti~t M~nchen (F.R.G.)
(Received 20 September 1"984; accepted 19 November 1984)
Keywords: isolation of thyroid follicles; cell polarity; cyclic AMP, thyrotropin
Summary One of the main problems in establishing isolated thyroid follicles in vitro is their tendency to form inside-out follicles. The reason for this change in polarity is unknown. We describe here a method for the preparation of stable thyroid follicles with preserved polarity for at least 6 days. Isolated follicles were obtained by infusion of collagenase (1.5 mg/ml) dissolved in minimal essential medium into the artery of intact thyroid glands. The morphological and functional properties of these follicles were compared to inside-out follicles. These inside-out follicles were obtained by digestion of minced thyroid tissue in a collagenase (1 mg/ml) solution. The polarity of follicles was proved by morphological criteria. Follicles with preserved polarity did not change polarity for at least 6 days in the presence of 1% or 5% fetal calf serum. As the culture conditions for both preparation methods were identical, we conclude that the preparation method rather than the culture condition is responsible for the preservation of cell polarity of isolated thyroid follicles in our system. Increases in cyclic AMP levels induced either by bovine thyrotropin (10 -3 U/ml) or by isoproterenol (10 -6 M) as well as iodide uptake and organification were rapid and significant in right-side-right follicles but not in inside-out follicles. Therefore the TSH receptor and the fl-adrenergic receptor appear to be exclusively located at the basal membrane of follicular cells. In addition, iodide uptake apparently is unidirectional.
About 20 years ago, Kerkof et al. (1964) first succeeded in inducing follicle-like structures in thyroid cell culture during TSH stimulation. Thereafter various methods have been reported for the isolation of thyroid follicles from different species such as rat, pig and man (Mauchamp et al., 1979; Nitsch and Wollman, 1980b; Denef et al., 1980; Herzog and Miller, 1981; Karlsson et al., * Supported by the Deutsche Forschungsgemeinschaft. A preliminary report of this work was presented at the 14th Annual Meeting of the European Thyroid Association, Rotterdam, 1984.
1982; Hanafusa et al., 1982). One of the main difficulties in the preparation and cultivation of thyroid follicles in vitro is their tendency to form follicles with inverted polarity (inside-out follicles). A change from normal to inverted polarity appears to occur when the serum concentration in the culture medium is raised from 0.5% to 5% fetal calf serum in cultured rat follicles (Nitsch and Wollman, 1980a) or in human follicles (Hanafusa et al., 1982). In addition, isolated cells from hog thyroids cultured in plastic dishes not treated for tissue culture appear to reorganize into 'hollow spheres', which have the morphological character-
0303-7207/85/$03.30 © 1985 Elsevier Scientific Publishers Ireland, Ltd.
10 istics of follicles with inverted polarity (Mauchamp et al., 1979). Inversion of cell polarity in isolated thyroid follicles seems to be dependent on the preparation method and culture conditions, but the reason for this phenomenon is still unclear. Because the main function of the thyroid gland, namely synthesis, storage and secretion of thyroid hormones, apparently is closely related to the morphological properties of the follicular structure, preparation and cultivation conditions have to be well defined for revealing stable follicles with definite polarity for functional investigations in vitro. We describe here a method for preparing and cultivating porcine thyroid follicles with a stable preserved polarity (right-side-right follicles). Some characteristic functional properties of these rightside-right follicles, iodide uptake and organification, thyroid hormone secretion, as well as increase in intracellular cyclic AMP (cAMP) levels induced by TSH and isoproterenol, were compared with the properties of the inside-out follicles, which were prepared by the method described by Herzog and Miller (1981). Right-side-right follicles showed a rapid response on cAMP formation with low doses of bTSH and showed iodine uptake and organification. In contrast, inside-out follicles responded to bTSH with a delayed and reduced increase in cAMP levels and did not trap iodine. The results strongly suggest that the TSH and fl-adrenergic receptors are located at the basal membrane of the thyroid cells and that iodide uptake is unidirectional from the basal to the apical membrane. Furthermore, the preparation method described here for porcine follicles with preserved polarity resulted in a primary culture of thyroid follicles with unchanged polarity over a culture period for at least 6 days.
obtained from Sigma Chemie (Taufkirchen, F.R.G.). Cyclic adenosine monophosphate (cAMP) radioimmunoassay (RIA) kit was obtained from Becton and Dickinson (Heidelberg, F.R.G.). Petriperm culture dishes were obtained from Heraeus (Hanau, F.R.G.). All other substances were from Merck (Darmstadt, F.R.G.).
Preparation of thyroid follicles with preserved polarity (right-side-right follicles) Pig thyroid glands were obtained from a local slaughterhouse immediately after sacrifice. The time interval between death and preparation of the thyroid glands was less than 1 h. The main thyroid arteries and veins, which are located at the caudal pole, were prepared, cannulated and infused with collagenase (1.5 m g / m l MEM). About 5-10 ml of collagenase solution, depending on the weight of the thyroid glands, were infused into the artery, and the main veins were occluded, so that the capillary system of the gland was filled with collagenase solution. After incubation for 1 h at 37°C the infused glands had a smooth consistency. The capsule and surrounding connective tissue was removed and the total glands carefully dissected into small pieces. Follicles were liberated by gently shaking in MEM. The follicles were then washed by repeated sedimentation for 2.5 min in 12 ml glass tubes, where most of the closed follicles sedimented, whereas broken follicles and single cells remained in the supernatant. Finally, the isolated follicles were washed under sterile conditions. The yield of closed follicles isolated from 10 thyroid glands was in the range of 1.5 2.5 ml follicle sediment.
Preparation of inside-out follicles Materials and methods
Materials Minimal essential medium (MEM), collagenase
(Clost. histolyticum), streptomycin, penicillin and amphotericin B were obtained from Boehringer (Mannheim, F.R.G.). Fetal calf serum (FCS) was obtained from Serva (Heidelberg, F.R.G.). Soybean trypsin inhibitor, bovine TSH (bTSH), 3-isobutyl1-methylxanthine (IBMX), isoproterenol, epinephrine, norepinephrine and propranolol were
Preparation of porcine thyroid follicles with inverted polarity followed the method described by Herzog and Miller (1981). In brief, thyroid glands were minced with razor blades into small pieces, connective tissue was mechanically removed, and the tissue fragments were incubated with collagenase (1 m g / m l MEM) for 30 min at 37°C. Follicle segments were obtained by disrupting the digested tissue with pasteur pipettes of decreasing diameters (1500-600 /~m). The segments were then filtered through nylon gauze (250
11 and 200/~m) to remove collagen. The follicle segments were then washed as described above.
Culture conditions Follicles and follicle segments were incubated under identical conditions. Unless otherwise indicated, culture medium contained MEM with 1% FCS, penicillin (100 U/ml), streptomycin (100 ~g/ml), amphotericin B (0.5 ~ g / m l ) and soybean trypsin inhibitor (0.1 mg/ml). Culture dishes were supplied with a hydrophobic membrane (Petriperm, Heraeus) to prevent attachment of the follicles to the bottom of the dishes. Incubation was performed for 24 h (resting period) in a humidified atmosphere with 5% CO 2 at 37°C before further functional investigations.
Morphological techniques Isolated follicles were fixed for 1 h in 6.25% (v/v) glutaraldehyde in a 0.1 M cacodylate buffer, pH 7.4, rinsed with 0.2 M sucrose and postfixed with 2% OsO 4. Fixed follicles were embedded in Epon and sections (1-2 /zm) were cut with a Reichart Ultramicrotom. Staining was done with Azur-II-Methylene blue.
Iodine uptake and organification After a resting period of 24 h, follicles were washed twice with MEM, resuspended, and aliquots (400 /~1, corresponding to about 10 ~g DNA) were incubated in siliconized plastic tubes for 2 min. The reaction was started by adding KI (0.1 /~M) and 0.1 /~Ci 125I (final concentrations). The reaction was stopped by centrifugation (15 sec) and washing the cells with ice-cold medium containing KI (0.1 /~M). The pellet was counted for radioactivity (iodide uptake), then precipitated with 10% trichloroacetic acid (TCA), centrifuged again, and the pellet was again counted for radioactivity (iodide organification).
Determination of DNA Follicles were incubated with proteinase K (200 /lg/ml) in 10 mM Tris-HC1, p H 7.0, containing 100 mM NaCI, 10 mM EDTA and 0.1% SDS (w/v) for 24 h. DNA content was measured by the method of Burton (1956) using calf thymus DNA as a standard.
Determination of cAMP Intra- a n d / o r extracellular cAMP levels were determined after stimulation of the follicles with bTSH or catecholamines. The suspension of follicles (450/~1) in MEM containing 1% FCS with or without IBMX was preincubated for 2 min in small siliconized plastic tubes. TSH or catecholamines at various concentrations were then added. After incubation at 37°C the follicles were centrifuged (30 sec, 10000 x g) and 500/~1 of 6% TCA were added to the pellet. The extracts were then shaken 3 times with water-saturated ether, evaporated under a nitrogen atmosphere at 65°C and dissolved in water, cAMP was measured either in the cell extracts (intracellular cAMP) or directly in the medium (extracellular cAMP) by a commercial RIA kit.
Determination of thyroxine secretion Thyroxine (T4) liberated into the medium was measured by a double-antibody radioimmunoassay, as described previously (G~irtner et al., 1980). The lower limit of detection was about 1 nmole/ml. Results
Morphological characteristics The method described here for preparation of porcine thyroid follicles with preserved polarity resulted in a suspension of follicles (Fig. 1A). Most of them had a closed structure, and only some fragments were disrupted out of the follicles. All the smaller segments were removed by the sedimentation procedure. The size of most follicles was in the range of 100-200/xm in diameter; a few of them were smaller. After an incubation period of 24 h in MEM containing 1% FCS, all the follicles were closed and showed some aggregation. The nuclei were located on the outer side of the follicular cells (Fig. 1B). In addition, microvilli were found exclusively at the luminal membrane indicating right-side-right polarity (Fig. 1B, C). When follicles were fixed immediately after isolation, about 20% of the closed follicles contained colloid, as demonstrated by the staining procedure. After incubation for 24 h, the number of colloid-containing follicles increased to more than 50%. By immunohistochemical investigation using pig thyroglobulin antibody, we were able to dem-
12
Fig. 1. Morphology of isolated porcine thyroid follicles. (A) Freshly prepared porcine thyroid follicles in suspension representing a mainly intact follicular structure (× 100). (B) Section through an isolated right-side-right follicle after 1 day in culture. Nuclei are located near the membrane, facing the medium (x640). (C) Under higher magnification, microvilli are seen, facing the follicular lumen (x 1000). (D) Section of an inside-out follicle after 1 day in culture. In contrast to right-side-right follicles, the nuclei are located near the membrane, facing the follicular lumen; microvilli are facing the medium (× 640).
onstrate that this protein was thyroglobulin (data not shown). The polarity of the follicles did not change, even during an incubation period of 6 days in 1% or 5% FCS. The isolation procedure described by Herzog and Miller (1981) resulted in follicle segments. After an incubation period of 20 h in M E M containing 1% FCS, all follicles were closed and showed an inverted polarity with the nuclei located close to the luminal m e m b r a n e and the microvilli facing the medium (Fig. 1D). Some of the follicles had a bilayer of cells or segments enclosed. After incubation for 6 days the follicles were enlarged and the cells had a very small cytoplasm. N o protein could be detected inside the lumen of these follicles.
Functional characteristics Iodide uptake and organification.
Iodide uptake of right-side-right follicles was very rapid. 10% and 30% of the activity was found within the follicles after 2 and 10 min, respectively (Fig. 2A). N o difference was found in iodide uptake with or without b T S H (1 m U / m l ) . Organification of iodide, however, depended on TSH stimulation. 50% of the a m o u n t of iodide which had been trapped by the follicle cells was b o u n d to protein within 2 min during TSH stimulation in rightside-right follicles, compared to 25% without TSH stimulation (Fig. 2B). In contrast to right-side-right follicles, no significant iodide uptake was observed in inside-out follicles even after 10 min (Fig. 2A).
13
301
A Right-side-right-
75
BT~ (rig/dish) /~bTSH
foLlicles
{lOOmU/ml} bTSH (1 mU/mL)
50
zo' "N
25
~bTSH
0
fotlic~
t
0 I
0
t
0
Z
5
I
min.
10
B
60 FoLLicLes e~w
40
~
20
o,i
Medium
I
I
I
0
2
5
I
min. 10
Fig. 2. Iodide uptake (A) and organification (B) of isolated follicles with preserved (right-side-right) and inverted (insideout) polarity. Iodide uptake is expressed as percentage of total activity of 125I. Organification of iodide is expressed as perventage of TCA-precipitable material compared to the total amount of 125I trapped by the follicles at the times indicated.
Thyroid hormone release. During TSH stimulation, 40 ng T4 per dish were released into the incubation medium after 14 h from right-side-right follicles; only 5 ng T4 per dish were released in control experiments without TSH (Fig. 3). Inside-out follicles did not release thyroid hormones into the medium. However, they pro-
I
I
17 14
;
ZO
;
~=
24 h
Fig. 3. Release of T4 into the medium by suspended right-sideright follicles after stimulation with bTSH over 24 h.
duced T4 and T3 when porcine thyroglobulin (Tg) (5 m g / m l ) was added to the incubation medium. 15 ng T4 per dish could be detected after 24 h incubation with 5 m g / m l Tg within the lumen of inside-out follicles after disrupting follicles by adding 0.1 mM EDTA to the medium (results not shown). Changes of cAMP levels induced by b TSH. The formation of cAMP appeared to be dose-dependent to bTSH in right-side-right follicles (Fig. 4). A significant increase in cAMP levels was found with 0.1 m U / m l bTSH. Under these conditions (incubation for 1 h with 1 mM IBMX), a 10-20fold increase of total cAMP (intra- and extracellular cAMP) was found with 1 m U / m l or 100 m U / m l bTSH, respectively, compared to the control value without bTSH. When isolated right-side-right follicles were preincubated in the presence of I/~M KI for 24 h, the increases in cAMP levels were reduced by more than 50% after stimulation with bTSH (10 m U / m l ) , as compared to controls without iodide during the preincubation period (data not shown). A comparison of right-side-right and inside-out follicles on cAMP levels after stimulation with bTSH (100 m U / m l ) is shown in Fig. 5. In insideout follicles only 5 pmoles cAMP per /zg DNA were found after 30 min in comparison to 25 pmoles cAMP per /~g DNA in right-side-right follicles (Fig. 5).
14
pMolcAMP II jug ONA I0
pMolcAMP 30 jug ONA
1.:,; :::::
i',ii
9
Rig ht - side - right-
::iii:i:E:
~
:.;:!::
B
70
fo[licIes
b
TSH
/
(I00mUlrn[)
iiE!i !!iii
7 @
/
L
5
Inside - out - fo{licIes
iiii
4
1::1,
3
iiiii! ,,.-.:,
2
:i:i:
[
I0
~
b TSH • ( 100mUlml)
::::: ?. ii
1;:?,
l
:i:i:i
0
0
iilli:iii
m
,,
10.7 10.6 I0s 10.4 10.3"10I bTSH ( u/ml )
0
Fig. 4. Dose-response curve of bTSH on cAMP formation in isolated fight-side-fight follicles. Follicles were incubated in M E M / I % FCS, containing 1 m M IBMX, for 1 h at 37°C. c A M P levels are the total of intra- and extracellular cAMP (mean + range).
pMolcAMP
jug DNA
1,4
T
1,2
~--
--~.~H
1,0 ~
( 1 mU/mt)
i
0,6
0.4
r
~
0,2 0
lsoproterenol ( 1 juM
~ I
I
!
0,51 Z
I
.5
Control
]~' ¢
rain lO
Fig. 6. Time-course of intracellular cAMP levels in right-sideright follicles after stimulation with bTSH (1 m U / m l , × ) or isoproterenol (1 /~M, O). Conditions for incubation were: M E M / I % FCS with 0.5 m M IBMX.
~'
"
!
I
I
I
0
30
60
~'b TSH
min
Fig. 5. Effects of bTSH (100 m U / m l ) on cAMP levels of right-side-right (C)) and inside-out follicles (O). Follicles were preincubated in M E M / I % FCS with 1 m M IBMX for 2 min at 37°C and the intracellular levels of cAMP were determined 30 and 60 min after addition of bTSH.
Short-term incubation of right-side-right follicles with bTSH (1 m U / m l ) in the presence of IBMX (0.5 mM) induced a significant increase in cAMP levels from 0.3 pmoles cAMP per ~g D N A to 0.75 pmoles cAMP per ~g D N A within 30 sec. A maximum of 1.25 pmoles cAMP per #g D N A was obtained after 2 min. The cAMP levels remained at this high level for about 5 min and decreased slightly after 10 min (Fig. 6). In the absence of IBMX, cAMP levels were increased by bTSH (1 m U / m l ) from 0.2 pmoles cAMP per/~g D N A to 0.32 p m o l e s / # g D N A within 5 rain (data not shown). In short-term experiments with insideout follicles no increase in cAMP levels was found.
Changes of cAMP levels induced by catecholamines. Isoproterenol increased intracellular c A M P levels of right-side-right follicles in a dosedependent manner (Fig. 7). A significant increase in cAMP level was found with 0.1 ~M isoproterenol after a 2 min incubation. The cAMP concentration increased from 0.38 pmoles cAMP per
15 pMot c AMP
Hg ONA 1.~ 1.2 10 0.8
0,6 0.~ 0,2 0
n
1mU/ml TSH Co
licles in MEM containing either 1% or 5% FCS did not change the polarity during a 6 day culture. This is in contrast to the findings of Hanafusa et al. (1982) and Nitsch and Wollman (1980), who demonstrated a change of polarity from rightside-right to inside-out by raising the FCS concentration from 0.5% to 5% FCS, In addition, Mauchamp et al. (1979) demonstrated the influence of culture conditions on cell polarity in hog thyroid cells. One explanation for this discrepancy may be the maintenance of the basal lamina at the follicle cells prepared by our method. The absence of a basal lamina in follicles which were prepared by mincing thyroid lobes and subsequent treatment with collagenase has been demonstrated by Herzog and Miller (1981) and also by Garbi and Wollman (1982). It is known from other cell systems that extracellular matrix (basal lamina) is very important not only for preserving the polarity of the cells but also for maintaining differentiation and growth potency (Rojkind et al., 1980). The comparison of functional properties clearly demonstrates that right-side-right follicles respond very rapidly to TSH stimulation. Although a crude bTSH preparation was used in all experiments, the sensitivity of these isolated follicles in terms of cAMP formation was relatively low (100 t~U/ml). Inside-out follicles, however, showed a delayed and reduced response even with high amounts of bTSH in comparison to right-side-right follicles. In addition, rapid iodide uptake and organification were demonstrated only in righ-side-right follicles, indicating that iodide uptake is unidirectional from the basal to the apical membrane. These data represent indirect evidence that TSH receptor, as well as the iodide 'pump', are located exclusively at the basolateral part of the plasma membrane. This is in agreement with the findings of Chambard et al. (1983). As inside-out follicles, however, respond to TSH in a diminished manner in comparison to right-side, right follicles, it may be assumed that TSH is transported by transcytosis (Herzog, i983) from apical to basolateral plasma membrane, reaching in this way its own receptor. Catecholamines are known to stimulate the adenylate cyclase system in thyroid cells by /3adrenergic receptors. It has been demonstrated that/~-adrenergic substances such as isoproterenol
nNnl,nHnn ~m o.lo 1.o lo loo 1~,4 ~O#M UJM I~oM ljuM ] E NE Isoproterenot(~N)" 1(~H P
Fig. 7. Dose-response curve of intracellular cAMP formation after stimulation for 2 min with catecholamines (isoproterenol, I; epinephrine, E; norepinephrine, NE) or bTSH (1 mU/ml). The increase in cAMP level induced by isoproterenol could be blocked by preincubation (1 min) with 10/~M propranolol (P).
/~g DNA to 0.72 pmoles cAMP per #g DNA when the isoproterenol concentration was increased from 0.1 to 100/~M. This rise in the cAMP concentration could be blocked by the addition of propranolol (10 ~M). Epinephrine and norepinephrine induced only a slight increase in cAMP levels, which was significantly lower than the increase in cAMP levels with isoproterenol. Inside-out follicles did not respond to isoproterenol, epinephrine and norepinephrine with a change in cAMP levels. The time-course of cAMP concentration in right-side-right follicles during stimulation with isoproterenol was different from that of TSH (Fig. 6). After stimulation with isoproterenol (1 #M) the cAMP levels reached a maximum after 1 min and then decreased to control values within 10 min, whereas with bTSH (1 m U / m l ) the curve showed a plateau between 2 and 5 min and decreased slightly after 10 min. Discussion
We describe here a new preparation procedure for thyroid follicles with preserved polarity. We demonstrate that incubation of right-side-right fol-
16
stimulate cAMP levels in dog thyroid slices (Dumont et al., 1978), in thyroid membrane preparations (Marshall et al., 1975) and in human thyroid cells (Toccafondi et al., 1983). We therefore investigated the responsiveness of right-sideright follicles to isoproterenol in comparison to inside-out follicles. Because only in right-side-right follicles was a response of cAMP accumulation found, these findings strongly suggest that the /3-adrenergic receptor, too, is located only at the basolateral part of the plasma membrane of follicle cells. In summary, these functional characteristics of porcine thyroid follicles with preserved polarity in comparison to inside-out follicles reveal indirect evidence that TSH and fl-adrenergic receptors are located at the basolateral part of the plasma membrane, and that the iodide uptake also occurs only at this part of the plasma membrane. The method described here for the preparation of thyroid follicles with preserved polarity reveals stable follicles which do not change their polarity and may be an important tool for further investigation of thyroid follicle function and growth. As the incubation conditions for inside-out follicles and right-side-right follicles were identical, we conclude that in our system the preparation method and probably the preservation of the basal lamina is more important for the maintenance of cell polarity than the culture conditions.
Acknowledgements We wish to thank Mrs. R. Buric and P. Rank for their excellent technical assistance and Miss E. Marton for preparation of the manuscript. We thank Dr. V. Herzog (Institut f~r Zellbiologie,
Mianchen) for his help in the preparation of insideout follicles. We also thank Dr. R. Gerzer for helpful suggestions during the preparation of this manuscript.
References Burton, K. (1956) Biochem. J. 62, 315-323. Chambard, M., Verrier, B., Gabrion, J. and Mauchamp, J. (1983) J. Cell Biol. 96, 1172-1177. Denef, J.-F., BjOrkman, U. and Ekholm, R. (1980) J. Ultrastruct. Res. 71, 185-202. Dumont, J.E., Boeynaems, J.M., Decoster, C., Erneux, C., Lamy, F., Lecocq, R., Mockel, J., Unger, J. and van Sande, J. (1978) Adv. Cyclic Nucleotide Res. 9, 723-734. Fayet, G., Michel-Brchet, M. and Lissitzky, S. (1971) Eur. J. Biochem. 24, 100-111. Filetti, S. and Rapoport, B. (1983) Endocrinology 113, 1608-1615. Garbi, C. and Wollman, S.H. (1982) J. Cell Biol. 94, 489-492. Gartner, R., Kewenig, M., Horn, K. and Scriba, P.C. (1980) J. Clin. Chem. Clin. Biochem. 18, 571-577. Hanafusa, T., Pujol-Burrell, R., Chiovato, L., Hammond, L.J. and Bottazzo, G.F. (1982) Ann. Endocrinol. 43, 30. Herzog, V. (1983) J. Cell Biol. 97, 607-617. Herzog, V. and Miller, F. (1981) Eur. J. Cell Biol. 24, 74-84. Karlsson, F.A., Westermark, K. and Westermark, B. (1982) Mol. Cell. Endocrinol. 28, 99-112. Kerkof, P.R., Long, P.J. and Chaikoff, I.L. (1964) Endocrinology 74, 170-179. Marshall, N.J., von Borcke, S. and Malan, P.G. (1975) Endocrinology 96, 1520-1524. Mauchamp, J., Margotat, A., Chambard, M., Charrier, B., Remy, L. and Michel-Brchet, M. (1979) Cell Tissue Res. 204, 417-430. Nitsch, L. and Wollman, S.H. (1980a) J. Cell Biol. 86, 875-880. Nitsch, L. and Wollman, S.H. (1980b) Proc. Natl. Acad. Sci. (U.S.A.) 77, 472-476. Rojkind, M., Gatmaitan, Z., Mackensen, S., Giambrone, M.-A., Ponce, P. and Reid, L.M. (1980) J. Cell Biol. 87, 255-263. Toccafondi, R.S., Brandi, M.L., Rotella, C.M. and Zonefrati, R. (1983) Acta Endocrinol. (Kbh.), 102, 62-67.