Characterization of adenohypophysial polypeptides by two-dimensional gel electrophoresis

~o~~&~~r and Ceffular ~ndocri~~olo~, 24 (1981) 165-179 Elsevier/North-Holland Scientific Publishers, Ltd.

CHA~CTERIZATION TWO-DIMENSIONAL

OF ADENOHYPOPHYSIAL

165

POLYPE~IDES

BY

GEL ELECTROPHORESIS

I. L-[ 3H]LEUCINE-LABELED

POLYPEPTIDES

A. ZANINI and P. ROSA Department of Pharmacology and CNR Center of Cytopharmacology, Universityof Milan, Milan (Italy) Received 26 March 1981; revision received 1 July L981; accepted 2 July 1981

Homogenates of cow and rat anterior pituitary slices, labeled in vitro with L-[ 3H]leucine, were analyzed by high-resolution twodimensional polyacrylamide gel electrophoresis. This technique was also applied to the materials released into the chase medium from bovine anterior pituitary slices. The pattern of both total polypeptides (reveaLed by Coomassie-Blue staining) and L-13H]leucine-labeled polypeptides (revealed by fiuorography) was found to be more complex than previously demonstrated by different techniques. In particular, the GH band separated by one~~ension~ Na-dode~yisulfate-polya~ryl~ide gel electrophoresis was resolved into 3 -5 components; 2 of these, which were highly labeled by L-[ 3H]leucine, were both identified as GH by immunoprecipitation with specific anti-GH bodies. In addition, we found evidence in favor of the existence of some, previously unsuspected, ‘putative’ secretory proteins. In fact, besides GH and PRL, several minor components (2 with apparent M, 70000-62000, pI - 4.8; others with M, - 50000, pl between -5.8 and -6.8; and 1 with M, - 26 000, pZ - 5.7) were found to be synthesized at high rates and to accumulate in the medium, with different kinetics, during chase incubation. Keywords: adenohypophysis;

fluorography; GH; secretory proteins.

Much of the present knowledge on pituitary polypeptides stems from the analysis, of either homogenates of the glandular tissue or hormone preparations purified by conventional chromatographic procedures, carried out by polyacrylamide gel electrophoresis (under non-denaturing or denaturing conditions) and by isoelectrofocusing. These techniques, however, are not entirely adequate to study gland homogenates. In fact, by electrophoresis under non-denaturing conditions only few, poorly resolved bands are separated. The major 2 of them were shown by bioassay Abbreviations: lD, oned~ension~; 2D, twod~ension~; KRB, Krebs-Ringer bicarbonate; NEPHGE, nonequilibrium pHgradiknt electrophoresis; PAGE, polyacrylamide gel electrophoresis; SDS, Nadodecylsulfate; TCA, trichloroacetic acid. 030-7207/81/0000-0000/$02.75

Q 1981 ~sevier/North-Holland

Scientific Publishers, Ltd.

A. Zanini, P. Rosa

166

and radioimmunoassay to contain GH and PRL, respectively (Kurcz et al., 1969; Nicoll et al., 1969; Nicholson, 1969; Yanai and Nagaswa, 1969; Lewis et al., 1969; Zanini and Giannattasio, 1972), but these hormones smeared also in other regions of the gel (Russel et al., 1978; Asawaroengchai et al., 1978). By the introduction of Na-dodecylsulfate (SDS)-polyacrylamide gel electrophoresis (PAGE), numerous polypeptides (-50) could be separated besides GH and PRL (Zanini et al., 1974, 1979, 1980). However, by this technique, as well as by polyacrylamide gel isoelectrofocusing, it was not possible to achieve complete polypeptide characterization. In fact, the bands obtained by these 2 procedures are expected to contain not individual molecules but groups of polypeptides with similar molecular weights in the first case, and similar isoelectric points in the latter. On the other hand, the study by PAGE and/or isoelectrofocusing of purified pituitary hormone preparations obtained from various animal species showed that these hormones, in particular GH and PRL, are not molecularly homogeneous, but composed of families of closely homologous, yet heterogeneous, peptides (Yadley et al., 1973; Lorenson and Ellis, 1975; Hummel et al., 1975; Wallis and Davies, 1976; Lewis et al., 1980). However, of the various types of heterogeneity so far described, only a few have been unambiguously shown to exist in fresh homogenates by studies at the molecular level (Lewis et al., 1971; Sinha and Baxter, 1978). In other cases the possibility still exists that they are artifactually produced during the isolation of the hormone preparations (Wallis and Davies, 1976). The above considerations point to the need of a methodology which could be applied to freshly prepared preparations and yield a high degree of analytical resolution. Twodimensional (2D)-PAGE combines the separations according to charge by isoelectrofocusing and according to size by SDS-PAGE and therefore allows the study of individual polypeptides. In our studies we analyzed, by 2D-PAGE, homogenates of rat and cow pituitary glands, as well as the material released into the medium during incubation of bovine pituitary slices in vitro. Our evidence indicates that, especially when used in conjunction with immunochemical and other revelation procedures, this technique represents a powerful tool for the separation and identification of pituitary polypeptides. Here we report a study on both the Coomassie-Blue-stained pattern and the polypeptide labeling with L-[3H]leucine. Analogous labeling results, obtained by using [35S]sulfate and D-[6-3H]glucosamine, are reported in our companion article.

MATERIALS

AND METHODS

Animals Female Sprague-Dawley rats (200-250 g), random with respect to stage of estrus cycle, were housed in a controlled temperature room (21-22”C), with 14-h light and 10-h dark periods. Standard laboratory food and water were given ad

2D-PAGE of L-/3Hlleucine-labeled

pituitary polypeptides

167

libitum. The animals were killed by decapitation, their pituitaries excised and the posterior and intermediate lobes removed. Holstein-Friesian cows (2-3 years old) were obtained from the Milan Abbatoir: anterior pituitary glands were removed soon after slaughter. Rat and bovine adenohypophyses were quickly transferred to ice-cold, oxygenated Krebs-Ringer bicarbonate solution (pH 7.4) containing glucose and amino acids (KRB), as previously described (Zanini et al., 1980). Incubation procedures Rat glands were cut with a razor blade into -6 slices, and bovine glands were sectioned into slices about 0.5 mm thick, with a Stadie-Riggs hand microtome. Slices from 4 rat glands or -l/l0 bovine hypophysis were transferred to lo-ml erlenmeyer flasks containing 1 ml of KRB medium plus L-[3H]leucine (100 PCi; spec. act., 1 mCi/pmole). Incubation was carried out for 10 min at 4°C to allow even penetration of the isotope into the cells and then for 180 min at 37”C, under 95% 02-5% CO?, in a shaking bath operated at 60 cycles/mm. At the end of the incubations, pituitary tissue fragments were collected and homogenized in 0.32 M sucrose; cell debris and erythrocytes were removed by a short low-speed centrifugation. In the experiments designed to analyze the adenohypophysial products discharged into the medium, bovine pituitary slices were labeled with L-[3H]leucine for 90 min, rinsed carefully with warm chase medium (KRB containing an excess (2 mM) of unlabeled L-leucine) and then re-incubated in chase medium for 30,90 and 150 min. At the end of the incubations, the media were centrifuged (100 000 g, 60 min) to remove tissue fragments and cell debris. The proteins of the media were concentrated by trichloroacetic acid (TCA) precipitation (final TCA concentration, 5%) followed by centrifugation, and then analyzed by 2D-PAGE. 2D-PAGE The high resolution 2D-PAGE method of O’Farrell et al. (1977), which allows the separation of acidic as well as basic proteins, was used with some modifications. Samples of pituitary homogenates (-170 pg protein/20 pl), or TCA pellets of medium polypeptides were supplemented with 21 mg of solid urea and 8 ~1 of a solution containing 1% SDS, 10% /3-mercaptoethanol and 0.4% ampholines (pH 3.510). After 10 min at 28°C the samples were mixed with 40 ~1 of a lysis buffer containing 9.5 M urea, 0.4% ampholines (pH 3.5-lo), 5% /3-mercaptoethanol and 8% Triton X100. The Triton XlOO/SDS ratio in the final mixture was therefore 8, as recommended by Ferro-Luzzi Ames and Nikaido (1976). After a further 10 min at 28”C, the solubilized samples (50 ~1) were layered onto disk gels (120 mm X 3 mm) prepared with a mixture containing 9.2 M urea, 2% Triton X100, 4% acrylamide/ bisacrylamide and 2% ampholines (pH 3.5-10). Non-equilibrium pH gradient electrophoresis (NEPHGE) was carried out with reversed polarity at 400 V for 5 h. At

168

A. Zanini, P. Rosa

the end of the electrophoresis some gels were fixed in 10% TCA and stained with Coomassie Brilliant Blue R (Ferro-Luzzi Ames and Nikaido, 1976) some were cut into l-cm-thick slices to monitor the pH gradient, and others were used for 2DPAGE. These last gels were first equilibrated for 30 mm in 6 ml SDS sample buffer (Wilson et al., 1977) and then placed on top of a 10% polyacrylamide gel slab, perpendicular to the axis of migration, embedded into a soft agarose gel (1% agarose, 10% glycerol, 5% fl-mercaptoethanol, 2.3% SDS). Electrophoresis in the 2nd dimension was performed at 15 mA for 4-5 h, as previously described for one-dimension (1Dj SDS-PAGE (Zanini et al., 1980). Slabs were fixed with 10% TCA in 25% isopropanol and stained with Coomassie Brilliant Blue R as previously described (Zanini et al., 1974). The apparent molecular weight of the polypeptides was determined by comparing their mobility with that of standards run in the same gel in both dimensions. To reveal 3H-labeled polypeptides, gels were processed for fluorography according to Bonner and Laskey (1974), dried, stapled to RP Royal-X-Omat films, which had been preexposed as described by Laskey and Mills (1975), and stored at -70°C for various times. Immunochemical

analysis

Development of anti-GH immunoglobulins. Antisera against bovine GH (NIHGH-B18) were developed in rabbits. To increase the antigenic activity, the hormone was coupled to ovalbumin by glutaraldehyde as described by Larsson (1979). Four 200~pg doses of the antigen (the first two emulsified in complete Freund adjuvant) were injected S.C. at weekly intervals. Rabbits were bled 5 days after the last injection. Sera were purified by affinity chromatography on small columns of Sepharose 4B bearing attached the NIH-GH-B18 preparation (Meldolesi et al., 1980). Eluted immunoglobulins were tested by immunodiffusion in the presence of 1% Lubrol PX according to Chua and Blomberg (1979): By this technique a single precipitation line was obtained. Immunoprecipitation of GH. Immunoprecipitation of GH from homogenates obtained from bovine adenohypophysis labeled with L-[3H]leucine was performed by the method of Bancroft (1973), with anti-CH antibodies purified by affinity chromatography. Control experiments were carried out with pre-immune sera. The immune precipitate was analyzed by 2D-PAGE as described for the homogenate.

Biochemical assays

Protein was determined by the method of Lowry et al. (1951), with bovine serum albumin as standard. Total protein radioactivity was measured on washed TCA precipitates as described precipitates as described previously (Meldolesi et al., 1972).

Materials L-[3HJLeucine was purchased from the Radiochemical Centre, Amersham (Great Britain); ampholines from LKB, Stockholm (Sweden); acrylamide and SDS-PAGE low M, standards from BIO-RAD, Richmond, CA (U.S.A.); N,N’-methylene bis acrylamide and ~,~~~‘,~‘-tetramethylenedia~~ide (TEMED) from Eastman Kodak, Rochester, NY (U.S.A.}; Lubrol PX, Triton Xl00 and rabbit actin from Sigma, St. Louis, MO (U.SA.1; Sepharose from Pharmacia, Upsala (Sweden); complete Freund’s adjuvant from Difco Laboratories, Detroit, MI (U.S.A.); Agarose from Seakem, Rockland, MA (U.S.A.); bovine GH (NIH-GH-B18) and bovine PRL (NIHP-B4) were kindly given by Dr. A. Parlow, Harbor General Hospital, Torrance, CA (U.S.A.). All other chemicals were reagent-grade.

RESULTS 2D-electrophoretic pattern of bovine and rat anterior pituitav tissue polypeptides Polypeptides stained with Coomassie Blue. Typical 2D-gel electrophoretograms of homogenates from bovine and rat adenohypophyses are shown in Fig. 1. The patterns in the 2 species are similar, even if some qualitative and qu~titative differences do emerge. At least 200 spots are well separated in both species and -70 appear prominent. Clearly evident spots are separated at the positions expected for PRL (apparent M, - 23 000) and GH (apparent M, - 20 000); in the bovine pituitary the identification of these spots with the hormones was confirmed by the co-migration with the authentic standards. The pattern revealed is not the same for the 2 hormones. In fact, only 1 major spot is present in the PRL region of both bovine (pi - 5.8) and rat (~1 - 5.5) glands, In contrast, in the GH region multiple discrete spots are separated: 3 in the rat and 5 in the cow. The major spot of the cow has pi of -7.0, while 2 other prominent spots have pi - 6.8 and -7.4. The major spot of the rat is focused at pH - 6.5; the others at pH - 6.3 and -6.8. Both the PRL and the major GH spots appear larger in the bovine than in the rat pattern. Most of the pituitary polypeptides other than GH and PRL are acidic (~1 - 4.56.5), with apparent molecular weight ranging from 140000 to 40000 dalton. Among these polypeptides, 2 (apparent M, - 70 000 and 62 000 and pZ - 4.8, arrowheads in Fig. 1, I) are more prominent in the bovine than in the rat gland. Another poiypeptide present in fair amount (apparent M, - 43 000, p1- 5.5 ~A in Fig. 1) is probably actin, as it co-migrates with the purified protein standard (not shown). L-[3HJLeucine-labeled polypeptides. The fluorography of ZD-polyacrylamide slab gels obtained with homogenates of anterior pituitary gland slices labeled in vitro for 3 h with L-fJH]leucine is shown in Fig. 2. Most spots stained by Coomassie

170

A. Zonini, P. Rosa

a

b

HOC WT

.70K

.43K

.23K .20K

PM

1 8.5

I 7.5

I

6.5

I

5.5

b

1 4.5

MOL

.43K

I_

GH

r)

.23K .2OK

WT

2D-PAGE of L-/3H]leucine-labeled

pituitary polypeptides

171

Blue are also labeled by L-[3H]leucine. However, the labeling level varied greatly for the different polypeptides. The extremes are polypeptides, well represented in the stained pattern which are very pale in the fluorography (for example some minor peptides in the region of M, comprised between -43 000 and 2.5 000) and polypeptides heavily labeled in the fluorography which have only faint Coomassie-Blue counterparts and which therefore seem to have high specific radioactivity (labeled by arrowheads in Fig. 2). The major spots of the fluorography pattern are PRL (which is more labeled in the rat than in the bovine pattern, where it is accompanied by 2 smaller radioactive spots) and 2 components in the GH zone with pZ - 6.8-7.0 (bovine gland) and -6.3-6.5 (rat gland). In contrast, the third prominent stained spot of the GH area (PI- 7.4 and 6.8 in the 2 species, respectively) is unlabeled. Among the other labeled polypeptides, some are present in the pituitary gland of both species. This is true of the component with apparent M, - 70 000, pZ - 4.8; of the 3 components of this same apparent size and more alkaline pI(l,2 and 3 in Fig. 2), as well as of the polypeptide tentatively identified as actin (A). In contrast, other spots present in the bovine pattern do not appear in the rat pattern (for example, the -62 000 dalton, 4.8 pZ polypeptide) and vice versa. Immunoprecipitation of 3H-labeled polypep tides with an ti-GH antibodies. Our 2D-gels show that the ID-SDS-PAGE band known to contain GH is in fact accounted for by several polypeptides. To establish which components correspond to the hormone, L-[3H]leucine-labeled homogenates of bovine pituitary were immunoprecipitated with anti-GH antibodies, and the washed precipitate analyzed by 2D-PAGE. As can be seen in Fig. 3, the 2 highly radioactive spots of the GH region (the major component, pZ - 7.0 and that with pI - 6.8) are both found in the immunoprecipitate pattern. Other minor components of higher molecular weight were also present in the immunoprecipitate. They could be polypeptides which crossreact with the anti-GH antibodies, whose nature (precursors?) as yet is unknown. When pre-immune serum was used, no radioactivity was precipitated from Ix[~H]leucine-labeled anterior pituitary homogenate.

Fig. 1. Coomassie-Blue staining of ZD-PAGE patterns of bovine (I) and rat (II) anterior pituitary homogenates (100 fig protein) obtained from pituitary slices incubated in vitro with Lj3H]leucine for 180 min. Here and in Fig. 4 the lD-NEPHGE pattern is shown in a; the 2D pattern, obtained by running the NEPHGE gel polypeptides in an SDS-lo% polyacrylamide slab gel, in b. The reproducibility of the separation is excellent when gels are run under standard conditions. The location of individual spots in successive runs was easily checked by simple superposition of negative films or by reference to other surrounding spots. In both animal species, the large spot with apparent M, - 23000 is pro&tin (PRL). No other spots of this apparent size are present. In contrast, several spots are present in the growth hormone region (GH): 5 in the cow and 3 in the rat pattern, with pI ranging between -5.8 and -7.4; -6.3 and -6.8, respectively. A, actin. Arrowheads point to the 2 peptides with apparent M, - 70000 and 62OOO;pl- 4.8.

A. Zanini, P. Rosa

L

20K

I

PH

a.5

PH

8.5

1

I

-1

1

7.5

I

7.5

I 6.5

I

I

5.5

4.5

I

1

I

6.5

5.5

4.5

I

P”

8.5

I

7.5

I

I

I

6.5

5.5

4.5

Fig. 3. ID-PAGE fluorography of ‘H-labeled polypeptides immunoprecipitated with antiGH antibodies, purified by affinity chromatography, from a homogenate obtained from bovine anterior pituitary slices labeled in vitro for 180 min with L-f3H]leucine. Film exposure time, 7 days. Mote that the 2 highly radioactive polypeptides revealed in the GH region (PI- 6.8 and -7 8, Fig. 2>appear both in the ~mmunopr~~ipitate.

2D-electrophoretic pattern of poiypeptides discharged in vitro. To identify the secreted by the gland, the polypeptides released into the incubation media by bovine pituitary slices pulse labeled for 90 min and then chased for 30,90 and 150 min were analyzed. As shown in Figs. 4 and 5, the 2D pattern of the products discharged into the medium is different from that of the tissue, as revealed by both products

Fig. 2. 2%PAGE fluorography of homogenates prepared from bovine (1) and rat (11) anterior pituitary slices labeled in vitro for 180 mm with L-[3H]leucine. Film exposure time: 3 days. Note that the large PRL radioactivity spot is flanked by 2 small spots not visible in the Coomassic-Blue pattern (Fig. 1). In the GH region 2 af the spots appearing in the stained gel are labeled. A, actin; arrowheads, -70 000-62 000 dalton, PI- 4.8 polypeptides. 1, 2, 3, mark poiypeptides with apparent M, - 70 000-62 000 and pl ranging between -S .2 and -5.8 which are present in both animal species and have high specific and have high specific radioactivity.

a MOL

b

1 t

Wl

70K

2 *

=: -43K

_23K, GH

_2OK

*r

-

815

PH

I I

-1

515

615

1.5

4.5

-

Y

a b

-

hOL

WT

. 1 e*

2

-

1)=

70K

P

I

e

43K

.23K .20K

Fig. 4. 2D-PAGE (I) and 150 min polypeptides that tions as in Figs. 1

polypeptides released into the medium from bovine pituitary slices during 30 (II) of chase incubation in vitro. Coomassie-Blue staining. Arrows point to have no visible counterparts in the tissue pattern (see Fig. 1). Other indicaand 2.

17.5

MOL

WT

.7OU

.43K

.23K .20K

PH

I

I

I

I

I

8.5

1.5

6.5

5.5

4.5

Fig. 5. 2D-PAGE fluorography of JH-labeled polypeptides released into the medium during chase incubation from bovine pituitary slices labeled in vitro for 90 min with L-[3H]leucine. I, 30 mm of chase; II, 150 mm of chase. Film exposure time, 4 days. The boxed spots have no visible counterparts in the tissue 3H-labeled pattern (Fig. 2). Arrowheads, peptides with apparent M, - 70 000-62 000, pl - 4.8. 1,2, polypeptides with high specific radioactivity also present in the tissue (Fig. 2, I).

176

A. Zanini, P. Rosa

Coomassie Blue staining and fluorography of L-[3H]leucine-labeled components. In polypeptides stained with Coomassie Blue (Fig. 4) the differences are mostly quantitative. By this procedure, in fact, we analyze not only the real secretion products, but also polypeptides whose release into the medium is due to leakage from disrupted cells. Nevertheless, this level of resolution, clear differences also exist between the medium and tissue patterns. In particular, many components are more prominent in media and their relative concentration progressively increases wtih respect to the others during chase incubation (PRL, the 2 forms of GH, as well as the peptides labeled by arrowheads in Fig. 4: apparent M, - 70 000-62 000, p1- 4.8). Moreover, the spots labeled by arrows in the medium pattern are not detected in the tissue. Even larger medium-versus-tissue differences appear in the L-[3H]leucine fluorography patterns (Fig. 5). This was expected because the cells already disrupted at the beginning of the incubation, which are the major source of the material artifactu~ly leaked out to the medium, do not synthesize protein to a significant extent, and therefore do not contribute to the released polypeptide radioactivity pattern. Among the labeled spots of the medium, PRL is the largest. It is already prominent after 30 min of chase incubation and greatly increases thereafter. This indicates that the labeled hormone is secreted throughout the entire chase period. Labeled peptides which show a similar behavior are the 2 focused at pH - 4.8 (apparent M, - 62 000 and -70 000) the group of peptides with apparent M, around SOOOO, p1 ranging from -5.8 to 6.8, and the peptide that migrates close to PRL (apparent M, - 26 000, p1- 5.7) (::;and n in Fig. 5). It is interesting that the 2 latter groups have detectable counterparts neither in the labeled, nor in the Coomassie Blue-stained, pattern of the tissue. The appearance of labeled GH in the medium is delayed. The hormone is not detected in the 30-min chase sample, but at 90 min the major GH spot appears and is further increased at 150 mm. Several highly labeled spots of the tissue remain either completely undetected in the medium or are present in trace amounts (spots 3 and A).

DISCUSSION The adenohypophysial polypeptide pattern revealed by 2D-PAGE is much more complex than that previously revealed by the 1D techniques. Owing to its complexity, the interpretation of this pattern requires detailed work, which so far has been carried out only in part. This paper deals with the identi~cation of only some polypeptides and the ch~acterization of others on the basis of their presence and labeling with L-[3H]leucine in the tissue and/or among the secretion products discharged into the incubation medium. Detailed characterization and identification of other polypeptides, including the known hormones and precursors, awaits future work. Several discrete spots (5 in the bovine and 3 in the rat homogenate patterns) were resolved by 2D-PAGE. However, the polypeptides of only 2 such spots (the major and one of the others) have fast turnover (high labeling with L-[3H]leucine)

and are precipitated by affinity-purified anti-GH antibodies, so that they could be unambigously attributed to the hormone. This behavior was not shared by the other spots, which therefore either have nothing to do with CH, or are accounted for by hormone molecules with very slow turnover and which do not react, or react poorly, with the purified antibodies (Sinha and Baxter, 1979). The possibility that the minor form of GH identified in the present study is a degradation artifact (de~idation, hydrolysis) is unlikely in view of the mild solubilization treatment of the samples and because the same pattern was obtained in fresh samples as well as in samples stored at -20°C for up to 3 months. In addition, these different forms of GH cannot be due to interactions of GH molecules with other components of the homogenates because the analyses were carried out under fully dissociating conditions: they are therefore probably due to true heterogeneity of the hormone molecule. In this respect it is worth noting that recent results obtained by ZD-PAGE by Evans et al. (1978) revealed that the primary translation product of the rat GH messenger RNA, pre-GH, studied in cell-free conditions, is also heterogeneous (3 forms). However, in these studies, the maturation of the preforms to the mature hormone was not followed. A question that might be asked is whether the minor form of GH now revealed by 2D-PAGE in pituitary homogenates corresponds to one of the forms of different p1 previously separated from purified preparations of the hormone. Because our experimental conditions were markedly different from those used in the previous studies, an answer to this question is not available and would require a systematic analysis by 2D-PAGE of the previously reported forms. The only possibility which can be excluded at the moment is that the 2 forms appearing in our 2D electrophoretograms correspond to the 2 allelic forms described by Seavey et al. (1971), because these latter are present in a ratio of 1 : 1 in individual glands, while our forms are present in a different ratio in individual as well in pooled glands. As far as PRL is concerned, this hormone was found by several authors to be polymorphic with respect to p1 (Lewis et al., 1971; Hwang et al., 1972; Hummel et al., 1975; Sinha and Baxter, 1979; Nyberg et al., 1980). However, in our experimental conditions it appeared homogeneous in the Coomassie Blue pattern. Moreover, the 2 spots revealed by fluorography in the bovine gland are too small to account for a significant proportion of the radioactivity existing in the PRL region. Besides GH and PRL, the pituitary tissue is expected to contain several other major polypeptides. Of these, only actin was identified (at least tentatively) on the basis of its M,, p1 and co-migration with the authentic protein. The putative actin spot was present in the tissue (and not in the medium) and was considerably labeled by L-[3H]leucine. Actin had recently been recognized as a major protein in other secretory cells (Trifarb, 1978; Lee et al., 1979), but not yet identified in the pituitary (Carrels and Gibson, 1976), even though its interaction with pituitary secretory granules has been observed (Ostlund et al., 1977). In addition, we have provided evidence in favor of the existence of previously unsuspected secretory proteins. In fact, besides GH and PRL, the 2 polypeptides

178

A. Zanini, P. Rosa

focused at pH - 4.8 with apparent M, of -70 000 and -62 000 exhibit a behavior typical of specific secretory proteins (Palade, 1975; Meldolesi et al., 1978). First, they are synthesized at rates higher than those of non-exportable proteins, so that after short pulse-labeling their specific radioactivity is high, second, they appear in the extraceilular space (the incubation medium in studies in vitro), not immediately after their synthesis, but with a delay, which corresponds to the time needed for their intracellular transport from rough endoplasmic reticulum to secretion granules; thereafter, they continue to be released, so that they progressively leave the tissue and accumulate in the medium. As reported in our companion article, the larger of these ‘putative’ secretory polypeptides is sulfated and glycosylated and is contained together with PRL within the secretion granules of mammotroph cells, even if it is not excluded that it is also present in other granule types. Other highly radioactive polypeptides (the group with apparent M, - 50 000 and p1 between 5.8 and 6.8; the component close to PRL, apparent M, - 26000, pI 5.7) also accumulated in the medium during chase incubation. However, their discharge kinetics was somewhat different, since they are already prominent in the 30mm medium fluorography pattern, suggesting that their intracellular transport might be faster than that of the other secretory proteins. Moreover, these polypeptides were visible only in the medium and were absent from the Coomassie Blue and fluorography patterns of the tissue. This surprising observation could have been made because that components either are not stored intrace~ularly before discharge or, more likely, originate by processing of tissue precursors occurring shortly before their discharge. Clearly, much work remains to be done to clarify the nature, the physiological significance and the cellular origin of these proteins.

ACKNOWLEDGEMENTS We thank Dr. J. Meldolesi for encouragement and suggestions and Dr. N. Borgese for helpful discussions. The photographic assistance of Mr. P. Tinelli and Mr. F. Crippa and the technical assistance of Miss Alma Nifantani are gratefully acknowledged.

REFERENCES Asawaroengchai, H., Russel, S.M., and Nicoll, C.S. (1978) Endocrinology

102,407-414. Bancroft, F.C. (1973) Endocrinology 92,1014-1021. Banner, M., and Laskey, R.A. (1974) Eur. J. Biochem. 46,83-88. Chua, N.H., and Blomberg, F. (1979) J. Biol. Chem. 254,215-223. Evans, G.A., David, D.N., and Rosenfeld, M.G. (1978) Proc. Natl. Acad. Sci. (U.S.A.) 75, 1294-1298. Ferro-Luzzi Ames, G., and Nikaido, K. (1976) Biochemistry l&616-623.

2D-PAGE of r;-f31flleucine-labeled pi?~irary~lypep~~de~

179

Garrels, J.I., and Gibson, W. (1976) Cell 9,793-805. Giannattasio, G., Zanini, A., Rosa, P., Meldolesi, J., Margolis, R.K., and Margolis, R.V. (1980) J. Cell Biol. 86, 273-279. Hummel, B.C.W., Brown, G.M., Hwang, P., and Friesen, H.G. (1975) Endocrinology 97, &IS-867. Hwang, P., Guyda, II., and Friesen, H.G. (1972) J. Biol. Chem. 247,1955-1958. Kurcz, M., Nagy, I., and Baranyai, P. (1969) Acta Biochim. Biophys. Acad. Sci. Hung. 4, 287295. Larsson, L.I. (1979) Nature (London) 282,743-746. Laskey, R.A., and Mills, A.D. (1975) Eur. J. Biochem. 56,335-341. Lee, R.W.H., Mushynski, W.E., and Trifaro, J.M. (1979) Neuroscience 4,843-852. Lewis, U.J., Litteria, M., and Cheever, E.V. (1969) Endocrinology 85,690-697. Lewis, U.J., Sir@, R.N.P., and Seavey, B.K. (1971) Biochem. Biophys. Res. Commun. 44, 1169-1176. Lewis, U.J., Sin& R.N.P., Tutwiler, G.F., Sigel, M.B., Vanderlaan, E.F., and Vanderlaan, W.P. (1980) in: Recent Progress in Hormone Research, Vol. 36, Ed. R.O. Greep (Academic Press, New York) pp. 477-508. Lorenson, M.G., and EBis, S. (1975) Endocrinology 96,833-838. Lowry, O.H., Rosebrough, N.J., Farr, A.L., and Randall, R.J. (1951) J. Biol. Chem. 193, 265275. Meldolesi, J., Marini, D., and De Monte-Marini, M.L. (1972) Endocrinology 91, 802-808. Meldolesi, J., Borgese, N., De Camilli, P., and CeccareBi, B. (1978) in: Cell Membrane Fusion, Eds.: G. Poste and G.L. Nicolson (Elsevier/North-Holland, Amsterdam) pp. 509-627. Meldolesi, J., Corte, G., Pietrini, G., and Borgese, N. (1980) J. Cell Biol, 85,516-526. Nicholson, P,M. (1969) J. Endocrinol. 44,4.59-460. Nicoll, C.S., Parsons, J.A., Fiorindo, R-P., and Nichols Jr., C.W. (1969) J. Endocrinol. 45,183196. Nyberg, F., Roos, P., and Wide, L. (1980) Biochim. Biophys. Acta 625, 255-265. O’Farrell, P.Z., Goodman, H.M., and O’Farrell, P.H. (1977) Cell 12, 1133-1142. Ostlund, R.E., hung, J.T., and Kipnis, D.M. (1977) J. Cell Biol. 73,78-87. Palade, G.E. (1975) Science l89,347-358. Russel, S.M., Vodiau, M.A., Hughes, J.P., and Nicoll, C.S. (1978) Life Sci. 23, 2373-2382. Seavey, B.K., Sir& R.N.P., and Lewis, U.J. (1971) Biochem. Biophys. Res. Commun. 43, 189-195. Sinha, Y.N., and Baxter, S.R. (1979) Biochem. Biophys. Res. Commun. 86,325-330. Trifaro, J.M. (1978) Neuroscience 3,1-24. Wallis, M., and Davies, R.V. (1976) in: Growth Hormone and Related Peptides, Int. Congr. Ser. No. 381. Eds.: A. Pecile and E.E. Miiller (Excerpta Medica, Amsterdam) pp. 1-13. Wilson, D.L., Hall, M.E., Stone, G.C.,and Rubin, R.W. (1977) Anal. Biochem. 83, 33-44. Yadley, R.A., Rodbard, D., and Chrambach, A. (1973) Endocrinology 93,866-873. Yanai, R.,and Nagaswa, H. (1969) Proc. Sot. Exp. Biol. Med. 131,167-171. Zanini, A., and Giannattasio, G. (1972) J. Endocrinol. 53,177-178. Zanini, A., Giannattasio, G., and Meldolesi, J. (1974) Endocrinology 94,594-598. Zanini, A., Giannattasio, G., De Cam& P., Panerai, A.E., Miiller, E.E., and Meldole~, J. (1979) Endocrinology 104,226-236. Zanini, A., Giannattasio, G., Nussdorfer, G., Margolis, R.K., Margolis, R.U., and Meldolesi, J. (1980) J. Cell Biol. 86,260-272.