| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Laboratoire de Biopathologie Nerveuse et Musculaire, Institut de Biologie du Développement de Marseille, Université de la Méditerranée (C.B., D.F.-B.), and Groupe de Recherches sur les Glandes Exocrines (N.B., C.F.), Faculté de Médecine La Timone, 13385 Marseille; and Laboratoire Interactions Cellulaires Neuroendocriniennes, Unité Mixte de Recherche 9941 Centre National de la Recherche Scientifique, Université d’Aix-Marseille 2, Faculté de Médecine Nord (I.P.), 13916 Marseille, France
Address all correspondence and requests for reprints to: D. Figarella-Branger, Laboratoire de Biopathologie Nerveuse et Musculaire, Institut de Biologie du Développement de Marseille, Université de la Méditerranée, Faculté de Médecine La Timone, 13385 Marseille Cedex 05, France.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Homologous rat PAP and Reg genes have been found in humans with a similar tissue distribution as that in the rat (8). The sequences of their complementary DNAs (cDNAs) present around 50% homology between them and also with encoded rat proteins (9). In addition, human PAP messenger RNA (mRNA) has been found to be expressed in 25% of primary liver cancer (and is called HIP); thus, the gene is now named PAP/HIP (10, 11). In human pancreas a Reg-like gene (RegL) has been also identified as well as a Reg-related sequence described as a pseudogene (9, 12, 13).
No study has been performed until now on the human pituitary gland. We then decided to look for a peptide homologous to the PAP/Reg family in human normal pituitary gland as well as in different adenomas, especially in GH-producing adenomas. We compared their potential expression to that in other normal adults and developing tissues known to express some of the PAP and Reg family genes.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Three plurihormonal adenomas GH-PRL and 30 monosecretant pituitary adenomas were selected: 8 chromophobes, 6 gonadotrophs (FSH, LH, or both), 8 prolactinomas (PRL), 4 somatotrophs (GH), 2 thyreotrophs (TSH), and 2 corticotrophs (ACTH). Tumor samples were collected in the surgical room. One part was frozen in liquid nitrogen and stored at -80 C before RNA extraction. The other part was put in 10% formalin for histological and immunohistochemical studies. All tumors were macroadenomas devoid of normal pituitary tissue. Normal adult and fetal tissues were provided from autopsies. Fragments from adult pancreas (3), stomachs (2), jejunums (2), colons (2), pituitary glands (2), frontal lobes (2), and cerebellum (2) were studied. Fragments from fetuses were pancreas (20, 34, or 37 weeks gestation), antrum (34 and 37 weeks), and 1 fragment each of jejunum, colon, pituitary gland, frontal lobe, and cerebellum (34 weeks gestation). In addition, total adult brain mRNAs and total fetal brain mRNAs (pools of mRNA from fetuses aged 14–25 weeks) were provided by Clontech (Palo Alto, CA).
Total RNA extraction
Total cellular RNA was extracted using the guanidinium thiocyanate method (14) followed by deoxyribonuclease I (20 U/sample) and RNAsin (40 U/sample) for 15 min at 37 C. Five micrograms were loaded in a 1% agarose-formaldehyde gel to determine RNA integrity.
Detection of Reg, RegL, and PAP/HIP mRNAs in the different tissues
RT was carried out on 1 µg total RNA (or 20 ng mRNA) using oligo(deoxythymidine) primer (Pharmacia, Saint Quentin en Yvelines, France) and the SuperScript (Life Technologies, Cergy Pontoise, France) reverse transcriptase according to the recommendations of the manufacturer.
First, relative RT-PCR amplification of the glyceraldehyde-3-phosphate
dehydrognase (GAPDH) transcript was performed to homogenize the
amount of cDNA to be used in each of the PCR reactions. Indeed, a
limited number of cycles were used, and this allowed us to be
positioned in the linear portion of the amplification curve. Intensity
analysis of the amplified bands was performed with an imager
(Appligene, Illkirch, France) and the appropriate computer program. The
volume of RT reactions (or cDNA amount) was then adjusted to obtain
equal amounts of amplification product for each case. The
glyceraldehyde-3-phosphate dehydrogenase primers used (15) are
described in Table 1
, using the following
program for 28 cycles: 1) denaturation at 94 C for 45 s, 2)
annealing at 60 C for 45 s, and 3) DNA synthesis at 72 C for 2
min. An aliquot of the final amplification solution was analyzed after
ethidium bromide staining of 1% agarose gel to assess the size of the
amplified fragment (933 bp).
|
were
localized in different exons separated by almost one intron and
selected so that their 3'-extremities did not cross-hybridize with the
related sequences. Reg and RegL PCR product
specificities were verified by enzyme digestions and transfer
hybridization with a Reg cDNA probe. Thirty amplification
cycles, ended by a last extension at 72 C for 5 min, were performed for
each gene as follows: PAP expression, for 30 cycles: 1) denaturation at
94 C for 1 min, 2) annealing at 65 C for 1 min, and 3) DNA synthesis at
72 C for 45 s; and Reg and RegL expression,
for 30 cycles: 1) denaturation at 94 C for 1 min, 2) annealing at
55 C for 1 min, and 3) DNA synthesis at 72 C for 1 min. Controls included samples lacking cDNA as template, and RT reactions without reverse transcriptase were performed.
cDNA probes
Reg and PAP probes were synthesized by RT-PCR as described above and subcloned into a pGEM-T vector (Promega Corp., Lyon, France). They were used for dot and Southern blots. The Reg probe presents almost 90% identity with the Reg-related sequences described previously (9, 12); therefore this probe is not a specific Reg probe, but it recognizes all mRNA sequences related to the Reg gene.
Southern hybridization
Twelve microliters of PCR reactions were submitted to electrophoresis in a 2% NuSieve agarose gel (Tebu, Le Perray en Yvelines, France) in 1 x TAE buffer and transferred onto a Hybond-N membrane (Amersham, Aylesbury, UK). Probes were labeled with digoxigenin-11-deoxy-UTP using the digoxigenin oligonucleotide 3'-end labeling kit (Boehringer, Meylan, France). Hybridization was carried out overnight with about 300 ng labeled probe in 5 x SSPE, 0.5% blocking reagent, 0.1% lauryl sarcosine, and 0.02% SDS at 65 C. Washes were performed as follows: twice for 5 min in 2 x SSC (standard saline citrate) at room temperature and twice for 40 min in 0.5 x SSC at 65 C. The chemiluminescence was detected following the instructions of the digoxigenin luminescent detection kit for nucleic acid (Boehringer, Meylan, France).
Quantitative analysis of mRNA by dot blot hybridization
Sequential dilutions of total RNA were spotted onto
nitrocellulose membranes using a manifold apparatus (Minifold I,
Schleicher & Schuell, Inc., Ecquevilly, France). For
hybridization with PAP/HIP and Reg probes, the dilutions
were 15:0.468 µg for pituitary adenomas and the pituitary gland and
1:0.031 µg for other tissues. For hybridization with the 28S cDNA
probe, the total amount of RNA was decreased 100-fold. After spotting,
the filters were baked for 2 h at 80 C before hybridization.
Prehybridization was performed at 42 C for 4 h in 50% formamide,
5 x SSPE, and 5 x Denhardt’s solution containing 250
µg/mL denatured salmon sperm DNA. Hybridization was carried out for
36 h at 42 C in the above-mentioned solution using 1 x, instead
of 5 x, Denhardt’s solution and supplied with a
[
-32P]deoxy-CTP-labeled probe (4.7 x
108 cpm/mg). Washes were performed twice with a 2 x
SSC-0.1% SDS solution for 5 min and twice with a 0.1 x SSC-0.1%
SDS solution for 30 min each time at 55 C. Quantification was performed
as previously described (16). Briefly, mRNA concentrations were
estimated from the slopes of the linear regression curves of the dots
after scanning the autoradiographs at 490 nm using an optical
densitometer (MR 5000, Dynatech Corp., Chantilly, VA).
Differences in total RNA loading were eliminated by correcting the
values obtained for PAP and the REG family, with quantification of the
28S RNA for each test. Results were calculated as the PAP/28S or
REG/28S ratio and were expressed in arbitrary units
(AU).
Immunocytochemistry
Immunohistochemistry was performed on six different adenomas (cases 52, 56, 57, 66, 178, and 213) and in adult and fetal pituitary glands using a monoclonal antibody (1:50 dilution) directed against the Reg protein (formerly called pancreatic stone protein; clone AbD4, Immunotech, Marseille, France). Normal adult pancreas was used as a positive control. Two serial sections from formalin-fixed, paraffin-embedded specimens were made. The first was used to react with the primary antibody, and the second served as a control and was incubated with irrelevant IgG mouse monoclonal antibody. After rinsing, the sections were incubated for 15 min with biotinylated antimouse Ig and then visualized by streptavidin-biotin peroxidase (LSAB K 680 kit, Dako Corp., Santa Barbara, CA).
Primary culture of human pituitary adenomas and RNA isolation
Three human pituitary adenomas [one chromophobe and two somatotroph (GH)] were removed, as described above, and sectioned. One part was immediately frozen; the cells of the other part were dispersed and resuspended in DMEM (Life Technologies) supplemented with 10% calf serum, 100 U/mL penicillin, 100 µg/mL streptomycin, and 50 µg/mL gentamicin (17). The cells were plated in tissue culture dishes (Costar, Brumath, France) coated with extracellular matrix from bovine endothelial corneal cells. After 10 or 30 days of culture, the cells and frozen tissues were harvested for RNA extraction by the method of Chomczynski and Sacchi (14) and used for RT-PCR and Southern blot analysis.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
All of the different pituitary adenomas (n = 33) were tested
for the expression of PAP/HIP, Reg, and RegL by
RT-PCR analysis followed by Southern hybridization with specific
probes. The results are summarized in Table 2. Weak expression of
PAP/HIP was observed in every type of tumor, but not in each individual
tumor (in 5 of 8 chromophobes, 3 of 6 gonadotrophs, 5 of 8
prolactinomas, 2 of 4 GH, 1 of 3 GH-PRL adenomas, and 1 of 2 TSH and
ACTH pituitary adenomas). RegL transcripts were also
observed in some chromophobe, gonadotroph, PRL, and GH-PRL adenomas,
but not in GH, TSH, and ACTH adenomas. Only 3 adenomas of the 33
studied [1 chromophobe (case 213), 1 prolactinoma (case 56), and 1
GH-PRL (case 214)] presented a concomitant expression of the PAP/HIP
and RegL transcripts. In contrast, Reg expression
was never observed in the tissues tested. The absence of Reg
expression was further documented by immunohistochemical studies; no
immunostaining for the Reg protein was observed in the 6
different pituitary adenomas tested [1 chromophobe (case 213, Table 2), 2 gonadotrophs (cases 57 and 66), 2 PRL-producing adenomas (cases
56 and 52), 1 PRL- and GH-producing adenoma (case 178), and 2 normal
pituitary glands (1 fetal and 1 adult)].
|
Cultured cells and freshly frozen corresponding tissues of one chromophobe and two somatotroph adenomas were tested for PAP/HIP and Reg gene family expression. No expression was found in tumoral tissue, and primary cultures of tumoral cells did not induce the expression of any gene tested.
Gene expression of PAP/HIP and Reg family in fetal and adult tissues
The expression of PAP/HIP and Reg family has been
looked for in different normal fetal and adult tissues, and Fig. 1 shows the results obtained after
specific Southern hybridization.
|
). Nevertheless, the level of expression depended upon the type
of tissue studied; a high level was observed in adult jejunum and lower
levels in other tissues. There were differences in Reg
expression between fetal and adult tissues (Fig. 1B). In fetal tissues,
Reg transcript was only observed in pancreas. In adult
tissues, Reg transcript was also observed at a high level in
pancreas and was present at a lower level in jejunum, stomach, and
colon. As previously noticed for pituitary adenomas, there was no
expression of the Reg gene in the normal fetal or adult
pituitary gland. As shown in Fig. 1C, RegL expression was
high in fetal pancreas and was present in fetal colon and total fetal
brain. There was no expression in cerebellum or frontal lobe extract of
the 34-week-old fetus, probably due to the older age of this fetus
compared to those from which total brain mRNAs were obtained (14–24
weeks). RegL was also observed in all adult tissues
tested, except stomach and brain. Expression of the
Reg-related sequence was only found in fetal and adult
pancreas (data not shown). Comparison between PAP/HIP and Reg family mRNA levels in fetal and adult pancreas, normal pituitary gland, and adenomas
The different qualitative patterns obtained by Southern hybridization led us to quantify by dot blot analysis the mRNA levels of PAP/HIP and Reg family in tumoral pituitary glands and to compare the levels of expression with those in pancreatic tissues, where the genes are the most expressed.
The results are shown in Fig. 2. As
expected, Reg family mRNAs were present at a high level in
the adult pancreas (n = 3; 1632 ± 95 AU) and at an even
higher level in the fetal pancreas (n = 4; 4894 ± 1430 AU).
In contrast, Reg mRNA levels were more than 100 times lower
in normal and tumoral pituitary glands, which presented roughly the
same level (20.15 AU).
|
As shown in Fig. 3 in an extended scale,
there is a great variability between the different levels of PAP/HIP
and Reg in normal pituitary gland and different
hormone-producing adenomas.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Our results presented here using human tissues are different from those that Friesen found in the rat. From all 33 pituitary adenomas tested, a weak expression was only shown by RT-PCR followed by Southern hybridization with a specific PAP/HIP probe and not in each tumor. Specifically, in four GH-producing adenomas and in three GH- and PRL-producing adenomas, only three of seven were positive with the probe. In addition, no difference regarding clinical, biochemical, and pathological data was observed between adenomas expressing the P23/PAP/HIP gene and those that do not express this gene. In the subgroup of PRL-producing adenomas, however, PAP/HIP expression was preliminarily shown to be correlated with the resistance to dopaminergic agonists. A more extensive study, performed in 10 additional PRL-producing adenomas in which the sensibility for dopaminergic agonist was known, failed to confirm these findings. Moreover, in contrast to cultured rat pituitary cells (3), expression of the PAP/HIP gene in cultured human pituitary cells was negative (data not shown).
This suggests that the expression of PAP/HIP and the Reg gene family is not up-regulated when tumoral cells are placed in primary culture. Moreover, only a few GH-producing adenomas express P23/PAP despite the accumulation of GHRH mRNA transcripts in almost all GH-producing pituitary tumors as reported previously (21). All of these data show that in contrast to peptide 23 in rat, PAP/HIP in human is not GHRH-inducible pituitary protein.
The significant homology observed between the different genes of the human PAP/Reg family has led us to try to identify in the same pituitary adenomas the presence of the related Reg genes. We observed a weak expression of the RegL gene in 9 of the 33 pituitary tumors. In 6 of 9, RegL gene expression was found despite the lack of PAP gene expression, leading us to suspect a mutual exclusion of these genes. In the 3 remaining cases, however [1 chromophobe (case 213), 1 PRL-producing tumor (case 56), and 1 GH- and PRL-producing tumor (case 214)], a weak expression of both PAP and RegL genes was observed. Interestingly, using the same technique, the expression of the Reg gene (which is the more extensively studied Reg gene) was undetectable in all tumors tested. A quantification by dot blot analysis of the normal pituitary gland and various adenomas showed that the PAP/HIP and Reg mRNA levels in these tissues were effectively about 100 times less than those in fetal and/or adult pancreas and were irregularly dispersed.
As our data on human pituitary tumors represented the first example of a careful comparison among the expressions of the three members of the PAP/Reg gene family, we decided to pursue our study in different normal fetal and adult tissues. Interestingly, the different transcripts are unevenly distributed. The most striking result is the lack of expression of the Reg gene in all of the fetal tissues studied, except pancreas. In contrast, PAP was widely expressed in all tissues. The RegL gene was also observed in fetal pancreas, but unlike the Reg gene it was also expressed in fetal colon and fetal brain. The lack of expression of the Reg gene in fetal brain allows us to speculate that the enhanced expression of the Reg transcript previously observed by de la Monte et al. in the developing human brain (as well as in Alzheimer’s disease) (22, 23) was not due to an overexpression of the Reg gene, but was probably caused by an overexpression of the RegL transcript.
In conclusion, our results show that all human tissues examined express at least one of the different transcripts of the PAP/Reg family, suggesting that the regulation of these homologous genes is coordinately controlled. In pituitary adenomas (n = 33), only 9 of the 33 tumors tested were negative for all transcripts. Thus, the roles of these genes, if any, in human pituitary glands and adenomas will be clearer and more precise when there is a better understanding of the function of the encoded proteins, particularly at the physiological, cellular, and molecular levels.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received May 6, 1998.
Revised July 1, 1998.
Accepted July 14, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  S. Vasseur, E. Folch-Puy, V. Hlouschek, S. Garcia, F. Fiedler, M. M. Lerch, J. C. Dagorn, D. Closa, and J. L. Iovanna p8 Improves Pancreatic Response to Acute Pancreatitis by Enhancing the Expression of the Anti-inflammatory Protein Pancreatitis-associated Protein I J. Biol. Chem., February 20, 2004; 279(8): 7199 - 7207. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J Carrere, O Guy-Crotte, E Gaia, and C Figarella Immunoreactive pancreatic Reg protein in sera from cystic fibrosis patients with and without pancreatic insufficiency Gut, April 1, 1999; 44(4): 545 - 551. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
