| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
University of Arkansas for Medical Sciences Little Rock, Arkansas 72205-7199
Address all correspondence and requests for reprints to: Gwen V. Childs, Ph.D., Professor and Chair, Department of Anatomy, University of Arkansas for Medical Sciences, Little Rock, Arkansas 72205-7199. E-mail: childsgwenv{at}exchange.uams.edu
 |
Introduction |
|---|
 Top Introduction References |
|---|
Over the years, attempts to fully understand this complexity, let alone individual cell types have been hampered by the diversity in any preparation of anterior pituitary cells. The diversity has added caveats to the interpretation of many findings from studies of regulation of individual cell types. For example, a study that stimulates a group of cells may also stimulate a chain of paracrine interactions that in turn affect other cell types (17, 18, 19, 20, 21).
Many neuroendocrine laboratories who want to study the regulation of individual pituitary cell types have used tumor cell lines as models for pituitary cell function (11, 14). Or, if they wanted to use primary pituitary cells, they have either separated and purified them (reviewed in Refs. 19, 20, 21, 22), or applied techniques that identify the cells, especially in the living state (4, 5, 6, 7, 8, 9). Some of these methods are laborious and have a low yield (20); others may alter the cells either temporarily or permanently in a way that limits their usefulness (4, 5, 6, 7, 8, 9). Neuroendocrinologists also recognize that they do not know if removal of a group of cells from the pituitary network has deleterious effects on their function, although many basic secretory and signal transduction functions remain intact. Nevertheless, ongoing studies have reported that pituitary cells behave differently in vitro when they are allowed to "reaggregate" (19, 20). Hence it may be important to keep them in their environment, however difficult the studies may be to interpret.
Therefore, it has become important to develop methods that might identify individual pituitary cells in their native environment. The ideal method would enable the neuroendocrinologist to identify the cell either in situ or in a mixed culture of pituitary cells without altering key regulatory processes or expression of the genes in question. Such a marker would essentially work in the background, allowing the pituitary cell types to be identified in the living state when needed. Ideally, neither the process of identification, nor the marker itself would interfere with normal cell function, its analysis, or with the physiology of the animal. Also, the marker might be a tool for the further purification of the cell population if needed.
Green fluorescent protein (GFP) has been widely used by cell biologists all over the world as a maker that can be attached to proteins without alterations in its fluorescence. In many cases, the function of the protein itself is not altered. The resulting "fusion product" becomes a "reporter molecule" that allows cell biologists to trace the intracellular distribution, secretion, and storage sites of the protein linked to GFP. Transgenes have been constructed that code for GFP linked to the protein in question. These constructs can be used to transfect cells or make transgenic mice. Specific promoter regions are inserted in the transgene to ensure expression by the right specific cell type. The cell then transcribes the GFP-protein transgene, translates the mRNA, and makes the protein linked in sequence with the GFP. This endogenous cellular "fusion product" essentially "lights the way" to the cell in question and "reports" or "signals" the subcellular domain of the protein. Immunolabeling for GFP and the protein is generally used to confirm its ultrastructural site and its coexistence with the protein.
However, the GFP-protein constructs have applications that go beyond mere identification. One can trace the trafficking of the GFP-protein product in real time, in the living cell, with video fluorescence microscopy and observe sorting and movement to different subcellular domains. For example, Jordan et al. (23) recently produced gap junctions in cells that do not normally make them by transfecting the cells with a GFP-connexin43 construct. Not only did the cells begin to produce normal gap junctions (as proved by appropriate physiological and cytochemical tests), the pathway and storage sites of connexin43 molecules could be traced in real time by video imaging.
Similarly, Gaidarov et al. (24) transfected cells with a construct encoding the mRNA for GFP linked to clathrin, a protein vital to receptor mediated endocytosis processes. Once translated, the GFP-clathrin moved to the surface and continued to form functional clathrin-coated pits (used to bring in ligands and receptors). The workers could thereby learn more about the route and turnover of clathrin during the secretory process and receptor mediated endocytosis. They discovered that the route these pits take to and from the plasma membrane is actually a fairly well defined pathway, suggesting an organized involvement of the cytoskeletal system. These studies provide only two of many examples showing how GFP can be used to address critical basic questions in cell biology.
A transgene coding for GFP linked to a pituitary hormone might make a particularly good identifying marker for the cells that express that gene. Of course, it would be critically important to produce a sequence that would target itself to the right cell types, but not alter the physiology of the animal or the cells themselves. Thus, one might want to construct a transgene with as complete a promoter region as possible but not enough of the genetic sequence to make a complete biologically active hormone. However, if one wanted to study sorting pathways, enough of the hormone should be translated to move it to the right subcellular sorting domain. For example, the signal peptide would be critical to allow entry into the rough endoplasmic reticulum. Eventually one might design studies that alter the sequences to learn which regions are important for further trafficking and sorting. Finally, there are variants of these fluorescent proteins that can be distinguished by different wavelengths, including yellow fluorescent protein (YFP) and cyan fluorescent protein (CFP), which potentially allow for the tracing of multiple proteins.
Recently, Magoulas et al. (26) produced a construct encoding the gene for enhanced GFP (eGFP) linked to sequences of human GH (hGH). This technique may represent a major breakthrough in our search for noninterfering cellular markers in the pituitary. However, this elegant paper takes the field several steps beyond the identification of GH cells.
One phase of their study used two different lengths of the 5' coding sequence of the hGH gene fused with an enhanced variant of the GFP gene. Cells transfected with the gene for the longer sequence produced eGFP attached to the first 48 amino acids of hGH. Those transfected with the gene for the shorter sequence produced eGFP linked to the first 8 amino acids of the hGH signal peptide (needed for entry to the rough endoplasmic reticulum). Both constructs were then transfected into GC cells (a tumor cell line that produces GH normally).
Stably transfected GC cells translated both the smaller and larger constructs. The shorter construct that contained the GH signal peptide had a relatively uniform distribution of fluorescence. However, the longer construct showed a more punctate distribution suggesting that the longer GH protein had been sorted through the Golgi complex to a granule compartment. Thus the first phase of this study showed how the length and content of the translated sequence drove the trafficking and sorting of the eGFP-GH fusion product.
The longer version of the eGFP-GH construct was then used to generate transgenic mice expressing eGFP linked to GH sequences. This was driven by the hGH locus control region (LCR), which is a relatively large, complete DNA promoter element in the GH gene. They chose this LCR to ensure the highest somatotropic specific expression. The promoter had to be mutated slightly to insert the hGH-eGFP sequence, but the mutation involved only 2 bp. The test of the founders and their progeny showed high level specific expression of eGFP that was isolated to subsets of anterior pituitary cells.
As stated in the beginning of this editorial, one criterion for a successful marker is that it will work "in the background" and not interfere with the physiology of the animal. This study showed a significant reduction in GH stores in both transgenic males and females (26). However, the animals appeared phenotypically normal, (including a normal growth rate). The reason for the reduction in GH stores is unclear. The translated "fusion product" of eGFP and GH did not contain intact or biologically active GH. However, enough of the GH sequence could have been present to signal short loop feedback or an autocrine regulatory loop.
Fluorescent fields from the pituitaries of the transgenic line chosen for study showed that nearly half of the population contained eGFP fluorescence, and all of these cells also contained GH. This suggests that the potential for expression of GH in the pituitary may be broader than that seen if only GH proteins are detected. Most workers report that GH cells detected by immunolabeling represent about a third of the pituitary cell population. Our studies show that cells with GH mRNA represent no more than 27–33% of the rat pituitary cell population (15). However, the potential for the production of GH by other cell types has been reported by us and others (6, 7, 8, 15). If the lower GH expression stems from an autoregulatory loop, it is possible that the transgenic pituitaries are compensating for the lower GH reserves by adding more cells. They may be thus expressing their full potential. This compensation may also have allowed for the normal growth rate. Further studies of the physiology of these animals (levels of IGF-1, other pituitary hormones) would be warranted. At this point, the changes seen in these animals do not appear to be a deterrent for further use and study of these identified GH cells.
In the transgenic animals, the longer eGFP-GH construct yielded a translated product that was sorted to granules, as shown by immunogold labeling. In fact, GFP was only found in secretory granules in these cells indicating that it was able to be moved through the Golgi complex and that recognizable product did not accumulate in the rough endoplasmic reticulum. eGFP itself is not normally targeted to the rough endoplasmic reticulum. The GH signal peptide would have been needed to allow it to enter this compartment.
The longer sequence also promoted the passage of the product along the secretory pathway (26). When they designed the longer construct, the authors cleverly chose to add sequences along the first 22 residues that include two N-terminal histidines. They reasoned that this would contribute significant Zn+2 binding activity to allow progress in the pathway to secretory granules. Their pioneering studies thus set the stage for further work that could differentially detect sequences in the GH gene that, when translated, actually promote advancement along this pathway to secretion granules.
Their findings also showed that the GH-eGFP protein was secreted in a dose-dependent manner following stimulation with a GH secretagogue, thus introducing new ways to study sorting pathways and regulated secretion in pituitary cells (26). Challenges yet to be met include assays for the eGFP that might be used to detect luminescent secreted product (in place of RIAs). In addition, they showed that the eGFP fluorescence could be used to separate the cells by fluorescence-activated cell sorting (FACS), producing a nearly pure population of GH cells. The advantage of this protocol is that it may lend itself to the separation and purification of larger batches of cells than are possible with the elutriation methods used in our laboratory (20). This would provide more material for biochemical analyses or product purification.
To summarize, this pioneering study designed an eGFP-GH transgene that, when translated in transfected cells or pituitary cells of transgenic mice, lighted a population of GH cells that could be identified in preparations of living cells, in situ. These cells will be potentially valuable for use in situ or in vitro, by cell biologists, neuroendocrinologists, and electrophysiologists (25). To show how useful the preparation is, the presentation included data from an experiment that recorded spontaneous changes in intracellular calcium in the identified GH cells. In this group of experiments, they used tissue slices that maintained some of the network of interacting cells seen in the pituitary. Therefore, the method fulfills most of the criteria stated in the beginning of this editorial. It allows for the rapid identification of a pituitary cell type, with minimal apparent affects on the physiology of the cell or the animal. The overall approach shows the feasibility of studies that use constructs of GFP genes linked to pituitary hormone genes to transfect cells or make transgenic animals. Studies of the identified cells may greatly improve our understanding of how pituitary hormones or growth factors are sorted and processed and how these processes may be regulated.
Received October 13, 2000.
|
References |
|---|
|
TopIntroductionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  E. Itakura, K. Odaira, K. Yokoyama, M. Osuna, T. Hara, and K. Inoue Generation of Transgenic Rats Expressing Green Fluorescent Protein in S-100{beta}-Producing Pituitary Folliculo-Stellate Cells and Brain Astrocytes Endocrinology, April 1, 2007; 148(4): 1518 - 1523. [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 |
