These studies were done in Gwen V. Childs, Ph.D's lab, The University of Texas Medical Branch, Galveston, TX. Growth hormone cells are one of the six hormone-bearing cell types in the anterior pituitary. They are the most abundant cell type and secrete growth hormone to increase growth in length of long bones. Other developmental roles have been postulated for growth hormone, however. Our research project is related to one of these. It is linked to infertility studies that show that the addition of growth hormone may be effective in promoting pregnancy in some clinical cases that do not respond to more classical treatment. Physician scientists have suggested that growth hormone may play a permissive role in facilitating ovulation and other events leading up to implantation and successful pregnancy. We are interested in its potential interactions in the pituitary itself. The following menu catalogs the evidence we have thus far for a role for GH at the pituitary level. 

GH cells express gonadotropin beta subunit mRNA during diestrus and proestrus (just before the preovulatory surge)

We began this work after we discovered that a subpopulation of GH cells expressed messenger ribonucleic acid (mRNA) for the beta subunits of gonadotropins just before ovulation. We used dual-labeling that detected mRNA for gonadotropins by in situ hybridization and growth hormone antigens by immunolabeling.

A photograph of a dual-labeled growth hormone cell is shown in this figure. The growth hormone cell (G) is labeled dark orange for growth hormone antigens. The cell also contains messenger RNA for one of the gonadotropins, follicle stimulating hormone. This is labeled dense gray-black (F) U=unlabeled cell. We suggest that it has become a transitional gonadotrope to augment the ovulatory surge of gonadotropins.

Counts of cells bearing GH antigens and gonadotropin mRNAs are shown in this graph. These data illustrate that significant percentages of GH cells express LH or FSH mRNA on the day of ovulation. After ovulation, these percentages drop to minimal levels.

For more information, see:

Childs, G.V. , Unabia G., Rougeau D. Cells that Express Luteinizing Hormone (LH) and Follicle Stimulating Hormone (FSH) Beta ( ) Subunit mRNAs during the Estrous Cycle: The major contributors contain LH , FSH and/or Growth Hormone, Endocrinology, 134: 990-997 1994.

Hypotheses to test: 

There are two working hypotheses to explain this labeling pattern. The first states that subsets of GH cells may become transitional gonadotropes to support the LH surge. They may be the medium-sized subset that appears before ovulation and secretes better than the largest gonadotropes.

The second hypothesis states that GH antigens are binding to GH receptors in gonadotropes. This would point to GH as a regulatory molecule for the gonadotropes themselves. The fact that gonadotropes have GH receptors suggests that they may be able to receive such regulation. However, electron microscopic data shown below indicates that the gonadotropin molecules are being translated by somatotropes which supports the first hypothesis.

These hypotheses are being tested in an NIH grant: R01 HD 33915 "Novel Somatotrope functions during ovulation" 1996-2000

Questions that address Hypothesis 1:

Test for GnRH binding: Do GH cells bind the neuropeptide releasing hormone that binds gonadotropes? (Gonadotropin Releasing hormone)

One way to test the first hypothesis would be to determine if the transitional "somatogonadotrope" binds the neuropeptide Gonadotropin releasing hormone (GnRH). We used dual affinity cytochemistry labeling protocols to detect GnRH on subsets of pituitary cells and found that cells with GH antigens did increase binding (to about 40% of GH cells) just before ovulation. At the same time, there was an increase in the percentages of gonadotropes that bound GnRH to 90% of gonadotropes. This supports the hypothesis that these GH cells may be transitional gonadotropes. The figure to the left illustrates the dual-labeling data. The figure was reproduced as a negative image to show the labeling better. Growth hormone cells (G) are immunolabeled blue for growth hormone antigens. The white label (arrows) shows that this cell also binds the neuropeptide, gonadotropin releasing hormone (GnRH). The labeling is mostly on the cell periphery. This releasing hormone may help convert these cells to transitional gonadotropes just before ovulation.

For more information, please see:

Childs, G.V., Unabia, G and Miller, BT Cytochemical detection of GnRH binding sites on rat pituitary cells with LH, FSH and GH antigens during diestrous upregulation. Endocrinology 134: 1943-1951 1994.

Can expression of GnRH receptors by GH cells be regulated by inhibin or activin?

It is well known, from our earlier studies, that percentages of GnRH receptive cells increase 4-fold during diestrus to reach a peak in early proestrus, on the morning before ovulation.  Expression of GnRH receptors is seen in over 90% of LH or 88% of FSH cells during this peak.  In addition, as shown above, GH cells express GnRH receptors (over 38% of the cells) only during this peak period. 

Can this expression be regulated by factors known to stimulate (activin) or inhibit (inhibin) GnRH receptors?  Studies recently completed and published show that GnRH receptive cells can be increased in diestrous rat pituitary cells by overnight incubation in 60 ng/ml activin. When the cell types were identified by dual labeling affinity cytochemistry, both LH and FSH gonadotropes were affected.  In addition, activin increased GnRH receptor expression by GH cells from either diestrous or proestrous rats.   The increase was from 38-60% of GH cells, which is a higher expression than normally seen in proestrus.  At the same time, inhibin causes a decrease in expression of GnRH receptors by LH or FSH cells and also GH cells from populations taken from proestrous rats.  These data suggest that expression of GnRH receptors by GH cells can be regulated by the reproductive hormones that normally regulate them in gonadotropes. 

This photo shows GH cells bound to Biotinylated GnRH.  They were taken from Proestrous rats. As in the above photo, the binding is in dense gray or lavendar-black patches on the cells (noted by arrows).   The immunolabel for GH is orange-amber. 

The following photograph shows the results of treatment with activin for 15 hours.  The cells were taken from diestrus rats and treated with 60 ng/ml activin during the culture period. The labeling for Biotinylated GnRH has expanded on the GH cells and is seen to cover a larger part of the surface area (arrows). The GH cells are often in clusters which may suggest that activin has stimulated mitosis.   GH is detected by the orange-amber label.   This can be measured by image analysis and the increase in the average area of label/cell was statistically significant.   The references below point to papers that show these data.

For more information, please see:

Childs GV, Miller, B, and Miller W. 1997 Differential effects of inhibin on gonadotropin stores and gonadotropin releasing hormone binding to pituitary cells from cycling female rats. Endocrinology 138: 1577-1584.

Childs GV and Unabia G 1997 Cytochemical studies of the effects of activin on gonadotropin releasing hormone (GnRH) binding by pituitary gonadotropes and growth hormone cells. J Histochem Cytochem 45: 1603-1610.

Storage of Gonadotropins:  Do Growth hormone cells translate and store gonadotropins in secretory granules?

The most recent studies have shown that the multipotential somatogonadotrope appears to translate and store gonadotropin molecules in secretory granules. Our dual labeling studies with immunogold techniques demonstrate that, just before ovulation, the female rat stores LH and GH in either the same or separate storage granules. Similarly, FSH is stored with GH in granules in these cells. Thus, the cells appear ready to release gonadotropins in support of the surge activity. The following illustrations show the dual-labeling for LH or FSH and GH.

Electron micrograph showing dual labeling for FSH (F, 10 nm gold) in a storage granule in the midst of granules containing growth hormone (G, 5 nm gold). For a higher magnification of the FSH-containing granule, please see the Cytochemistry Web page .

Labeling for LH (L, 10 nm gold) in a granule near granules containing GH (G, 5 nm gold) antigens. In the dual-labeled LH-GH fields, there are also cells labeled for LH only. Therefore, these data show that the changes in multihormonal storage are occurring in somatotropes, not gonadotropes.

Test for GHRH binding: Do gonadotropes bind the releasing hormone that stimulates Growth hormone cells, during the period of transition? (Growth hormone releasing hormone)

If the GH cells retain their ability to bind GHRH (their normal phenotype), then we should be able to use dual labeling affinity cytochemistry to detect biotinylated GHRH and either LH or FSH antigens in the transitional "somatogonadotropes".  This binding should be highest during proestrus, when the GH cells express LH and FSH mRNAs.

As will be presented at the upcoming meetings of the Endocrine Society (June, 1998),  proestrous female rats have significant numbers of gonadotropes (70-80%) that bind GHRH compared with 20-30% of gonadotropes from male rats.    This transitional expression correlates well with the hypothesis that growth hormone cells contribute to the gonadotrope population, but appear to retain their GH phenotype.  (click here to see a copy of the abstract)

The photograph below shows dual labeling for biotinylated GHRH (Bio-GHRH; seen as dense purple-black, arrows) and LH (orange) in gonadotropes from proestrous female rats. Note that the labeling for Bio-GHRH is in dense patches or lines.   The right figure shows a cell (lowest cell in the figure) that is labeled only for LH and does not appear to express binding sites for Bio-GHRH.

The following figure shows FSH gonadotropes that are dual labeled for Biotinylated GHRH (arrows) and FSH. Note the difference in the labeling pattern from cell to cell.  Some cells have label covering the surface. Other cells have labeling for Bio-GHRH only in patches.  Label for FSH is orange-amber.

Do gonadotropes express growth hormone mRNA (GH mRNA)?

If GH cells retain their phenotype as they express gonadotropin mRNAs, then they may also retain the ability to express GH mRNA.  Thus, we should see expression of GH mRNA by gonadotropes.  This characteristic can be tested by dual labeling for GH mRNA and gonadotropin antigens with in situ hybridization histochemistry followed by immunochemistry.

Ongoing studies show that expression of GH mRNA changes with the stage of the cycle.  It is lowest in GH cells early in the cycle and then begins to rise.  Nearly all of the GH cells express GH mRNA by proestrus.  Also, there is a significant increase in expression of GH mRNA by cells with LH and FSH antigens during this rise.  Counts are continuing to validate the pattern seen in the early experiments.  Thus far, the hypothesis that GH cells retain their phenotype is supported by these data.

The following figures show dual labeling for GH mRNA and GH antigens.  The labeling for GH mRNA is in linear patches or spots distributed in the cytoplasm.  The GH antigen label is pale amber or dense orange.

Labeling for GH mRNA and LH-beta subunits is seen in proestrous rats.  At peak expression in proestrous, as many as 60% of LH cells contain GH mRNA. The arrows show the linear or vacuolar distribution of label for GH mRNA in the gonadotropes. U= unstained cell.

Labeling for GH mRNA and FSH is shown in the next photographs.  Sometimes the labeling appears to circle the nucleus, as if it is focused on mRNA over the nuclear envelope (the outer membrane of the envelope is rough endoplasmic reticulum).  Sometimes the labeling in gonadotropes is polarized and shown in only one pole of the cell.  These photos show examples of these labeling patterns.  The arrows point to the dense gray-black labeling for GH mRNA and the FSH label os orange-amber.

Do gonadotropes from proestrous rats secrete GH?  Do GH cells from proestrous rats secrete gonadotropins?

Stay tuned, this study is just beginning.  We are still completing the study that addresses the previous question.  However, our early evidence shows that a population of nearly pure somatotropes (90% GH cells by immunolabeling) taken from diestrus or proestrous rats will secrete LH or FSH and secretion is stimulated by GnRH.  They store little LH or FSH unless they are stimulated by GnRH.  After stimulation, labeling for LH or FSH appears in cellular processes in the GH cells.  This is depicted in the photograph below: The black label is for FSH antigens and the orange label is for LH antigens.  Most of the label is in cellular processes.

Publications describing this work

Childs, G.V. , Unabia G., Rougeau D. Cells that Express Luteinizing Hormone (LH) and Follicle Stimulating Hormone (FSH) Beta ( ) Subunit mRNAs during the Estrous Cycle: The major contributors contain LH , FSH and/or Growth Hormone, Endocrinology, 134: 990-997 1994.

  Childs, G.V., Unabia, G and Miller, BT Cytochemical detection of GnRH binding sites on rat pituitary cells with LH, FSH and GH antigens during diestrous upregulation. Endocrinology 134: 1943-1951 1994.

Childs, G.V. Division of Labor among Gonadotropes, Vitamins and Hormones, 50: 217- 283 1995

Childs, GV, Cytochemical studies of multifunctional gonadotropes. Microscopy Research and Techniques. 39: 114-130, 1997

Childs GV, Miller, B, and Miller W. 1997 Differential effects of inhibin on gonadotropin stores and gonadotropin releasing hormone binding to pituitary cells from cycling female rats. Endocrinology 138: 1577-1584.

Childs GV and Unabia G 1997 Cytochemical studies of the effects of activin on gonadotropin releasing hormone (GnRH) binding by pituitary gonadotropes and growth hormone cells. J Histochem Cytochem 45: 1603-1610.

Other pages:

Protocols and related web pages on gonadotropes:

| Affinity Cytochemistry | In situ hybridization histochemistry | Go to Childs Home Page |

Number of visitors to this site beginning 08/17/97